San Lucas: Latest News

Strategies to Enable a Stable Warfarin Maintenance Dose

Wed, 28 Jul 2010 05:20:45 +0100

INTRODUCTION

In North America, warfarin is the current standard for oral anticoagulation therapy in the treatment and/or prophylaxis of venous thrombosis, atrial fibrillation, pulmonary embolism, and acute myocardial infarction [1]. However, a significant proportion of patients on long-term warfarin therapy fail to stabilize within their target therapeutic range leading to a resultant increased risk of thromboembolism or bleeding. A systematic review by Oake et al analyzing results from 71,065 patients on long-term oral anticoagulant therapy found that 44% of all major bleeding events (95% CI 39–49%) occurred when patient International Normalized Ratios (INRs)—a measure of a patient's coagulation ability—were above target therapeutic range and 48% of all thromboembolic events (95% CI 41–55%) occurred when patient INRs were below target therapeutic range [2]. In data derived from “real world” practice patients were in therapeutic range only 55% of the time, reducing the efficacy of treatment [3]. A recent study in patients with atrial fibrillation showed that in patients with less than 55% of their treatment time in range, warfarin was no better than aspirin plus clopidrogrel for the prevention of stroke [4]. Lower time in range is also associated with increased chances of major bleeding, mortality, and thromboembolic events [1, 3, 5].

Given these statistics, the development of interventions to help patients stay within their target therapeutic range (ie, achieve their maintenance dose) would contribute greatly to improved patient safety and decreased adverse events while on oral anticoagulants.

Ensuring patients receive an appropriate dose is complicated by the fact that numerous clinical variables can impact dose requirements. With regards to the impact of clinical variables, several are frequently cited in the literature. Age has an independent negative correlation with dose with decreases of 0.5–0.7 mg for every decade [6, 7]. Low warfarin dose requirements have also been observed with low body weight, white race, liver disease, tobacco use, lack of physical exercise, certain concomitant medications, and low dietary Vitamin K intake [8–15]. While it is well documented, changes in concurrent medication, comorbidities, and patient adherence to warfarin therapy can affect anticoagulation control in a predictable way, a large portion of the intra-individual variability in response to warfarin is unexplained [9]. A large proportion of the inter-individual variability has been shown to be attributable to genetic polymorphisms.

There are many publications that have reported on strategies to improve INR control and enable a stable maintenance dose of warfarin. In this review we will discuss these in the following categories: Genetics, Nomograms, Computerized dosing tools, Models and Settings of Care, Supplementation with Vitamin K, Education, and frequency of monitoring.

Genetics

Pharmacogenetics plays an important role in the response to warfarin and authors have suggested the use of genotype-based algorithms for the dosing of patients starting therapy with warfarin and predict maintenance dose [22]. Cytochrome P450 2C9 *2 and *3 allelic variants are strongly associated with sensitivity to warfarin, resulting in a lower warfarin dose requirement, a greater risk of bleeding at initiation and maintenance, and a longer time to achieve a stable dose [16–18]. Variations in CYP2C9 have been found to account for 5–22% of inter-individual variability of warfarin dose requirements [19]. A matched case control study also found that Cytochrome P450 CYP2C9 variants *1/*3 or *2/*3 or *3/*3 were more frequent among cases with unstable coagulation control than stable controls (29% vs. 15%; P = 0.042) [20]. These mutations have been widely studied in different trials. Polymorphisms in the gene coding for CYP2C9, the principal enzyme (responsible for metabolizing the S enantiomer), alter the half-life of warfarin, affecting both the rapidity of initial effect and the dose required to maintain a therapeutic INR. Two mutations CYP2C9*2, and CYP2C9*3 showed they are associated with lower doses needed (19% and 30% reduction, respectively) [23, 24] and along with gender justify 40% of the variations in the maintenance. They also might increase the risk of major and life threatening bleeding according to two recent studies [25, 26].

Another important gene codes for VKORC1—an enzyme responsible for converting Vitamin K from its oxidized to reduced form. In addition to being a crucial enzyme in Vitamin K metabolism (especially relevant given that the nature of this intervention is to provide supplemental Vitamin K), polymorphisms in the gene that codes for this enzyme account for over 37% of the pharmacological variability associated with warfarin dosing [21]. Mutations in the gene coding for the VKORC1 enzyme will lead to a protein that is more sensitive or resistant to warfarin inhibition affecting the doses required for initiation and maintenance of the INR on range [23, 24]. There is suggestion that mutations of the VKORC1 have a higher impact than mutations CYP2C9*2 and CYP2C9*3 when time in range after initiation was compared [24].

A 1347 C-T polymorphism in the gene encoding CYP450 4F2 (V433M, rs2108622) has recently been found to be significantly associated with warfarin dose requirements, particularly among Caucasian individuals. Patients possessing the TT genotype require, on average, 1 mg/day more warfarin than those with the CC genotype [27]. The CYP4F2 polymorphism was found to independently contribute to the variability in warfarin dose even after clinical variables and polymorphisms in CYP2C9 and VKORC1 were accounted for. The incorporation of polymorphisms in CYP2C9, VKORC1, CYP4F2, as well as clinical variables was found to explain approximately 56% of the inter-individual variability with respect to warfarin dose in this study [27]. Despite the fact that the T allele has a frequency of over 30% in Caucasians and Asians; the CYP4504F2 1347 C-T polymorphism has not yet been commonly adopted in dosing algorithms.

Several genetic algorithms [18, 23, 29, 30] have been described in the literature, but to this date none of them have shown to predict the dosing of warfarin better than already known clinical algorithms [31]. In a study we performed three existing models were found to not be predictive of the maintenance dose [32]. The mean clinical maintenance dose in the study patient population was 5.58 mg/day, while the mean doses predicted according to the models by Gage et al, Kamali et al, and Sconce et al were 5.48 mg/day, 5.09 mg/day, 4.32 mg/day, and 4.34 mg/day, respectively [6, 7, 18]. For patients in this study the aforementioned models would have correctly predicted the dose to within ±0.5 mg in only 22%, 21%, and 21% of patients, respectively. However, as was reported by Caldwell et al, we confirmed a significant association between maintenance dose requirement and CYP4F2 polymorphisms. Our study, to the best of our knowledge, is the first to rigorously evaluate performance against existing clinical practice and other frequently cited models. Recent studies have also concluded that pharmacogenetics should not be used to predict dose [33–35].

Finally the real cost-effectiveness of pharmacogenetic strategies was recently questioned by an economical analysis showing that [based on current data and cost of testing (about US$400)], there is only a 10% chance that genotype-guided dosing is likely to be cost-effective (

Nomograms

Genetic nomograms do not appear to be the answer at this point. However, other nomograms have been evaluated, usually for warfarin initiation. One of the first studies trying to predict warfarin maintenance dosing after the first week was published in 1984 [36]. They calculated the predicted maintenance dose of warfarin by using the prothrombin time measured on day 4. Predicted and actual maintenance doses were closely related. Initiation nomograms are more useful if they also provide maintenance dosing information. Another nomogram, proven effective in a randomized trial has recently been further validated and analysis has enabled prediction of the maintenance dose from the data acquired during initiation of therapy [37–39]. The nomogram originally published by Kovacs et al

[38] uses a fixed warfarin dose of 10 mg during the first 2 days of treatment and subsequent doses are adjusted according to the INR values. One of the studies, in patients starting warfarin for venous thromboembolism, validated three logistic regression models for patients completing the 10 mg nomogram. For those patients who followed it to day 3: warfarin dose (mg per week) = Exp[2.737 + 1.896(INR3)–0.008(Age)]; R2 adj = 0.462); for those completing to day 5 warfarin dose (mg per week) = Exp[2.261 + 2.412(INR3)×0.285(DINR5–3)]; R2 adj = 0.603); and to day 8 warfarin dose (mg per week) = Exp[1.574 + 1.788(INR8) + 0.024 (cumulated warfarin dose until nomogram day 7)]; R2 adj = 0.643), where Exp is the exponential function; INR3 and INR8 are the INR on days 3 or 8 of the nomogram, and DINR5–3 is the difference in the INR on days 5 and 3. These models predicted the actual dose within 25% in more than 60% of patients [39]. The second one, on all outpatients starting warfarin according to 10 mg nomogram, developed a clinical prediction rule [37]. They found the daily warfarin maintenance dose to be equal to 2.5 + 10% of the first-week cumulative dose—INR value at day 8 + 1.5 if INR was below 2.0 at day 5. Both models have the advantage of being cheaper, easier to use, and readily available than genetic models but needed further validation.

SETTINGS AND MODELS OF CARE

The options for management of warfarin include community practice, specialized anticoagulation clinics, and self-management. The health care provider can also vary with nurses, pharmacists, family physicians, and specialists providing care. In one study patients followed by community practices had significantly worse anticoagulation control than those from anticoagulation clinics or from randomized clinical trials [40]. A more recent study confirmed this and demonstrated 11% more time in range if care was provided by an anticoagulation clinic [41]. Interestingly, despite the fact anticoagulation services have been shown to increase time in range, changes in major outcomes such as bleeding or stroke were not demonstrated [42, 43]. In a Canadian study, INR values of patients who were managed by anticoagulation clinics were within therapeutic range 82% of the time compared to 76% of the time for those managed by family physicians. In this study careful communication of INR information from the hospitalization was provided to the family physicians. More INR measurements were performed by family physicians than by anticoagulation clinics, but no differences in major bleeding events, thromboembolic events, or deaths were found. Still, anticoagulation clinics increase patient satisfaction and knowledge of INR range [44]. Anticoagulation clinics have also proven to be cost-effective, in different settings, by a reduction of the number of patients admitted for high INR or related emergency department visits [45, 46].

A meta-analysis on point of care devices (POCD) suggested self-management with POCD provided even better control than anticoagulation clinics but the self-managed patients received better education and more frequent testing. The self-managed POCD patients statistically had significantly fewer thromboembolic events and deaths [47]. As we discuss later these factors appear to be very important in maintaining accurate INR control.

Computerized dosing tools

Use of computerized dosing and tracking systems can improve dosing and time on range, by allowing an easier follow up of the patient, and providing algorithms for dosing [48–50]. A recent multicenter randomized study showed that the number of clinical events with two computer-assisted dosage systems was lower in all patients when compared to manual dosing, especially in the ones treated for venous thromboembolism [48, 49]. Time in target INR range was significantly improved by computer assistance when compared to management at the majority of centers [51]. A group from Iceland made clinically and statistically significant increases in patient time in range after the introduction of the Dawn AC software [52]. Others have demonstrated that software programs are cost-effective in different scenarios [53].

Use of Vitamin K to improve control

A recent hypothesis is that patient Vitamin K status (whether the patient is getting a sufficient and consistent dietary intake of Vitamin K) may strongly influence how stable a patient's INRs will be while on oral anticoagulant therapy. Sconce and colleagues showed that mean daily Vitamin K intake for patients with unstable anticoagulation control was 2.5 times lower than that for patients with stable control during a 2-week study period, assessing 42 patients [54]. Couris and colleagues confirmed this result in a 5 week study measuring the dietary changes, warfarin dose, and INRs of 60 patients—clearly demonstrating that as dietary Vitamin K intake increased, patient INRs became more stable [55]. Conversely, they also demonstrated that as Vitamin K intake decreased, patient INRs became more variable. The findings led to a number of retrospective and prospective studies being initiated as well as two randomized, controlled trials to assess the impact of low-dose Vitamin K supplementation on improving patient stability on oral anticoagulants.

Two observational studies utilized Vitamin K supplementation of 100 µg/day and 500µg/day, respectively, and both studies found that the time patients were in therapeutic range increased and the overall variability of patient INRs decreased [56, 57]. Since Vitamin K counteracts the effect of warfarin (a Vitamin K antagonist) a requisite increase in warfarin dose was required to bring back the INRs within therapeutic range, this dose increase ranged from 6% to 95%. However, while these observational studies showed benefit there are some that did not. A case study involving supplementation with 80 µg/day of Vitamin K showed no benefit with respect to percentage of time within therapeutic range [57]. Earl and colleagues [58] published a series of case reports assessing Vitamin K supplementation in varying doses from 100 to 300 µg/day and they found no improvement in INR stabilization and observed that increases in warfarin dose were required in most cases.

Currently, only one of the two RCTs to date has assessed the efficacy of low dose Vitamin K supplementation in unstable patients on warfarin. Sconce and colleagues demonstrated that INR stability did improve with Vitamin K supplementation in a double blind study of 70 unstable warfarin patients randomized to either 150 µg/day Vitamin K or placebo [59]. An 87% reduction in the standard deviation of patient INR values was seen relative to a 59% reduction found in the control group. In addition, the time within therapeutic range increased by 15% in the placebo group but by 28% in the Vitamin K group. Just as was seen in the Ford [60] study, Sconce and colleagues documented that the warfarin dosage among Vitamin K supplemented patients needed to be increased in order to maintain INR values within therapeutic range—the mean increase was 16% in the Vitamin K group compared to 1.5% with the placebo group. Unfortunately no clinically relevant outcomes were demonstrated. A second, larger RCT assessing 100 µg/day Vitamin K supplementation in 200 patients on phenprocoumon anticoagulation found an adjusted increase in time in therapeutic range of only 3.6% (95% CI 0.8–8.0%). The study was only powered to detect a 10% difference meaning that one cannot say whether this difference was statistically significant [61].

In summary the data is not conclusive and the only large study, in addition to being underpowered, may not be applicable to patients on warfarin due to the fact phenprocoumon (the anticoagulant assessed in the trial) has a longer half-life than warfarin. Therefore, it is not surprising the recent evidence-based clinical practice guidelines from the American College of Chest Physicians makes a grade 2B recommendation (weak recommendation, based on moderate quality evidence) that for patients receiving long-term warfarin therapy and who have variable INRs not attributable to known causes of instability a trial of low dose Vitamin K (100–200 µg/day) might be considered for stabilizing INRs [62]. However, no indication is given as to what criterion should be used to assess stability, nor are there guidelines with respect to frequency of testing and dose adjustment to ensure INRs do not remain subtherapeutic, nor any assessment as to whether there are certain patients who may be more successful with this therapy than others. Further study is required before widespread use, but low dose vitamin K can be considered part of the arsenal for patients with poor control.

Education

Our meta-analysis on POCD devices suggested patient education and frequency of monitoring may be very important in achieving better INR control. This fact has now been validated in a randomized trial. Pernod and colleagues randomized patients to a specific one-on-one education program or standard care with unstructured information and standard booklet provided by the French Heart Association [63]. During the three month follow-up, hemorrhage (virtually all non-major) or thromboembolic events were 20 times more frequent in the standard arm. Although it can be argued unstructured information provision should never be the “standard,” this study demonstrates the value of education. Decision aids have also been demonstrated to increase patient knowledge [64].

Frequency of monitoring

Guidelines recommend a first measurement 48 h after the initiation of therapy, followed by daily INR in hospitalized patients or every 48 h until INR is greater than 2. Then every 2–3 days for the next 2 weeks [65]. After this stage INR testing can be performed at 4-week intervals and clear guidance on frequency are not available. Computerized dosing programs do provide this information to the user. However, evidence suggests that periods shorter than 4 weeks might be better. One study demonstrated an increase of 9% of the time on range can be achieved, when testing intervals are reduced from 5 to 3 weeks [66]. Self-management studies have generally shown and suggested that INRs be performed weekly and the studies have demonstrated increase time in range (73% vs. 62% in anticoagulant clinics in which testing frequency was on average every 4 weeks] [47]. In one review of 350 patients only higher warfarin doses and perhaps more frequent monitoring were associated with better INR control [67]. In patients treated with stable doses, monitoring should be increased with certain conditions or drugs. Hospitalization, per se, leads to an overall decrease of the time on range by 15%, and an increased time with a critically low INR or critically high INR [68].

SUMMARY AND SUGGESTED PRACTICE GUIDELINE

Although several strategies have been demonstrated to increase time in therapeutic range, their adoption has been slow in clinical practice. A recent survey from Canada proved that the utilization of methods that improve care (anticoagulation services, data tracking software, and self-monitoring) of patients is low. Primary care physicians more often managed warfarin based on their experience only or with a manual dosing algorithms and less often used one or more evidence-based dosing methods. Only a small number of specialists or family doctors acknowledged having problems with warfarin dosing, suggesting lack of insight as part of the problem [69].

In our daily practice we tend to use a single tablet strength in all our patients (2.5 mg), this allows better understanding between the members of our team (nurses, doctors, and pharmacists), increased patient knowledge of their therapy, and facilitates the implementation of algorithms. This method has similar outcomes to multidose regimens [70]. shows the 10 mg nomogram we use during the initiation of therapy, utilizing 2.5 mg tablets, and the formula we use as a maintenance dose calculator performed after the day 8 INR [37]. After the first 2–3 weeks, patients are transferred to our clinic pharmacist who follows patients using the Dawn AC version 8 software program. Patients obtain their INR at any certified lab, before 9 am so the results can be faxed to us the same day and allow dose adjustments the same day. This ensures out of range INR problems are not compounded. Our large anticoagulation clinic model has been chosen since this model of care increases patient satisfaction, knowledge of INR range, time in range, and is cost-effective. Although self-monitoring/self-dosing improves time in range and outcomes due to cost constraints, we do not employ this widely and only 5% of our 1600 patients perform self-monitoring. These patients were carefully selected and, as done in the studies evaluating self-monitoring, the patients receive structured education sessions on warfarin therapy [47]. The utilization of this approach has been further limited by the number of patients who are good candidates, are willing to use this approach, and can afford it. If our Canadian health care system paid for self-monitoring it would be cost-effective from a societal perspective, but sadly they are not paid for [71].

New drugs, such as dabigatran and rivaroxaban, could eventually reduce our utilization of warfarin. Monitoring is not recommended, or indeed reliable, for either drug, making them potentially more suitable for the management of patients, but their effectiveness with widespread use remains to be proven. To date only two non-inferiority studies have been published with dabigatran showing a similar outcome profile as with warfarin in patients treated for atrial fibrillation or venous thromboembolism (VTE).

If we can utilize best evidence in the management of patients on warfarin, we could potentially reduce adverse events and enable a continued safer use of warfarin, thus, reducing societal costs of anticoagulation therapy.

Disclosure: The authors have no financial interests to disclose in relation to the contents of the article.[AQ7][AQ8]

References

1.     Oake N, Jennings A, Forster AJ, Fergusson D, Doucette S, van Walraven C. Anticoagulation intensity and outcomes among patients prescribed oral anticoagulant therapy: a systematic review and meta-analysis. CMAJ. 2008;179:235-244.
2.     Oake N, Fergusson DA, Forster AJ, van Walraven C. Frequency of adverse events in patients with poor anticoagulation: a meta-analysis. CMAJ. 2007;176:1589-1594.
3.     van Walraven C, Oake N, Wells PS, Forster AJ. Burden of potentially avoidable anticoagulant-associated hemorrhagic and thromboembolic events in the elderly. Chest. 2007;131:1508-1515.
4.     Connolly SJ, Pogue J, Eikelboom J, et al. Benefit of oral anticoagulant over antiplatelet therapy in atrial fibrillation depends on the quality of international normalized ratio control achieved by centers and countries as measured by time in therapeutic range. Circulation. 2008;118(20):2029-2037.
5.     Jones M, McEwan P, Morgan CL, Peters JR, Goodfellow J, Currie CJ. Evaluation of the pattern of treatment, level of anticoagulation control, and outcome of treatment with warfarin in patients with non-valvar atrial fibrillation: a record linkage study in a large British population. Heart. 2005;91:472-477.
6.     Sconce EA, Khan TI, Wynne HA, et al. The impact of CYP2C9 and VKORC1 genetic polymorphism and patient characteristics upon warfarin dose requirements: proposal for a new dosing regimen. Blood. 2005;106:2329-2333.
7.     Kamali F, Khan TI, King BP, et al. Contribution of age, body size, and CYP2C9 genotype to anticoagulant response to warfarin. Clin Pharmacol Ther. 2004;75:204-212.
8.     Shibata Y, Hashimoto H, Kurata C, Ohno R, Kazui T, Takinami M. Influence of physical activity on warfarin therapy. Thromb Haemost. 1997;80:203-204.
9.     Holbrook AM, Pereira JA, Labiris R, et al. Systematic overview of warfarin and its drug and food interactions. Arch Intern Med. 2005;165:1095-1106.
10.     Cushman M, Booth SL, Possidente CJ, Davidson KW, Sadowski JA, Bovill EG. The association of vitamin K status with warfarin sensitivity at the onset of treatment. Br J Haematol. 2001;112:572-577.
11.     Routledge PA, Chapman PH, Davies DM, Rawlins MD. Factors affecting warfarin requirements. A prospective population study. Eur J Clin Pharmacol. 1979;15(5):319-322.
12.     Blann A, Hewitt J, Siddiqui F, Bareford D. Racial background is a determinant of average warfarin dose required to maintain the INR between 2.0 and 3.0. Br J Haematol. 1999;107(1):207-209.
13.     Fergusson RJ, Eade OE, Logie AW, Gaddie J. A flexible loading dose schedule for warfarin therapy. Scott Med J. 1987;32(6):169-171.
14.     Mungall DR, Ludden TM, Marshall J, Hawkins DW, Talbert RL, Crawford MH. Population pharmacokinetics of racemic warfarin in adult patients. J Pharmacokinet Biopharm. 1985;13(3):213-227.
15.     Lubetsky A, Dekel-Stern E, Chetrit A, Lubin F, Halkin H. Vitamin K intake and sensitivity to warfarin in patients consuming regular diets. Thromb Haemost. 1999;81(3):396-399.
16.     Margaglione M, Colaizzo D, D'Andrea G, et al. Genetic modulation of oral anticoagulation with warfarin. Thromb Haemost. 2000;84:775-778.
17.     Higashi MK, Veenstra DL, Kondo LM, et al. Association between CYP2C9 genetic variants and anticoagulation-related outcomes during warfarin therapy. JAMA. 2002;287:1690-1698.
18.     Gage BF, Eby C, Milligan PE, Banet GA, Duncan JR, McLeod HL. Use of pharmacogenetics and clinical factors to predict the maintenance dose of warfarin. Thromb Haemost. 2004;91:87-94.
19.     Yin T, Miyata T. Warfarin dose and the pharmacogenomics of CYP2C9 and VKORC1 - rationale and perspectives. Thromb Res. 2007;120:1-10.
20.     Palareti G, Legnani C, Guazzaloca G, et al. Risk factors for highly unstable response to oral anticoagulation: a case-control study. Br J Haematol. 2005;129(1):72-78.
21.     Bodin L, Verstuyft C, Tregouet DA, et al. Cytochrome P450 2C9 (CYP2C9) and vitamin K epoxide reductase (VKORC1) genotypes as determinants of acenocoumarol sensitivity. Blood. 2005;106:135-140.
22.     Leey JA, McCabe S, Koch JA, Miles TP. Cost-effectiveness of genotype-guided warfarin therapy for anticoagulation in elderly patients with atrial fibrillation. Am J Geriatr Pharmacother. 2009;7(4):197-203.
23.     International Warfarin Pharmacogenetics Consortium, Klein TE, Altman RB, Eriksson N, et al. Estimation of the warfarin dose with clinical and pharmacogenetic data. N Engl J Med. 2009;360:753-764.
24.     Schwarz UI, Ritchie MD, Bradford Y, et al. Genetic determinants of response to warfarin during initial anticoagulation. N Engl J Med. 2008;358:999-1008.
25.     Wadelius M, Chen LY, Lindh JD, et al. The largest prospective warfarin-treated cohort supports genetic forecasting. Blood. 2009;113(4):784-792.
26.     Aithal GP, Day CP, Kesteven PJ, Daly AK. Association of polymorphisms in the cytochrome P450 CYP2C9 with warfarin dose requirement and risk of bleeding complications. Lancet. 1999;353(9154):717-719.
27.     Caldwell MD, Awad T, Johnson JA, et al. CYP4F2 genetic variant alters required warfarin dose. Blood. 2008;111:4106-4112.
28.     Eckman MH, Rosand J, Greenberg SM, Gage BF. Cost-effectiveness of using pharmacogenetic information in warfarin dosing for patients with nonvalvular atrial fibrillation. Ann Intern Med. 2009;150:73-83.
29.     Gage BF, Eby C, Johnson JA, et al. Use of pharmacogenetic and clinical factors to predict the therapeutic dose of warfarin. Clin Pharmacol Ther. 2008;84:326-331.
30.     Zhu Y, Shennan M, Reynolds KK, et al. Estimation of warfarin maintenance dose based on VKORC1 (-1639 G>A) and CYP2C9 genotypes. Clin Chem. 2007;53(7):1199-1205.
31.     Shaw PB, Donovan JL, Tran MT, Lemon SC, Burgwinkle P, Gore J. Accuracy assessment of pharmacogenetically predictive warfarin dosing algorithms in patients of an academic medical center anticoagulation clinic. J Thromb Thrombolysis 2010.[AQ9][AQ10]
32.     Wells PS, Rodger MA, Fleming J, Kassem S. Current models for predicting warfarin maintenance dose do not work in an ambulatory venous thromboembolism patient population [Abstract]. Blood. 2007;110(11):95a-96a.
33.     Kangelaris KN, Bent S, Nussbaum RL, Garcia DA, Tice JA. Genetic testing before anticoagulation? A systematic review of pharmacogenetic dosing of warfarin. J Gen Intern Med. 2009;24(5):656-664.
34.     Mannucci PM, Spreafico M, Peyvandi F. Dosing anticoagulant therapy with coumarin drugs: is genotyping clinically useful? No. J Thromb Haemost. 2008;6:1450-1452.
35.     Rosove MH, Grody WW. Should we be applying warfarin pharmacogenetics to clinical practice? No, not now. Ann Intern Med. 2009;151(4):270-273.
36.     Fennerty A, Dolben J, Thomas P, et al. Flexible induction dose regimen for warfarin and prediction of maintenance dose. Br Med J. 1984;288:1268-1270.
37.     Le Gal G, Carrier M, Tierney S, Majeed H, Rodger M, Wells PS. Prediction of the warfarin maintenance dose after completion of the 10mg initiation nomogram: do we really need genotyping? J Thromb Haemost. 2010;8(1):90-94.
38.     Kovacs MJ, Rodger M, Anderson DR, et al. Comparison of 10-mg and 5-mg warfarin initiation nomograms together with low-molecular-weight heparin for outpatient treatment of acute venous thromboembolism. A randomized, double-blind, controlled trial. Ann Intern Med. 2003;138:714-719.
39.     Lazo-Langner A, Monkman K, Kovacs MJ. Predicting warfarin maintenance dose in patients with venous thromboembolism based on the response to a standardized warfarin initiation nomogram. J Thromb Haemost. 2009;7:1276-1283.
40.     van Walraven C, Jennings A, Oake N, Fergusson D, Forster AJ. Effect of study setting on anticoagulation control. A systematic review and metaregression. Chest. 2006;129:1155-1166.
41.     Baker WL, Cios DA, Sander SD, Coleman CI. Meta-analysis to assess the quality of warfarin control in atrial fibrillation patients in the United States. J Manag Care Pharm. 2009;15(3):244-252.
42.     Wallvik J, Sjalander A, Johansson L, Bjuhr O, Jansson JH. Bleeding complications during warfarin treatment in primary healthcare centres compared with anticoagulation clinics. Scand J Prim Health Care. 2007;25(2):123-128.
43.     Connock M, Stevens C, Fry-Smith A, et al. Clinical effectiveness and cost-effectiveness of different models of managing long-term oral anticoagulation therapy: a systematic review and economic modelling. Health Technol Assess. 2007;11(38):iii-66.[AQ11]
44.     Wilson SJ, Wells PS, Kovacs MJ, et al. Comparing the quality of oral anticoagulant management by anticoagulation clinics and by family physicians: a randomized controlled trial. CMAJ. 2003;169:293-298.
45.     Gray DR, Garabedian-Ruffalo SM, Chretien SD. Cost-justification of a clinical pharmacist-managed anticoagulation clinic. Drug Intell Clin Pharm. 1985;19(7-8):575-580.
46.     Chiquette E, Amato MG, Bussey HI. Comparison of an anticoagulation clinic with usual medical care. Anticoagulation control, patient outcomes, and health care costs. Arch Intern Med. 1998;158:1641-1647.
47.     Wells PS, Brown A, Jaffey J, McGahan L, Poon M-C, Cimon K. Safety and effectiveness of point-of-care monitoring devices in patients on oral anticoagulant therapy: a meta-analysis. Open Med. 2007;1(3):131-146.
48.     Poller L, Keown M, Ibrahim S, et al. An international multicenter randomized study of computer-assisted oral anticoagulant dosage vs. medical staff dosage. J Thromb Haemost. 2008;6:935-943.
49.     Poller L, Shiach CR, MacCallum PK, et al. Multicentre randomised study of computerised anticoagulant dosage. Lancet. 1998;352:1505-1509.
50.     Poller L, Shiach CR, MacCallum PK, et al. The European concerted action on anticoagulation randomised multicentre study on computerised anticoagulant dosage. Int Cong Thromb Program Book (Turkey) 1998; Program book:115-120.
51.     Poller L, Keown M, Ibrahim S, et al. A multicentre randomised assessment of the DAWN AC computer-assisted oral anticoagulant dosage program. Thromb Haemost. 2009;101:487-494.
52.     Onundarson PT, Einarsdottir KA, Gudmundsdottir BR. Warfarin anticoagulation intensity in specialist-based and in computer-assisted dosing practice. Int J Lab Hematol. 2008;30:382-389.
53.     Jowett S, Bryan S, Poller L, et al. The cost-effectiveness of computer-assisted anticoagulant dosage: results from the European action on anticoagulation (EAA) multicentre study. J Thromb Haemost. 2009;7(9):1482-1490.
54.     Sconce E, Khan T, Mason J, Noble F, Wynne H, Kamali F. Patients with unstable control have a poorer dietary intake of vitamin K compared to patients with stable control of anticoagulation. Thromb Haemost. 2005;93:872-875.
55.     Couris R, Tataronis G, McCloskey W, et al. Dietary vitamin K variability affects International Normalized Ratio (INR) coagulation indices. Int J Vitam Nutr Res. 2006;76(2):65-74.
56.     Reese AM, Farnett LE, Lyons RM, Patel B, Morgan L, Bussey HI. Low-dose vitamin K to augment anticoagulation control. Pharmacotherapy. 2005;25(12):1746-1751.
57.     Leong WA. Case studies in anticoagulation management. J Thromb Thrombolysis. 2008;25(1):78-84.[AQ12]
58.     Earl L, Pena R, Hutchinson J. Utilization of dietary vitamin K supplementation to aid in INR stabilization for patients on chronic anticoagulant therapy: a series of case reports. J Thromb Thrombolysis. 2008;25(1):91.[AQ12][AQ13]
59.     Sconce E, Avery P, Wynne H, Kamali F. Vitamin K supplementation can improve stability of anticoagulation for patients with unexplained variability in response to warfarin. Blood. 2007;109:2419-2423.
60.     Ford SK, Moll S. Vitamin K supplementation to decrease variability of International Normalized Ratio in patients on vitamin K antagonists: a literature review. Curr Opin Hematol. 2008;15:504-508.
61.     Rombouts EK, Rosendaal FR, van der Meer FJ. Daily vitamin K supplementation improves anticoagulant stability. J Thromb Haemost. 2007;5:2043-2048.
62.     Ansell J, Hirsh J, Hylek E, Jacobson A, Crowther M, Palareti G. Pharmacology and management of the vitamin K antagonists: ACCP evidence-based clinical practice guidelines. Chest. 2008;133:160S-198S.
63.     Pernod G, Labarere J, Yver J, et al. EDUC'AVK: reduction of oral anticoagulant-related adverse events after patient education: a prospective multicenter open randomized study. J Gen Intern Med. 2008;23(9):1441-1446.
64.     Holbrook A, Labiris R, Goldsmith CH, Ota K, Harb S, Sebaldt RJ. Influence of decision aids on patient preferences for anticoagulant therapy: a randomized trial. CMAJ. 2007;176:1583-1587.
65.     Kearon C, Kahn SR, Agnelli G, Goldhaber SZ, Raskob GE, Comerota AJ. Antithrombotic therapy for venous thromboembolic disease: American college of chest physicians evidence-based clinical practice guidelines (8th edition). Chest. 2008;133(Suppl.):454-545.
66.     Shalev V, Rogowski O, Shimron O, et al. The interval between prothrombin time tests and the quality of oral anticoagulants treatment in patients with chronic atrial fibrillation. Thromb Res. 2007;120(2):201-206.
67.     McGriff-Lee NJ, Csako G, Chen JT, et al. Search for predictors of nontherapeutic INR results with warfarin therapy. Ann Pharmacother. 2005;39:1996-2002.
68.     van Walraven C, Forster AJ. Anticoagulation control in the peri-hospitalization period. J Gen Intern Med. 2007;22:727-735.
69.     Nieuwlaat R, Barker L, Kim YK, et al. Underuse of evidence-based warfarin dosing methods for atrial fibrillation patients. Thromb Res. 2009.[AQ9][AQ10]
70.     Hixson-Wallace JA, Dotson JB, Blakey SA. Effect of regimen complexity on patient satisfaction and compliance with warfarin therapy. Clin Appl Thromb Hemost. 2001;7(1):33-37.
71.     Brown A, Wells PS, Jaffey J, et al. Point-of-care monitoring devices for long-term oral anticoagulation therapy: clinical and cost effectiveness. CADTH Technol Report. 2007;72:1-77.

Use of Human Prothrombin Complex Concentrate in Patients with Acquired Deficiency and Active or in High-Risk Severe Bleeding

Wed, 28 Jul 2010 04:57:35 +0100

INTRODUCTION

Prothrombin complex concentrates (PCC), marketed in several European countries [1], are the preferred options to prevent or treat life-threatening bleeding [2–4]. PCC are virus-inactivated, blood-derivative products that contain vitamin K-dependent purified clotting factors (II, VII, IX, and X) and are also balanced with heparin, protein C, protein S, and antithrombin III [2]. A single intravenous (iv) 600 IU PCC dose has been considered to have the factor concentration equivalent to a bag of fresh frozen plasma (FFP).

PCC, initially indicated for the restoration of clotting factor levels in patients with congenital deficits, especially hemophilia B, are currently mainly used for the restoration of clotting factor levels in combined acquired deficits.

PCC [1, 5] and factor VIIa [6] have been used for critical and massive life-threatening bleeding in exceptional circumstances. PCC have been stated for treatment of intracerebral hemorrhage and are also the preferred agent to treat oral anticoagulant treatment (OAT) acute hemorrhages [1, 5, 7, 8]. PCC are considered for use in severe life-threatening bleeding related to secondary deficits in patients who had previously been treated only with FFP [9–16]. Furthermore, it has been suggested to use PCC, factor VIIa, or fibrinogen concentrates to reduce the use of other hemoderivatives such as red blood cell concentrates, platelets, FFP, and cryoprecipitate [5, 6, 17–19]. Moreover, the use of PCC is more feasible, faster, and even safer than the administration of the primary, less fractionated hemoderivatives [1, 5, 11].

Therefore, in secondary deficiencies, the use of PCC includes a wide range of clinical applications characterized by combined deficits of clotting factors and major bleeding [5]. However, there are few controlled studies that demonstrate the clinical efficacy of PCC in each of those situations [11].

The aim of this study is to ascertain the true role of PCC in clinical practice, in which situations and what doses are they used. And a secondary goal is to measure the effect ofPCC on international normalized ratio (INR), hematocrit (Hct), and hemoglobin (Hgb), and ultimately, it is intended to establish PCC efficiency, based on survival rates at the end of the procedure.

METHODS

An observational, partially retrospective, non-comparative study initially included all patients with life-threatening hemorrhages who received PCC (Prothromplex®; Baxter Healthcare, Deerfield, IL, USA) in the Vall d'Hebron General Hospital, since January 2008 and for a period of 6 months. Treatments were detected in the Pharmacy Service following the appropriate PCC medical prescription. Data were obtained systematically from electronic medical records and also verified from primary paper records.

Previously and following ethics regulations, the study was evaluated and then approved by the Institutional Review Board (Ethics Committee) of the institution.

Life-threatening hemorrhage was defined as critical or acute major bleeding that requires transfusion. Critical hemorrhages are central nervous system hemorrhages with high mortality rates and, although not usually related to INR, they are more frequent in OAT patients, and require emergency treatment [19]. Acute major hemorrhages occur with blood losses >20% of volume, a fall of 2 g/dL Hgb and/or Hct values <30% are life threatening, and blood product transfusion is required [1, 3].

Massive uncontrolled hemorrhage is one of the causes of unexpected death in trauma and surgical patients. Excess transfusion may lead to hemodilution, which causes coagulation factor diminution and coagulation alterations. Thus, interventions such as cardiac bypass surgery and other valve substitutions or general major surgery interventions that will lead to high expected blood losses have to be considered as life-threatening bleeding [1, 5, 20].

Once patients with life-threatening bleeding who received PCC were identified, clinical and demographic data were collected retrospectively and then followed up prospectively until the clinical acute phase ended. The data collected are as follows: demographic data, diagnosis at hospital admission, PCC dose prescribed and number, concrete treatment indication; INR, Hgb, Hct before and after PCC treatment; previous or current OAT; vitamin K treatment, blood and platelet concentrates; FFP and prophylactic enoxaparin administration were recorded if documented.

Survival at the end of the procedure was also recorded and, if presented, the adverse events were also registered. Data sources were the original medical prescription, the medical and laboratory records, and the blood bank register.

Statistical analysis was performed using the SPSS v.12.0 (SPSS Inc., USA). For demographic purposes, results are expressed as mean and standard deviation for continuous variables, and as frequencies for discontinuous variables. Survival rates at the end of the procedure were stated in percentages. The paired Student's t test was used to compare means (Mann-Whitney U test, when necessary). The Chi-square test (Fisher's exact test, when necessary) was used to compare proportions. To evaluate the association between the INR after PCC administration and survival, a logistic regression was performed. Subgroup analysis was made by OAT and no-OAT and indication for treatment: general surgery, cardiac surgery, gut hemorrhage, cerebral hemorrhage, emergency and intensive care, and neoplasia; OAT and no-OAT subgroups in general and cardiac surgery and gut hemorrhage were also evaluated separately. For single-outcome comparisons, the treatment effect was considered significant if P values were 0.05 or less.

RESULTS

Initially, 124 patients received a prescription of PCC from January to June 2008. Of these, 21 patients (16%) were excluded because PCC was not finally administered or there were insufficient clinical or analytical data. Therefore, 103 patients were included in the follow-up and analysis.

Patients’ mean age was 63.73 years (SD 17.82 years); 67 (66%) were men. PCC mean dose administered was 2720 IU expressed as factor IX (1551). Mean INR before treatment was 2.92 (2.54) and INR after treatment was 1.47 (0.44). Global survival at the end ofthe episode was 63/103 (61.2%). Logistic regression indicates that survival is significantly related to INR after PCC administration (P=0.003).

PCC indication

The main clinical indications for PCC administration are presented in : 31.9% were involved in cardiac surgery with extracorporeal circulation (ECC) for valve exchanges (23), aorta aneurysm dissection (3), or other major bleeding in complications ofcardiac surgery (9); 35.0% were involved in general surgery, major and digestive surgery (24) including two cases of liver transplantation, thoracic surgery (7), including three pulmonary transplantation, neurosurgery (3), and vascular surgery (2); 14.5% received PCC for gastrointestinal bleeding, 10.7% for central nervous system bleeding; 1.9% received PCC for bleeding complications of tumors.

Patients with OAT numbered 41 (39.8%). Of these, 29.3% were involved in general surgery; 26.8% in cerebral hemorrhage; 22.0% in cardiac surgery; 17.1% in digestive bleeding; and 4.2% in other bleeding in emergency or intensive care. On the other side, no-OAT patients numbered 62 (60.2%). Of these, 38.7% were in general surgery; 38.7% in cardiac surgery; 12.9% in digestive bleeding; 6.5% for other emergencies or intensive care bleeding; and 3.2% for bleeding complications of tumors.

INR change

Mean INR value previous to the PCC administration was 2.92 (2.54), whereas mean INR value after PCC was 1.47 (0.44). Mean difference was −1.54 (2.89), which reached statistical significance (P<0.005). Significance was maintained when general surgery patients were analyzed separately (P<0.001), but not in cardiac surgery (P=0.36). In other clinical indications, such as gut hemorrhage (P=0.055), cerebral hemorrhage (P=0.036), and other emergency hemorrhages (P=0.053), significance was almost reached, although the number of patients was limited. IfOAT and no-OAT patients were considered separately, significance was maintained in no-OAT patients (P=0.001) but not in OAT patients (P=0.07). Moreover, significance was maintained in general surgery, no-OAT patients (P=0.003), and in cardiac surgery, OAT patients (P=0.31); conversely, lack of significance was observed in cardiac surgery, no-OAT patients (P=0.07).

Survival

Patients’ observed overall survival rate was 6i.2%. If considered separately, 65.9% of OAT patients and 58.1% of no-OAT patients survived (). Following clinical indications, the survival rate was 87.9% of cardiac patients, 61.1% of general surgery patients, 45.5% of cerebral hemorrhage patients, 20.0% ofgastrointestinal hemorrhages, and 50.0% in tumor and other emergency and critical care bleeding, albeit the number ofpatients in the last two groups was clearly limited (). When general surgery was analyzed separately by OAT and no-OAT patients, 91.7% survival was observed in OAT patients and 45.8% in no-OAT patients. This difference was not observed in cardiac surgery patients, with values of 88.9% and 87.5%, for OAT and no-OAT patients respectively. Conversely, 28.6% ofOAT patients with gastrointestinal bleeding survived, compared with 12.5% survival in no-OAT patients with gastrointestinal bleeding ().

Hct and Hgb

There were no significant differences between Hgb and Hct (before and after PCC). Nevertheless, a trend to better values was observed: Hgb mean difference was −0.54 (2.84) and Hct mean difference −1.41 (8.50).

Hemoderivative use

Some 84% of patients received at least one bag of concentrated blood red cells, 51% of patients received at least one FFP bag, and 47% received platelet concentrates.

Adverse effects

No adverse drug reactions were observed directly related to PCC administration. Of special interest, no thrombotic events were registered in the clinical records, directly or indirectly related to PCC administration.

DISCUSSION

The use of PCC, which includes a wide range of clinical applications characterized by combined deficits of clotting factors and major bleeding, is still controversial, as there are limited studies that demonstrate the clinical efficacy of PCC [5]. Moreover, in different countries, factor VIIa or the use of primary, less fractionated hemoderivatives such as red blood cell concentrates, platelets, FFP, and cryoprecipitate is preferred, in spite of the use of PCC being more feasible and even safer [1, 5, 11].

This study describes the clinical use of PCC in a third-level, general hospital, with the aim to establish role for PCC in clinical practice. As the study is descriptive, it captures the use of PCC in routine clinical practice. For instance, 41 (39.8%) patients were previously OAT patients, and emergency reversal was needed because of bleeding or urgent surgical intervention. In this use, PCC has been compared with FFP [21] and FFP plus PPC in intracerebral hemorrhage [22], and with vitamin K [23] (). In all studies, the clinical efficacy is considered good, and a significant improvement in INR has been observed. Pivotal and non-controlled studies show that PCC provide a rapid decrease in INR and good clinical responses, with 72% overall survival () [24–32]. This indication is widely accepted over Europe, (or european countries) and these products have been recommended for reversal of OAT in severe hemorrhage worldwide [7, 14, 33–35]. Currently, two studies are being conducted comparing the efficacy of PCC with standard treatment on previous OAT patients in the United States [36, 37], and even a dose-escalating study of PCC in cerebral hemorrhage [38].

The use of PCC has been recommended in several guidelines for cases of massive hemorrhage. PCC provide more rapid and complete vitamin K-dependent factor replacement, are infused at lower volume, and have enhanced safety because of viral inactivation [39]; thus, PCC appear to offer several advantages over FFP. Nevertheless, none of these advantages has been proved in controlled and randomized studies in comparison with the standard treatment. Massive hemorrhage in non-OAT patients is more common in surgery. There are limited controlled studies that demonstrate PCC efficacy in this group of patients [11], but clinical guidelines on major trauma include the use of PCC in OAT patients or patients with coagulation alteration [35]. Moreover, the use of PCC has been recommended in spontaneous hemorrhage with underlying coagulation changes [18]. Nevertheless, a greater use of PCC in no-OAT patients was observed before [40]. European critical care physicians also recommend the use of PCC as an alternative therapy in patients with severe bleeding that does not respond to other therapeutic measures or other hemoderivative products [3]. An experimental study demonstrated the efficacy of the use of fibrinogen and PCC in pigs in a situation of dilutional coagulopathy and uncontrolled hemorrhage [17]. In our series, 62 (60.2%) PCC-treated patients were no-OAT. Of these, 24 were undergoing general and 24 cardiac surgery. Only Bruce and Nokes [5] have previously reported a few cases of PCC-treated patients in general and cardiac surgery no-OAT patients. Two series of cases on cardiac surgery with liver dysfunction were also published [41]. Eight cases in our series (13%) had severe liver damage and bleeding. Only a single series of 22 patients with liver damage and hemostatic defects has been reported, showing an improvement in Quick's test that was considered highly efficacious in 76% of patients after the first dose [42].

Although INR is a good parameter for OAT control, in no-OAT patients, the specificity and, consequently, the use of INR could be discussed. Nevertheless, we have considered its usefulness as a single coagulation parameter for the whole population studied. In terms of INR change, our data showed that it was significant (−1.54 (2.89)), with a final mean value of 1.47 (0.44). Thus, PCC almost corrects INR. In a subgroup analysis, stratified by OAT and no-OAT patients, significance was maintained in no-OAT (P=0.001) but not in OAT patients (P=0.07). The loss of significance can be explained because of the smaller INR difference observed in cardiac surgery patients (−0.65 (1.56)). This smaller difference can be explained by the fact that, in cardiac surgery, patients’ INR determination is conditioned by the hemodilution that occurs during the surgical process. Conversely, INR changes remained significant in the general surgery subgroup (P=0.003). Stratifying by clinical episode, INR changes are significant for general surgery and cerebral hemorrhage (P=0.036), but not for cardiac surgery; the lack of significance in gut hemorrhage (P=0.055) and other emergency hemorrhages (P=0.053) can be explained because of the small number of patients evaluated, as crude INR changes are wide. The changes in INR values could be of great interest, because improved INR final values were associated with greater survival rates in an initial statistical evaluation.

The observed mortality in our study was similar or lower compared with other series in general or cardiac surgery or gastrointestinal life-threatening bleeding [5]. It should be considered that, in OAT patients, 11 cerebral, 8 gastrointestinal, 2 emergency non-specified severe hemorrhages were evaluated; and 12 general and 9 cardiac emergency surgeries were also registered. Thus, the severity of bleeding cases that required treatment with PCC could justify a 62% overall survival rate. The OAT patients’ survival in our study was 67%, whereas the overall survival of PCC in pivotal studies was 72% (). In no-OAT patients, critical bleeding treated with PCC had a 58% survival rate. Similarly, the series of Bruce and Nokes [5], which included both no-OAT and warfarin reversal patients, showed an overall survival of 50%. For general emergency surgery, a 46% survival was found at the end of the episode. The group of emergency bleeding on general surgery in Bruce and Nokes's series showed a survival of 33% [5]. An 88% survival was observed among cardiac surgery patients, whereas similar clinical series [5] showed lower survival rates varying between 40% and 50%.

It has been stated that PCC can lead to thrombosis but, on the other hand, a large amount of FFP has led to transfusion-related acute lung injury (TRALI). In pivotal studies (), an overall 3% of possible thrombotic events after PCC administration were reported, although not directly related to drug use. In our study, no cases of thrombosis, directly or indirectly related to PCC administration, were detected, even though, outside the study but during the same period, a thrombotic event was recorded in the context of a type A aortic dissection. Thus, thrombotic events are usually recorded in our clinical records. In addition, tolerance to administration was good. PCC administration is easier, faster, thawing free, and does not require the administration of a large volume of protein-rich solutions, as happens with FFP or cryoprecipitate.

In summary, PCC are widely used for severe bleeding in OAT patients, and PCC are also widely used for emergency bleeding in certain situations, in no-OAT patients, as recommended by international guidelines [43]. Nevertheless, in most cases, no published evidence is available to support their use. Our data show that routine practice involves the use of PCC in a wide range of clinical situations. The observed significant improvement in INR after PCC administration, together with the fact that survival seems to be related to the final INR, empirically supports the use of PCC in life-threatening bleeding situations. Nevertheless, further prospective controlled studies are needed to establish the true role of PCC in this context, and meanwhile, with the absence of more evidence, the use of PCC should be closely regulated in each center by mandatory guidelines.

Disclosure: The authors declare no conflict of interest.

Acknowledgements: We wish to thank Jesika Gómez López for her kind contribution to collecting and editing data.

References

1. Vigué B. Bench-to-bedside review: optimising emergency reversal of vitamin K antagonists in severe haemorrhage—from theory to practice. Crit Care. 2009;13(2):209 (online).

2. Hanley JP. Warfarin reversal. J Clin Pathol. 2004;57:1132–1139.

3. Quintana-Diaz M, Carvalho M. Hemorragia por anticoagulantes orales y tratamiento con complejos protrombinicos. Intensivos. 2008;27. http://intensivos.uninet.edu/27/2701.html.

4. Mannucci PM, Levi M. Prevention and treatment of major blood loss. N Engl J Med. 2007;356:2301–2311.

5. Bruce D, Nokes TJC. Prothrombin complex concentrate (Beriplex P/N) in severe bleeding: experience in a large tertiary hospital. Crit Care. 2008; 12(4):R105.

6. Mallarkey G, Brighton T, Thomson A, Kaye K, Seale P, Gazarian M. An evaluation of eptacog alfa in nonhaemophiliac conditions. Drugs. 2008;68(12):1665–1689.

7. Appelboam R, Thomas EO. The headache over warfarin in British neurosurgical intensive care units: a national survey of current practice. Intens Care Med. 2007;33(11):1946–1953.

8. Schick KS, Fertmann JM, Jauch KW, Hoffmann JN. Prothrombin complex concentrate in surgical patients: retrospective evaluation of vitamin K antagonist reversal and treatment of severe bleeding. Crit Care. 2009;13(6):R191.

9. Prothomplex(R). Ficha técnica Febrero 2007. Available from https://sinaem4.agemed.es/consaem/especialidad.do?metodo=verFichaWordPdf&codigo=54938&formato=pdf&formulario=FICHAS (accessed October 2009).

10. Octaplex(R). Ficha técnica 05.11.07. Available from https://sinaem4.agemed.es/10/09.

11. Beriplex(R). Ficha técnica Julio 2008. Available from https://sinaem4.agemed.es/ (accessed October 2009).

12. Dentali F, Ageno W, Crowther M. Treatment of coumarin-associated coagulopathy: a systematic review and proposed treatment algorithms. J Thromb Haemost. 2006;4(9):1853–1863.

13. Schulman S, Beyth RJ, Kearon C, Levine MN; American College of Chest Physicians. Hemorrhagic complications of anticoagulant and thrombolytic treatment: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edn). Chest. 2008;133(6 Suppl):257S–298S.

14. Broderick J, Connolly S, Feldmann E, et al; American Heart Association/ American Stroke Association Stroke Council; American Heart Association/American Stroke Association High Blood Pressure Research Council; Quality of Care and Outcomes in Research Interdisciplinary Working Group. Guidelines for the management of spontaneous intracerebral hemorrhage in adults: 2007 update: a guideline from the American Heart Association/American Stroke Association Stroke Council, High Blood Pressure Research Council, and the Quality of Care and Outcomes in Research Interdisciplinary Working Group. Circulation. 2007;116(16):e391–413.

15. Spahn DR, Cerny V, Coats TJ, et al; Task Force for Advanced Bleeding Care in Trauma. Management of bleeding following major trauma: a European guideline. Crit Care. 2007;11(1):R17. Erratum in Crit Care. 2007;11(2):414.

16. Leissinger CA, Blatt PM, Hoots WK, Ewenstein B. Role of prothrombin complex concentrates in reversing warfarin anticoagulation: a review of the literature. Am J Hematol. 2008;83(2):137–143.

17. Fries D, Haas T, Klinger A, et al. Efficacy of fibrinogen and prothrombin complex concentrate used to reverse dilutional coagulopathy—a porcine model. Br J Anesth. 2006;97(4):460–467.

18. Paramo JA, Monedero P, Hidalgo F, Hernandez M. Fundamentos básicos para el empleo de hemoderivados y estrategias de ahorro de sangre en cirugía. Reu Med Uniu Nauarra. 2008;53(3):9 14.

19. Farriols A, Gallur L, Rodriguez-Bueno L, Mendarte L, Montoro JB. Efficacy and tolerability of human fibrinogen concentrate administration to patients with acquired fibrinogen deficiency and active or in high-risk severe bleeding. Vox Sang. 2008;94(3):221–226.

20. Villar Camacho A, Juárez Giménez JC, Montoro Ronsano JB. Tratamiento de la hemorragia grave con fármacos hemoderivados. Aten farm. 2005;7(2):127–133.

21. Makris M. Emergency oral anticoagulant reversal: the relative efficacy of infusions of fresh frozen plasma and clotting factor concentrate on correction of the coagulopathy. Thromb Haemost. 1997;77:477–480.

22. Boulis NM, Bobek MP, Schmaier A, Hoff JT. Use of factor IX complex in warfarin-related intracranial hemorrhage. Neurosurgery. 1999;45:1113–1119.

23. Taberner DA. Comparison of prothrombin complex concentrate and vitamin K1 in oral anticoagulant reversal. BMJ. 1976;2(6027):83–85.

24. Pabinger I. Prothrombin complex concentrate (Beriplex P/N) for emergency anticoagulation reversal: a prospective multinational clinical trial. J Thromb Haemost. 2008;6(4):622–631.

25. Lorenz R. Successful emergency reversal of phenprocoumon anticoagulation with prothrombin complex concentrate: a prospective clinical study. Blood Coagul Fibrinolysis. 2007;18(6):565–570.

26. Riess HB. Prothrombin complex concentrate (Octaplex) in patients requiring immediate reversal of oral anticoagulation. Thromb Res. 2007;121(1):9–16.

27. Vigué B, Ract C, Tremey B, et al. Ultra-rapid management of oral anticoagulant therapy-related surgical intracranial hemorrhage. Intens Care Med. 2007;33:721–725.

28. Lavenne-Pardonage E, Itegwa MA, Kalaai M, et al. Emergency reversal of oral anticoagulation through PPSB-SD: the fastest procedure in Belgium. Acta Anaesthesiol Belg. 2006;57(2):121–125.

29. Lankiewicz MW, Hays J, Friedman KD, Tinkoff G, Blatt PM. Urgent reversal of warfarin with prothrombin complex concentrate. Thromb Haemost. 2006;4(5):967–970.

30. Lubetsky A, Hoffman R, Zimlichman R, et al. Efficacy and safety of a prothrombin complex concentrate (Octaplex) for rapid reversal of oral anticoagulation. Thromb Res. 2004;113(6):371–378.

31. Preston FE, Laidlaw ST, Sampson B, Kitchen S. Rapid reversal of oral anticoagulation with warfarin by a prothrombin complex concentrate (Beriplex): efficacy and safety in 42 patients. Br J Haematol. 2002;116(3):619–624.

32. Evans G, Luddington R, Baglin T. Beriplex P/N reverses severe warfarin-induced overanticoagulation immediately and completely in patients presenting with major bleeding. Br J Haematol. 2001;115(4):998–1001.

33. Broderick J, Connolly S, Feldmann E, et al; American Heart Association; American Stroke Association Stroke Council; High Blood Pressure Research Council; Quality of Care and Outcomes in Research Inter-disciplinary Working Group. Guidelines for the management of spontaneous intracerebral hemorrhage in adults: 2007 update: a guideline from the American Heart Association/American Stroke Association Stroke Council, High Blood Pressure Research Council, and the Quality of Care and Outcomes in Research Interdisciplinary Working Group. Circulation. 2007;116(16):e391–413. Republished in Stroke. 2007;38(6):2001–2023.

34. Levine MN, Raskob G, Beyth RJ, Kearon C, Schulman S. Hemorrhagic complications of anticoagulant treatment: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest. 2004;126(3 Suppl):287S–310S.

35. Spahn DR, Cerny V, Coats TJ, et al; Task Force for Advanced Bleeding Care in Trauma. Management of bleeding following major trauma: a European guideline. Crit Care. 2007;11(1):R17.

36. Program director, Clinical R&D CSL Behring. Efficacy and safety study of BERIPLEX(r) P/N compared with plasma in patients on anticoagulant therapy who require emergency surgery or invasive intervention. NCT00803101 (online).

37. Frenze, IMD. Octapharma USA. Study of OCTAPLEX and FFP in patients under vit K therapy antagonist needing urgent surgery or invasive procedures. NCT00618098.

38. Bradley SD. Factor VII, prothrombin complex concentrate, and fresh frozen plasma in warfarin-related intracranial hemorrhage. Available from http://clinicaltrials.gov/ct2/show/NCT00770718.

39. Leissinger CA, Blatt PM, Hoots WK, Ewenstein B. Role of prothrombin complex concentrates in reversing warfarin anticoagulation: a review of the literature. Am J Hematol. 2008;83:137–147.

40. Mendarte L, Factor IX complex drug use evaluation in a university general and orthopedic surgery 1000-bed hospital. ASHP Midyear Clinical Meeting 2000. Am J Health Syst Pharm. 2000;57(20):1908.

41. Stuklis RG, O'Shaughnessy DF, Ohri SK. Novel approach to bleeding in patients undergoing cardiac surgery with liver dysfunction. Eur J Cardiothorac Surg. 2001;19(2):219–220.

42. Lorenz R, Kienast J, Otto U, et al. Efficacy and safety of a prothrombin complex concentrate with two virus-inactivation steps in patients with severe liver damage. Eur J Gastroenterol Hepatol. 2003;15(1):15–20.

43. Böhrer H. Prothrombin complex concentrate substitution during liver transplantation. Thromb Res. 1999;95(4 Suppl 1):S71–74.

Warfarin Use and Risk of Valvular Calcification

Tue, 06 Jul 2010 05:58:30 +0100

INTRODUCTION

Warfarin, a vitamin K antagonist (VKA), has been the mainstay of oral anticoagulant therapy for many years. In the United States alone, about 2.5 million patients are estimated to be on long-term treatment with warfarin, principally for thromboprophylaxis in the presence of atrial fibrillation or a mechanical heart valve. Warfarin exerts its effect by inhibiting vitamin K epoxide reductase (VKOR) in the liver, which is needed for the synthesis of functional clotting factors II, VII, IX, and X. Warfarin also affects the synthesis and function of the matrix Gla protein (MGP), a vitamin K-dependent protein, which is a potent inhibitor of tissue calcification. Animal studies have demonstrated that warfarin can cause calcification in vivo and ex vivo. Recently, there have been human studies in which it has been implicated as one of the factors associated with calcification. In this review, we summarize the current data on the role of warfarin in vascular and valvular calcification.

VITAMIN K-DEPENDENT PROTEINS

There are many vitamin K-dependent proteins that need y-carboxylation for their physiologic activity. The clotting factors: II, VII, IX and X [1-4] are y-carboxylated in the liver to be functionally active. The regulatory factors protein C, protein S, and protein Z [2, 5] are y-carboxylated predominantly in the liver and to some extent in extrahepatic tissues. The other proteins are bone Gla protein osteocalcin [6-8], the calcification-inhibiting MGP [9-11], growth arrest-specific gene 6 protein (Gas6) [12-15], and transmembrane Gla proteins (TMGPs). The precursors of these factors require carboxylation of their glutamic acid residues to allow the coagulation factors to bind to phospholipid surfaces. This carboxylation is linked to oxidation of vitamin K to form vitamin K epoxide, which is in turn recycled back to the reduced form by the enzyme VKOR. Warfarin inhibits the epoxide reductase (specifically the VKORCi subunit), thereby diminishing available vitamin K stores and inhibiting the production of functioning coagulation factors.

ANIMAL STUDIES ON WARFARIN-INDUCED CALCIFICATION

Animal experimentation has demonstrated that either the lack of MGP or the use of VKAs has effects on cardiovascular calcification. An initial study done on mice lacking MGP showed that they developed spontaneous calcification of arteries and cartilage and died of vascular rupture 2 months later. This identified MGP as the first recognized inhibitor of calcification in vivo [16].
Howe and Webster [17] designed a study to cause extrahepatic vitamin K deficiency with a warfarin treatment regimen. Rats were treated from birth for 5-i2 weeks with daily doses of warfarin and concurrent vitamin Ki. This treatment was presumed to cause extrahepatic vitamin K deficiency without affecting the vitamin K-dependent blood clotting factors. At the end of treatment, examination of the vascular system of these rats revealed extensive arterial calcification.
A study by Price et al [18] showed that warfarin causes focal calcification of the elastic lamellae in the tunica media of major arteries and in aortic valves in the rat. These investigators found that the calcification of arteries induced by warfarin was similar to that seen in the MGP-deficient mouse, and suggested that warfarin induces arterial calcification by inhibiting carboxylation of MGP, thereby inactivating the putative calcification-inhibitory activity of the protein. Later, these investigators also showed that concurrent warfarin administration increased the extent of calcification in the media of vitamin D-treated rats [19]. It was hypothesized that living arteries secrete the calcification inhibitor MGP; inactivation of MGP with warfarin causes living arteries to calcify; and that the addition of MGP to medium containing warfarin prevents this calcification. In a subsequent study [20], they demonstrated that addition of warfarin to culture medium caused extensive Alizarin red staining for calcification in the living carotid artery segment, whereas no staining could be detected in living carotid arteries incubated in the same medium without warfarin. No calcification could be detected if the living arteries were incubated in culture medium containing warfarin but not serum, which confirms the serum requirement for artery calcification in this system. Purified bovine MGP also prevented warfarin-induced calcification of devitalized arteries in the same medium.
Sweatt et al [21] elucidated the interaction of MGP and bone morphogenetic protein-2 (BMP-2). Using immunohistochem-istry, these investigators showed that calcified lesions in the aortic wall of aging rats contained elevated concentrations of MGP that was poorly y-carboxyiated and did not bind BMP-2. They demonstrated the existence of a BMP-2/MGP complex in vivo, consistent with a roie for MGP as a BMP-2 inhibitor. They postuiated that age-reiated arteriai caicification may be a consequence of under-y-carboxyiation of MGP, allowing unopposed BMP-2 activity.

HUMAN STUDIES ON WARFARIN-INDUCED CALCIFICATION

In humans, various studies have shown the association of warfarin use with vascuiar and cardiac vaivuiar caicification.

Vascular calcification

A series of 16 patients with cutaneous necrosis from calcific uremic arterioiopathy identified warfarin as a risk factor [22]. In two of these patients, substitution of iow-moiecuiar-weight heparin for warfarin therapy resuited in ciinicai improvement.
Spronk et al [23] found that MGP accumuiated at the borders of vascuiar caicification in human tissue specimens. These investigators suggested that undercarboxyiated MGP is bioiogicaiiy inactive, and that poor vascuiar vitamin K status may be a risk factor for vascuiar caicification. Schori and Stungis [24] reported a case of arteriai caicification in a person who had iong-term treatment with warfarin, and cautioned that physicians prescribing warfarin for iong-term treatment shouid consider arteriai caicification as one of its potentiai consequences.
Verdaiies Guzman et al [25] did a retrospective anaiysis on eight femaie patients on hemodiaiysis who deveioped caiciphyiaxis. Six patients were receiving anticoaguiation therapy with warfarin. The patients did not have severe aiterations in caicium metaboiism; aii had a caicium-phosphorus product,55.7, and the patients had proximai iesions in fatty regions such as the abdomen and thighs. Histopathoiogic examination reveaied caicium deposits in arterioie-sized and smaii vesseis with vascuiar thrombosis. They conciuded that anticoaguiant therapy was one ofthe risk factors for deveioping caiciphyiaxis, in the absence of severe disorders of caicium metaboiism.
In a cross-sectionai anaiysis of 70 patients (46 men, mean age 68 + 13 years) on warfarin therapy without known coronary artery disease, a non-significant trend to increased coronary artery caicification (CAC) with increasing warfarin exposure (P=o.i8) was observed in univariate anaiysis. However, after adjustment for cardiovascuiar risk factors, no correiation between warfarin duration and CAC score (r=o.075, P=o.537) was observed on muitivariate anaiysis. They conciuded that warfarin exposure does not appear to piay a significant roie in potentiating arteriai caicification, as measured by eiectron beam computed tomography (CT) in a middie-aged to oider screening popuiation [26].

Valvular calcification

Schurgers et al [27] investigated whether iong-term orai anticoaguiant treatment may induce caicification in humans. These investigators measured the grade of aortic vaive caicification in vaives removed from patients undergoing surgicai vaive repiacement. Caicifications in vaives from patients receiving preoperative orai anticoaguiant treatment were significantiy (twofoid) iarger than in patients not receiving preoperative orai anticoaguiants. These investigators conciuded that orai anticoaguiants may induce cardio-vascuiar caicification as an adverse side-effect.
Koos et al [28] reported the association of warfarin use with aortic vaive caicification (AVC) and CAC assessed by muitisiice spirai CT. They studied 86 patients (53 men, mean age 7i+ 8 years) with caicific aortic vaive disease, 23 patients on iong-term warfarin therapy (mean duration 88+ ii3 months), and 66 patients without anticoaguiation. Patients on warfarin therapy had increased CAC (coronary Agatston score 1561 + 1141 vs 738 + 978; P=o.024) and AVC (vaivuiar Agatston score 2410 +1759 vs 1070 +1085; P=0.002) compared with patients without anticoaguiation treatment. Subsequentiy, they investigated the effect ofiong-term warfarin treatment on circuiating MGP ieveis in humans and on MGP expression in mice, and the association between circuiating inactive MGP (ucMGP) ieveis and the presence and severity ofAVC in patients with aortic vaive disease (AVD) [29]. They anaiyzed the circuiating ucMGP ieveis in i9i patients with echocardiogra-phicaiiy proven caicific aortic vaive disease (CAVD) and 35 controi subjects. They found that MGP ieveis were significantiy iower in patients with AVD (348.6+ i23.i nM) than in the controi group (571.6+ 153.9 nM, P<0.00i). In mice on warfarin, mRNA expression of MGP in the aorta was down-reguiated. They conciuded that patients with caicific AVD had significantiy iower ieveis of circuiating ucMGP compared with a reference popuiation, free from coronary and vaivuiar calcifications. In addition, warfarin treatment may decrease the local expression of MGP, resulting in decreased circulating MGP levels and subsequently increased aortic valve calcifications as an adverse side-effect.
In a retrospective cohort study among hemodialysis patients by Holden et al [30], the odds ratio of falling into a higher category of AVC following 18 months of warfarin use was not statistically significant (P=o.055). However, there was an association between lifetime months of warfarin exposure and severity of AVC (P= 0.004), which was independent of dialysis use, calcium and calcitriol intake. These investigators suggested that warfarin use may be associated with the severity of AVC in hemodialysis patients.
In the Japanese Aortic Stenosis Study (JASS) [31], a retrospective observational study of 556 subjects aged .50 years and with calcification in any aortic valve leaflet or peak aortic jet velocity $2 m/s, the use of warfarin was identified as a prognostic factor in early-stage disease (peak aortic jet velocity of $2 m/s), but not for late-stage disease (peak aortic jet velocity of $3 m/s). These investigators concluded that we should be vigilant about progression of CAVD in patients treated with warfarin.
We had previously reported the incidence of mitral valve calcium (MVC), mitral annular calcium (MAC), and AVC with two-dimensional echocardiograms in ii55 patients with non-valvular atrial fibrillation. Of these ii55 patients, mean age 74 years, 725 (63%) were treated with warfarin and 430 (37%) without warfarin. MVC, MAC, or AVC was present in 473 of 725 patients (65%) on warfarin vs 225 of 430 patients (52%) not on warfarin (P<0.000i). On stepwise logistic regression analysis, there was a significant association between the use of warfarin and the risk of calcification (unadjusted odds ratio= i.7i, 95% CI i.34-2.i8) after adjustment for confounding risk factors. We concluded that the use of warfarin in patients with non-valvular atrial fibrillation (AF) is associated with an increased prevalence of MVC, MAC, or AVC [32].

Treatment

Until recently, calcification was thought to be an irreversible phenomenon. But current understanding dictates that calcification is an active process with inhibitors and stimulators of calcification. Being a recently recognized phenomenon, current guidelines do not discuss the screening, prevention, or treatment of warfarin-induced calcification. Stopping warfarin and using an alternative anticoagulant can help. This has been shown in patients with cutaneous necrosis from calcific uremic arteriolopathy [22]. In another case report, a patient with biopsy-proven calciphylaxis thought to be attributable to warfarin was treated with a therapeutic substitution of anticoagulant and hyperbaric oxygen therapy leading to resolution ofcutaneous lesions [33]. In a study on rats, atorvastatin use was demonstrated to protect aortic media from warfarin-induced calcification [34]. Alendronate [35] has also been shown to have a dose-dependent protective effect on vitamin D and warfarin-induced calcification in rats, but there are no data on its use in humans. Osteoprotegerin, a secreted protein of the tumor necrosis factor family that inhibits osteoclast differentiation and activation, has been shown to protect against warfarin-induced calcification in rats [36]. Warfarin-induced vascular calcification in rats was prevented, and in some cases reversed, by high vitamin K intake [37]. Vascular smooth muscle cell (VSMC)-derived apoptotic vesicles were loaded with calcification inhibitors, including MGP, and these vesicles had pro-mineralizing properties when MGP function was impaired [38]. High vitamin K intake was associated with significantly less VSMC apoptosis and with significant regression of arterial calcification. Calcium deposits were removed by phagocytosis carried out by the surrounding VSMCs [39]. Vitamin K is present in different forms. Vessel walls specifically accumulate K2, even when the diet contains exclusively Ki [40]. Spronk et al [4i] showed that K2 (MK4) is more effective than Ki in preventing calcification during warfarin treatment. However, Schurgers et al [37] proposed that, in the high-Ki group, Ki had been converted to K2 to such an extent that arterial K2 had comparable tissue concentrations as in the K2 (MK4)-treated group. They concluded that, at very high intakes of Ki (200fold the daily requirement of the liver), both these forms of vitamin K might help to decrease arterial calcification.

CONCLUSION

Although animal studies have proved that warfarin causes calcification, and human studies have shown this strong association, still a large-scale, prospective, randomized, controlled trial is needed to establish this effect and to determine the rate of progression. It is still controversial whether warfarin-induced calcification is clinically significant enough to affect the morbidity and mortality associated with vascular and valvular calcification. Warfarin still remains the least expensive and most widely available mode of anticoagulation with long-term expertise in managing it and an inexpensive test for monitoring the dose. It has a proven benefit in preventing strokes in AF patients, recurrent deep venous thrombosis, and pulmonary embolism, and hence, reducing mortality in patient populations at risk for thromboembolic disease. Warfarin is also fairly quickly reversible in case of bleeding, and can be safely used in patients with low creatinine clearance. However, the use of warfarin has always been difficult because of complex pharmacodynamics, a narrow therapeutic window, numerous drug–drug and drug–food interactions, and multiple adverse effects. New oral anticoagulants with different mechanisms of action than warfarin have been developed: the direct thrombin inhibitor dabigatran etexilate and the factor Xa inhibitors rivaroxaban and apixaban have shown promise in early trials. With the advent of these newer anticoagulants, this association of warfarin and calcification needs to be examined urgently. Further studies are needed to determine the target population that needs to be screened for calcification, an appropriate and cost-effective screening modality, prevention strategies, and treatment of warfarin-induced calcification.

Disclosures: The authors declare no conflict of interest.

References

Berkner KL. Vitamin K-dependent carboxyiation. Vitam Horm. 2008178: i3i-i56.
Furie B, Furie BC. The moiecuiar basis of biood coaguiation. Cell. i988;
53:505-5i8.
Shearer MJ, Newman P. Metaboiism and ceii bioiogy of vitamin K. Thromb Haemost. 2008;i00:530-547.
Bristoi JA, Ratciiffe JV, Roth DA, Jacobs MA, Furie BC, Furie B. Biosynthesis of prothrombin: intraceiiuiar iocaiization of the vitamin K-dependent carboxyiase and the sites of gamma-carboxyiation. Blood. i996; 88:2585-2593.
Sadier JE. Medicine: K is for koaguiation. Nature. 2004;427:493-494.
Bugei S. Vitamin K and bone heaith in aduit humans. Vitam Horm. 2008; 78:393-4i6.
Heiss C, Hoesei LM, Wehr U, et al. Diagnosis of osteoporosis with vitamin K as a new biochemicai marker. Vitam Horm. 2008;78:4i7-434.
Hauschka PV, Lian JB, Coie DE, Gundberg CM. Osteocaicin and matrix Gia protein: vitamin K-dependent proteins in bone. Physiol Rev. i989;69:
990-i047.
9.Schurgers LJ, Cranenburg EC, Vermeer C. Matrix Gia-protein: the
caicification inhibitor in need of vitamin K. Thromb Haemost. 2008;i00:
593-603.
Cranenburg EC, Schurgers LJ, Vermeer C. Vitamin K: the coaguiation vitamin that became omnipotent. Thromb Haemost. 2007;98:i20-i25.
Proudfoot D, Shanahan CM. Moiecuiar mechanisms mediating vascuiar caicification: roie of matrix Gia protein. Nephrology (Carlton). 2006;ii:455-46i.
Beiiido-Martin L, de Frutos PG. Vitamin K-dependent actions of Gas6. Vitam Horm. 2008;78:i85-209.
Borgei D. Gas6 inflames ceii interactions. Blood. 2008;iii(8):39i5.
Hafizi S, Dahiback B. Gas6 and protein S. Vitamin K-dependent iigands for the Axi receptor tyrosine kinase subfamiiy. FEBS J. 2006;273(23):523i-
5244.
Waizer T, Vivier E. NK ceii deveiopment: gas matters. Nature Immunol. 2006;7(7):702-704.
Luo G, Ducy P, McKee MD, et al. Spontaneous caicification ofarteries and cartiiage in mice iacking matrix GLA protein. Nature. i997;385:78-8i.
Howe AM, Webster WS. Warfarin exposure and caicification of the arteriai system in the rat. Int J Exp Pathol. 2000;8i(i):5i-56.
Price PA, Faus SA, Wiiiiamson MK. Warfarin causes rapid caicification of the eiastic iameiiae in rat arteries and heart vaives. Arterioscler Thromb Vasc Biol. i998;i8:i400-i407.
Price PA, Faus SA, Wiiiiamson MK. Warfarin-induced artery caicification is acceierated by growth and Vitamin D. Arterioscler Thromb Vasc Biol. 2000;
20:3i7-327.
Price PA, Chan WS, Joison DM, Wiiiiamson MK. The eiastic iameiiae of devitaiized arteries caicify when incubated in serum: evidence for a serum caicification factor. Arterioscler Thromb Vasc Biol. 2006;26(5):i079-i085.
Sweatt A, Sane DC, Hutson SM, Waiiin R. Matrix Gia protein (MGP) and bone morphogenetic protein-2 in aortic caicified iesions of aging rats. J Thromb Haemost. 2003;i:i78-i85.
Coates T, Kirkiand GS, Dymock RB, et al. Cutaneous necrosis from caicific uremic arterioiopathy. Am J Kidney Dis. i998;32:5i4-5i8.
Spronk HMH, Soute BAM, Schurgers LJ, et al. Matrix Gia protein accumuiates at the border ofregions ofcaicification and normai tissue in the media of the arteriai vessei waii. Biochem Biophys Res Commun. 200i; 289:485-490.
Schori TR, Stungis GE. Long-term warfarin treatment may induce arteriai caicification in humans: case report. Clin Invest Med. 2004;27:i07-i09.
Verdaiies Guzman U, de ia Cueva P, Verde E, et al. [Caiciphyiaxis: fatai compiication of cardiometaboiic syndrome in patients with end stage kidney disease]. Nefrologia. 2008;28(i):32-36.
Viiiines TC, O'Maiiey PG, Feuerstein IM, Thomas S, Tayior AJ. Does proionged warfarin exposure potentiate coronary caicification in humans? Resuits of the warfarin and coronary caicification study. Calcif Tissue Int. 2009;85(6):494-500.
Schurgers LJ, Aebert H, Vermeer C, Biiitmann B, Janzen J. Orai anticoaguiant treatment: friend or foe in cardiovascuiar disease? Blood. 2004;i04:323i-3232.

Koos R, Mahnken AH, Muehienbruch G, et al. Reiation of orai anticoaguiation to cardiac vaivuiar and coronary caicium assessed by muitisiice spirai computed tomography. Am J Cardiol. 2005;96:747-749.
Koos R, Krueger T, Westenfeid R, et al. Reiation of circuiating Matrix Gia-Protein and anticoaguiation status in patients with aortic vaive caicification. Thromb Haemost. 2009;i0i(4):706-7i3.
Hoiden RM, Sanfiiippo AS, Hopman WM, Zimmerman D, Gariand JS, Ross A. Warfarin and aortic vaive caicification in hemodiaiysis patients. J Nephrol. 2007;20:4i7-422.
Yamamoto K, Yamamoto H, Yoshida K, Kisanuki A, Hirano Y, Ohte N, Akasaka T, Takeuchi M, Nakatani S, Ohtani T, Sozu T, Masuyama T. Prognostic factors for progression of eariy- and iate-stage caicific aortic vaive disease in Japanese: the Japanese Aortic Stenosis Study (JASS) Retrospective Anaiysis. Hypertens Res. 20i0 Mar;33(3):269-74.
Lerner RG, Aronow WS, Sekhri A, et al. Warfarin use and the risk of vaivuiar caicification. J Thromb Haemost. 2009;7(i2):2023-2027.
Banerjee C, Woiier SC, Hoim JR, Stevens SM, Lahey MJ. Atypicai caiciphyiaxis in a patient receiving warfarin then resoiving with cessation ofwarfarin and appiication ofhyperbaric oxygen therapy. Clin Appl Thromb Hemost. 2009 Dec i6. [Epub ahead of print]
Liu C, Wan J, Yang Q, Qi B, Peng W, Chen X. Effects of atorvastatin on warfarin-induced aortic mediai caicification and systoiic biood pressure in rats. J Huazhong Univ Sci Technolog Med Sci. 2008^8(5^535-538.
Li H, Jia GL, Wang HC, et al. [The effect of aiendronate on arteriai caicification in rat modei]. Zhonghua Nei Ke Za Zhi. 2006;45(6):489-492.
Price PA, June HH, Buckiey JR, Wiiiiamson MK. Osteoprotegerin inhibits artery caicification induced by warfarin and by vitamin D. Arterioscler Thromb Vasc Biol. 200i;2i(i0):i6i0-i6i6.
Schurgers LJ, Spronk HM, Soute BA, Schiffers PM, DeMey JG, Vermeer C. Regression of warfarin-induced mediai eiastocaicinosis by high intake of vitamin K in rats. Blood. 2007;i09(7):2823-283i.
Reynoids JL, Joannides AJ, Skepper JN, et al. Human vascuiar smooth muscie ceiis undergo vesicie-mediated caicification in response to changes in extraceiiuiar caicium and phosphate concentrations: a potentiai mechanism for acceierated vascuiar caicification in ESRD. J Am Soc Nephrol. 2004^5:2857-2867.
Proudfoot D, Davies JD, Skepper JN, Weissberg PL, Shanahan CM. Acetyiated iow-density iipoprotein stimuiates human vascuiar smooth muscie ceii caicification by promoting osteobiastic differentiation and inhibiting phagocytosis. Circulation. 2002^06:3044-3050.
Ronden JE, Drittij-Reijnders MJ, Vermeer C, Thijssen HH. Intestinai fiora is not an intermediate in the phyiioquinone-menaquinone-4 conversion in the rat. Biochim Biophys Acta. i998;i379:69-75.
Spronk HM, Soute BA, Schurgers LJ, Thijssen HH, De Mey JG, Vermeer C. Tissue-specific utiiization of menaquinone-4 resuits in the prevention of arteriai caicification in warfarin-treated rats. J Vasc Res. 2003;40:53i-
537.

Genetic Testing for the Diagnosis of Von Willebrand Disease: Benefits and Limitations

Mon, 28 Jun 2010 09:41:02 +0100

INTRODUCTION

von Willebrand disease and von Willebrand factor

von Willebrand disease (VWD) is the most common inherited bleeding disorder and arises from deficiencies and/or defects in the plasma protein von Willebrand factor (VWF) [] (). VWD is classified into six different types, with Type 1 identified as a (partial) quantitative deficiency of VWF, Type 3 a (virtually) total (quantitative) deficiency of VWF, and Type 2 identifying four separate types (2A, 2B, 2M, 2N) characterized by qualitative defects [] (). The classification is currently based on phenotypic assays, supplemented by multimeric analysis of the von Willebrand protein, and genetic analysis is not required in order to diagnose VWD or to define a patient's VWD classification type.

VWF is a large and complex multimeric protein that has two main recognized functions to facilitate arrest of bleeding following injury (ie, promotes adhesion of platelets to each other and to the vasculature [“primary hemostasis”], and binds and stabilizes factor VIII [FVIII; for use in “secondary” hemostasis]) []. To accomplish these functions, VWF contains several functional binding sites in addition to those that are classically recognized as facilitating binding of VWF to its major platelet receptor glycoprotein Ibα (GPIBA) and to FVIII, including binding sites for other platelet receptors (eg, αIIbβ3) and subendothelial matrix components such as collagen. It is also important to recognize that a complex process of manufacture, storage, secretion, proteolysis, and clearance controls the steady-state VWF plasma concentration and composition [–]. Only some of these mechanisms have been elucidated.

Phenotypic characterization and diagnosis of VWD

The diagnosis of VWD is normally based on clinical evidence of a bleeding or bruising tendency (typically mucocutaneous; both personal and family histories are considered), supported by laboratory evidence of low levels of VWF and/or abnormal VWF function [, ]. In diagnostic laboratories, VWF is typically measured using a panel of tests that evaluate the total plasma VWF protein (VWF “antigen”; VWF:Ag), as well as VWF “activities” (eg, ristocetin cofactor (VWF:RCo); collagen binding (VWF:CB)). FVIII coagulant activity (FVIII:C) is also evaluated. In specialist centers, additional functional assays may be undertaken (eg, FVIII binding (VWF:FVIIIB); ristocetin-induced platelet agglutination (RIPA)), and the multimeric structure analyzed by protein electrophoresis.

In Type 1 VWD, there is a partial deficiency of VWF and the various assays show similar (concordant) reductions in VWF:Ag, FVIII:C, VWF:RCo, and VWF:CB. Multimeric analysis if performed would show a normal multimeric structure. In Type 3 VWD, VWF is virtually absent.

In Type 2 VWD, VWF:Ag is usually low, but may be normal; however, functional tests will show an abnormality, helping to identify the type of VWD [, ] (). In Type 2A VWD, the presiding defect is represented by a loss of high molecular weight (HMW) VWF, either due to faulty assembly or enhanced proteolysis []; this deficiency can be identified either using VWF multimer analysis or by assessing VWF activity, where VWF:RCo and VWF:CB will be lower than VWF:Ag (since HMW VWF is lacking, the residual low molecular weight [LMW] VWF will be identified by the VWF:Ag assay, but not by the functional VWF assays). This test pattern is recognized as VWF functional , and expressed numerically as a ratio of functional VWF (VWF:RCo or VWF:CB) to VWF:Ag. Type 2A VWD is therefore typically characterized by VWF:RCo/VWF:Ag (RCo/Ag) VWF:CB/VWF:Ag (CB/Ag) ratios below around 0.7 [, , ] (). Type 2B VWD is characterized by enhanced binding of VWF (ie, hyper-adhesive activity) to its platelet receptor (GP1BA), and this property is identified by elevated responsiveness in the RIPA assay [, , ] (). This enhanced binding typically causes some clearance or loss of HMW VWF (the most adhesive forms), as well as clearance of VWF-bound platelets (ie, mild thrombocytopenia). However, some “atypical” forms of 2B VWD will not show these latter features [, ]. Like 2A VWD, loss of HMW VWF can be identified using VWF multimer analysis, or surrogate markers such as VWF:CB and VWF:RCo. Phenotypically, then, Type 2B VWD cases typically express RCo/Ag and CB/Ag ratios under around 0.7, particularly under periods of stress, and this pattern may cause some 2B VWD patients to be misidentified as 2A VWD if RIPA is not performed. Type 2M VWD is recognized as an inherent VWF defect, where loss of VWF function is not associated with a loss of HMW VWF. Thus, although multimer analysis may show some structural abnormalities, HMW multimers will still be present, and the pattern may more closely resemble a Type 1 VWD. Most cases of Type 2M can be shown to have defective binding to the VWF platelet receptor GP1BA; hence, VWF:RCo, which functionally detects this binding, tends to show a lower value (ie, VWF functional ) than VWF:Ag, with RCo/Ag ratios below around 0.7. However, since collagen binding does not seen to be as affected as platelet receptor binding in these individuals, VWF:CB values are similar to VWF:Ag (ie, show concordance), and CB/Ag ratios are therefore typically above 0.7 [, , ] (). Conversely, some rare forms of 2M VWD show the opposite pattern, since the defect in VWF affects collagen binding but not GP1BA binding, phenotypically leading to low CB/Ag ratios, with normal RCo/Ag ratios [, , , , ]. 2N VWD is recognized as an inherent VWF defect causing loss of FVIII binding [, , ] (). Thus, plasma FVIII is more labile, prone to proteolysis, and FVIII:C tends to be lower than VWF (ie, a low FVIII/VWF ratio is evident). Phenotypically, patients with 2N VWD tend to mimic hemophilia A or hemophilia A carriers. 2N VWD is phenotypically identified using the VWF:FVIIIB assay, which will show low FVIIIB/VWF ratios in 2N VWD, but these will be normal in hemophilia A (/carriers) [].

Phenotypic assay limitations and additional useful investigative approaches

Although the phenotypic characterization of patients with presumed VWD should be straightforward and enable the effective diagnosis or exclusion of VWD, as well as its characterization (ie, the typing of VWD), several problems remain. The usual methods for establishing reference ranges for phenotype-based assays (ie, mean ±2×standard deviation) by definition ascribes ∼2.5% of the normal population as having a low level of VWF. There exists a considerable overlap of reference intervals for normal individuals and patients with apparent VWD. The phenotypic assays all have lower limits of sensitivity, and that for the VWF:RCo assay may be as high as 20 IU/dL []. This poses a difficulty in VWD diagnostics since the most severe forms of VWD are those most difficult to fully characterize, having levels of VWF:RCo usually below 20 IU/dL.

Two other significant problems with phenotypic characterization are assay reproducibility and the use of small or inappropriate test panels. Intra-assay, inter-assay, and inter-laboratory variability tends to be considerable, especially for VWF:RCo [, ]. Thus, different assay values will be obtained with different VWF tests on different occasions or in different laboratories, leading to the misidentification of non-VWD individuals as VWD (false positive), the misidentification of VWD as not VWD (false negative), or the misidentification of VWD as variable types of VWD (Type 1, 2A, or 2M, etc.), depending on the presiding test results and test panel employed on that testing occasion. Diagnostic problems are thus exacerbated when laboratories use small test panels, and of some significant but under-recognized importance, the VWF test panels used in laboratories are insufficient to properly identify phenotypically all cases of VWD.

Some diagnostic problems can be minimized by the use of improved methodology (eg, enzyme-linked immunosorbant assay [ELISA] in place of electro-immuno diffusion [Laurel gel]), by extending the panel of tests used, and by repeat testing of patients using a fresh sample collected on another occasion for confirmation. The addition of a VWF:CB assay to a test panel of VWF:RCo, VWF:Ag, and FVIII:C, for example, will reduce phenotypic-based diagnostic error rates by around half [, ].

Another way to improve VWD diagnostics is to use the desmopressin (DDAVP) challenge test. DDAVP is a non-transfusional form of VWD therapy that results in the release of endogenous (endothelial cell stored) VWF, and is given to select patients with VWD. Notably, DDAVP is (nearly invariably) effective in cases of Type 1 VWD with VWF levels above 30 IU/dL, is (usually) effective in other cases of Type 1, and is sometimes effective in Type 2 VWD []. It is usual to conduct a trial of DDAVP in VWD patients to assess clinical utility, since although responsiveness varies individuals, responsiveness is typically stable over time individuals. There is additional but under-recognized utility in assessing data from DDAVP trials for the diagnosis of VWD, because the response profile is often characteristic of a given form of VWD [, , ]. In brief, FVIII and VWF test parameters will rise in individuals with Type 1 VWD, and RCo/Ag and CB/Ag ratios will remain above 0.7 (). In Type 2A VWD, FVIII and VWF:Ag will primarily rise, but VWF:RCo and VWF:CB will not rise substantially, or their response will be transient, and RCo/Ag and CB/Ag ratios will therefore typically remain below 0.7. In platelet dysfunction binding Type 2M VWD, FVIII and VWF:Ag will again primarily rise, but so too will VWF:CB; however, VWF:RCo will not rise substantially, therefore CB/Ag ratios will typically remain above 0.7, but RCo/Ag ratios will remain below 0.7. Thus, in a patient where the VWD type is unclear, the DDAVP test pattern response can help to assign the VWD type. The potential use of the PFA-100 in this setting should also be considered, as initially prolonged closure times will tend to shorten and normalize in Type 1 VWD, but will not normally correct in Types 2A and 2M VWD [, , ] ().

GENETIC TESTING IN VWD DIAGNOSTICS—BENEFITS AND LIMITATIONS

There are a number of (largely theoretical) benefits to genetic testing in VWD, but these need to be tempered by genuine practical limitations [] (). Genetic testing is sometimes considered diagnostically superior to phenotypic testing, because when successful, it may provide a distinct answer (ie, evidence of a discrete mutation or not), whereas, as highlighted above, phenotypic testing in any individual will generate a continuum of test values, with overlap of results from normal individuals and those with VWD, and different values obtained with different procedures in different laboratories on different occasions. However, experience with genetic testing for VWD has also identified considerable limitations to its broader application in VWD. These limitations were recently highlighted in some detail [], and only a few are summarized here to help provide a balanced perspective (see ).

Genetic testing provides only one piece of the jigsaw puzzle that a VWD diagnosis represents. In family studies, the heritability of VWF levels across the population distribution is low (only 25–30%) []. Factors influencing VWF variance includes, for example, the ABO blood group, which is believed to account for 20–30% of the total heritable variance of VWF []. There are many other influences on, or modifiers of, plasma VWF level and function, including epigenetic events, platelet and endothelial cell activity, hormonal influences (eg, menstrual and pregnancy related), and ADAMTS13 level and activity [], and there are probably many other influences and modifiers that we simply have yet to identify. The latter could include various unknown influences related to the manufacture, storage, secretion, proteolysis, and clearance of VWF. Various additional environmental factors can also contribute to alter VWF levels, including stress, exercise, medications, illness, and inflammation []. In total, these events mean that there may be a large disconnection between the genetic situation (ie, mutations, polymorphisms) and the measured phenotype.

The molecular analysis of the gene is also complex and problematic []. The gene is located on chromosome 12, and analysis of the gene is complicated by several factors in addition to its large size []. There is a partial pseudogene copy on chromosome 22 that recapitulates the sequences of several exons with minor sequence variance. This necessitates the use of carefully chosen primers to ensure that the “true” gene is evaluated. The complexity of whole gene sequencing also carries considerable cost.

There is also increasing evidence that the gene is highly polymorphic, with over 150 polymorphisms so far identified, more than 50 of which alter the coding sequence. The distinction between neutral sequence variation and pathologic mutations at this locus may require functional expression studies and often remains challenging.

All these issues have been found problematic in genetic studies of VWF. In VWD, especially Type 1 VWD, mutations are spread across the whole gene, which necessitates whole gene sequencing. In the thorough genetic studies of Type 1 VWD, only in a proportion of patients was a mutation found. Experts have differentially assigned several single mutations to various subtypes of VWD in the international database and the literature []. Thus, our view remains that genetic testing will sometimes provide a useful adjuvant to clinical evaluation and phenotypic testing, but this is not likely to be the general case. We will explore the utility of genetic testing in the following section, using various examples to highlight both the strengths and limitations of a genetic approach. Readers particularly interested in genetic testing are also directed to other reports dealing with this issue for similar, alternate, or additional viewpoints [, –].

WHEN IS AND WHEN SHOULD GENETIC TESTING BE CONSIDERED?

We believe there are a variety of situations in which genetic testing could be considered potentially useful (), although the greatest clinical utility lies within the proper identification of Type 2 VWD.

Type 1 VWD

For Type 1 VWD: (i) most genetic investigations will prove fruitless (no mutation detected); (ii) the plasma VWF level is particularly influenced by non- gene factors; and (iii) since mutations can appear anywhere on the gene, testing will also be very expensive (∼$2000). Many identified mutations require confirmation by expression studies to demonstrate the mutation to be significant or causal. Accordingly, genetic testing in Type 1 VWD is not considered diagnostically useful, particularly where VWF levels are >30 IU/dL. The potential exceptions for genetic testing in Type 1 VWD are (i) in family studies, where a mutation has already been found; or (ii) where VWF levels fall below around 20 IU/dL, since there is a good likelihood here of identifying a causal mutation (between 85% and 100%)—however, many of these will turn out to be Type 2 VWD “masquerading” as Type 1 []. The clinical utility of genetic testing related to therapy of VWD is also low, since the choices (DDAVP, VWF concentrate, anti-fibrinolytics) are typically associated with the level of presenting VWF rather than the presenting genetic mutation.

Type 3 VWD

For Type 3 VWD, phenotypic testing is usually fairly clear (eg, VWF:Ag and VWF:CB both <2 IU/dL, FVIII:C <10 IU/dL; although low-level assay sensitivity issues often lead VWF:RCo data to be uninformative, ie, “ < 10%”). In Type 3 VWD, the gene “solely” determines the plasma VWF level, and mutation analysis will likely identify a causal mutation(s). However, as in Type 1 VWD, mutations can appear anywhere on the gene, thus testing will also be very expensive (∼$2000) unless the causal mutation is already known. There are, however, some situations in which genetic testing may be justified, notably related either to prenatal diagnosis and/or family studies, or to assessment of potential alloantibody development risk [–, , ]. Importantly, Type 3 VWD is a rare and profoundly significant disorder. Thus, while justifiable, this is likely to represent only a small level of usage for genetic testing in Western society. Indeed, Type 3 VWD has an incidence rate of around 1–3 per million inhabitants in Western countries, although the incidence is substantially higher in developing countries []. Moreover, the incidence of alloantibody development in this VWD group is likely to represent a minor proportion of all Type 3 VWD patients [].

Type 2 VWD

Genetic testing is likely to have greatest utility for the investigation of Type 2 VWD, or the discrimination of various Type 2 VWD types, either from each other, from Type 1 VWD, or from other phenotypically similar disorders. While all VWD experts seem to agree on this point [, , , , ], the perception of relative utility in specific situations, of a genetic vs a more thorough phenotypic approach, will differ between experts, based on local populations, local practice, or personal biases. Of major relevance to supporting the application of genetic testing in Type 2 VWD is the high success rate (up to 100% in the French cohort experience []), the restriction of most mutations to discrete regions of the gene (supporting focused and less expensive genetic investigations), and much clearer clinical utility. Nevertheless, a small proportion of investigations (perhaps 10% or so) may not identify a causal mutation, even if the molecular search is extended beyond the genetic regions normally associated with Type 2 VWD defects. Nevertheless, the associated costs of genetic testing may be reasonable, particularly if phenotypic testing is unclear, as is sometimes the case. The following section looks at several of the more common situations, both from the perspective of the theoretical benefit, as well as the practical limitations.

Identification of Type 2B VWD (and discrimination from PT-VWD)

Although genetic testing can be applied to help diagnose Type 2B VWD, this is best initially achieved using a phenotypic approach and evaluation by RIPA. The greatest perceived utility of genetic testing in Type 2B VWD is to help distinguish this from platelet-type (PT-) or pseudo-VWD. PT-VWD phenotypically mimics Type 2B VWD (ie, typically presents with some loss of HMW VWF, VWF functional discordance expressed by RCo/Ag and CB/Ag ratios below around 0.7, [mild] thrombocytopenia, and enhanced responsiveness in a RIPA assay []). Also like 2B VWD, these phenotypic features arise because of enhanced binding of VWF to its platelet receptor (GP1BA); however, unlike 2B VWD, where this is due to a mutation in the gene, in PT-VWD, the hyper-adhesive activity results from a defect or mutation in the gene. There is clear clinical utility in discriminating PT-VWD from 2B VWD, because of differential therapeutic support (platelets in PT-VWD, VWF concentrate or possible DDAVP therapy in 2B VWD). There are two ways in which discrimination can be achieved—phenotypically using RIPA mixing assays [, ] or genetically through an evaluation of the and/or genes. The main decider for the approach undertaken then becomes a question of which methodology is employed in a particular locality.

There is also a question of relative prevalence to consider here. The approximate prevalence of 2B VWD is known [] and is believed to represent 1–5% of all cases of all VWD, or to affect between one and five individuals per million inhabitants. The prevalence of PT-VWD is largely unknown, since most cases of PT-VWD are incorrectly identified as either 2B VWD or idiopathic thrombocytopenia (ITP) [, ]. In our own limited experience, PT-VWD appears to occur at a rate of around 10% of 2B VWD. Also in our limited experience, RIPA mixing assays have proven to be 100% sensitive and specific for both 2B VWD and PT-VWD. Accordingly, RIPA mixing assays are always performed as the second-line test process for patients who are identified with a 2B/PT-VWD phenotype, with an enhanced RIPA assay being a mandatory requirement for both the provisional diagnosis of either 2B/PT-VWD and subsequent performance of the RIPA mixing assay.

Nevertheless, genetic testing can then be used to confirm the findings of the phenotypic test approach, or can be considered as an alternative to this approach where RIPA mixing is unavailable. Our own experience with genetic testing of 2B VWD and PT-VWD, and of RIPA mixing is largely reported elsewhere [–], and has also recently been extended with unpublished findings. Based on these data, we have identified approximately 30 cases of 2B VWD and three genetically confirmed cases of PT-VWD (two unrelated families, one in Australia and the other in New Zealand). The PT-VWD cases were all found to have the G249S mutation (previously identified as G233S) within

[]. This is a supposedly rare mutation, only described in two other studies to our knowledge (in one Japanese family [] and in one French [] family). The 2B VWD cases instead represented a variety of mutations. In total, we have genetically investigated 16 cases of presumed 2B VWD from nine unrelated families. The R1306W mutation (reportedly a common mutation) was identified in 7/16 (44%) cases and 3/9 (33%) of families. The R1308C mutation (also reportedly a common mutation) was identified in 3/16 (19%) cases and 2/9 (22%) families. Interestingly, the R1306L and R1341W mutations, both considered “rare” mutations, were, respectively, identified in 5/16 (31%) cases (2/9 [22%] families) and 1/16 (6%) cases (1/9 [11%] families). We have also recently identified a novel mutation (L1340P) in 1/16 (6%) cases (1/9 [11%] families).

To conclude 2B/PT VWD, genetic testing was successfully applied in all cases that we had provisionally assigned either as 2B or PT-VWD based on an initial thorough phenotypic workup (ie, 100% local success rate). Moreover, there is an interesting diversity of genetic mutations identified by such studies, and in time, we will build an understanding of the phenotypic–genetic relationships. The recent experience from the Italian group [] can perhaps best highlight this potential. On the other hand, genetic analysis in our patient cohort largely acted as a confirmatory procedure to a full phenotypic analysis including RIPA mixing studies, which otherwise provided for 100% sensitivity and specificity for 2B/PT-VWD.

In theory, the same genetic testing approach can be used to discriminate 2B VWD from other subtypes such as 2A, 2M, or 1 VWD, since limited phenotypic evaluation means that 2B VWD patients may be misidentified as 2A, 2M, or 1 VWD [, ]. However, genetic testing should not be applied as a surrogate for poor phenotypic workups, and the pre-genetic phenotypic evaluation for all cases of likely 2B VWD should be as complete as possible.

Identification of Type 2N VWD (and discrimination from hemophilia A or hemophilia A carrier)

Although genetic testing can be applied to help diagnose 2N VWD, initially this is again best achieved using a phenotypic approach including an evaluation of VWF:FVIIIB. Nevertheless, there is clear potential utility for genetic testing to help distinguish 2N VWD from hemophilia A or hemophilia A carrier presentations, because of similar phenotypic presentations (ie, all these patient types typically present with low levels of plasma FVIII:C and VWF vs FVIII:C discordance expressed by FVIII/VWF ratios under 0.7). In the case of 2N VWD, the defect is in the gene, causing loss of VWF-mediated binding to FVIII, and consequent loss of plasma FVIII:C due to enhanced proteolysis. In the case of hemophilia A (/carrier), the defect is in the gene. There is clear clinical utility in discriminating 2N VWD from hemophilia A (/carrier), because differential therapeutic support may be applied (VWF concentrate in 2N VWD, FVIII concentrate in hemophilia A; DDAVP therapy sometimes applied in both cases). There are also inheritance differences and implications related to counseling to consider.

There are two ways in which 2N VWD and hemophilia A (/carrier) can be distinguished—phenotypically using the specific VWF:FVIII binding (VWF:FVIIIB) assay [] or genetically through an evaluation of the and/or genes. The main decider for the approach, like that for 2B vs PT-VWD, again becomes the question of which methodology is employed within a particular locality. There is also a similar question related to prevalence. The approximate prevalence of 2N VWD is known [], and believed to represent (like 2B VWD) around 1–5% of all cases of VWD (or around 1–5 individuals per million inhabitants). In our experience, the prevalence of hemophilia A is likely to exceed that of 2N VWD by about 10:1, or contra-analogous to the case for 2B vs PT-VWD, 2N VWD appears to occur at a rate around 10% that of hemophilia A. Also in our limited experience, the VWF:FVIIIB assay has proven 100% sensitive and specific for 2N VWD. Accordingly, this assay is always performed as the second-line test process for patients who are identified with a 2N VWD/hemophilia A (/carrier) phenotype.

Nevertheless, genetic testing can be used to confirm the findings of the phenotypic test approach, or can be considered as an alternative to this approach where the VWF:FVIIIB assay is unavailable. Our own experience with genetic testing of 2N VWD, and the VWF:FVIIIB assay is largely reported elsewhere [, ], and has also recently been extended with unpublished findings. Based on these data, over 500 patients have been phenotypically evaluated for possible 2N VWD, of which 24 have provided evidence of low or equivocal VWF:FVIIIB-based findings consistent with potential 2N VWD (including heterozygotes, homozygotes, or compound defects). Genetic evaluation was performed in 16 of these cases, of which 15 were either homozygous (=5) or heterozygous (=10) for the common R854Q mutation. No mutation was found in one case definitely identified as 2N VWD by phenotypic testing, therefore, this is a presumed false negative by mutation analysis.

To conclude 2N VWD, a genetic mutation was found in most genetically investigated cases (15/16 [94%] local success rate). Moreover, all cases of genetic mutations identified a single common mutation (R854Q). Thus, although several other 2N VWD mutations have been identified in the literature and in the VWD database [], this would seem less likely within our geography. Thus, on balance, genetic testing can be used to discriminate 2N VWD from hemophilia A (/carrier), but phenotypic testing incorporating the VWF:FVIIIB assay has otherwise proved more valuable locally, identifying more patients with 2N VWD than were identified by genetic testing. Again, genetic testing should not be used as a surrogate for poor phenotypic workups, and the pre-genetic phenotypic assessment should be as comprehensive as possible. Additionally, in the case of 2N VWD, some occasional false negatives seem likely. Nevertheless, according to UK guidelines [], “A diagnosis of type 2N VWD based upon VWF–FVIII binding studies does not exclude the need for genetic analysis; family members may be carriers of a recessive type 2N allele and genetic diagnosis provides a means to clearly establish inheritance.”

Identification, confirmation or discrimination of 2M (or 2A) VWD

Unless a comprehensive phenotypic assessment has been undertaken, 2M VWD is sometimes phenotypically difficult to identify and/or discriminate from Types 1 and 2A VWD. Type 2M VWD is characterized by VWF dysfunction not due to loss of HMW VWF. Typically, 2M VWD cases show a relative VWF functional discordance characterized by low relative VWF:RCo activity, since most of the mutations interfere with platelet GP1BA binding, and VWF:RCo mimics this property of VWF. Thus, RCo/Ag ratios are typically ≤0.7 in these cases. By contrast, these mutations do not generally affect collagen binding; hence, CB/Ag ratios tend to be normal, or >0.7. Unfortunately, in typical test practice incorporating limited test panels (ie, no VWF:CB), single or duplicated phenotypic testing often fails to provide clarity because of the high variability in test results (particularly VWF:RCo), and the poor sensitivity of VWF assays (particularly VWF:RCo) at low levels of VWF.

Thus, 2M VWD cases will often provide phenotypic test results similar to those of Type 1 VWD or 2A VWD. Performing multimer analysis may not help if the intention is to discriminate 2M and 1 VWD, since neither type should show a loss of HMW VWF, and unless the multimers are of sufficient high quality to enable identification of abnormal features, Type 2M VWD could still be misidentified as Type 1. That 2M VWD is often misidentified as Type 1 has most clearly be shown in Type 1 VWD genetic studies, where upward of 25% or so of initially presumed Type 1 VWD could alternatively be identified as Type 2 post-study analysis, of which most would be 2M []. 2M and 2A VWD will also be difficult to differentiate should the laboratory perform a limited VWF test panel that excludes VWF:CB, as both types will show reduced RCo/Ag ratios.

Thus, one theoretical way to firm up a diagnosis of Type 2M (as opposed to Type 1 or 2A), is to undertake genetic testing. However, this is not very straightforward in practice, since different experts will often ascribe the same mutation to Type 1, 2A, and 2M VWD []. Therefore, another way to improve the identification of Type 2M (or 2A) VWD, or their discrimination, is to improve the phenotypic characterization of the patient under investigation. This would include four primary aspects [, , , ], namely, (i) expanding the test panel to include VWF:CB (ie, VWF:Ag, VWF:RCo, VWF:CB, and FVIII:C as the minimum test panel); (ii) repeating all tests at least once (and sometimes more often) on a separate occasion using a fresh blood sample for confirmation (to overcome assay variability issues); (iii) utilizing tests capable of low assay value sensitivity (ie, sensitive to VWF levels <5 IU/dL); and (iv) consideration of a DDAVP trial ().

Nevertheless, genetic analysis may be helpful in a proportion of cases, and this should be evaluated on a case-by-case basis. However, be warned that this approach may sometimes confuse rather than clarify the diagnosis, and in most cases will actually provide little additional clinical benefit since treatment options in these cases are limited (DDAVP [if effective; accordingly a trial is mandatory] and/or VWF concentrate). Our experience of genetic testing in these cases is also limited, given that we believe that an effective diagnosis can usually be achieved using a thorough phenotypic approach (as explained above), and because genetic testing will not invariably help. Nevertheless, we will summarize our recent experience () to highlight some strengths and limitations of the genetic approach in this setting.

In brief, over the last 6 months, we have genetically assessed, using a site-specific focused approach (ie, exon 28 of the gene), a total of 12 cases that, based on various levels of phenotypic evidence, we felt were probable or possible 2A or 2M VWD (). Of the five cases that we considered, based on phenotypic testing to be highly probable 2A VWD, in four cases we identified three distinct mutations (R1597W×2 cases, C1272Y, I1628T), all of which were consistent with a diagnosis of 2A VWD. Genetic testing failed to identify a mutation in one case (ie, presumed false negative).

In seven other cases, we identified four mutations (1546-1548del3, G1417W, R1315C, R1334W). Interestingly, however, 1546-1548del3 has alternatively been identified as Type 1 VWD by James [, ], and the VWD database identifies R1315C differentially as being Type 1, 2A, 2M, 3, and unclassified [, ]; hence, although these genetic evaluations have provided clarity about the defect, the final diagnosis remains somewhat clouded (ie, what VWD type should these two patients be “labeled”?). While we would identify them as being 2M VWD based on phenotypic testing, other experts might disagree.

We also failed to identify a genetic defect in 4/12 of these cases. Importantly, this outcome matched our phenotypic evidence base; thus, two of these cases had poor and incomplete phenotypic characterization, were originally identified as only “possible” 2M VWD, and on balance are therefore likely not to have Type 2M VWD. The remaining two patients have probable 2A or 2M VWD, and the failure of genetic testing to identify a causal mutation most probably represents false negative results.

In conclusion, for 2A and 2M VWD, a genetic mutation was found in most cases (8/12; 67%), particularly those where 2A or 2M VWD was considered likely, but not found in four cases, of which two are probably false negatives and two are likely not to have 2A or 2M VWD. Thus, on balance, from our experience, genetic testing can be used to confirm 2A or 2M VWD, but the yield will not be 100%, and the relative success can also be somewhat predicted from the phenotypic findings. Furthermore, some false negatives seem likely.

CONCLUSION

Genetic testing in VWD diagnostics is at a crossroads. There has been recent and intense activity in this field within a research setting, particularly in Type 1 VWD, with a resurgence seen also in Type 2 VWD. Most probably related to this research activity, we are also seeing increasing requests and enquiries into genetic testing within the diagnostic arena. This is matched by published interest, including a recent debate on this topic [, , , ], a recent report of rapid diagnostic test approach [], and a variety of reports from experts and expert groups [, , , , –, , , ].

Increasing reports are also becoming available of the potential utility of genetic testing to predict the DDAVP response. However, there are really only limited data available relating to genotype-DDAVP response profiles. For example, there is some information building on this relationship in Type 2B VWD [, ], where certain mutations are associated with lack of thrombocytopenia, even after DDAVP; hence, DDAVP may be useful in these patients and genetic information can be useful for optimizing therapy.

However, although much is known about genotype–phenotype correlations, much more data still need to be ascertained. In some cases, a genetic test approach will assist the diagnosis. However, at other times, this (i) may be a frustrating and an expensive and fruitless exercise (eg, for most cases of presumed VWD, which tend to be “mild type 1”), (ii) merely lead to additional confusion (since experts will disagree on what VWD type a particular mutation represents), and (iii) some “mutations” will prove to be non-causal or innocent polymorphisms.

Therefore, all requests for genetic testing should be assessed on a case-by-case basis—will it help the clinical diagnosis, the potential therapeutic support, or assist in family counseling? Also, does the cost of testing justify the end result, or alternatively, will the testing and/or the genetic finding increase anxiety, or cause confusion and frustration?

Genetic testing is not an approach that should be universally applied to all cases of VWD under investigation, but seems supportable in discrete situations (), in order to confirm or assist the diagnosis in, namely, (i) Type 2N VWD (as an aid to discriminate this from hemophilia A/carrier), (ii) Type 2B VWD (primarily as an aid to discriminate this from PT-VWD), (iii) Type 3 VWD (for prenatal assessment/family studies and alloantibody risk assessment), and perhaps in (iv) 2A/2M VWD, and (v) Type 1 VWD where VWF levels are <20–30 IU/mL.

In general, genetic testing in Types 2N and 2B VWD is supported because the testing is focused, is perhaps a reasonable and justifiable cost, and because there are therapeutic implications to an incorrect diagnosis. In Types 2A/2M, focused genetic testing can be applied, is probably cost-effective, and is often successful when reasonable phenotypic testing has been applied and the likelihood of 2A/2M defects is considered high; nevertheless, clinical benefit is lower, since the choices are similar and limited (DDAVP and/or VWF concentrate).

In Type 3 VWD, there are potential management and family implications, and testing can also potentially represent a reasonable cost benefit (focused genetic testing in family studies; significant adverse risk in alloantibody formation).

In Type 1 VWD, the probability of successful genetic testing is likely to be high only where VWF levels are <20–30 IU/dL, but this may require an expensive and exhaustive evaluation of the entire gene, except perhaps in family studies where a mutation has already been found. Clinical utility is usually low, since the choices are similar and again limited (DDAVP and/or VWF concentrate). In Type 1 VWD, where VWF levels are >30 IU/dL or so, there is no obvious clinical benefit to genetic testing, since the likelihood of test success is low, except perhaps in family studies where a mutation has already been found, and since DDAVP is generally useful, and the only other option remains VWF concentrate.

Acknowledgments: We would like to thank members of our diagnostics staff for performing the routine phenotypic testing of our VWD patients (Soma Mohammed, Jane McDonald, Ella Grezchnik).

Disclosure: The authors declare that they have no conflict of interest related to the publication of this article and no funding has been received.

References

1. Sadler JE, Budde U, Eikenboom JC, et al.; Working Party on von Willebrand Disease Classification. Update on the pathophysiology and classification of von Willebrand disease: a report of the Subcommittee on von Willebrand Factor. J Thromb Haemost. 2006;4:2103-2114.
2. Reininger AJ. Function of von Willebrand factor in haemostasis and thrombosis. Haemophilia. 2008;14(Suppl. 5):11-26.
3. Sadler JE. von Willebrand factor assembly and secretion. J Thromb Haemost. 2009;7(Suppl. 1):24-27.
4. Favaloro EJ. Identification and functional characterisation of von Willebrand Disease. US Hematology. 2009;2(1):16-23.
5. Favaloro EJ. Towards a new paradigm for the identification and functional characterisation of von Willebrand disease. Semin Thromb Hemost. 2009;35:60-75.
6. Othman M, Favaloro EJ. Genetics of type 2B von Willebrand disease: "true 2B," "tricky 2B," or "not 2B." What are the modifiers of the phenotype? Semin Thromb Hemost. 2008;34:520-531.
7. Federici AB, Mannucci PM, Castaman G, et al. Clinical and molecular predictors of thrombocytopenia and risk of bleeding in patients with von Willebrand disease type 2B: a cohort study of 67 patients. Blood. 2009;113(3):526-534.
8. Ribba AS, Loisel I, Lavergne JM, et al. Ser968Thr mutation within the A3 domain of von Willebrand factor (VWF) in two related patients leads to a defective binding of VWF to collagen. Thromb Haemost. 2001;86(3):848-854.
8. Riddell AF, Gomez K, Millar CM, et al. Characterization of W1745C and S1783A: 2 novel mutations causing defective collagen binding in the A3 domain of von Willebrand factor. Blood. 2009;114(16):3489-3496.
10. Favaloro EJ, Mohammed S, Koutts J. Identification and prevalence of von Willebrand disease type 2N (Normandy) in Australia. Blood Coag Fibrinolysis. 2009;20:706-714.
11. Favaloro EJ, Bonar R, Marsden K (on behalf of the RCPA QAP Haemostasis Committee). Lower limit of assay sensitivity: an under-recognised and significant problem in von Willebrand disease identification and classification. Clin Lab Sci. 2008;21:178-185.
12. Favaloro EJ, Bonar R, Meiring M, Street A, Marsden K (on behalf of the RCPA QAP in Haematology). 2B or not 2B? Disparate discrimination of functional VWF discordance using different assay panels or methodologies may lead to success or failure in the early identification of Type 2B VWD. Thromb Haemost. 2007;98:346-358.
13. Favaloro EJ, Bonar R, Kershaw G, et al (on behalf of the RCPA QAP in Haematology). Reducing errors in identification of von Willebrand disease: the experience of the Royal College of Pathologists of Australasia Quality Assurance Program. Sem Thromb Haemost. 2006;32:505-513.
14. Rodeghiero F, Castaman G, Tosetto A. Optimizing treatment of von Willebrand disease by using phenotypic and molecular data. Hematology Am Soc Hematol Educ Program. 2009:113-123.
15. Favaloro EJ, Thom J, Patterson D, et al. Desmopressin therapy to assist the functional identification and characterisation of von Willebrand disease: differential utility from combining two (VWF:CB and VWF:RCo) von Willebrand factor activity assays? Thromb Res. 2009;123:862-868.
16. Favaloro EJ, Thom J, Patterson D, et al. Potential supplementary utility of combined PFA-100 and functional VWF testing for the laboratory assessment of desmopressin and factor concentrate therapy in von Willebrand disease. Blood Coag Fibrinolysis. 2009;20:475-483.
17. Favaloro EJ. Genetic testing for von Willebrand disease: the case against. J Thromb Haemost. 2010;8:6-12.
18. Souto JC, Almasy L, Borrell M, et al. Genetic determinants of hemostasis phenotypes in Spanish families. Circulation. 2000;101:1546-1551.
19. Souto JC, Almasy L, Soria JM, et al. Genome-wide linkage analysis of von Willebrand factor plasma levels: results from the GAIT project. Thromb Haemost. 2003;89:468-474.
20. James P, Lillicrap D. The role of molecular genetics in diagnosing von Willebrand disease. Semin Thromb Hemost. 2008;34:502-508.
21. Peake IR, Goodeve AC. Genetic testing for von Willebrand disease: the case for. J Thromb Haemost. 2010;8:13-16.
22. Nichols WL, Hultin MB, James AH, et al. von Willebrand disease (VWD): evidence-based diagnosis and management guidelines, the National Heart, Lung, and Blood Institute (NHLBI) Expert Panel report (USA). Haemophilia. 2008;14:171-232.
23. Keeney S, Bowen D, Cumming A, Enayat S, Goodeve A, Hill M. The molecular analysis of von Willebrand disease: a guideline from the UK Haemophilia Centre Doctors' Organisation Haemophilia Genetics Laboratory Network. Haemophilia. 2008;14:1099-1111.
24. Lillicrap D. Genotype/phenotype association in von Willebrand disease: is the glass half full or empty? J Thromb Haemost. 2009;7(Suppl. 1):65-70.
25. Favaloro EJ. Detailed von Willebrand factor multimer analysis in patients with von Willebrand disease in the European study, molecular and clinical markers for the diagnosis and management of type 1 von Willebrand disease (MCMDM-1VWD)-a rebuttal. J Thromb Haemost. 2008;6:1999-2001.
26. Mannucci PM. Genetic testing in von Willebrand disease: a rebuttal. J Thromb Haemost. 2010;8(4):860.
27. Favaloro EJ. Genetic testing in von Willebrand disease: reply to rebuttal. J Thromb Haemost. 2010;8(4):861-862.
28. Goudemand J, Fressinaud E, Boisseau P, et al. Distribution of von Willebrand disease (VWD) types in 94 families from the French National Reference Center for VWD (CRMW): type 1 is not the most frequent within 'true' VWD. J Thromb Haemost. 2009;7(Suppl. 2): Abstract PP-TH-623.
29. Favaloro EJ. Phenotypic identification of platelet type-von Willebrand disease and its discrimination from Type 2B von Willebrand disease: a question of 2B or not 2B? A story of non-identical twins? Or two-sides of a multi-denominational or multi-faceted primary haemostasis coin? Sem Thromb Hemost. 2008;34:113-127.
30. Favaloro EJ, Patterson D, Denholm A, et al. Differential identification of a rare form of platelet-type (pseudo-) von Willebrand disease from Type 2B VWD using a simplified ristocetin-induced-platelet-agglutination (RIPA) mixing assay and confirmed by genetic analysis. Br J Haematol. 2007;139:623-626.
31. Facey DA, Favaloro EJ, Maxwell E, Baker R, Hertzberg MS. Type 2B von Willebrand's disease in thirteen individuals from five unrelated Australian Families: phenotype and genotype correlations. Am J Hematol. 2000;63:197-199.
32. Matsubara Y, Murata M, Sugita K, Ikeda Y. Identification of a novel point mutation in platelet glycoprotein I a, Gly to Ser at residue 233, in a Japanese family with platelet-type von Willebrand disease. J Thromb Haemost 2003;1:2198-2205.
33. Nurden P, Lanza F, Bonnafous-Faurie C, Nurden A. A second report of platelet-type von Willebrand disease with a Gly233Ser mutation in the GP1BA gene. Thromb Haemost. 2007;97:319-321.
34. Rodgers SE, Lerda NV, Favaloro EJ, et al. Identification of von Willebrand's disorder type 2N (Normandy) in Australia: a cross-laboratory investigation using different methodologies. Am J Clin Pathol. 2002;118:269-276.
35. VWD database. http://www.vwf.group.shef.ac.uk/ [cited 3 March 2010].
36. James PD, Notley C, Hegadorn C, et al. The mutational spectrum of type 1 von Willebrand disease: results from a Canadian cohort study. Blood. 2007;109:145-154.
37. Corrales I, Ramirez L, Altisent C, Parra R, Vidal F. Rapid molecular diagnosis of von Willebrand disease by direct sequencing. Detection of 12 novel putative mutations in VWF gene. Thromb Haemost. 2009;101:570-576.
38. Sadler JE. Low von Willebrand factor: sometimes a risk factor and sometimes a disease. Hematology Am Soc Hematol Educ Program. 2009:106-112.
39. Rodeghiero F, Ruiz-Sáez A, Bolton-Maggs PH, Hayward CP, Nair SC, Srivastava A. Laboratory issues in bleeding disorders. Haemophilia. 2008;14(Suppl. 3):93-103.

The Molecular Basis of Type 3 von Willebrand Disease

Mon, 28 Jun 2010 09:10:06 +0100

INTRODUCTION

Von Willebrand disease (VWD), the most common inherited bleeding disorder in humans, is a heterogeneous disorder caused by partial quantitative (type 1 VWD), qualitative (type 2 VWD) or severe quantitative (type 3 VWD) deficiencies of von Willebrand factor protein (VWF) []. VWF has a central role in primary hemostasis where it functions at sites of vascular injury in an adhesive matrix between platelets and sub-endothelial components. There is an absolute requirement for VWF in primary hemostasis in the high shear stress conditions that are present in the microvasculature []. VWF also acts as a carrier for coagulation factor VIII (FVIII) in the circulation, protecting FVIII from proteolytic degradation and localizing it to the site of vascular injury. Type 3 VWD is a moderate to severe autosomal recessive inherited bleeding disorder, often requiring replacement therapy with VWF/FVIII concentrates. It is caused by mutations, which give rise to a severe quantitative deficiency of VWF in plasma and platelets [, ]. VWF antigen (VWF:Ag) is absent (or present only in trace amounts) and VWF activity (for example, ristocetin cofactor activity, VWF:RCo, and collagen binding, VWF:CB) is also absent. FVIII levels are usually very low, less than 20 IU dL []. Hemorrhagic symptoms in patients with type 3 VWD include severe epistaxis, menorrhagia, postoperative bleeding, and arthropathy []. Type 3 VWD has a prevalence of about 0.5–1 individuals per million in the general population [], although this may be as high as 6 per million in populations where consanguinity is common []. Approximately 80% of the reported type 3 VWD mutations in the von Willebrand factor gene () are nonsense mutations, splice-site mutations, small (one or only a few nucleotides) or large (one or more exons) deletions, and insertions associated with null alleles.

The gene is located at the short arm of chromosome 12 (12p13.2). It is a large gene comprising approximately 178 kb of genomic DNA, including 52 exons varying in size from 40 to 1379 bases [, ]. transcribes a mRNA product of 8.7 kb. This is translated to produce a 2813 amino acid product, including a signal peptide of 22 amino acids, a large propeptide of 741 amino acids, and a mature VWF molecule containing 2050 amino acids [, ]. A non-coding partial pseudogene () spanning approximately 25 kb of DNA is located at chromosome 22q11.2 []. shows 97% sequence homology with exons 23–34 of ; however, the presence of multiple splice site and nonsense mutations in the pseudogenic sequence indicates that it is not a functional gene []. The 3% divergence in sequence between and the pseudogene is indicative of a relatively recent origin of by partial gene duplication. is present in humans and great apes but not in more distantly related primates [], also suggesting a recent origin. The presence of the pseudogene can complicate laboratory genetic studies in VWD, as there is the potential to inadvertently amplify nucleic acid sequence from in PCR-based sequencing reactions. Careful design of PCR primers and protocols are required to avoid this.

The VWF mutation spectrum described herein is from a review of the literature associated with type 3 VWD published on PubMed and reported on the International Society on Thrombosis and Hemostasis (ISTH) VWF database (, accessed on 30 March 2010). For a comprehensive listing of the reported VWF mutations in type 3 VWD the reader is referred to the latter database.

The numbering system used in this article to describe VWF mutations is in accordance with Human Genome Variation Society (HGVS) recommendations. At the cDNA level, +1 is the A of the transcription initiator methionine codon (RefSeq NCBI accession number NM_000552.3) found in exon 2 of . At the protein level +1 is the initiator methionine (RefSeq NCBI accession number NP_000543.2).

THE MOLECULAR GENETIC BASIS OF TYPE 3

The severe quantitative deficiency of VWF in type 3 VWD is the direct result of homozygosity or compound heterozygosity for recessive mutations at the locus that give rise to null alleles. This is distinct from the partial quantitative deficiency of VWF seen in type 1 VWD where the plasma level of VWF is substantially influenced by a wide range of biological and environmental factors, resulting in incomplete penetrance and markedly variable expression of the type 1 VWD phenotype in affected individuals and between affected members of the same family.

Mutations in patients with type 3 VWD have been identified throughout the 178 kb length of the VWF gene locus, including the promoter region, coding and non-coding regions, and the 5′ and 3′ untranslated regions (). The overall detection rate for homozygous or compound heterozygous mutations of in molecular genetic studies of type 3 VWD ranges from 81% (22/27 index cases) [] to 100% (40/40 index cases) [].

Large deletions

Deletion mutations result in the removal of essential sequences of . Type 3 VWD was originally assumed to be due to large deletions of , however, only 11 of 109 type 3 VWD mutations reported on the ISTH VWF database, accessed on 30 March 2010, are large gene deletions. Large deletions are defined here as ranging from the deletion of one or more exons of [–], to the deletion of the entire gene [–]. One study of 40 multiethnic type 3 VWD patients identified 13 deletions involving , of which only one was large [, ], supporting the observation that large deletions are overall a relatively infrequent cause of type 3 VWD. A recently reported exception to this, identified in a cohort of type 3 VWD patients from the north-west of England [], is a large deletion of (c.221-977_532 + 7059del [p.Asp75_Gly178del], or ex4-5del) that was present in 7 of 12 Caucasian type 3 VWD patients from 6 unrelated families []. This deletion of exons 4 and 5 was absent in nine patients of Asian origin. Expression studies of the ex4-5del mutation showed markedly decreased secretion and defective multimerization of the mutant VWF protein. Haplotype analysis indicated a possible founder origin of the mutation among apparently unrelated kindreds who all geographically originated in Britain so far as could be ascertained. The 5′ and 3′ breakpoints of the deletion were shown to lie between two regions of AluY repetitive elements. Alu repeats are known to promote unequal homologous recombination, resulting in deletion mutations, with at least three Alu-mediated deletions in type 3 VWD identified [, , ]. Interestingly, the ex4-5del mutation was also identified in several type 1 VWD families, segregating with VWD in an autosomal dominant fashion and exerting a dominant-negative effect to produce a type 1 VWD phenotype. The ex4-5del mutation represents a previously unreported cause of both type 1 and type 3 VWD.

Despite the above observation, heterozygous carriers of deletions are generally clinically asymptomatic and have normal levels of VWF, demonstrating that a single locus is able to provide adequate amounts of functional VWF. Some carriers do, however, have reduced levels of VWF:Ag and VWF:RCo [, , , ], and there are also reports in the literature of a mild type 1 VWD phenotype in heterozygous carriers of type 3 VWD [, –].

Alloantibodies to VWF develop in a small proportion of type 3 VWD patients, particularly those with large gene deletions, as a consequence of the transfusion of VWF replacement therapy products. Antibodies have been reported in 2.6–9.5% of type 3 VWD patients [, ]; however, the exact incidence remains uncertain. Anti-VWF alloantibodies can inhibit the hemostatic effect of VWF-replacement therapy and may also cause life-threatening anaphylactic reactions to treatment. Knowledge of the defect(s) in patients with type 3 VWD may inform the assessment of the risk for inhibitor development; however, it will not significantly influence routine clinical management of the affected individual.

Null alleles

Approximately 70% of reported mutations in type 3 VWD are nonsense mutations or small (one or only a few nucleotides) deletions or insertions associated with a null allele. Transcripts with premature stop codons, if translated, may give rise to potentially deleterious truncated proteins. In order to prevent this, mRNA surveillance mechanisms ensure that only RNA transcripts with full coding potential are available for translation. Aberrant mRNAs, such as those containing premature stop codons are recognized by the mRNA surveillance machinery and subjected to nonsense-mediated decay (NMD) []. Hence in patients where premature stop codons are present in in the homozygous or compound heterozygous state there is a complete quantitative deficiency of VWF, leading to a type 3 VWD phenotype. Frame shift changes, produced by deletions, insertions or splicing errors, also generally result in the production of premature stop codons, therefore the resulting “null” RNA transcripts may be predicted to be similarly subject to decay by NMD.

Splice-site mutations

Splice-site mutations, which change the conserved GT … AG nucleotides flanking most introns, usually abolish gene function. A number of splice-site mutations have been described in association with type 3 VWD. Expression studies have demonstrated exon skipping in affected VWF mRNA transcripts [, , ]. Alternatively, splice-site mutations may lead to rapid clearance of the resulting defective VWF RNA transcript [].

Missense changes

Missense mutations in , inherited in the homozygous or compound heterozygous state, account for 15–20% of sequence variants associated with type 3 VWD. In particular, missense mutations located at cysteine amino acids are predicted to have a deleterious effect on mature VWF. Cysteine residues are involved in the formation of disulphide bonds, which are important for the correct function and tertiary structure of the mature VWF protein. Unpaired cysteine residues are known to prevent the intracellular transport of many proteins, and cotransfection studies of normal and mutant VWF have demonstrated that these mutations can interfere with dimerization of VWF monomers [, , –], impairing VWF multimerization and secretion through a dominant-negative mechanism [, , , , ]. frame shift mutations have also been predicted to cause defective protein folding, interfering with normal propeptide processing [].

Gene conversion

Gene conversion events between and the () have been shown to cause type 3 VWD, by the introduction of non-functional pseudogenic sequence into the locus [, , , –]. Gene conversions occur as a result of the high degree of sequence homology between and (97%), and usually involve the region extending from the 3′ part of intron 27 to the 5′ part of exon 28 of the gene and pseudogene. This is due to the presence in this region of chi sequences (CCTGGTGG) or chi-like sequences (GCTGGTGG) that can promote DNA recombination []. Gene conversion events can be identified by the presence of multiple base substitutions within the region of corresponding to the sequence (). As a consequence of the presence of multiple stop codons in the sequence, the majority of gene conversion events result in the introduction of a premature termination codon into the sequence, and hence give rise to a non-functional allele. The nonsense mutation p.Gln1311X in exon 28 of has been reported in many type 3 VWD patients and this mutation is often observed in association with a gene conversion event [, , , , ]. It is worth noting that Surdhar described a patient with type 3 VWD arising from a gene conversion mechanism in whom a high-titer inhibitor of VWF was present []. A second patient with type 3 VWD associated with a gene conversion mutation was also reported to have developed a VWF inhibitor []. The frequency of gene conversion mutations in VWD is not known.

POPULATION STUDIES OF TYPE 3 VON WILLEBRAND DISEASE (VWD)

While the mutation spectrum associated with type 3 VWD is broad, there is no commonly occurring mutation or mutation mechanism, other than gene conversion mediated events, which has been described in a wide range of populations. A number of mutations of have been reported to be commonly associated with type 3 VWD in populations located in distinct geographical regions, suggesting a founder origin of some mutations.

Type 3 VWD in the “original” VWD family [] from the Åland Islands, situated in the Gulf of Bothnia between Sweden and Finland, was associated with a single cytosine deletion in exon 18 of , c.2435delC []. This mutation predicts the substitution of proline at position 812 of VWF by arginine. The resulting shift in the reading frame introduces a premature stop codon (p.Pro812ArgfsX31), resulting in a null allele. The c.2435delC mutation has been found in other families from the Åland Islands and is also prevalent in the Swedish, Finnish [, ], and Dutch populations []. A study of German patients found this single cytosine deletion to be the most frequent type 3 VWD mutation, being present on 12.5% of type 3 VWD-associated chromosomes in a study population of 32 type 3 VWD patients from 28 unrelated families []. This study also identified one complete gene deletion, one large gene deletion, one missense change, two frame shifts (including c.2435delC), and two nonsense mutations. A genetic founder effect for c.2435delC is likely in the Swedish and Dutch populations [, ], whereas the German population was shown to be genetically more heterogeneous []. A population study performed in patients of Italian origin did not identify c.2435delC [], although a subsequent study of multiethnic patients did find this mutation in an Italian patient []. The c.2435delC mutation was not identified in Indian or Greek [], or Mexican Mestizo [] type 3 VWD populations, and we also are unaware of any reports of this mutation in patients of British origin. Further studies in the Swedish and Finnish populations identified nonsense mutations in a large proportion of patients with type 3 VWD [, , ]. Small deletions, splice site, and missense mutations were also found in these patients [].

A study of type 3 VWD in the Netherlands [] identified heterozygosity for a nonsense mutation in three type 3 VWD patients (p.Arg2535X) and heterozygosity for a p.Asn2546Tyr missense change in another three patients. This study restricted the search for mutations to the 11 CGA codons in ; it is therefore likely that compound heterozygosity for a second unidentified mutation was present in these patients.

In a study of 24 unrelated Hungarian patients [] a large partial deletion of the 5′-region of (exons 1–3) was detected in 12 of 48 alleles (25% of all type 3 VWD alleles). Analysis of the deletion breakpoints showed Alu Y and Alu SP repetitive elements at the ends of the deletion, indicative of a recombination event leading to the deletion of a 35 kb fragment. This mutation was reported to be unique among Hungarian patients, and probably represents a founder effect originating within Hungary.

In a recent UK study, the molecular basis of type 3 VWD was investigated in 11 Caucasian patients and in nine individuals of Asian origin []. Fifteen different mutations were identified at the genomic DNA level: two gene conversion events, three nonsense, three frame shift, one missense, two splice site, one insertion–deletion, and three deletion mutations, reflecting the range of reported mutations in type 3 VWD (). Homozygosity or compound heterozygosity for mutations was present in 15 of the 20 patients. In the remaining five individuals, heterozygosity for a single mutation was identified in four cases and one patient had no detectable mutation. Analysis of platelet-derived VWF RNA from these five individuals revealed heterozygosity for a deletion of exons 4 and 5 (ex4-5del, as described previously) in four cases []. The remaining patient was heterozygous for a three base deletion that had already been identified at the DNA level. Only heterozygosity for a mutation was found in two cases and the genetic basis of type 3 VWD could not be explained in these individuals. Overall the observed genotype explained the phenotype in 18 of the 20 patients investigated. haplotype analysis was performed for five mutations that were seen on more than one allele in the patient study group. Each of these mutations was inherited in association with a specific haplotype, suggesting possible founder origins of these mutations. The wide spread of mutation location and mechanisms witnessed in the population studies described illustrates the diverse nature of type 3 VWD associated VWF allelelic defects.

ARE TYPE 3 AND TYPE 1 VON WILLEBRAND DISEASE (VWD) ALWAYS GENETICALLY DISTINCT DISORDERS?

In most cases type 3 and type 1 VWD are genetically distinct disorders. Approximately 80% of mutations reported in association with type 3 VWD give rise to null alleles, whereas in type 1 VWD about two-thirds of mutations are missense in nature (). The majority of mutations associated with type 3 VWD affect the production of VWF from one allele only, hence mutations associated with this recessive disorder are present in the homozygous or compound heterozygous state in affected individuals. In contrast, a number of mutations associated with type 1 VWD have been shown to exhibit an autosomal dominant mode of inheritance, whereby inheritance of a single defective allele affects the normal allele via a dominant-negative mechanism (ie, the mutation results in a mutant gene product that interferes with the wild-type gene product in heterozygotes, for example, affecting dimerization and/or multimerization of VWF) [, , , , ].

Different molecular pathogenic mechanisms in type 1 and type 3 VWD have been further illustrated in a recent study []. The risk of significant bleeding in obligatory carriers of type 1 VWD was demonstrated to be significantly higher than in obligatory carriers for type 3 VWD, indicating that heterozygosity for type 1 VWD mutations may affect the production of VWF to a greater extent than heterozygosity for type 3 VWD mutations. The authors of this study proposed that carriers of type 3 VWD represent a distinct population from individuals with type 1 VWD, even though both groups of patients are expected to carry one mutated allele. To complicate matters, however, marked variability of plasma VWF levels, ranging from normal to mildly reduced, has been described in heterozygous carriers for type 3 VWD [, ], and Castaman and colleagues [] reported the bleeding risk in type 3 VWD obligatory carriers to be significantly higher than in normal controls. Earlier studies had found that some heterozygous carriers of type 3 VWD had a mild type 1 VWD phenotype [, , , ] and, more recently, the previously referred to ex4-5del mutation of has been described in association with both type 3 and type 1 VWD []. To some extent, therefore, there is a remaining question as to whether or not there is significant overlap between the molecular pathologies of the two quantitative variants of VWD, and to what extent modifying factors have an effect on this overlap.

GENETIC DIAGNOSIS IN TYPE 3 VON WILLEBRAND DISEASE (VWD)

Type 3 VWD is associated with a severe bleeding diathesis and a clear autosomal recessive inheritance pattern. Homozygous or compound heterozygous mutations in , which explain the VWD phenotype, can usually be identified in affected individuals, and genetic diagnosis and associated family studies are therefore important tools to inform genetic counseling of affected families. Genetic diagnosis allows the identification of asymptomatic carriers of type 3 VWD, as this diagnosis cannot be made either by phenotypic testing or, frequently, by pedigree analysis. It also provides valuable information to help family members with family planning decisions, including making informed decisions about the option for prenatal diagnosis based on knowledge of the familial mutation(s). Furthermore, awareness of the type 3 VWD-carrier status of the parents, together with knowledge of the likelihood of the birth of an affected child, provides important information to clinicians involved with the management of childbirth and the postnatal care of the neonate. Although the molecular pathogenesis of type 3 VWD is generally well understood, there are a number of cases in which mutations to explain the disease phenotype have not been found. Historically this was no doubt at least in part due to the technical challenges faced when attempting analysis of the large gene, which restricted researchers to the examination of mutation “hotspots,” for example, CG dinucleotides [, , ]. There have been cases more recently however where, despite sequencing the essential regions of , homozygosity or compound heterozygosity for type 3 VWD mutations was still not found [, , ]. These observations indicate that analysis of the essential regions of genomic DNA may be insufficient to elucidate the underlying molecular pathology in a proportion of type 3 VWD cases. Therefore, we suggest that RNA studies should be performed to search for intronic mutations, heterozygous deletions or aberrant splicing/posttranscriptional events in type 3 VWD patients in whom DNA studies fail to identify a genotype which is considered to explain the disease phenotype. Alternative methods to investigate partial gene deletions at the genomic DNA level, such as multiplex ligation-dependent probe amplification (MLPA) may also be considered []. However, there remains the possibility that this comprehensive approach to genetic diagnosis may still not explain all cases of previously diagnosed type 3 VWD. This may indicate other, as yet unexplained, mechanisms underlying a small number of cases.

CONCLUSION

Type 3 VWD is a genetically heterogeneous autosomal recessive disorder mediated by a variety of mutation mechanisms. Mutations are spread throughout the entire locus, or involve homologous recombination events between repetitive sequence elements or the sequence. Other than gene conversion events no common mutation mechanisms across a wide range of population groups have been identified. mutation types in type 3 VWD, which give rise to null alleles, are generally distinct from those found in type 1 VWD, where heterozygous missense mutations predominate. The molecular basis of type 3 and type 1 VWD is different in most cases.

Disclosure: The authors have no financial interests to disclose in relation to the contents of the article.

References

1. Sadler JE, Budde U, Eikenboom JCJ, et al. Update on the pathophysiology and classification of von Willebrand disease: a report of the subcommittee on von Willebrand factor. J Thromb Haemost. 2006;4(10):2103-2114.
2. Ruggeri ZM. Structure of von Willebrand factor and its function in platelet adhesion and thrombus formation. Best Pract Res Clin Haematol. 2001;14(2):257-279.
3. Castaman G, Rodeghiero F, Tosetto A, et al. Hemorrhagic symptoms and bleeding risk in obligatory carriers of type 3 von Willebrand disease: an international, multicenter study. J Thromb Haemost. 2006;4(10):2164-2169.
4. Federici AB. Clinical diagnosis of von Willebrand disease. Haemophilia. 2004;10(Suppl. 4):169-176.
5. Mannucci PM, Bloom AL, Nilsson IM, West RR. Atherosclerosis and von Willebrand factor I. Prevalence of severe von Willebrand's disease in western Europe and Israel. Br J Haematol. 1984;57(1):163-169.
6. Sadler JE, Mannucci PM, Berntorp E, et al. Impact, diagnosis and treatment of von Willebrand Disease. Thromb Haemost. 2000;84(2):160-174.
7. Ginsburg D, Handin RI, Bonthron DT, et al. Human von Willebrand factor: isolation of complementary DNA (cDNA) clones and chromosomal location. Science. 1985;228(4706):1401-1406.
8. Mancuso DJ, Tuley EA, Westfield LA, et al. Structure of the gene for human von Willebrand factor. J Biol Chem. 1989;264(33):19514-19527.
9. Bonthron D, Orr EC, Mitsock LM, Ginsburg D, Handin RI, Orkin SH. Nucleotide sequence of pre-pro-von Willebrand factor cDNA. Nucleic Acids Res. 1986;14(17):7125-7127.
10. Verweij CL, Diergaarde PJ, Hart M, Pannekoek H. Full-length von Willebrand factor (vWF) encodes a highly repetitive protein considerably larger than the mature vWF subunit. EMBO J. 1986;5(8):1839-1847.
11. Patracchini P, Calzolari E, Aiello V, Palazzi P, Banin P, Marchetti GB. Sublocalization of von Willebrand factor pseudogene to 22q11.22-q11.23 by in situ hybridization in a 46,X,t (X;22) (pter; q11.21) translocation. Hum Genet. 1989;83(3):264-266.
12. Mancuso DJ, Tuley EA, Westfield LA, et al. Human von Willebrand factor gene and pseudogene: structural analysis and differentiation by polymerase chain reaction. Biochemistry. 1991;30(1):253-269.
13. Sadler JE. Von Willebrand factor. J Biol Chem. 1991;266(34):22777-22780.
14. Gupta PK, Saxena R, Adamtziki E, et al. Genetic defects in von Willebrand disease type 3 in Indian and Greek patients. Blood Cells Mol Dis. 2008;41(2):219-222.
15. Baronciani L, Cozzi G, Canciani MT, et al. Molecular defects in type 3 von Willebrand disease: updated results from 40 multiethnic patients. Blood Cells Mol Dis. 2003;30(3):264-270.
16. Peake IR, Liddell MB, Moodie P, et al. Severe type III von Willebrand's disease caused by deletion of exon 42 of the von Willebrand factor gene: family studies that identify carriers of the condition and a compound heterozygous individual. Blood. 1990;75(3):654-661.
17. Mancuso DJ, Tuley EA, Castillo R, de Bosch N, Mannucci PM, Sadler JE. Characterization of partial gene deletions in type III von Willebrand disease with alloantibody inhibitors. Thromb Haemost. 1994;72(2):180-185.
18. Mohl A, Marschalek R, Masszi T, et al. An Alu-mediated novel large deletion is the most frequent cause of type 3 von Willebrand disease in Hungary. J Thromb Haemost. 2008;6(10):1729-1735.
19. Xie F, Wang X, Cooper DN, et al. A novel Alu-mediated 61-kb deletion of the von Willebrand factor (VWF) gene whose breakpoints co-locate with putative matrix attachment regions. Blood Cells Mol Dis. 2006;36(3):385-391.
20. Eikenboom JCJ, Castaman G, Vos HL, Bertina RM, Rodeghiero F. Characterization of the genetic defects in recessive type 1 and type 3 von Willebrand disease patients of Italian origin. Thromb Haemost. 1998;79(4):709-717.
21. Ngo KY, Glotz VT, Koziol JA, et al. Homozygous and heterozygous deletions of the von Willebrand factor gene in patients and carriers of severe von Willebrand disease. Proc Natl Acad Sci USA. 1988;85(8):2753-2757.
22. Schneppenheim R, Castaman G, Federici AB, et al. A common 253-kb deletion involving VWF and TMEM16B in German and Italian patients with severe von Willebrand disease type 3. J Thromb Haemost. 2007;5(4):722-728.
23. Schneppenheim R, Krey S, Bergmann F, et al. Genetic heterogeneity of severe von Willebrand disease type III in the German population. Hum Genet. 1994;94(6):640-652.
24. Shelton-Inloes BB, Chehab FF, Manucci PM, Federici AB, Sadler JE. Gene deletions correlate with the development of alloantibodies in von Willebrand disease. J Clin Invest. 1987;79(5):1459-1465.
25. Baronciani L, Cozzi G, Canciani MT, et al. Molecular characterization of a multiethnic group of 21 patients with type 3 von Willebrand disease. Thromb Haemost. 2000;84(4):536-540.
26. Sutherland MS, Keeney S, Bolton-Maggs PHB, Hay CRM, Will A, Cumming AM. The mutation spectrum associated with type 3 von Willebrand disease in a cohort of patients from the North West of England. Haemophilia. 2009;15(5):1048-1057.
27. Sutherland MS, Cumming AM, Bowman M, et al. A novel deletion mutation is recurrent in von Willebrand disease types 1 and 3. Blood. 2009;114(5):1091-1098.
28. Anvret M, Blombäck M, Lindstedt M, Söderlind E, Tapper-Persson M, Thelander AC. Genetic and blood coagulation characterization of "Swedish" families with von Willebrand's disease types I and III: New aspects of heredity. Hum Genet. 1992;89(2):147-154.
29. Lethagen S, Isaksson C, Schaedel C, Holmberg L. Von Willebrand's disease caused by compound heterozygosity for a substitution mutation (T1156M) in the D3 domain of the von Willebrand factor and a stop mutation (Q2470X). Thromb Haemost. 2002;88(3):421-426.
30. Zhang ZP, Lindstedt M, Falk G, Blomback M, Egberg N, Anvret M. Nonsense mutations of the von Willebrand factor gene in patients with von Willebrand disease type III and type I. Am J Hum Genet. 1992;51(4):850-858.
31. Zhang ZP, Blomback M, Nyman D, Anvret M. Mutations of von Willebrand factor gene in families with von Willebrand disease in the Åland Islands. Proc Natl Acad Sci USA.1993;90(17):7937-7940.
32. Zhang ZP, Blomback M, Egberg N, Falk G, Anvret M. Characterization of the von Willebrand factor gene (VWF) in von Willebrand disease type III patients from 24 families of Swedish and Finnish origin. Genomics.1994;21(1);188-193.
33. Lak M, Peyvandi F, Mannucci PM. Clinical manifestations and complications of childbirth and replacement therapy in 385 Iranian patients with type 3 von Willebrand disease. Br J Haematol. 2000;111(4):1236-1239.
34. Mannucci PM, Federici AB. Antibodies to von Willebrand factor in von Willebrand disease. Adv Exp Med Biol. 1995;386:87-92.
35. Lykke-Andersen J, Shu M, Steitz JA. Human Upf proteins target an mRNA for nonsense-mediated decay when bound downstream of a termination codon. Cell. 2000;103(7):1121-1131.
36. Castaman G, Eikenboom JC, Missiaglia E, Rodeghiero F. Autosomal dominant type 1 von Willebrand disease due to G3639T mutation (C1130F) in exon 26 of von Willebrand factor gene: description of five Italian families and evidence for a founder effect. Br J Haematol. 2000;108(4):876-879.
37. Schneppenheim R, Budde U, Obser T, et al. Expression and characterization of von Willebrand factor dimerization defects in different types of von Willebrand disease. Blood. 2001;97(7):2059-2066.
38. Tjernberg P, Vos HL, Castaman G, Bertina R, Eikenboom JCJ. Dimerization and multimerisation defects of von Willebrand factor due to mutated cysteine residues. J Thromb Haemost. 2004;2(2):257-265.
39. Allen S, Abuzenadah AM, Hinks J, et al. A novel von Willebrand disease-causing mutation (Arg273Trp) in the von Willebrand factor propeptide that results in defective multimerization and secretion. Blood. 2000;96(2):560-568.
40. Baronciani L, Federici AB, Cozzi G, et al. Expression studies of missense mutations p.D141Y, p.C275S located in the propeptide of von Willebrand factor in patients with type 3 von Willebrand disease. Haemophilia. 2008;14(3):549-555.
41. Mohlke KL, Nichols WC, Rehemtulla A, et al. A common frameshift mutation in von Willebrand factor does not alter mRNA stability but interferes with normal propeptide processing. Br J Haematol. 1996;95(1):184-191.
42. Zhang ZP, Falk G, Blomback M, Egberg N, Anvret M. A single cytosine deletion in exon 18 of the von Willebrand factor gene is the most common mutation in Swedish vWD type III patients. Hum Mol Genet. 1992;1(9):767-768.
43. Eikenboom JCJ, Vink T, Briet E, Sixma JJ, Reitsma PH. Multiple substitutions in the von Willebrand factor that mimic the pseudogene sequence. Proc Natl Acad Sci USA. 1994;91(6):2221-2224.
44. Gupta PK, Adamtziki E, Budde U, et al. Gene conversions are a common cause of von Willebrand disease. Br J Haematol. 2005;130(5):752-758.
45. Surdhar GK, Enayat MS, Lawson S, Williams MD, Hill FGH. Homozygous gene conversion in von Willebrand factor gene as a cause of type 3 von Willebrand disease and predisposition to inhibitor development. Blood. 2001;98(1):248-250.
46. Kenter AL, Birshtein BK. Chi, a promoter of generalized recombination in λ phage, is present in immunoglobulin genes. Nature. 1981;293(5831):402-404.
47. Casaña P, Martinez F, Haya S, Lorenzo JI, Espinós C, Aznar JA. Q1311X: a novel nonsense mutation of putative ancient origin in the von Willebrand factor gene. Br J Haematol. 2000;111(2):552-555.
48. von Willebrand EA. Hereditar pseudohemofil. Finska Lakar Hand.1926;68:87-112.
49. Eikenboom JCJ, Ploos van Amstel HK, Reitsma PH, Briet E. Mutations in severe, type III von Willebrand's disease in the Dutch population: candidate missense and nonsense mutations associated with reduced levels of von Willebrand factor messenger RNA. Thromb Haemost. 1992;68(4):448-454.
50. Melo-Nava BM, Benitez H, Palacios JJ, et al. Molecular study of VWF gene from Mexican Mestizo patients with von Willebrand disease, and the finding of three new mutations. Blood Cells Mol Dis. 2007;39(3):361-365.
51. Zhang ZP, Falk G, Blomback M, Egberg N, Anvret M. Identification of a new nonsense mutation in the von Willebrand factor gene in patients with von Willebrand disease type III. Hum Mol Genet. 1992;1(1):61-62.
52. Mannucci PM, Lattuada A, Castaman G, et al. Heterogeneous phenotypes of platelet and plasma von Willebrand factor in obligatory heterozygotes for severe von Willebrand disease. Blood. 1989;74(7):2433-2436.
53. Kakela JK, Friedman KD, Haberichter SL, et al. Genetic mutations in von Willebrand disease identified by DHPLC and DNA sequence analysis. Mol Genet Metab. 2006;87(3):262-271.
54. Gille J., Hogervorst FB, Pals G, et al. Genomic deletions of MSH2 and MLH1 in colorectal cancer families detected by a novel mutation detection approach. Br J Cancer. 2002;87(8):892-897.

Blood Coagulation Factors V and VIII: Molecular Mechanisms of Procofactor Activation

Mon, 28 Jun 2010 08:53:58 +0100

INTRODUCTION

The majority of blood coagulation factors are synthesized as inactive precursors that only express activity following discrete and limited proteolysis. A defined conformational change ensues that allows these proteins to assemble on cellular surfaces localized at the site of injury where they optimally function. This molecular strategy allows for a high level of temporal and spatial regulation as well as protection against naturally circulating inhibitors that typically target the active protein conformation. It is well established that the serine protease zymogens of coagulation (eg, FVII, FIX, FX, prothrombin, etc.) follow a general activation strategy that is shared by all serine proteases and is typified by trypsinogen and chymotrypsinogen []. The process requires cleavage following Arg (the bond between Arg and Ile) that generally removes an activation peptide and exposes a new N-terminus in the catalytic domain beginning with Ile. The new N-terminal sequence then folds back into the catalytic domain and inserts into the N-terminal binding cleft in a sequence-specific manner forming a salt bridge between the α-NH group of Ile and Asp in the interior of the catalytic domain. This transition is associated with numerous changes in the structure and ultimately leads to the maturation of the active serine protease. In contrast to this molecular strategy, the mechanism underlying the activation of the two major procofactors in blood coagulation, factor V (FV) and factor VIII (FVIII) are substantially different.

FV and FVIII are structurally and functionally homologous proteins and play a central role in the hemostatic process. Once activated, these proteins serve analogous functions as cofactors in the blood coagulation system []. Activated FV (FVa) assembles with the serine protease factor Xa (FXa) while activated FVIII (FVIIIa) is a cofactor for factor IXa (FIXa); these protein complexes assemble in the presence of Ca ions on a negatively charged membrane surface to form the prothrombinase and intrinsic tenase complexes, respectively [, ]. The prothrombinase complex catalyzes the conversion of prothrombin to thrombin, while intrinsic Xase catalyzes the proteolytic conversion of FX to FXa; both pivotal steps in the coagulation cascade []. The individual contribution of the serine proteases FIXa and FXa to overall thrombin generation is relatively minor, as incorporation of FVa and FVIIIa into the macromolecular enzyme complexes enhances the reaction rate by several orders of magnitude []. The importance of these cofactors is further underscored by clinical findings, which indicate that FV and FVIII deficiency states lead to parahemophilia and hemophilia A, respectively [, ].

Following the discovery of FV and FVIII in the 1930s to 1940s, it was quickly recognized that they require proteolytic activation to fully participate in coagulation [–]. Over the past three decades considerable effort and progress has been made defining their mode of activation [–]; however, key mechanistic details remain to be uncovered. This review focuses on our current understanding of FV and FVIII activation and discusses the various structural elements that assist in keeping these proteins in an inactive procofactor state.

THE FACTOR V (FV) PROCOFACTOR TO COFACTOR TRANSITION

FV is a large (=330,000), multidomain (A1-A2-B-A3-C1-C2), single chain glycoprotein that circulates in blood at a concentration of ∼20 nM (∼10 µg/mL) [, ]. Of the total FV pool in whole blood, ∼20% is stored in the α-granules of platelets and secreted upon platelet activation []. At physiological plasma concentrations, the procofactor FV cannot assemble or function in the prothrombinase complex; proteolytic processing within the B domain is an absolute requirement for the expression of cofactor function [, –]. Thrombin is considered the key physiological activator of FV and cleaves three peptide bonds (Arg, Arg, and Arg) within the B domain, thereby facilitating B domain removal () [, –]. While not completely defined, thrombin appears to interact with FV through the heavy and light chains [, ]. The binding site within the light chain is thought to reside within the C2 domain and appears important for proteolysis at all three thrombin cleavage sites []. The resulting active cofactor species, FVa, is a heterodimer composed of a heavy chain (A1-A2; =105,000) and a light chain (A3–C1–C2; =71/74,000), which are associated through Ca ions () [, –, ]. The heavily glycosylated B domain, spanning amino acids 710–1545, is released as two large fragments (=71,000 and =150,000) [, , , ].

In addition to thrombin, various proteases have been identified that cleave FV to generate a cofactor species with variable amounts of activity. For example, several groups have established the FXa activates FV in a membrane- and Ca-dependent fashion following cleavage at or near Arg and Arg, and possibly at other sites in the light chain depending on reaction conditions [, , ]. Other proteases include, to name a few, activators from (Russell's viper), venoms [, –], calpain [], plasmin [], platelet proteases [, ], meizothrombin [], as well as elastase and cathepsin G [–]. These later two proteases are of potential significance as they are released from polymorphonuclear leukocytes at extravascular tissue sites and could amplify thrombin generation through FV activation. This in combination with FV released from activated platelets may play a major role in the initiation phase of cell-based coagulation events.

While previously appreciated, the laboratories of Mann and Esmon firmly established that FV requires proteolytic processing at multiple sites to effect activation [, ]. In the following decades, numerous studies have attempted to define the contribution of the individual cleavage sites to the development of FV cofactor activity. Although somewhat conflicting results have been obtained, it is generally acknowledged that proteolysis by thrombin follows a kinetically preferred order of bond cleavage: Arg is cleaved first, followed by cleavage at Arg, and Arg. Furthermore, most data support the finding that maximal activation of FV requires proteolysis at Arg. Cleavage at Arg and Arg yields a FV derivative with significant, but partial cofactor activity [, , , ], whereas individual cleavage at these sites does not lead to any substantial increase in cofactor activity [, , ]. Support for the contribution of cleavage at Arg to maximal FV cofactor activity came from mutagenesis studies, which demonstrated that isolated cleavage at this site is sufficient for complete activation [, , ]. This is also consistent with the observation that proteolysis by Russell's viper venom (RVV-V) and venom (LVV-V), which cleave FV at Arg, results in full activation [, , –]. Thus release of the B domain from the FV light chain appears to be a necessary requirement for the expression of cofactor activity, which is facilitated by proteolysis of the two preceding activation sites in the B domain. Collectively, the results suggest that the FV B domain somehow keeps FV in an inactive state and its removal by proteolysis contributes to the activation mechanism.

THE B DOMAIN PRESERVES THE PROCOFACTOR STATE OF FACTOR V (FV)

The human FV B domain is 836 amino acids long, comprises ∼50% of the mass of the protein, and has no homology to any other known protein, including the FVIII B domain () [, ]. The B domain is heavily glycosylated and has unusual regions of tandem repeats, of which the function remains to be elucidated. Electron microscopy and physical studies have suggested that the B domain appears as a bulky extension to a globular core, assumed to be the heavy/light chains [, ]. Because of these unusual properties and lack of importance to FVa procoagulant activity, less attention has been paid to investigating its functional significance. Some studies have suggested that the B domain may play a role in the anticoagulant function of FV by stimulating the activated protein C (APC)-mediated inactivation of FVIIIa (reviewed here []). As suggested above, one role must be to maintain FV as an inactive procofactor. Recent work from our laboratory as well as others has shed some light on how the FV B domain regulates the function of FV. An important first observation came from the Kane laboratory who showed that a B-domainless derivative of FV (FVdes811-1491 or FV-810) has constitutive, but partial activity compared to FVa [, ]. These studies suggested that the B domain somehow prevents expression of procoagulant activity prior to proteolytic processing []. More recently, our laboratory has further investigated the molecular properties of this B domain-deleted FV variant. We found that purified FV-810 as well as a thrombin-resistant derivative interact with membrane-bound FXa with high affinity and are functionally equivalent to FVa in the absence of intentional proteolysis []. These findings indicate that proteolysis within the B domain, while necessary, is incidental to the mechanism by which cofactor function is actually realized. Instead, proteolytic activation of FV simply eliminates steric and/or conformational constraints imposed by the B domain that interfere with discrete binding interactions essential to the FVa cofactor activity, for example, the FXa-binding site. Removal of these inhibitory constraints through recombinant truncation bypasses the requirement for proteolysis to activate the molecule. Using a panel of progressively finer B domain-truncated variants, we were able to identify a discrete region of the B domain that appears to play a critical role in stabilizing the procofactor state []. Part of this B domain region (residues 963–1008) is unusually basic with 18 out of 46 residues being Arg or Lys and is well conserved across the vertebrate lineage []. As expected, disruption of this B domain region by mutagenesis or through deletion yielded derivatives with cofactor-like properties in the absence of intentional proteolysis, indicating that the length of the B domain per se is not a primary factor in preserving the procofactor state. Thrombin-mediated proteolysis of FV facilitates removal of these inhibitory B domain sequences; however, it is likely that other, as yet to be identified components of the B domain also play a role in preserving the procofactor state.

INSIGHT INTO B DOMAIN FUNCTION FROM NON-MAMMALIAN FORMS OF FACTOR V (FV)

Whereas most of the B domain sequence is highly variable throughout vertebrate evolution, several short motifs are strongly conserved, including the basic region 963–1008 detailed above [–]. An exception to these findings has been found in an unusual form of FV derived from the venom of some Australian Elapidae family members (, and ), which are among the most venomous snakes in the world []. A unique feature of these snakes is that approximately 5–40% of their venom consists of a large prothrombin activating complex comprising a cofactor FVa-like subunit and a serine protease FXa-like subunit, which share high sequence homology with mammalian FVa and FXa (55–60%) [–]. Remarkably, the FV homologues expressed in the elapid venom have extraordinarily short B domains: 46 vs. ∼600–800 residues in mammals, and they lack the basic region. This intriguing observation prompted us to assess the functional properties of purified recombinant venom-derived FV (-FV) []. Consistent with our previous observations, we were able to show that the absence of the basic region correlates with the expression of procoagulant FV activity, indicating that venom FV is expressed as a constitutionally active FV variant []. As such, this is the first FV species observed thus far that exists as an active cofactor. Notably, this protein can also function in the absence of anionic membranes and is completely resistant to inactivation by APC, despite APC-mediated proteolysis within the heavy chain at the equivalent Arg and Arg sites []. We speculate that a unique disulfide bond between the A2 and A3 domains of -FV enhances its structural stability and prevents dissociation of the A2 domain upon APC cleavage []. Thus, -FV represents an exceptional example of a protein that has adapted into a potent biological weapon for host defense and envenomation of prey.

MUTATIONS IN FACTOR V (FV) ASSOCIATED WITH DISEASE STATES

Some mutations in FV have been reported to increase the risk of developing thrombosis (for a review see Ref. []), whereas others can lead to a bleeding disorder (reviewed in []). No mutations or polymorphisms associated with disease states have been observed at the FV thrombin cleavage sites [].

ACTIVATED P C (APC)-RESISTANT FACTOR V (FV)

One of the most commonly observed genetic risk factors for thrombosis is FV, which is partially resistant to inactivation by APC due to a Gln substitution at the APC cleavage site Arg –]. Individuals who are heterozygous for this mutation have a 3–8-fold increased risk of developing thrombosis, whereas the risk for homozygous individuals is 50–80-fold higher as compared to individuals with normal FV [, ]. Several other mutations and polymorphisms that have been shown to also result in some degree of APC resistance are reviewed elsewhere [].

PARAHEMOPHILIA

FV deficiency (parahemophilia) is an autosomal recessive bleeding disorder that was first described in the 1940s by Paul Owren in Norway []. It is a rare bleeding disorder that affects one in a million individuals and is characterized by low or undetectable FV activity and/or antigen levels (reviewed in []). Patients with undetectable levels of FV (<1%) due to homozygous non-sense, frameshift, or missense mutations exhibit the gamut of phenotypes from asymptomatic to severe bleeding. Two-thirds of all mutations causing FV deficiency are non-sense mutations in the FV gene []. Even though mRNA containing a premature stop codon is normally degraded by non-sense-mediated decay, it has been suggested that trace amounts of FV may be expressed following ribosomal slippage or somatic inversion events []. Of the reported missense mutations, most are clustered in the A and C domains, while none are found in the B domain []. Deficiency in FV presents a conundrum because the phenotype is variable and unexpectedly correlates poorly with FV levels in plasma [, –]. These observations are generally not consistent with the fundamental role played by FV in coagulation and with findings in the FV knockout mice that have a lethal phenotype []. However, observations made in the past 2 years point to an important role for platelet FV that may help explain these observations.

Approximately one-fifth of the total FV pool is stored in the α-granules of platelets from which it is secreted upon platelet activation []. While megakaryocytes can synthesize FV [–], the vast majority of platelet FV is endocytosed from the plasma pool by megakaryocytes [–]. Following endocytosis via a specific receptor-mediated process [, ], FV is modified intracellularly such that it is functionally unique compared to its plasma-derived counterpart [, ]. Recently, Duckers were able to correlate the levels of platelet FV with thrombin generation in FV-deficient patients. They showed that patients deficient in FV resulting from missense mutations have sufficient functional FV in their platelets to guarantee thrombin generation and protect them against major bleeding []. Furthermore, the same group demonstrated that the FV requirement for thrombin generation is considerably lower in these patients due to markedly reduced levels of the anticoagulant protein tissue factor pathway inhibitor (TFPI) in FV-deficient plasma []. Whether the disease phenotype correlates with the platelet FV levels and whether a similar mechanism may explain the relatively mild phenotype in patients that have an introduced stop codon in the FV gene remains to be determined.

THE PROCOFACTOR TO COFACTOR TRANSITION OF FACTOR VIII (FVIII)

FVIII circulates as a large (=330,000), multidomain (A1-A2-B-A3-C1-C2) heterodimer resulting from limited proteolysis at the B-A3 junction and at additional sites in the B domain []. This heterodimer consists of a variably sized heavy chain (A1-A2-B; 200–90 kDa) and a light chain (A3-C1-C2; 80 kDa) that are non-covalently associated (). The A domains are bordered by short segments (∼30–40 amino acids) of negatively charged residues, known as the acidic regions a1 (337–372), a2 (711–740), and a3 (1649–1689) (). Whereas regions a2 and a3 are more or less well conserved in FV (), the a1 region of FVIII is absent from both FV and ceruloplasmin [–], the latter being a ferroxidase with an A1-A2-A3 domain structure that originates from the same ancestral protein as FV and FVIII [, ]. The FVIII acidic regions are thought to function, in part, as binding sites for thrombin [, , ]. In addition, the a3 region and regions within both C domains have been suggested to mediate the tight interaction of the FVIII heterodimer with its carrier protein von Willebrand factor (VWF) [–]. This interaction serves various important roles in FVIII physiology, as it has been reported to stabilize the heterodimeric structure of FVIII as well as to prevent proteolysis by FXa and APC [].

The FVIII heterodimer is an inactive procofactor and must be subjected to limited proteolysis to effect full cofactor activity [, –]. FVIII can be proteolytically activated by both thrombin and FXa []. Thrombin cleaves three peptide bonds at Arg, Arg, and Arg, thereby generating FVIIIa, which is a heterotrimer composed of the A1 (50 kDa; 1–372), A2 (43 kDa; 373–740), and the light chain (A3-C1-C2; 73 kDa; 1689–2332) () [, ]. Activation of FVIII results in a transient ∼20–50-fold increase in biological activity that decays over a short period of time due to A2 domain dissociation from A1/A3-C1-C2, a mechanism that contributes to the regulation of FVIIIa cofactor activity [–].

Numerous studies have examined the role of the individual thrombin cleavage sites in the expression of FVIIIa cofactor activity. Similar to thrombin-mediated activation of FV, the evidence obtained supports an ordered cleavage pathway, with cleavage at Arg occurring first, followed by cleavage at Arg, and subsequently at Arg. The specific role of the individual cleavage sites in the expression of FVIIIa cofactor activity will be discussed in the following sections.

CLEAVAGE AT ARG AND ARG

Although cleavage at Arg appears to be of little consequence to the development of cofactor function [], it is thought to facilitate subsequent proteolysis at Arg and Arg []. Surprisingly however, its significance in FVIII activation is not reflected in the hemophilia A patient population, as there have been no reports of non-sense mutations at position 740 (). Cleavage of the light chain at Arg results in dissociation of FVIII from VWF, which allows for association of the cofactor with anionic phospholipids and FIXa [, ]. Whether proteolysis at this site directly contributes to the potentiation of FVIIIa cofactor activity remains controversial. There is some evidence that this cleavage partially increases cofactor activity [, ]; however, Pipe and Kaufman have shown using a single chain FVIII derivative (IR8) that cofactor activity can be obtained even in the absence of the Arg cleavage site [].

CLEAVAGE AT ARG

The results obtained with various FVIII derivatives as well as other biochemical studies and naturally occurring mutations indicate that cleavage at Arg is essential to procofactor activation [, –]. Biochemical data suggest that cleavage at this site exposes a functional FIXa-binding site that promotes rapid FX activation by cofactor-bound FIXa []. Based on these and other observations, it was suggested that acidic region a1 and possibly a portion of the a3 region (1649–1689) could obscure functionally important surface areas such as a FIXa-binding site; results that are in line with functional studies []. Interestingly, acidic region a1 of FVIII is noticeably absent from FV (missing from exon 7) possibly pointing to a unique function in FVIII () [, ]. Furthermore, recent structural data on B domain-deleted FVIII indicate that this part of FVIII is highly flexible as no electron density was observed in this region [, ]. Alternatively, cleavage at Arg could induce a change in conformation that is critical for the expression of FVIIIa cofactor function. Evidence for this comes from studies employing cross-linking agents and apolar probes as well as circular dichroism experiments. These studies support the idea that there are subtle, yet measureable changes in conformation in the vicinity of the A2 domain when FVIII is activated to FVIIIa [–]. Future biochemical and structural studies are needed to resolve the precise mechanism by which cleavage at Arg facilitates the FVIII procofactor to cofactor transition.

THE FACTOR VIII (FVIII) B DOMAIN

The FVIII B domain, like that of FV, is very large (908 residues), encoded by a single exon, heavily glycosylated, and is also removed following thrombin-mediated proteolysis; however, it does not share sequence homology with the FV B domain. Yet, unlike FV, several groups have established that removal of most of the B domain yields a derivative that remains as an inactive procofactor [, –]. Thus, the FVIII B domain does not appear to play a role in preserving FVIII as a procofactor. As such, the molecular mechanisms that regulate or prevent the potential cofactor activities of FV and FVIII are surprisingly different.

MUTATIONS IN FACTOR VIII (FVIII) ASSOCIATED WITH DISEASE STATES

In contrast to the dual effects that mutations in FV can have on the hemostatic balance (eg, procoagulant vs. anticoagulant), no FVIII mutations have been described thus far that are linked to a prothrombotic state. Even though APC resistance is strongly correlated with an enhanced risk of developing thrombosis, no mutations at the equivalent APC inactivation sites in FVIII have been observed in a cohort of patients with venous thrombosis []. In addition, recombinant FVIII variants carrying substitutions at one of the two APC cleavage sites did not show an APC resistant phenotype []. It was found, though, that high levels of FVIII are associated with an increased risk of venous thrombosis (reviewed here []). To date, no genetic variation in the FVIII gene has been identified that might account for this phenotype.

HEMOPHILIA A

A deficiency or functional defect in FVIII is at the basis of the X-linked congenital bleeding disorder known as hemophilia A (for a review see Ref. []), which has an incidence of one in 5000 in the general population. Hemophilia A is categorized as severe (<1%), moderate (1–5%), or mild (5–20%), according to the relative amount of FVIII activity in the patient's plasma. The classic symptoms of hemophilia include bleeding episodes that affect joints, muscles, internal organs, and the brain. Whereas approximately one-third of all hemophilia A cases are due to intron 22 inversions [, ], missense mutations have been observed to be the most frequent mutation type and are almost exclusively responsible for the mild/moderate hemophilia A phenotypes []. In addition, stop codons account for ∼10% of all hemophilia A mutations, and one-third of the identified FVIII gene defects other than intron inversions represent new mutations []. Most reported mutations have been registered in the hemophilia A mutation database (), and a graphic representation of the number of missense mutations per FVIII region is given in .

As expected, missense mutations at or near some of the thrombin activation sites have been reported to result in hemophilia A (). Two different substitutions have been observed at position 1689 in a total of 35 mild to severe hemophilia A patients. Furthermore, there has been one report of a substitution at the P1′ position Ser resulting in mild hemophilia. At position 372, four different missense mutations have been observed in a total of 22 reported cases of mild to severe hemophilia A, and missense mutations at P1′ Ser have been found in two mild hemophilia A patients. Remarkably, there have been no reports of missense mutations at or near position 740, which may suggest that mutations at this site do not result in hemophilia A.

CONCLUDING REMARKS

The molecular process of maintaining FV and FVIII as inactive procofactors plays a critical regulatory role that has evolved to limit the expression of cofactor activity. Despite its significance, clear mechanistic insight by which the various proteolytic events lead to expression of FVa and FVIIIa procoagulant activity has proven difficult to pinpoint. Despite the similarities in structure and function of FV and FVIII, their molecular mechanisms of activation have been shown to differ substantially. For FV, the B domain plays a fundamentally important role as discrete conserved B domain sequences are involved in the mechanism by which FV persists as an inactive procofactor. The data suggest that the FV B domain serves an inhibitory function that, under normal physiological conditions, is efficiently removed upon proteolytic processing. In contrast, the FVIII B domain does not appear to be involved in regulating cofactor activity. Rather, cleavage between the A1 and A2 domains at position Arg is critical for the procoagulant activity of FVIII. The precise mechanism by which cleavage at Arg facilitates the transition to the active cofactor state remains to be determined. Collectively these studies lay the groundwork for further uncovering the precise molecular mechanism by which FV and FVIII transition from the procofactor to cofactor state.

Conflicts of interest: RMC receives research support and royalties from Pfizer Pharmaceuticals for technology related to FXa. MHAB has no financial interest to disclose related to the contents of this article.

Funding disclosure: Funding was provided by the National Institutes of Health (HL-88010 and HL-74124, Project 2; to RMC) and the National Hemophilia Foundation (Judith Graham Pool Postdoctoral Research Fellowship to MHAB).

References

1.    Khan AM, James MNG. Molecular mechanisms for the conversion of zymogens to active proteolytic enzymes. Prot Sci. 1998;7:815-836.
2.    Mann KG, Jenny RJ, Krishnaswamy S. Cofactor proteins in the assembly and expression of blood clotting enzyme complexes. Annual Rev Biochem. 1988;57:915-956.
3.    Kane W, Davie EW. Blood coagulation factors V and VIII: structural and functional similarities and their relationships to hemorrhagic and thrombotic disorders. Blood. 1988;71:539-555.
4.    Mann KG, Jenny RJ, Krishnaswamy S. Cofactor proteins in the assembly and expression of blood clotting enzyme complexes. Ann Rev Biochem. 1988;57:915-956.
5.    Mann KG, Nesheim ME, Church WR, Haley P, Krishnaswamy S. Surface-dependent reactions of the vitamin K-dependent enzyme complexes. Blood. 1990;76:1-16.
6.    Jacquemin M, De Maeyer M, D'Oiron R, et al. Molecular mechanisms of mild and moderate hemophilia A. J Thromb Haemost. 2003;1:456-463.
7.    Mannucci PM, Duga S, Peyvandi F. Recessively inherited coagulation disorders. Blood. 2004;104:1243-1252.
8.    Ware AG, Guest MM, Seegers WH. Plasma accelerator factor and purified prothrombin activation. Science. 1947;106(2741):41-42.
9.    Ware AG, Murphy RC, Seegers WH. The function of Ac-globulin in blood clotting. Science. 1947;106(2764):618-619.
10.    Owren PA. Parahaemophila: haemorrhagic diathesis due to absence of a previously unknown clotting factor. Lancet. 1947;249(6449):446-448.
11.    Rapaport SI, Schiffman S, Patch MJ, Ames SB. The importance of activation of anti-haemophilic globulin and proaccelerin by traces of thrombin in the generation of intrinsic prothrombinase activity. Blood. 1963;21(2):221-236.
12.    Mann KG, Kalafatis M. Factor V: a combination of Dr. Jekyll and Mr. Hyde. Blood. 2002;101:20-30.
13.    Fay PJ. Activation of factor VIII and mechanisms of cofactor action. Blood Rev. 2004;18:1-15.
14.    Kane WH. Factor V. In: Hemostasis and Thrombosis. 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. p. 177-192.
15.    Tracy PB, Eide LL, Bowie EJW, Mann KG. Radioimmunoassay of factor V in human plasma and platelets. Blood. 1982;60:59-63.
16.    Suzuki K, Dahlbäck B, Stenflo J. Thrombin-catalyzed activation of human coagulation factor V. J Biol Chem. 1982;257:6556-6564.
17.    Esmon CT, Owen WG, Duiguid D, Jackson CM. The action of thrombin on blood clotting factor V: Conversion of factor V to a prothrombin-binding protein. Biochim Biophys Acta. 1973;310:289-294.
18.    Nesheim ME, Taswell JB, Mann KG. The contribution of bovine factor V and factor Va to the activity of prothrombinase. J Biol Chem. 1979;254:10952-10962.
19.    Foster WB, Nesheim ME, Mann KG. The Factor Xa-catalyzed activation of factor V. J Biol Chem. 1983;258:13970-13977.
20.    Jenny RJ, Pittman DD, Toole JJ, et al. Complete cDNA and derived amino acid sequence of human factor V. Proc Natl Acad Sci USA. 1987;84:4846-4850.
21.    Esmon CT. The subunit structure of thrombin-activated factor V. Isolation of activated factor V, separation of subunits and reconstitution of biological activity. J Biol Chem. 1979;254:964-973.
22.    Nesheim ME, Mann KG. Thrombin-catalyzed activation of single chain bovine factor V. J Biol Chem. 1979;254:1326-1334.
23.    Dharmawardana KR, Bock PE. Demonstration of exosite I-dependent interactions of thrombin with human factor V and factor Va involving the factor Va heavy chain: analysis by affinity chromatography employing a novel method for active-site selective immobilization of serine proteinases. Biochemistry. 1998;37:13143-13152.
24.    Suzuki H, Shima M, Nogami K, et al. Factor V C2 domain contains a major thrombin-binding site responsible for thrombin-catalyzed factor V activation. J Thromb Haemost. 2006;4(6):1354-1360.
25.    Krishnaswamy S, Russell GD, Mann KG. The reassociation of factor Va from its isolated subunits. J Biol Chem. 1989;264:3160-3168.
26.    Nesheim ME, Foster WB, Hewick R, Mann KG. Characterization of factor V activation intermediates. J Biol Chem. 1984;259:3187-3196.
27.    Monkovic DD, Tracy PB. Functional characterization of human platelet-released factor V and its activation by factor Xa and thrombin. J Biol Chem. 1990;265:17132-17140.
28.    Thorelli E, Kaufman RJ, Dahlbäck B. Cleavage requirements for activation of factor V by factor Xa. Eur J Biochem. 1997;247:12-20.
29.    Kane WH, Majerus PW. Purification and characterization of human coagulation factor V. J Biol Chem. 1981;256:1002-1007.
30.    Segers K, Rosing J, Nicolaes GA. Structural models of the snake venom factor V activators from Daboia russelli and Daboia lebetina. Proteins. 2006;64(4):968-984.
31.    Bakker HM, Tans G, Thomassen MCLGD, et al. Functional properties of human factor Va lacking the Asp683-Arg709 domain of the heavy chain. J Biol Chem. 1994;269:20662-20667.
32.    Siigur J, Aaspollu A, Tonismagi K, et al. Proteases from Vipera lebetina venom affecting coagulation and fibrinolysis. Haemostasis. 2001;31(3-6):123-132.
33.    Bradford HN, Annamalai A, Doshi K, Colman RW. Factor V is activated and cleaved by platelet calpain: comparison with thrombin proteolysis. Blood. 1988;71:388-394.
34.    Lee CD, Mann KG. Activation/Inactivation of human factor V by plasmin. Blood. 1989;73(1):185-190.
35.    Kane WH, Mruk JS, Majerus PW. Activation of coagulation factor V by a platelet protease. J Clin Invest. 1982;70:1092-1100.
36.    Tracy PB, Nesheim ME, Mann KG. Proteolytic alteration of factor Va bound to platelets. J Biol Chem. 1983;258:662-669.
37.    Tans G, Nicolaes GAF, Christella M, et al. Activation of human factor V by meizothrombin. J Biol Chem. 1994;269:15969-15972.
38.    Oates AM, Salem HH. The regulation of human factor V by a neutrophil protease. Blood. 1987;70:846-851.
39.    Allen DH, Tracy PB. Human coagulation factor V is activated to the functional cofactor by elastase and cathepsin G expressed at the monocyte surface. J Biol Chem. 1995;270:1408-1415.
40.    Samis JA, Garrett M, Manuel RP, Nesheim ME, Giles AR. Human neutrophil elastase activates human factor V but inactivates thrombin-activated human factor V. Blood. 1997;90:1065-1074.
41.    Camire RM, Kalafatis M, Tracy PB. Proteolysis of factor V by cathepsin G and elastase indicates that cleavage at Arg1545 optimizes cofactor function by facilitating factor Xa binding. Biochemistry. 1998;37:11896-11906.
42.    Esmon NL, Owen WG, Esmon CT. Isolation of a membrane-bound cofactor for thrombin-catalyzed activation of Protein C. J Biol Chem. 1982;257:859-864.
43.    Keller FG, Ortel TL, Quinn-Allen MA, Kane WH. Thrombin-catalyzed activation of recombinant human factor V. Biochemistry. 1995;34:4118-4124.
44.    Steen M, Dahlbäck B. Thrombin-mediated proteolysis of factor V resulting in gradual B-domain release and exposure of the factor Xa-binding site. J Biol Chem. 2002;277:38424-38430.
45.    Smith CM, Hanahan DJ. The activation of factor V by factor Xa or -chymotrypsin and comparison with thrombin and RVV-V action. An improved factor V isolation procedure. Biochemistry. 1976;15:1830-1837.
46.    Kane WH, Majerus PW. Purification and characterization of human coagulation factor V. J Biol Chem. 1981;256:1002-1007.
47.    Kalafatis M, Beck DO, Mann KG. Structural requirements for expression of factor Va activity. J Biol Chem. 2004;278:33550-33561.
48.    Kane WH, Ichinose A, Hagen FS, Davie EW. Cloning of cDNAs coding for the heavy chain region and connecting region of human Factor V, a blood coagulation factor with four types of internal repeats. Biochemistry. 1987;26:6508-6514.
49.    Mosesson MW, Church WR, DiOrio JP, et al. Structural model of factors V and Va based on scanning transmission electron microscope images and mass analysis. J Biol Chem. 1990;265(15):8863-8868.
50.    Mosesson MW, Nesheim ME, DiOrio JP, et al. Studies on the structure of bovine factor V by scanning transmission electron microscopy. Blood. 1985;65(5):1158-1162.
51.    Segers K, Dahlback B, Nicolaes GA. Coagulation factor V and thrombophilia: background and mechanisms. Thromb Haemost. 2007;98(3):530-542.
52.    Kane WH, Devore-Carter D, Ortel TL. Expression and characterization of recombinant human factor V and a mutant lacking a major portion of the connecting region. Biochemistry. 1990;29:6762-6768.
53.    Toso R, Camire RM. Removal of B-domain sequences from factor V rather than specific proteolysis underlies the mechanism by which cofactor function is realized. J Biol Chem. 2004;279:21643-21650.
54.    Zhu H, Toso R, Camire RM. Inhibitory sequences within the B-domain stabilize circulating factor V in an inactive state. J Biol Chem. 2007;282(20):15033-15039.
55.    Camire RM, Bos MH. The molecular basis of factor V and VIII procofactor activation. J Thromb Haemost. 2009;7(12):1951-1961.
56.    Vos HL. Alignment of factor V sequences in twelve mammalian species shows several strongly conserved motifs in the B-domain. J Thromb Haemost. 2005;3(Suppl. 1):P0041.
57.    Vos HL, van Wijngaarden A. Variation and conservation of the B-domain of factor V. J Thromb Haemost. 2009;7(2):PP-MO-151.
58.    Broad AJ, Sutherland SK, Coulter AR. The lethality in mice of dangerous Australian and other snake venom. Toxicon. 1979;17(6):661-664.
59.    Bos MH, Boltz M, St PL, et al. Venom factor V from the common brown snake escapes hemostatic regulation through procoagulant adaptations. Blood. 2009;114(3):686-692.
60.    Lavin MF, Masci PP. Prothrombinase complexes with different physiological roles. Thromb Haemost. 2009;102(3):421-423.
61.    Masci PP, Whitaker AN, de Jersey J. Purification and characterization of a prothrombin activator from the venom of the Australian brown snake, Pseudonaja textilis. Biochem Int. 1988;17(5):825-835.
62.    Rao VS, Kini RM. Pseutarin C, a prothrombin activator from Pseudonaja textilis venom: its structural and functional similarity to mammalian coagulation factor Xa-Va complex. Thromb Haemost. 2002;88(4):611-619.
63.    Speijer H, Govers-Riemslag JW, Zwaal RF, Rosing J. Prothrombin activation by an activator from the venom of Oxyuranus scutellatus (Taipan snake). J Biol Chem. 1986;261(28):13258-13267.
64.    Vos HL. Inherited defects of coagulation factor V: the thrombotic side. J Thromb Haemost. 2005;4:35-40.
65.    Asselta R, Tenchini ML, Duga S. Inherited defects of coagulation factor V: the hemorrhagic side. J Thromb Haemost. 2006;4(1):26-34.
66.    Vos HL. An online database of mutations and polymorphisms in and around the coagulation factor V gene. J Thromb Haemost. 2007;5(1):185-188.
67.    Bertina RM, Koeleman B.P.C., Koster T., et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature. 1994;369:64-67.
68.    Greengard JS, Sun X, Xu X, et al. Activated protein C resistance caused by Arg506Gln mutation in factor Va. Lancet. 1994;343:1362-1363.
69.    Sun X, Evatt B, Griffin JH. Blood coagulation factor Va abnormality associated with resistance to activated protein C in venous thrombophilia. Blood. 1994;83:3120-3125.
70.    Voorberg J., Roelse J., Koopman R., et al. Association of idiopathic venous thromboembolism with single point-mutation at Arg506 of factor V. Lancet. 1994;343:1535-1538.
71.    Zöller B, Svensson PJ, He X, Dahlbäck B. Identification of the same factor V gene mutation in 47 out of 50 thrombosis-prone families with inherited resistance to activated protein C. J Clin Invest. 1994;94:2521-2524.
72.    Koster T, Rosendaal FR, de Ronde H, et al. Venous thrombosis due to a poor anticoagulant response to activated protein C: Leiden Thrombophilia study. Lancet. 1993;342:1503-1506.
73.    Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH. High risk thrombosis in patients homozygous for factor V Leiden (activated protein C resistance). Blood. 1995;85:1504-1508.
74.    Segers O, Castoldi E. Factor V Leiden and activated protein C resistance. Adv Clin Chem. 2009;49:121-157.
75.    Duckers C, Simioni P, Rosing J, Castoldi E. Advances in understanding the bleeding diathesis in factor V deficiency. Br J Haematol. 2009;146(1):17-26.
76.    Lunghi B, Pinotti M, Maestri I, Batorova A, Bernardi F. Evaluation of factor V mRNA to define the residual factor V expression levels in severe factor V deficiency. Haematologica. 2008;93(3):477-478.
77.    Ellestad SC, Zimmerman SA, Thornburg C, et al. Severe factor V deficiency presenting with intracranial haemorrhage during gestation. Haemophilia. 2007;13(4):432-434.
78.    Guasch JF, Cannegieter S, Reitsma PH, van't Veer-Korthof E. Severe coagulation factor V deficiency caused by a 4 bp deletion in the factor V gene. Br J Haematol. 1998;101:32-39.
79.    Lak M, Sharifian R, Peyvandi F, Mannucci PM. Symptoms of inherited factor V deficiency in 35 Iranian patients. Br J Haematol. 1998;103(4):1067-1069.
80.    Salooja N, Martin P, Khair K, Liesner R, Hann I. Severe factor V deficiency and neonatal intracranial haemorrhage: a case report. Haemophilia. 2000;6(1):44-46.
81.    Van WR, Montefusco MC, Duga S, et al. Coexistence of a novel homozygous nonsense mutation in exon 13 of the factor V gene with the homozygous Leiden mutation in two unrelated patients with severe factor V deficiency. Br J Haematol. 2001;114(4):871-874.
82.    Cui J, O'Shea KS, Purkayastha A, Saunders TL, Ginsburg D. Fatal haemorrhage and incomplete block to embryogenesis in mice lacking coagulation factor V. Nature. 1996;384:66-68.
83.    Gewirtz AM, Keefer M, Doshi K, et al. Biology of human megakaryocyte Factor V. Blood. 1986;67:1639-1642.
84.    Chiu HC, Schick P, Colman RW. Biosynthesis of Factor V in isolated guinea pig megakaryocytes. J Clin Invest. 1985;75:339-346.
85.    Giampaolo A, Vulcano F, Macioce G, et al. Factor-V expression in platelets from human megakaryocytic culture. Br J Haematol. 2005;128(1):108-111.
86.    Camire RM, Pollak ES, Kaushansky K, Tracy PB. Secretable human platelet-derived factor V originates from the plasma pool. Blood. 1998;92:3035-3041.
87.    Christella M, Thomassen LG, Castoldi E, et al. Endogenous factor V synthesis in megakaryocytes contributes negligibly to the platelet factor V pool. Haematologica. 2003;88(10):1150-1156.
88.    Gould WR, Simioni P, Silveira JR, et al. Megakaryocytes endocytose and subsequently modify human factor V in vivo to form the entire pool of a unique platelet-derived cofactor. J Thromb Haemost. 2005;3(3):450-456.
89.    Bouchard BA, Williams JL, Meisler NT, Long MW, Tracy PB. Endocytosis of plasma-derived factor V by megakaryocytes occurs via a clathrin-dependent, specific membrane binding event. J Thromb Haemost. 2005;3(3):541-551.
90.    Bouchard BA, Meisler NT, Nesheim ME, et al. A unique function for LRP-1: a component of a two-receptor system mediating specific endocytosis of plasma-derived factor V by megakaryocytes. J Thromb Haemost. 2008;6(4):638-644.
91.    Gould WR, Silveira JR, Tracy PB. Unique in vivo modifications of coagulation factor V produce a physically and functionally distinct platelet-derived cofactor: characterization of purified platelet-derived factor V/Va. J Biol Chem. 2004;279(4):2383-2393.
92.    Duckers C, Simioni P, Spiezia L, et al. Residual platelet factor V ensures thrombin generation in patients with severe congenital factor V deficiency and mild bleeding symptoms. Blood. 2009.Vol:115;pp 879-886
93.    Duckers C, Simioni P, Spiezia L, et al. Low plasma levels of tissue factor pathway inhibitor in patients with congenital factor V deficiency. Blood. 2008;112(9):3615-3623.
94.    Cripe LD, Moore D, Kane WH. Structure of the gene for human coagulation factor V. Biochemistry. 1992;31:3777-3785.
95.    Gitschier J, Wood WI, Goralka TM, et al. Characterization of the human factor VIII gene. Nature. 1984;312:326-330.
96.    Koschinsky ML, Funk WD, van Oost BA, MacGillivray RT. Complete cDNA sequence of human preceruloplasmin. Proc Natl Acad Sci USA. 1986;83(14):5086-5090.
97.    Davidson CJ, Tuddenham EG, McVey JH. 450 million years of hemostasis. J Thromb Haemost. 2003;1(7):1487-1494.
98.    Jiang Y, Doolittle RF. The evolution of vertebrate blood coagulation as viewed from a comparison of puffer fish and sea squirt genomes. Proc Natl Acad Sci USA. 2003;100(13):7527-7532.
99.    Newell JL, Fay PJ. Acidic residues C-terminal to the A2 domain facilitate thrombin-catalyzed activation of factor VIII. Biochemistry. 2008;47(33):8786-8795.
100.    Nogami K, Zhou Q, Myles T, et al. Exosite-interactive regions in the A1 and A2 domains of factor VIII facilitate thrombin-catalyzed cleavage of heavy chain. J Biol Chem. 2005;280(18):18476-18487.
101.    Jacquemin M, Lavend'homme R, Benhida A, et al. A novel cause of mild/moderate hemophilia A: mutations scattered in the factor VIII C1 domain reduce factor VIII binding to von Willebrand factor. Blood. 2000;96(3):958-965.
102.    Leyte A, Verbeet MP, Brodniewicz-Proba T, Van Mourik JA, Mertens K. The interaction between human blood-coagulation factor VIII and von Willebrand factor. Characterization of a high-affinity binding site on factor VIII. Biochem J. 1989;257(3):679-683.
103.    Lollar P, Hill-Eubanks DC, Parker CG. Association of the factor VIII light chain with von Willebrand factor. J Biol Chem. 1988;263:10451-10455.
104.    Saenko EL, Scandella D. The acidic region of the factor VIII light chain and the C2 domain together form the high affinity binding site for von Willebrand factor. J Biol Chem. 1997;272(29):18007-18014.
105.    Lenting PJ, Van Mourik JA, Mertens K. The life cycle of coagulation factor VIII in view of its structure and function. Blood. 1998;92(11):3983-3996.
106.    Vehar GA, Davie EW. Preparation and properties of bovine factor VIII (antihemophilic factor). Biochemistry. 1980;19(3):401-410.
107.    Fass DN, Knutson GJ, Katzmann JA. Monoclonal antibodies to porcine factor VIII coagulant and their use in the isolation of active coagulant protein. Blood. 1982;59:594-600.
108.    Fulcher CA, Roberts J, Zimmerman TS. Thrombin proteolysis of purified FVIII procoagulant protein: Correlation of activation with generation of specific polypeptides. Blood. 1983;60:807-811.
109.    Lollar P, Knutson G, Fass D. Stabilisation of thrombin activated porcine Factor VIII: C by Factor IXa and phospholipid. Blood. 1984;63:1303-1308.
110.    Toole JJ, Knopf JL, Wozney JM, et al. Molecular cloning of a cDNA encoding human antihaemophilic factor. Nature. 1984;312:342-347.
111.    Wood WI, Capon DJ, Simonsen CC, et al. Expression of active human factor VIII from recombinant DNA clones. Nature. 1984;312:330-336.
112.    Eaton D, Rodriguez H, Vehar G. Proteolytic processing of human FVIII. Correlation of specific cleavages by thrombin FXa and activated protein C with activation and inactivation of Factor VIII coagulant activity. Biochemistry. 1986;25:505-512.
113.    Vehar G, Keyt B, Eaton D, et al. Structure of human factor VIII. Nature. 1984;312:337-342.
114.    Lollar P, Parker C. pH-dependent denaturation of thrombin-activated porcine factor VIII. J Biol Chem. 1990;265:1688-1692.
115.    Fay PJ, Haidari PJ, Smudzin TM. Human factor VIIIa subunit structure. Reconstitution of factor VIIIa from the isolated A1/A3-C1-C2 dimer and A2 subunit. J Biol Chem. 1991;266:8957-8962.
116.    Lollar P, Parker ET. Structural basis for the decreased procoagulant activity of human factor VIII compared to the porcine homolog. J Biol Chem. 1991;266(19):12481-12486.
117.    Pittman DD, Kaufman RJ. Proteolytic requirements for thrombin activation of anti-hemophilic factor (factor VIII). Proc Natl Acad Sci USA. 1988;85:2429-2433.
118.    Newell JL, Fay PJ. Proteolysis at Arg740 facilitates subsequent bond cleavages during thrombin-catalyzed activation of factor VIII. J Biol Chem. 2007;282(35):25367-25375.
119.    Hamer RJ, Koedam JA, Beeser-Visser NH, et al. Factor VIII binds to von Willebrand factor via its Mr-80,000 light chain. Eur J Biochem. 1987;166:37-43.
120.    Hill-Eubanks DC, Parker CG, Lollar P. Differential proteolytic activation of factor VIII-von Willebrand factor complex by thrombin. Proc Natl Acad Sci USA. 1989;86:6508-6512.
121.    Regan LM, Fay PJ. Cleavage of factor VIII light chain is required for maximal generation of factor VIIIa activity. J Biol Chem. 1995;270:8546-8552.
122.    Donath M-JSH, Lenting PJ, Van Mourik JA, Mertens K. The role of cleavage of the light chain at positions Arg1689 or Arg1721 in subunit interactions and activation of human blood coagulation factor VIII. J Biol Chem. 1995;270:3648-3655.
123.    Pipe SW, Kaufman RJ. Characterization of genetically engineered inactivation-resistant coagulation factor VIIIa. Proc Natl Acad Sci USA. 1997;94:11851-11856.
124.    Shima M, Ware J, Yoshioka A, Fukui H, Fulcher CA. An arginine to cysteine amino acid substitution at a critical thrombin cleavage site in dysfunctional factor VIII molecule. Blood. 1989;74:1612-1617.
125.    Arai M, Inaba H, Higuchi M, et al. Direct characterization of factor VIII in plasma: detection of a mutation altering a thrombin cleavage site (arginine-372 to histidine). Proc Natl Acad Sci USA. 1989;86:4277-4281.
126.    O'Brien DP, Pattinson JK, Tuddenham EGD. Purification and characterization of factor VIII 372-Cys: a hypofunctional cofactor from a patient with moderately severe haemophilia A. Blood. 1990;75:1664-1672.
127.    Nogami K, Zhou Q, Wakabayashi H, Fay PJ. Thrombin-catalyzed activation of factor VIII with His substituted for Arg372 at the P1 site. Blood. 2005;105(11):4362-4368.
128.    Fay PJ, Mastri M, Koszelak ME, Wakabayashi H. Cleavage of factor VIII heavy chain is required for the functional interaction of a2 subunit with factor IXa. J Biol Chem. 2001;276(15):12434-12439.
129.    Shen BW, Spiegel PC, Chang CH, et al. The tertiary structure and domain organization of coagulation factor VIII. Blood. 2008;111(3):1240-1247.
130.    Ngo JC, Huang M, Roth DA, Furie BC, Furie B. Crystal structure of human factor VIII: implications for the formation of the factor IXa-factor VIIIa complex. Structure. 2008;16(4):597-606.
131.    Curtis JE, Helgerson SL, Parker ET, Lollar P. Isolation and characterization of thrombin-activated human factor VIII. J Biol Chem. 1994;269(8):6246-6251.
132.    O'Brien LM, Huggins CF, Fay PJ. Interacting regions in the A1 and A2 subunits of factor VIIIa identified by zero-length cross-linking. Blood. 1997;90(10):3943-3950.
133.    Sudhakar K, Fay PJ. Exposed hydrophobic sites in factor VIII and isolated subunits. J Biol Chem. 1996;271(38):23015-23021.
134.    Toole JJ, Pittman DD, Orr EC, et al. A large region (~95 kDa) of human factor VIII is dispensable for in vitro procoagulant activity. Proc Natl Acad Sci USA. 1986;83:5939-5942.
135.    Eaton DL, Wood WI, Eaton D, et al. Construction and characterization of an active factor VIII variant lacking the central one-third of the molecule. Biochemistry. 1986;25:8343-8347.
136.    Pittman DD, Alderman EM, Tomkinson KN, et al. Biochemical, immunological, and in vivo functional characterization of B-domain-deleted FVIII. Blood. 1993;81(11):2925-2935.
137.    Lind P, Larsson K, Spira J, et al. Novel forms of B-domain deleted recombinant factor VIII molecules. Construction and biochemical characterization. Eur J Biochem. 1995;232:19-27.
138.    Roelse JC, Koopman MM, Buller HR, et al. Absence of mutations at the activated protein C cleavage sites of factor VIII in 125 patients with venous thrombosis. Br J Haematol. 1996;92(3):740-743.
139.    Amano K, Michnick DA, Moussalli M, Kaufman RJ. Mutation at either Arg336 or Arg562 in factor VIII is insufficient for complete resistance to activated protein C-mediated inactivation: implications for the APC resistance test. Thromb Haemost. 1998;79:557-563.
140.    Kamphuisen PW, Eikenboom JC, Bertina RM. Elevated factor VIII levels and the risk of thrombosis. Arterioscler Thromb Vasc Biol. 2001;21(5):731-738.
141.    Graw J, Brackmann HH, Oldenburg J, et al. Haemophilia A: from mutation analysis to new therapies. Nat Rev Genet. 2005;6(6):488-501.
142.    Lakich D, Kazazian HH, Antonarakis SE, Gitschier J. Inversions disrupting the factor VIII gene are a common cause of severe haemophilia A. Nature Genet. 1993;5:236-241.
143.    Naylor J, Brinke A, Hassock S, Green PM, Giannelli F. Characteristic mRNA abnormality found in half the patients with severe haemophilia A is due to large DNA inversions. Hum Mol Genet. 1993;2(11):1773-1778.
144.    Oldenburg J, Ananyeva NM, Saenko EL. Molecular basis of haemophilia A. Haemophilia. 2004;10(Suppl. 4):133-139.

Individually Tailored Prophylaxis in Patients with Severe Hemophilia

Mon, 28 Jun 2010 08:34:50 +0100

INTRODUCTION

Patients with severe hemophilia are at risk of life-threatening bleeding and development of joint destruction []. Bleeding into joints (hemarthrosis) occurs following imperceptible trauma and repeated hemarthroses lead to the development of synovial hypertrophy, cartilage and bone damage with chronic pain, limited range of motion, and fixed flexion contractures responsible for disability [, ]. This complication, known as hemophilic arthropathy, is associated with high morbidity [] and may be prevented by institution of primary prophylaxis in early childhood []. Prophylaxis involves regular infusions of factor VIII/IX (FVIII/IX) concentrate in order to prevent bleeding episodes. Starting from the clinical observations that patients with moderate and mild hemophilia rarely develop severe chronic arthropathy, the main objective of primary prophylaxis is therefore to reduce the number of joint bleeds by maintaining the patient's plasma FVIII-FIX levels ≥ 1 IU/dL [, ]. Numerous studies in hemophiliacs have confirmed the efficacy of prophylaxis in reducing the annual number of major bleeds, clinic visits, and hospitalizations, in addition to the development of arthropathy [, ]. There are also some data suggesting that regular prophylaxis may be an independent negative predictor associated with a 60% lower risk of inhibitor development than on-demand treatment []. Thus, severe hemophiliacs on prophylaxis and their families have a better quality of life than patients who receive on-demand treatment []. In the 1990s, major hemophilia organizations and societies therefore recommended long-term prophylaxis as an optimal treatment modality for severe hemophilia [, ]. In countries with large resources where prophylaxis is widely used, the main issue is determined by the optimal prophylactic regimen rather than the choice between on-demand treatment and prophylaxis. However, prophylaxis is a demanding medical regimen and recent surveys have reported that the international recommendations have not been uniformly adopted in several countries [, ].

In this review, we will focus on current prophylactic treatment strategies for severe hemophilia and highlight the need to develop more optimal, individually tailored prophylactic strategies.

CURRENT PRIMARY PROPHYLACTIC STRATEGIES

The Malmö model

Primary prophylaxis, originally practiced in Sweden more than 40 years ago, may have different objectives according to treatment choices. At one end of this large panel of prophylactic strategies, the aim is to prevent arthropathy and life-threatening bleeding, while at the other end the Swedish prophylaxis goal is that patients with severe hemophilia should be able to live as normal a life as possible without over protection []. According to Swedish protocol, prophylaxis is started around the age of 1 year, when children begin to walk and before the occurrence of the first joint bleed. The rationale for starting prophylaxis before the onset of joint bleeds is to avoid the risk of a target joint developing. Using this protocol, Nilsson

[, ] reported a large series of 60 patients on prophylaxis who received 25–40 IU/kg three times a week for hemophilia A and twice weekly for hemophilia B. Clinically, no bleeds were observed and the radiologic joint scores were zero, ie, absence of abnormality []. A follow-up report published 5 years later confirmed that all children treated with this regimen had perfect musculoskeletal status []. Over the years, the Malmö protocol has been refined and is currently 20–30 IU/kg every other day in hemophilia A and every third day in hemophilia B [, ]. Prophylaxis is aimed at keeping plasma FVIII-FIX activity above 1 IU/dL. All hemophilia patients receive prophylaxis and continue to use it into adulthood. Full-dose primary prophylaxis based on the Swedish experience is considered the gold standard of hemophilia therapy []. It has also been shown that high-dose, long-term prophylaxis also preserves bone mineral density []. Primary full-dose prophylaxis clearly reduces the rate of joint bleeding in most patients, but compliance with frequent injections, difficult venous access in young children, complications related to central venous access devices, the lack of access to safe clotting factor concentrates, and cost are the biggest barriers to overcome for primary prophylaxis to be adopted globally [–].

The Dutch intermediate-dose prophylaxis

An alternative approach to the full-dose Malmö prophylaxis regimen is the intermediate-dose prophylaxis favored by the Dutch group, which has been used in the Netherlands since the early 1970s []. According to this approach, prophylaxis is started in children after the occurrence of at least one joint bleed []. Prophylactic treatment is started with twice weekly infusions for hemophilia A, once weekly infusions for hemophilia B, and the frequency of infusions are intensified over the years according to the patient's bleeding pattern [, ]. Patients generally receive 15–25 IU/kg two or three times a week for hemophilia A and 30–50 IU/kg once or twice weekly for hemophilia B. Bleeding frequency is evaluated by clinical follow-up. Independently of plasma FVIII-FIX levels, prophylactic infusion dosages are increased in case of spontaneous breakthrough bleeds. In patients on long-term intermediate-dose prophylaxis, variability in bleeding phenotype is observed in terms of joint bleeds and the dose required to prevent bleeding []. In cohorts from the Netherlands and Denmark, treated with early intermediate-dose prophylaxis, 35% of patients exhibiting a milder bleeding phenotype discontinued prophylaxis in adulthood and experienced a median of three joint bleeds per year only [, ]. Compared with the high-dose prophylactic regimen, intermediate-dose prophylaxis shows a higher annual number of joints bleeds with slightly higher Pettersson X-ray scores and two-fold less clotting factor consumption after two decades of follow-up []. Thus, the Dutch group proposes that it is more cost effective to use the bleeding pattern instead of a residual plasma factor activity of 1 IU/dL as the criterion to increase prophylactic therapy.

The Canadian-tailored prophylaxis

An individualized approach by dose-escalation is used in Canada. All children with severe hemophilia A begin early prophylaxis with 50 IU/kg of FVIII. If a threshold of bleeding is reached, defined as (i) three or more hemorrhages into a single joint or (ii) four clinically determined soft tissue or joint hemorrhages affecting any site within a consecutive 3-month period or (iii) five or more bleeds into any single joint over any time period, the patient is escalated to 30 IU/kg twice weekly. Using the same threshold criteria, a child can be escalated further to 25 IU/kg on alternate days (three times a week) [, ]. The Canadian experience shows that after the first 5 years, 48% of patients remain on once-weekly therapy and only 16% of children need escalation to full-dose Malmö regimen [, ]. A systematic rapid increase of dosages may therefore be unnecessary for certain patients. Patients on the Canadian prophylactic regimen do not present life-threatening bleeds, but may develop target joints [–].

The French protocol

In 2006, the French group CoMETH (Medical Coordination for the study and treatment of the constitutional hemorrhagic diseases) proposed comprehensive recommendations for long-term primary prophylaxis dedicated to children with severe hemophilia, whatever their bleeding history and joint status [, ]. The French approach is very similar to the Canadian protocol, in that it stipulates that prophylactic regimen should be intensified only when unacceptable bleeding has occurred, but it tolerates less breakthrough bleeds. As a consequence, patients are escalated quicker in the French protocol compared with the Canadian regimen [].

Currently, there is no international recommendation on prophylactic therapy regimens and each hemophilia comprehensive care center decides the dose and frequency of infusions according to the locally established criteria. Because of the high cost and limited availability of clotting factor concentrates, dosing is an important issue in prophylactic therapy. Tailored prophylactic regimens favored in Canada, the Netherlands, and France beginning at a low frequency and escalating with repeated bleeding may prevent arthropathy and life-threatening bleeding at a lower cost than the high-dose Malmö protocol, but they carry the risk that certain patients may exhibit some joint bleeds before dose escalation occurs. The challenge is to optimize hemophilia therapy by individualizing prophylactic regimen to the needs of each patient. It would be highly relevant to find a reliable laboratory marker to identify individuals who can be maintained on once-weekly prophylaxis and to introduce a novel approach to the management of hemophilia based on “individual tailoring” of prophylactic therapies rather than the “same regimen for all.”

Individualization of prophylaxis according to the pharmacokinetics of FVIII-FIX

In the 1990s, the Swedish group described considerable inter-individual variation in required doses of FVIII-FIX and reported that tailoring the dose of clotting factor according to the pharmacokinetics of FVIII-FIX in the individual patients could raise trough levels with a saving of factor concentrates [–]. This data has been confirmed by other groups, emphasizing the importance of pharmacokinetic evaluations before considering prophylactic treatment []. However, in 2003, Ahnström demonstrated that pharmacokinetic data revealed a weak relationship between plasma clotting factor activity and clinical bleeding tendency in patients with hemophilia A and B []. In that study, some patients with very low plasma FVIII:C <1 IU/dL had no joint bleeds while others with higher FVIII-FIX levels exhibited spontaneous hemarthroses []. This data confirm that plasma FVIII-FIX activity does not fully predict the clinical bleeding tendency of patients. In fact, 10–15% of patients with severe hemophilia have a milder bleeding phenotype and experience no or only minor arthropathy []. The basis for this heterogeneity in the clinical expression of severe hemophilia is still poorly understood. Possible explanations include (i) the effect of co-inherited thrombophilia markers such as factor V Leiden or factor II G20210A mutations [, ] and (ii) the importance of the other coagulation factors and inhibitors in the determination of the overall hemostatic capacity [].

Can we use thrombin generation measurement for individualization of prophylaxis?

Classical clotting tests such as activated partial prothrombin time (aPTT) and prothrombin time assess only a coagulation time and do not reflect thrombin generation entirely. The thrombin generation test (TGT) is a global hemostasis assay reflecting the overall function of the blood clotting system []. The most important parameters that can be derived from the TGT are:Several groups reported a statistically significant correlation between plasma FVIII-FIX levels and ETP, peak and time to peak obtained by thrombin generation measurement [–]. Our group has published the mean values of the TGT parameters in severe, moderate, and mild hemophiliacs []. Moreover, independently of the plasma FVIII-FIX activity, a correlation was found between severe clinical bleeding phenotype and thrombin-generating capacity [–]. We demonstrated that patients with a severe clinical bleeding tendency usually had a low ETP <50% of normal, independently of FVIII:C/FIX:C plasma concentrations []. In a large series of severe hemophilia, using platelet rich plasma, Santagostino [] demonstrated an association between clinical phenotype of severe hemophilia and thrombin-generating capacity, showing significantly higher ETPs in patients with a mild bleeding tendency. Taken together, these data strongly suggest that thrombin generation assay may be a useful laboratory tool to reflect the individual bleeding tendency of patients with hemophilia independently of their FVIII-FIX levels.

Our group and others also showed that the TGT can be used for monitoring the efficacy of FVIII-FIX replacement therapy [, ]. Our results, obtained 24 h after FVIII concentrate administration, showed that in patients exhibiting similar plasma FVIII levels, thrombin generation capacity may be significantly different (). Recently, this information was confirmed by Lewis [] in a larger series of patients. This suggests that in hemophiliacs, the TGT could be useful for tailoring a prophylactic regimen to the individual patient. A multicenter, prospective study, named OPTIPHASE (optimizing prophylaxis in patients with severe hemophilia A by tailoring the infusions to individual patient's needs) has been designed to assess the TGT for tailoring prophylactic regimen to the individual needs of each patient. The study is currently on going in several French centers.

Can global hemostasis assays help physicians to individually tailor prophylactic regimen of hemophilia patients with inhibitors?

The development of an inhibitor is one of the most serious complications associated with the treatment of hemophilia. In patients with high titer inhibitors (≥ 5 Bethesda Units (BU)/mL), bypassing agents represent effective treatment strategies for the treatment or prevention of hemorrhages []. The two widely used bypassing agents are FEIBA® (Baxter, Vienna, Austria), which is an activated prothrombin complex concentrate (APCC) and recombinant factor VIIa (rFVIIa, NovoSeven®, NovoNordisk, Copenhagen, Denmark). Usually, patients on rFVIIa prophylaxis receive daily injections, whereas patients on FEIBA prophylaxis are treated either every 48 h or three times a week. It has been shown that the established benefits of prophylaxis observed in patients with severe hemophilia A can be extended to patients with high-titer inhibitors through the use of prophylactic therapy with the bypassing agent [–]. Prophylaxis with daily FEIBA has been used by European hematologists as a component of the Bonn immune tolerance induction (ITI) regimen for more than 30 years and the analysis of a large series of 22 cases placed on prophylaxis with FEIBA showed that the treatment was effective in preventing bleeding and avoiding arthropathy during ITI []. FEIBA prophylaxis usually starts at a dose of 50 U/kg administered three times a week. Patients exhibiting a suboptimal response have their dose increased to 85 U/kg administered three times a week. If the suboptimal response persists, the dose can be increased to 85 U/kg/day.

A recent meta-analysis including six studies and 34 inhibitor patients showed a reduction of 63.9% in bleeding episodes during FEIBA prophylaxis []. No thrombotic or other complications were reported.

The efficacy of prophylactic rFVIIa outside the surgical setting has been demonstrated in patients with hemophilia and inhibitors [, ]. Despite the short half-life of rFVIIa in the circulation, NovoSeven® prophylaxis reduces the number of hemorrhagic events. It has been hypothesized that rFVIIa may diffuse extravascularly and be available at the site of injury where it can increase thrombin generation []. In a recent randomized, prospective, double-blind study, Konkle

[] evaluated the efficacy and safety of rFVIIa prophylaxis. In 22 patients with inhibitors, they showed that bleeding frequency was reduced by 45 and 59% during prophylaxis with 90 or 270 µg/kg/day, respectively, but no statistically significant difference was observed between the two prophylactic doses. In addition, although the incidence of all types of bleeds was reduced to a similar extent, the effect was most pronounced for spontaneous joint bleeds. No thromboembolic events were reported.

However, clinical response to bypassing therapies may be variable between patients and also between bleeding episodes. The optimal use of bypassing agents is hampered by a lack of laboratory assays to monitor therapeutic efficacy and determine adequate dosing. As the final product generated by these molecules is thrombin, the TGT could theoretically be used to monitor the efficacy of these agents. It has been previously reported that the TGT enabled the monitoring of both FEIBA and rFVIIa [–]. Recently, our group described the first direct application of the TGT in a surgical setting, showing that the assay might be useful in assessing the efficacy of FEIBA and guiding the choice of the most effective therapeutical option in patients with inhibitors []. The capability of determining the most effective therapeutical option and the optimal prophylactic dose of bypassing agent for a given patient would represent a major advance.

Recently, in a prospective clinical trial studying the efficacy and safety of FEIBA prophylaxis, thrombin generation measurement was used to monitor treatment response []. ETP exceeded 80% of normal after FEIBA infusion in the majority of patients. Between regular prophylactic infusions, mean ETP was 2.6-fold of the inhibitor plasma control mean. Despite prophylaxis, one of the patients experienced repeated joint bleeding and a variable response to FEIBA was observed with the TGT.

In our center, dose tailoring of bypassing agents using the TGT and a standardized three-step protocol is currently under investigation. This protocol includes (i) spiking experiments evaluating the hemostatic effect of increasing concentrations of rFVIIa (0, 90, 180, 200, 240, 270, 300 µg/kg) and FEIBA (50, 75, and 100 U/kg) in order to determine the minimal dose of each bypassing agent normalizing thrombin-generating capacity on the basis of ETP values; (ii) confirmation step where thrombin generation is measured before and after the administration of the bypassing agent that gave the best hemostatic efficacy in the previous spiking experiment using the dose that fully normalized ETP values; and (iii) monitoring of the chosen dose of the bypassing agent in clinical settings such as surgery or prophylaxis. Thrombin generation is measured using a low tissue factor concentration of 1 pM and the calibrated automated thrombin generation assay (CAT; Thrombinoscope BV, Maastricht, the Netherlands). illustrates the clinical use of the TGT for individualization of prophylactic regimen in a boy with severe hemophilia A and high titer FVIII inhibitors of 21 BU/mL, on prophylaxis with rFVIIa 270 µg/kg/day. Despite high doses of rFVIIa, the patient had persisting joint bleeds. The prophylactic regimen was modified according to the TGT results and the patient had no further bleeding (Dargaud Y. unpublished data).

CONCLUSION

The optimum prophylactic regimen is still under discussion. Different barriers to the use of prophylaxis require careful consideration of the goals of prophylaxis; should joint bleeds and thus hemophilic arthropathy be completely prevented or can a limited amount of arthropathy be tolerated without loss of quality of life? Several reports have clearly shown that prophylaxis should be started at a young age and potentially continued indefinitely. Should all patients receive the same prophylactic regimen or should prophylaxis be tailored to individual patient needs, if so, how are those needs determined? These questions remain unresolved. However, it is also clear that some patients with severe hemophilia experience very few bleeding episodes. Laboratory markers to predict the phenotype are warranted. Correlations between the TGT parameters and clinically observed bleeding in patients with inherited bleeding disorders have been published and recent technical developments make the assay potentially applicable to clinical laboratories. When more reliable predictors are available, considering a less active prophylactic regimen in some individuals will be more feasible. Optimizing prophylactic therapy to a patient's phenotype with no loss of clinical effectiveness can significantly improve patients’ quality of life, protect hemophilic children against arthropathy, and possibly limit the cost of prophylactic therapy.

Disclosure: The authors have no financial interests to disclose in relation to the contents of the article.

References

1. Mannucci PM, Tuddenham EGD. The hemophilias - from royal genes to gene therapy. N Engl J Med. 2001;344:1773-1779.
2. Rodriguez-Merchan EC. The destructive capabilities of the synovium in the haemophilic joint. Haemophilia. 1998;4:506-510.
3. Gilbert MS. Musculoskeletal complications of hemophilia: the joint. Haemophilia. 2000;6(Suppl. 1):34-37.
4. Bohn RL, Avorn J, Glynn RJ, Choodnovsky I, Haschemeyer R, Aledort LM. Prophylactic use of factor VIII: an economic evaluation. Thromb Haemost. 1998;79:932-937.
5. Manco-Johnson MJ, Abshire TC, Shapiro AD, et al. Prophylaxis versus episodic treatment to prevent joint disease in boys with severe hemophilia. N Engl J Med. 2007;357:535-544.
6. Panicker J, Warrier I, Thomas R, Lusher JM. The overall effectiveness of prophylaxis in severe haemophilia. Haemophilia. 2003;9:272-278.
7. Santagostino E, Mancuso ME. Prevention of arthropathy in haemophilia: prophylaxis. Haemophilia. 2008;14(Suppl. 6):16-19.
8. Shapiro AD. A global view on prophylaxis: possibilities and consequences. Haemophilia. 2003;9(Suppl. 1):10-18.
9. Hay CRM. Prophylaxis in adults with haemophilia. Haemophilia. 2007;13(Suppl. 2):10-15.
10. Gouw SC, van der Bom JG, van den Berg HM. Treatment related risk factors of inhibitor development in previously untreated patients with haemophilia A: the CANAL cohort study. Blood. 2007;109:4648-4654.
11. Nilsson IM, Berntorp E, Loftqvist T, Petterson H. Twenty five years experience of prophylactic treatment in severe haemophilia A and B. J Intern Med. 1992;232:25-32.
12. Berntorp E, Boulyjenkov V, Brettler D, et al. Modern treatment of haemophilia. Bull World Health Organ. 1995;73:691-701.
13. National Haemophilia Foundation. MASAC recommendation concerning prophylaxis (prophylactic administration of clotting factor concentrate to prevent bleeding). Medical and scientific advisory council (MASAC) Recommendations no. 117, 2001.
14. Butler RB, McClure W, Wulff K. Practice patterns in haemophilia A therapy - a survey of treatment centers in the United States. Haemophilia. 2003;9:549-554.
15. Shapiro AD. A global view on prophylaxis: possibilities and consequences. Haemophilia. 2003;9(Suppl. 1):10-18.
16. Ljung RCR. Aspects of haemophilia prophylaxis in Sweden. Haemophilia. 2002;8(Suppl. 2):34-37.
17. Petrini P. Identifying overcoming barriers to prophylaxis in the management of haemophilia. Haemophilia. 2007;13(Suppl. 2):16-22.
18. Löfqvist T, Nilsson IM, Berntorp E, Petterson H. Haemophilic prophylaxis in young patients - a long term follow up. J Intern Med. 1997;241:395-400.
19. Ljung R. Prophylactic therapy in hemophilia. Blood Rev. 2009;23:267-274.
20. Astermark J. When to start and when to stop primary prophylaxis in patients with severe haemophilia. Haemophilia. 2003;9(Suppl. 1):32-37.
21. Carlsson KS, Hojgard S, Lindgren A, et al. Costs of on demand and prophylactic treatment for severe haemophilia in Norway and Sweden. Haemophilia. 2004;10:515-526.
22. Khawaji M, Akesson K, Berntorp E. Long-term prophylaxis in severe haemophilia seems to preserve bone mineral density. Haemophilia. 2009;15:261-266.
23. Skinner MW. Treatment for all: a vision for the future. Haemophilia. 2006;12(Suppl. 3):169-173.
24. Van Dijk K, Van Der Bom JG, Bax KN, Van Der Zee DC, Van Den Berg MH. Use of implantable venous access devices in children with severe haemophilia: benefits and burden. Haematologica. 2004;89:189-194.
25. Lindvall K, Colstrup L, Wolter IM, et al. Compliance with treatment and understanding of own disease in patients with severe and moderate haemophilia. Haemophilia. 2006;12:47-51.
26. Van Creveld S. Prophylaxis of joint haemorrhages in haemophilia. Acta Haemat. 1971;45:120-127.
27. Fischer K, Van Den Berg HM, Mauser-Bunschoten EP, et al. Changes in treatment strategies for severe haemophilia over the last 3 decades: effects on clotting factor consumption and arthropathy. Haemophilia. 2001;7:446-452.
28. Van Den Berg MH, Fischer K, Mauser-Bunschoten EP, et al. Long-term outcome of individualized prophylactic treatment of children with severe haemophilia. Br J Haematol. 2001;107:561-565.
29. Fischer K, Van Den Berg M. Prophylaxis for severe haemophilia: clinical and economical issues. Haemophilia. 2003;9:376-381.
30. Van Dijk K, Fischer K, ven der Bom JG, Grobbee DE, Van Den Berg HM. Variability in clinical phenotype of severe haemophilia: the role of the first joint bleed. Haemophilia. 2005;11:438-443.
31. Fischer K. Can we consider discontinuing prophylaxis in adults with severe haemophilia? Haemophilia. 2008;14(Suppl. 4):10.
32. Van Dijk K, Fischer K, van der Bom JG, Scheibel E, Ingerslev J, Van Den Berg MH. Can long prophylaxis for severe haemophilia be stopped in adulthood? Results from Denmark and The Netherlands. Br J Haematol. 2005;130:107-112.
33. Fischer K, Valentino L, Ljung R, Blanchette V. Prophylaxis for severe haemophilia: clinical challenges in the absence as well as in the presence of inhibitors. Haemophilia. 2008;14(Suppl. 3):196-201.
34. Blanchette VS, Manco-Johnson M, Sanatagostino E, Ljung R. Optimization factor prophylaxis for the haemophilia population: where do we stand? Haemophilia. 2004;10(Suppl. 4):97-104.
35. Blanchette VS, Rivard G, Israels S, et al. A survey of factor prophylaxis in the Canadian haemophilia A population. Haemophilia. 2004;10:679-683.
36. Blanchette VS, McCready M, Achonu C, et al. A survey of factor prophylaxis in boys with haemophilia followed in North American haemophilia treatment centres. Haemophilia. 2003;9(Suppl. 1):19-26.
37. Biss TT, Chan AK, Blanchette VS, et al. The use of prophylaxis in 2663 children and adults with haemophilia: results of the 2006 Canadian national haemophilia prophylaxis survey. Haemophilia. 2008;14:923-930.
38. Ota S, Mclimont M, Carcao MD, et al. Definitions for haemophilia prophylaxis and its outcomes: the Canadian consensus study. Haemophilia. 2007;13:12-20.
39. Carcao M, Chambost H, Ljung R. Devising a best practice approach to prophylaxis in boys with severe haemophilia: evaluation of current treatment strategies. Haemophilia. 2010;16(Suppl. 2):4-9.
40. Meunier S, Trossaert M, Berger C, et al. French guidelines. Long-term prophylaxis for severe haemophilia A and B children to prevent haemophiliac arthropathy. Arch Pediatr. 2009;16:1571-1578.
41. Carlsson M, Berntorp E, Björkman S, et al. Pharmacokinetic dosing in prophylactic treatment of hemophilia A. Eur J Haematol. 1993;51:247-252.
42. Carlsson M, Berntorp E, Björkman S, et al. Improved cost effectiveness by pharmacokinetic dosing of factor VIII in prophylactic treatment of haemophilia A. Haemophilia. 1997;3:96-101.
43. Carlsson M, Björkman S, Berntorp E. Multidose pharmacokinetics of factor IX: implications for dosing in prophylaxis. Haemophilia. 1998;4:83-88.
44. Kisker CT, Eisberg A, Schwartz B, et al. Prophylaxis in factor IX deficiency product and patient variation. Haemophilia. 2003;9:279-284.
45. Ahnstrom J, Berntorp E, Lindvall K, et al. A 6-year follow-up of dosing, coagulation factor levels and bleedings in relation to joint status in the prophylactic treatment of haemophilia. Haemophilia. 2004;10:689-697.
46. Aledort LM, Haschmeyer RH, Pettersson H; the Orthopaedic Outcome Study Group. A longitudinal study of orthopaedic outcomes for severe factor-VIII-deficient haemophiliacs. J Intern Med. 1994;236:391-399.
47. Lee DH, Walker IR, Teitel J, et al. Effect of factor V Leiden mutation on the clinical expression of severe haemophilia A. Thromb Haemost. 2000;83:387-391.
48. Escuriola Ettinghausen C, Halimeh S, Kurnik K, et al. Symptomatic onset of severe haemophilia A in childhood is dependent on the presence of prothrombotic risk factors. Thromb Haemost. 2001;85:218-220.
49. Butenas S, van't Veer C, Mann KG. Normal thrombin generation. Blood. 1999;94:2169-2178.
50. Hemker HC, Giesen P, Al Dieri R, et al. The calibrated automated thrombogram (CAT): a universal routine test for hyper- and hypocoagulability. Pathophysiol Haemost Thromb. 2002;32:249-253.
51. Hemker HC, Giesen P, AlDieri R, et al. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb. 2003;33:4-15.
52. Chantarangkul V, Clerici M, Bressi C, Giesen PL, Tripodi A. Thrombin generation assessed as endogenous thrombin potential in patients with hyper- or hypo-coagulability. Haematologica. 2003;88:547-554.
53. Siegemund T, Petros S, Siegemund A, et al. Thrombin generation in severe haemophilia A and B: the endogenous thrombin potential in platelet-rich plasma. Thromb Haemost. 2003;90:781-786.
54. Beltran-Miranda CP, Khan A, Jaloma-Cruz AR, Laffan MA. Thrombin generation and phenotypic correlation in haemophilia A. Haemophilia. 2005;11:326-334.
55. Dargaud Y, Beguin S, Lienhart A, et al. Evaluation of thrombin generating capacity in plasma from patients with haemophilia A and B. Thromb Haemost. 2005;93:475-480.
56. Trossaert M, Regnault V, Sigaud M, Boisseau P, Lecompte T, Fressinaud E. Mild hemophilia A with factor VIII assay discrepancy: potential utility of thrombin generation assays. J Thromb Haemost. 2008;6:486-493.
57. Santagostino E, Mancuso ME, Tripodi A, et al. Severe haemophilia with mild bleeding phenotype: molecular characterization and global coagulation profile. J Thromb Haemost. 2010; in press.
58. Bruemmel Ziedins KE, Whelihan MF, Gissel M, Mann KG, Rivard G. Thrombin generation and bleeding in haemophilia. Haemophilia. 2009;15:1118-1125.
59. Gomperts ED, Astermark J, Gringeri A, Teitel J. From theory to practice: applying current clinical knowledge and treatment strategies to the care of hemophilia A patients with inhibitors. Blood Rev. 2008;22(Suppl 1):S1-S11.
60. Salvagno GL, Astermark J, Lippi G, et al. Thrombin generation assay: a useful routine check-up tool in the management of patients with haemophilia. Haemophilia. 2009;15:290-296.
61. Lewis SJ, Stefens E, Florou G, et al. Measurement of global haemostasis in severe haemophilia A following FVIII infusion. Br J Haematol. 2007;138:775-782.
62. Astermark J. Treatment of the bleeding in inhibitor patient. Semin Thromb Haemost. 2003;29(Suppl. 1):77-89.
63. Leissinger CA, Becton DL, Ewing NP, et al. Prophylactic treatment with activated prothrombin complex concentrate (FEIBA) reduces the frequency of bleeding episodes in paediatric patients with haemophilia A and inhibitors. Haemophilia. 2007;13:249-255.
64. Valentino LA, Carcao M, Mathew P, et al. The application of bypassing agent prophylaxis in haemophilia A patients with inhibitors: a meeting report. Haemophilia. 2009;15:959-965.
65. Escuriola Ettinghausen C, Kreuz W. Early long-term FEIBA prophylaxis in haemophilia A patients with inhibitor after failing immune tolerance induction: a prospective clinical case series. Haemophilia. 2010;16:90-100.
66. Perry D, Berntorp E, Tait C, et al. FEIBA prophylaxis in haemophilia patients: a clinical update and treatment recommendations. Haemophilia. 2010;16:80-89.
67. Valentino LA. Assessing the benefits of FEIBA prophylaxis in haemophilia patients with inhibitors. Haemophilia. 2010;16:263-271.
68. Saxon BR, Shanks D, Jory CB, et al. Effective prophylaxis with daily recombinant factor VIIa in a child with high titre inhibitors and a target joint. Thromb Haemost. 2001;86:1126-1127.
69. Young G, McDaniel M, Nugent DJ. Prophylactic recombinant factor VIIa in haemophilia patients with inhibitors. Haemophilia. 2005;11:203-207.
70. Kreuz W, Escuriola Ettinghausen C, Mentzer D, et al. Factor VIII inhibitor bypass activity (FEIBA) for prophylaxis during immune tolerance induction in patients with high responding inhibitors. Blood. 2000;96:266a.
71. Hedner U. Potential role of recombinant factor VIIa in prophylaxis in severe hemophilia patients with inhibitors. J Thromb Haemost. 2006;4:2498-2500.
72. Konkle BA, Ebbesen LS, Erhardtsen E, et al. Randomized, prospective clinical trial of recombinant factor VIIa for secondary prophylaxis in haemophilia patients with inhibitors. J Thromb Haemost. 2007;5:1904-1913.
73. Varadi K, Negrier C, Berntorp E, et al. Monitoring the bioavailability of FEIBA with a thrombin generation assay. J Thromb Haemost. 2003;1:2374-2380.
74. Varadi K, Turecek PL, Schwarz HP. Thrombin generation assay and other universal tests for monitoring haemophilia therapy. Haemophilia. 2004;10(Suppl. 2):17-21.
75. Dargaud Y, Bordet JC, Lienhart A, Negrier C. Use of the thrombin generation test to evaluate response to treatment with recombinant activated factor VII. Semin Hematol. 2008;2(Suppl. 1):S72-S73.
76. Dargaud Y, Lambert T, Trossaert M. New advances in the therapeutic and laboratory management of patients with haemophilia and inhibitors. Haemophilia. 2008;14(Suppl. 4):20-27.
77. Dargaud Y, Lienhart A, Meunier S, et al. Major surgery in a severe haemophilia A patient with high titre inhibitor: use of the thrombin generation test in the therapeutic decision. Haemophilia. 2005;11:552-558.

Retrospective Evaluation of Secondary Episodic Prophylaxis with rFVIIa in Hemophilia Patients with Inhibitor

Mon, 28 Jun 2010 07:59:05 +0100

INTRODUCTION

Hemophilic arthropathy, which occurs as a consequence of repeated bleeding into the joints, is a significant cause of morbidity in hemophilia patients. The prevention of this condition is currently one of the most important and immediate objectives of treatment in these patients. Timely administration of on-demand therapy at adequate doses in response to a bleeding episode can delay the progression of joint disease but cannot avert it if several episodes have already occurred in the same joint [–].

It is well established that the best way to preserve joint integrity is to prevent the first and successive bleeds, and primary prophylaxis is the treatment of choice in young patients []. In patients with a certain degree of joint compromise, secondary prophylaxis can significantly delay the development of arthropathy but not reverse it [, ]. One approach in the application of secondary prophylaxis is the so-called episodic prophylaxis [], given in specific periods—generally after on-demand treatment, during periods of rehabilitation, or during an interval when there is a particular increase in the incidence of bleeding episodes.

When the hemophilia patient presents an inhibitor, particularly a high-responding inhibitor, management of joint bleeds and sequelae of advanced arthropathy becomes especially difficult []. Current therapy for this purpose, consisting of bypassing agents, is not fully effective, and the experience with their use is less extensive. Due to the short plasma half-life of these drugs and their high cost, short treatment regimens (1–3 doses) have been used. Prolonged preventive treatment that would avert new bleeds and rebleeding in the inflamed synovium and prevent progression of injury in the target joint is generally avoided.

The favorable results obtained with prophylactic treatment in hemophilia patients without inhibitors are an example to follow in patients with inhibitors. There is some consensus that these treatment models are also applicable to patients with inhibitors []. Until now, several reports—mainly isolated cases—and one prospective study have presented the outcome of prophylaxis experiences using bypassing agents, with promising results [–].

The objective of this study, carried out by the Working Group for the Treatment of Haemophilia Patients with Inhibitor of the Spanish Society of Thrombosis and Haemostasis (SETH, Sociedad Española de Trombosis y Hemostasia) in several hospitals in Spain, was to assess the efficacy and safety of rFVIIa as secondary episodic prophylaxis in severe hemophilia patients with high-responding inhibitors and arthropathy despite on-demand treatment.

PATIENTS AND METHODS

A retrospective, non-comparative study that initially included 11 patients with severe hemophilia A/B and high-responding inhibitors, who had received some type of secondary prophylaxis with rFVIIa, was conducted in Spain. Nine patients (eight with hemophilia A and one with hemophilia B) treated in nine hospitals were included in the study. Two cases were not included because the data did not arrive in time or the follow-up period was limited to 1.5 months.

The mean age of the study patients was 10.2 years (range, 2–21 years). Inhibitors had developed in all patients following 10–50 exposures of Factor VIII/IX concentrate. Their historical inhibitor titer was 18 to 6084 Bethesda units (BU/mL), and the titer at the time of prophylaxis with rFVIIa was 0 to 210 BU/mL (). All patients had experienced repeated hemarthroses in the target joints during the 3 months prior to initiation of prophylaxis, with a mean of 5.5 episodes (range, 3–9), which impeded the development of a normal, active lifestyle. In addition, all patients had experienced more than one bleeding episode in the same joint in the 3 months prior to the start of prophylaxis.

The rFVIIa prophylaxis dosing schedule was as follows: the dose used was 90 µg/kg or more in seven patients and 60 µg/kg in two, administered as a single daily dose (six patients), every two days (one patient) or three times a week (two patients).

Efficacy of the rFVIIa secondary prophylaxis regimen was assessed by the number of patients who had no hemarthroses during the 3 months after the start of prophylaxis; and secondarily, by evaluating the difference between the number of hemarthroses that occurred in the 3 months prior to and the 3 months after the start of prophylaxis for each patient. Safety was considered in terms of tolerance to rFVIIa administration and the development of adverse events, such as clinical or laboratory signs of thrombosis.

The influence of age, hemophilia A/B, inhibitor titer and dosing schedule on the efficacy of secondary prophylaxis was also investigated to detect sources of variability in the patients’ responses to rFVIIa.

RESULTS

Among the nine patients included in the follow-up (), prophylaxis lasted 3 months or more (range, 3–19 months) in seven patients. Two patients (Patients 8 and 9) quit the prophylaxis regimen, which was considered ineffective, since the number of hemarthroses was similar or even higher during prophylaxis than during the control period.

In the evaluation of the efficacy of secondary rFVIIa prophylaxis, four patients (44%) presented no hemarthroses in the first 3 months of the prophylaxis program versus a mean number of 4.8 hemarthroses (range, 3–6) that occurred in the 3 months prior to starting prophylaxis. One patient presented one joint bleed during the prophylaxis period versus eight during the observation period, and was included in the success group, thereby increasing the success rate to 56% (5/9). Another patient presented five joint bleeds before and three during prophylaxis and was considered a partial response, yielding a combined partial and total success rate of 67% (6/9).

The remaining three patients (33%) did not experience any improvement during the time on prophylaxis: the mean number of prior hemarthroses was 5.3 (range, 3–9) and the mean during prophylaxis was 6.0 (range, 4–10). In addition, these patients required on-demand treatment when bleeding occurred. Because of this lack of response, administration of rFVIIa prophylaxis was discontinued in two patients at 4 and 8 weeks, respectively.

None of the nine patients included in the study showed signs of adverse events related to the administration of rFVIIa, or lack of tolerance to the prophylaxis schedule.

In an attempt to find an explanation for the differing responses, the patients were divided into two groups, one including the five patients with a favorable response and the other, three with no response, to assess the influence of other variables. This evaluation excluded the patient with a partial response. Mean age was similar in both groups, 10.6 years (range, 2–21 years) and 11.0 years (range, 4–20 years), respectively. The highest historic inhibitor titer was in a patient with a favorable response (6084 BU/mL) and the lowest in a non-responder (18 BU/mL), indicating the lack of influence of the inhibitor titer on bleeding tendency. At the time of the study, the inhibitor titer was less than 10 BU/mL in three of five patients with a favourable response and in one of three patients with no response. Both the maximum (210) and minimum (0) inhibitor titers were in patients with no response.

The rFVIIa dosing schedule used in the two groups was similar: the dose was 90 µg/kg or more in all cases, except in two patients in whom the dose was 60 µg/kg. The dosing frequency in the favorable response group was once daily in three of five patients, thrice weekly in 1one of five, and once every 2 days in one of five, whereas in non-responders, administration was daily in two of three patients and thrice weekly in one of three. Mean duration of prophylaxis was 4.4 months (range, 3–8 months) and 2.0 months (range, 1–3 months), respectively. The duration was shorter in the non-responders because treatment was discontinued in two patients due to lack of response. The patient with partial response received 60 µg/kg for 19 months.

DISCUSSION

The benefit of secondary prophylaxis in hemophilia patients with high-responding inhibitors and a high rate of joint bleeds, and in those with arthropathy, has been less extensively investigated than in hemophilia patients without inhibitors. Thus, it is important to determine whether the favorable results obtained in this latter population are also applicable to patients with inhibitors. The use of bypassing agents is emerging as an interesting option, although it is known that their hemostatic effect is not always predictable, and a standardized treatment regimen has not yet been established.

The results of this retrospective study show that the use of rFVIIa for secondary prophylaxis resulted in clear benefits for five of nine (56%) patients evaluated. In these patients, the overall reduction in the total number of joint bleeds was 96.3% (27 hemarthroses during the on-demand period versus one episode during prophylaxis). Three patients (33%) had a similar number of episodes of hemarthroses, 16 versus 18, with no benefit. The remaining case presented a partial response, five versus three. The overall reduction in the total number of joint bleeds in the series was 54% (48 episodes versus 22). In addition, it should be considered that all had varying degrees of arthopathy, since all patients had experienced repeated bleeding episodes prior to prophylaxis.

There were no significant differences between the groups that would explain the differing response to prophylaxis; neither age nor historical or current inhibitor titer differed significantly. Nor could the differences in response be attributed to variations in the dosing regimen. A daily dosing schedule was used in four of five responders and two of three non-responders. In a prospective, double blind study, Konkle [] found no difference in the efficacy of prophylaxis when a daily dose of 90 µg/kg or 270 µg/kg was given. The different phenotypes (genetic factors, thrombophilic markers, blood group, and others) can act as modifiers of the patients’ bleeding pattern, as seen in patients without inhibitors [].

This study has the limitations inherent to a retrospective design with a small number of patients and substantial variations in the parameters involved (inclusion criteria, dose, dosing interval, duration, level of arthropathy); nonetheless, the difference in the clinical response obtained in these patients is noteworthy. In Konkle's study [], which was the first prospective randomized trial on the use of bypassing agents for secondary prophylaxis with clear and concise inclusion criteria, the number of patients with a lack of response was lower, 13.6% (3/22), than in the series reported here.

There is growing interest in the use of bypassing agents for preventive purposes. In the case of rFVIIa, the theory that this factor is redistributed at the subendothelial level and in platelets may explain the prolonged effect of its administration, which exceeds the or 3 hours of mean plasma half-life [, ]. To date, there are few reported experiences with the use of rFVIIa for prophylaxis. A study performed in Canada reported that approximately 13% of patients with inhibitors received prophylaxis with a bypassing agent in 2006 []. The study by Morfini [] demonstrated the efficacy of rFVIIa in different situations, although the cases included were quite heterogeneous. In the studies of Jiménez [, ], the effectiveness of prophylactic treatment with rFVIIa was documented even in patients who had not begun to experience repeated bleeding episodes (primary and secondary prophylaxis).

With regard to secondary prophylaxis in patients with inhibitors, the results of various studies seem to indicate that the joint injury incurred cannot be reversed once it has started []. Nonetheless, in the randomized trial by Konkle [], a clear reduction in bleeding episodes was documented during the prophylaxis period, and the effect was maintained in the 3 months following prophylaxis. That trial also found that 13.6% of patients had a lack of response. Hoots [] found a clear improvement in the quality of life of these patients which, in the author's opinion, justifies the use of a secondary prophylaxis approach. In an article based on an intention to treat survey, Mannucci [] theorized on the potential beneficial effects of secondary prophylaxis with bypassing agents in patients who present a high rate of joint bleeds.

In summary, the results of this study support the concept that secondary prophylaxis with rFVIIa is efficacious and safe in hemophilia patients with high-responding inhibitors and a high rate of joint bleeds, and in those in whom arthropathy has begun. These findings, together with the results of other studies, suggest that the use of bypassing agents as prophylaxis is an interesting emerging option for patients with high-responding inhibitors. Nevertheless, the limited number of patients involved and the variable pattern of response observed warrants further investigation to ensure efficacy and enable selection of patients who will benefit from prophylactic treatment. Prospective studies in which the pertinent variables are standardized (ie, dose, dosing interval, degree of arthropathy) are required to provide an accurate answer to the doubts regarding the true efficacy of prophylactic treatment in hemophilia patients with inhibitors.

ADDENDUM

This original was a multicentric collaborative study by the Working Group for the Treatment of Haemophilia Patients with Inhibitor from Sociedad Española de Trombosis y Hemostasia (SETH, Spanish Society of Thrombosis and Hemostasis). The study design was developed by Carmen Sedano, Carmen Altisent, Maria Eva Mingot, Ramiro Nuñez, María José Paloma, Inmaculada Soto, Ana Rosa Cid, Victor Jimenez, María Fernanda Fernández-Lopez, and Manuel Prieto. The statistical analysis and initial draft of the manuscript were developed by Carmen Sedano. The principal investigators were Carmen Sedano, Carmen Altisent, Ana Rosa Cid, and María Fernanda Fernández-Lopez.

References

1. Berntorp E. Methods of haemophilia care delivery: regular prophylaxis versus episodic treatment. Haemophilia. 1995:1(Suppl. 1): 3-7.
2. Molho P, Rolland N, Lebrun T, et al. Epidemiological survey of the orthopaedic status of severe haemophilia A and B patients in France. Haemophilia. 2000;6:23-32.
3. Aznar JA, Magallón M, Querol F, Gorina E, Tusell JM. The orthopaedic status of severe haemophiliacs in Spain. Haemophilia. 2006;6:170-176.
4. Manco-Johnson MJ, Abshire TC, Shapiro AD, et al. Prophylaxis versus episodic treatment to prevent joint disease in boys with severe haemophilia. N Engl J Med. 2007;357:535-544.
5. Manco-Johnson MJ, Nuss R, Geraghty S, Funk S, Kilcoyne R. Results of secondary prophylaxis in children with severe hemophilia. Am J Hematol. 1994;47:113-117.
6. Valentino LA. Secondary prophylaxis therapy: what are the benefits, limitations and unknowns? Haemophilia. 2004;10:147-157.
7. Luhtman-Jones L, Valentino LA, Manno C. Considerations in the evaluation of haemophilia patients for short-term prophylactic therapy: a paediatric and adult case study. Haemophilia. 2006;12:82-86.
8. Morfini M, Haya S, Tagariello G, et al. European study on orthopaedic status of haemophilia patients with inhibitors. Haemophilia. 2007;13:606-612.
9. Rodriguez-Merchan EC, Hedner U, Heijnen L, et al. Prevention of haemophilic arthropathy during childhood. May common orthopaedic management be extrapolated from patients without inhibitors to patients with inhibitors? Haemophilia. 2008;14(Suppl.6):68-81.
10. Leissinger CA, Becton DL, Ewing NP, Valentino LA. Prophylactic treatment with activated prothrombin complex concentrate (FEIBA®) reduces the frequency of bleeding episodes in paediatric patients with haemophilia A and inhibitors. Haemophilia. 2007;13:249-255.
11. Saxon BR, Shanks D, Jory CB, Williams V. Effective prophylaxis with daily recombinant factor VIIa (rVIIa-Novoseven) in a child with high titre inhibitors and target joint. Thromb Haemost. 2001;85:1126-1127.
12. Young G, Mcdaniel M, Nugent DJ. Prophylactic recombinant factor VIIa in haemophilia patients with inhibitors. Haemophilia. 2005;11:203-207.
13. Morfini M, Auerswald G, Kobelt RA, et al. Prophylactic treatment of haemophilia patients with inhibitors: clinical experience with recombinant factor VIIa in European Haemophilia Centres. Haemophilia. 2007;13:502-507.
14. Konkle BA, Ebbesen LS, Erhardtsen E, et al. Randomized, prospective clinical trial of recombinant factor VIIa for secondary prophylaxis in hemophilia patients with inhibitors. J Thromb Haemost. 2007;5:1904-1913.
15. Jimenez-Yuste V, Alvarez MT, Martin-Salces M, et al. Prophylaxis en 10 patients with severe haemophilia and inhibitor: different approaches for different clinical situations. Haemophilia. 2009;15:203-209.
16. Jimenez-Yuste V, Quintana M, Alvarez MT, Martin-Salces M, Hernandez-Navarro F. "Primary prophylaxis" with rVIIa in a patient with severe haemophilia and inhibitor. Blood Coagul Fibrinol. 2008;19:719-720.
17. Biss TT, Chan AK, Blanchette VS, et al. The use of prophylaxis in 2663 children and adults with haemophilia: results of the 2006 Canadian national haemophilia prophylaxis survey. Haemophilia. 2008;14:923-930.
18. Lopez-Fernandez M, Andon-Saavedra C, Amor-Otero MA, Batlle J. Primary prophylaxis treatment with recombinant activated factor VII (rVIIa) during immune tolerance in a haemophilic child (abstract). Abstract no.P-M-161, XXIst ISTH Congress, Geneva, July 6-11, 2007.
19. Lee CA. Prevention of haemophilic synovitis: prophylasis. Haemoplilia. 2007;13(Suppl. 3):20-25.
20. Hedner U. Potential role of recombinant factor VIIa in prophylaxis in severe hemophilia patients with inhibitors. J Thromb Haemost. 2006;4:2489-2500.
21. Galan A, Lopez-Vilchez I, Gracia J, Gines E. Redistribution of recombinant activated factor VIIa into platelets and vascular bed: implications on bioavailability and hemostatic mechanisms. Blood. (ASH Annual Meeting Abstracts) 2007;110:Abstract 3957.
22. Hilgartner MW, Makipernna A, Dimichele DM. Long-term FEIBA prophylaxis does not prevent progression of existing joint disease. Haemophilia. 2003;9:261-268.
23. Hoots WK, Ebbesen LS, Konkle BA,et al; Novoseven (F7HAEM-1505) Investigators. Secondary prophylaxis with recombinant activated factor VII improves health-related quality of life of haemophilia patients with inhibitors. Haemophilia. 2008;14:466-475.
24. Mannucci PM, Palhares de Miranda PA. International survey of attitudes toward secondary prophylaxis with recombinant factor VIIa in haemophilia A patients with inhibitors. Haemophilia. 2009;15:345-347.

Effectiveness of Bosentan in the Treatment of Chronic Thromboembolic Pulmonary Hypertension: A Systematic Review of Randomized and Nonrandomized Trials

Thu, 24 Jun 2010 08:56:10 +0100

INTRODUCTION

Chronic thromboembolic pulmonary hypertension (CTEPH) is a serious complication of pulmonary embolism. One large, long-term follow-up study found that 3.8% of patients with pulmonary embolism developed pulmonary hypertension within two years []. The pathogenesis of CTEPH remains unclear, but it is considered to involve failure to completely resolve pulmonary vascular obstruction after thromboembolism leading to major vessel vascular remodeling processes. In pulmonary arterial hypertension (PAH), overexpression of endothelin-1 (ET-1) is considered an important mechanism that leads to smooth muscle vasoconstriction, hypertrophy, fibrosis, and inflammation, which has also been demonstrated in CTEPH patients [].

CTEPH is associated with substantial morbidity and mortality, as it can lead to right ventricular failure and, consequently, death. It has a poor prognosis when left untreated, with more than half of patients diagnosed with CTEPH dying within one year when having a mean pulmonary arterial pressure higher than 50mmHg []. The treatment of choice is pulmonary endarterectomy (PEA). The surgical mortality is under 10% in expert centers, and the outcome is favorable, with pulmonary hemodynamics restored to normal in almost 80% of operated patients. Pulmonary hemodynamics and pulmonary vascular resistance are thought to be related to prognosis, as lower values are associated with increased survival. However, surgical removal of pulmonary obstructions is not possible in about 50% of patients, either because the clot is too distal in the lungs to be accessible for PEA or because patients have comorbidities that make them unsuitable for surgery. In addition, about 10% to 15% of patients who undergo PEA fail to respond and continue have recurrent or persistent pulmonary hypertension []. Thus, there is need for effective nonsurgical therapeutic interventions.

Standard therapy currently attempts to cure only the symptoms and usually consists of oral anticoagulation, diuretics, and supplemental oxygen in case of hypoxemia, but a clear policy on medication addressing the cause is lacking. There are several oral agents registered for the treatment of PAH, but for CTEPH, which medication should preferably be administered is not yet generally agreed on. The available oral agents include phosphodiesterase 5 inhibitors and ET-1 receptor antagonists. ET-1 binds to ET-A and ET-B receptors, and bosentan represents a competitive and specific antagonist to both ET-1 receptor types []. Bosentan was shown to be effective in the treatment of PAH, but its value for treating CTEPH has not yet been firmly established.

To investigate whether bosentan lowers pulmonary vascular resistance and is associated with a favorable outcome, randomized and nonrandomized trials that assessed the effectiveness of bosentan in comparison with placebo or alternative therapies in the treatment of adults diagnosed with CTEPH were systematically reviewed.

METHODS

Protocol

Review methods were not documented in a protocol in advance. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement was used for reporting [, ].

Eligibility criteria for considering studies

Types of studies

Randomized and nonrandomized controlled trials studying the effectiveness of bosentan in patients with CTEPH who were not eligible for PEA or had undergone PEA unsuccessfully were included. No publication date or publication status restrictions were imposed. Case studies were excluded.

Types of participants

Participants older than 18 years of age with CTEPH or resistant PAH after PEA were considered. Diagnosis of CTEPH had to be established by either pulmonary angiography or ventilation/perfusion lung scan showing typical abnormalities. In patients considered unsuitable for PEA, this had to be established by an experienced surgeon, and patients deemed fit to undergo PEA were excluded from this review.

Types of intervention

Included trials compared the efficacy and safety of bosentan administered twice daily in a dose of 125 mg for at least 3 months to treatment consisting of standard care alone or in combination with a placebo.

Types of outcome measures

Primary outcome measure was defined as change in pulmonary vascular resistance (PVR) and in exercise capacity as assessed by the 6-minute walking distance (6MWD) test. Secondary outcome measures were change in other hemodynamic parameters, such as mean pulmonary arterial pressure (mPAP), as well as improvement in quality of life and time to clinical worsening.

Search methods for identification of studies

Studies were identified by searching electronic databases and scanning reference lists of relevant studies. A systematic literature search was applied to PubMed, MEDLINE (1950–present), Embase (1989–present), and Cochrane databases.

The following search terms were used to search all databases: bosentan, chronic thromboembolic pulmonary hypertension. An example of a full electronic search strategy can be found in . The last search was run on February 8, 2010.

Of the articles identified by the search strategy, titles and abstracts were screened for eligibility by the first author under supervision of the last author. If no abstract was available or if it was unclear, the full text was obtained to make a decision. The included studies were evaluated by the first author using a checklist of 27 criteria assessing the study reporting and methodological quality of randomized and nonrandomized studies, which was developed and validated by Downs and Black [].

Data collection and analysis

From each included trial, the following information was extracted for data analysis by the first author:

The obtained information was evaluated qualitatively. No statistical method of analysis was done.

Risk of bias

In individual studies

The aforementioned checklist developed by Downs and Black for assessing the methodological quality of trials includes items that address bias related to the selection of study subjects and to the measurement of intervention and outcome. The included studies were individually evaluated for risk of bias based on the checklist by Downs and Black in an unblinded manner.

Across studies

By searching databases such as PubMed, only studies accepted for publication can be found. Therefore, it is not possible to exclude the possibility of publication bias that favors studies in which a significant positive effect is described, thereby overestimating the true effect of bosentan treatment.

Furthermore, not reporting outcomes that were not found to differ significantly between treatment groups or omitting data can lead to selective reporting bias. This is difficult to assess, as the protocols for trials were not available. However, for the included studies, outcome measures mentioned in the methods section were compared to those in the results section to check for any inconsistencies. No formal assessment of risk of bias across studies was performed.

RESULTS

Study selection

The search of the databases identified a total of 117 publications, of which 63 remained after adjusting for duplicates. After preliminary screening of title and abstract, 54 publications were excluded either because they covered a related, but distinct topic or because they concerned a publication other than a clinical trial, such as case studies or reviews. Of the remaining nine publications, one was in Spanish []. As the full text could not feasibly be translated into English, eligibility for inclusion was decided based on the abstract alone. The study did not meet the predetermined criteria, as it concerned a small, retrospective study without a control group. For the other eight publications, the full text was obtained and studied in detail in order to determine eligibility. Six studies did not meet the predetermined criteria, as they did not include any form of control treatment, such as placebo or alternative medication [–]. For details of the characteristics of excluded studies, see . The other two studies met the eligibility criteria and were included in the systematic review. No further studies were identified by checking the references of relevant studies, and no unpublished relevant studies were obtained (see ).

Study characteristics

Characteristics of the trials

The two studies finally selected for the review were one randomized clinical trial and one nonrandomized clinical trial. Both studies were published in English.

The RCT was a placebo-controlled, double-blind multicenter study with a total of 157 patients (77 receiving bosentan, 80 placebo) who were followed for 16 weeks []. Patients with WHO functional class II, III, or IV and a diagnosis of CTEPH determined by ventilation/perfusion lung scanning and pulmonary angiography were included. Furthermore, inoperability of the patients was established by an experienced surgeon and independently reviewed before unblinding by a committee of surgeons and pulmonologists. The main exclusion criteria were forms of pulmonary hypertension other than CTEPH and concomitant obstructive or restrictive lung disease.

The nonrandomized trial was a prospective, open-label multicenter study with a total of 34 participants, 17 each in the bosentan and control groups []. The length of follow-up was 12 months. Consecutive patients with CTEPH confirmed by echocardiography, ventilation/perfusion scanning, and/or computed tomography were enrolled. Pulmonary angiography was not performed as a standard. Inoperability of patients was determined by a panel of experts in case of distal thrombi, relevant comorbidity, or patient refusal. Exclusion criteria included any other lung diseases and liver laboratory abnormalities. Patients who previously underwent PEA were not included. For details, see .

Downs and Black's checklist has a maximum score of 32 points, of which the RCT scored 25 and the nonrandomized trial 20 (). Study reporting was generally well done in both studies. The evaluation of methodological quality showed that the RCT scored low on external validity, as the settings of the centers that participated in this multicenter study remained unclear, making it difficult to estimate whether participants in the trial were representative of the entire population from which they were drawn. The nonrandomized trial scored poorly for power, as no beforehand analysis was done to determine sample size. The scores for internal validity are discussed in Section “Risk of bias in included studies”.

Characteristics of trial participants

Participants of the RCT had a mean age of 63 years, and the bosentan and placebo groups consisted of 71.4% and 58.8% females, respectively. The study group consisted mainly of patients in WHO functional class III and a few in class IV. PVR, 6MWD, and mPAP were comparable in both groups ().

Participants of the nonrandomized trial were of similar age, with an average of 64.9 and 62.8 years in the bosentan and control groups, respectively. Again, most patients were in WHO functional class III, with a few in classes II and IV. The bosentan group contained more women (15 of 17) compared with the control group (8 of 17). PVR values are not mentioned in the article, and when the author was contacted, he confirmed that it was not assessed in most patients and therefore could not be compared between groups, so no data on this parameter is available. Concerning the PAP, only systolic values could be extracted from the article. Patients in the bosentan group had a higher 6MWD at baseline than patients in the control group, but this difference was not statistically significant. For details, see .

Intervention

In both trials, patients in the treatment group received a starting dose of 62.5 mg bosentan twice a day orally, which was increased to 125 mg twice a day after 4 weeks if no liver-function test abnormalities occurred. Elevation of liver transaminases is a common and known side effect of bosentan treatment, and therefore, a phased dosing regimen is recommended. Furthermore, liver function has to be monitored throughout administration of bosentan. In case of transaminases elevation, the values generally normalize after treatment is stopped, and then bosentan treatment can be commenced again with a low starting dose.

All patients in the RCT received anticoagulation, but other medications are not further specified. Patients in the control group of the RCT received a placebo. In the nonrandomized trial, the treatment group received standard therapy consisting of oral anticoagulation, diuretics, and supplemental oxygen in case of hypoxemia next to bosentan. The control group received only standard therapy.

Outcome measures

Both studies had as one of their primary endpoints change in 6MWD. The RCT used as independent coprimary endpoint change in PVR, whereas the nonrandomized trial assessed WHO functional class and baseline arterial oxygenation as further primary endpoints. WHO functional class was a secondary endpoint in the RCT, along with cardiopulmonary hemodynamics such as mPAP and time to clinical worsening. Furthermore, as exploratory endpoints, the N-terminal pro-brain natriuretic peptide (NT-proBNP), Borg dyspnoea index tested immediately following the 6MWD test, and quality of life as assessed by the Short Form 36 health survey questionnaire were measured. Both studies recorded adverse effects to assess the safety and tolerability of treatment. The RCT assessed outcome variables at baseline and at the end of the study after 16 weeks. The nonrandomized trial measured the outcome variables at baseline and after 3, 6, and 12 months of the study. For this review, the last measurements were considered, as this data was provided in the articles—thus, after 16 weeks for the RCT and after 12 months for the nonrandomized trial.

The RCT found a significant treatment effect for the change in PVR, which decreased in the bosentan group and increased in the placebo group. The slightly larger increase in 6MWD seen in the bosentan group was not significant. The nonrandomized trial, on the other hand, found a significant difference in 6MWD after 12 months, with an increased distance in the bosentan group and a decreased distance in the control group (see ). The mean treatment effect for mPAP of −2.5 mmHg (95% confidence interval (CI) −5.0 to 0.0) was not found to be significant in the RCT (=0.0652). In the nonrandomized trial, the increase in systolic PAP in the control group was found to be significant, with a systolic PAP of 80.4 mmHg compared with 71.5 mmHg at baseline (=0.029). In the bosentan group, a decrease in systolic PAP was seen after 12 months (71.0 mmHg compared with 76.2 mmHg at baseline), but this change was not significant (=0.221). The RCT found a larger improvement in WHO functional class for the bosentan group that was not significant, whereas the nonrandomized trial found a significant improvement in functional class for patients treated with bosentan (). Quality of life was not formally assessed in the nonrandomized trial, and the RCT did not show a difference in outcome between the bosentan and placebo groups (relative risk for improvement 1.07, 95% CI: 0.74 to 1.53). Concerning other exploratory endpoints, the RCT found that NT-proBNP significantly decreased in the bosentan group, whereas it increased in the placebo group (mean treatment effect −622 ng/L for bosentan treatment, =0.0034). For the Borg dyspnea index, the placebo-corrected treatment effect in favor of bosentan was −0.6 (95% CI: −1.2 to 0.0), which was statistically significant (=0.0386). The nonrandomized trial also assessed the Borg dyspnea index and found that patients receiving bosentan improved significantly compared with baseline (4.29 vs. 5.59, respectively; =0.048), whereas patients in the control group worsened significantly compared with baseline (7.06 vs. 5.77, respectively; =0.029).

Both studies recorded adverse events. The RCT lists in detail all recorded adverse events, including mild ones such as headache, nasopharyngitis, and palpitations. The nonrandomized trial, on the other hand, described as adverse events only abnormal liver function tests and one death in the control group, which was due to myocardial infarction and heart failure. That explains the large difference in number of adverse events as presented in . It was unfortunately not possible to extract clear data for only major adverse events from the RCT. Elevation of liver transaminases was monitored in both trials and found to occur more frequently in the bosentan group than in the control group (). In the RCT, two patients (one each from the bosentan and placebo groups) withdrew from the study due to elevated liver transaminases. In the nonrandomized trial, two patients from the bosentan group stopped taking bosentan temporarily due to this side effect until transaminases were no longer elevated.

Risk of bias in included studies

The assessment of the two included studies based on Downs and Black's checklist, which included a total of 13 items about internal validity, showed that the RCT scored 11 of 13 and the nonrandomized trial 7 of 13 points. In the checklist, 7 and 6 items were related to bias and confounding, respectively. Concerning bias, the RCT scored 6 of the possible 7 points, with the only major negative point being that no adjustments were made in the analyses for different lengths of follow-up, which varied considerably, from 4 to 21 weeks. The nonrandomized trial scored 5 of 7 points, because no attempt was made to blind either those receiving or those measuring the outcome of the intervention. Concerning confounding, the RCT scored 5 out of a maximum of 6 points. It remains unclear whether subjects were recruited over the same period of time, as it is not mentioned in the article. The nonrandomized trial scored 2 of 6 points. Patients in the two groups were only in the widest sense recruited from the same population, as the study was conducted in four hospitals in northern Italy. Whether patients received bosentan in addition to standard therapy or standard therapy alone depended on which hospital they attended, as two hospitals allowed the use of bosentan for patients with CTEPH, whereas the other two did not. This could possibly introduce selection bias. Furthermore, the nonrandomized trial scored low because patients were not randomized to intervention groups and, consequently, the assignment to different interventions was not concealed.

Next to the checklist, a comparison between the outcome measures as described in the methods and results section was done. The RCT showed no inconsistencies, since for all outcomes mentioned in the methods section, the results were reported. The nonrandomized trial reported results for all outcomes stated in the methods section as well, but in addition, reported results for Borg's dyspnea score, which was not specified in the methods section.

Synthesis of results

The systematic search of databases identified only two eligible studies, as most studies investigating the effectiveness of bosentan in the treatment of CTEPH were done in an open-label, noncontrolled setting. It is therefore difficult to generalize the findings. Furthermore, the two studies included in this review differ in that one was randomized, was double-blind, and had a reasonably large study population, but a follow-up of only 4 months, whereas the other one included a rather small group of patients, but followed them for a year, and administered bosentan open-label.

The qualitative analysis of both studies showed a significant effect of bosentan, but for different outcome measures, namely change in PVR in the RCT and change in 6MWD in the nonrandomized trial. PVR was assessed only in the RCT, so that a comparison between studies is not possible. The increase in 6MWD was found to be significant only in the nonrandomized trial. Other outcome measures, such as mPAP and quality of life, were assessed only in the RCT and did not show significant improvement. An outcome measure reported in both studies was the WHO functional class, which for patients receiving bosentan was shown to improve significantly in the nonrandomized trial but not significantly in the RCT.

In addition to the effectiveness of bosentan, both trials assessed the safety and tolerability of bosentan. A side effect more frequently observed in patients treated with bosentan compared to controls is elevation of liver transaminases, which normalizes when the drug is discontinued. Generally, bosentan treatment was found to be safe and well tolerated.

DISCUSSION

There is at present no approved medication for the treatment of inoperable CTEPH. This review tried to assess the level of evidence for the effectiveness of bosentan, an endothelin-1 receptor antagonist, in improving pulmonary hemodynamics and outcome in adult CTEPH patients who were either judged inoperable or had recurrent or resistant pulmonary hypertension following PEA.

Disappointingly, only nine studies have been conducted, and of these, seven lacked satisfactory methodological quality and were therefore excluded. It can be argued whether the inclusion criteria were too stringent, thereby limiting the number of studies considered in this review. However, studies must allow for a comparison between groups in order to assess whether addition of bosentan to standard treatment improves outcome. Ideally, only randomized, placebo-controlled studies would be included, but since only one study of this kind has been conducted so far, this was not possible. This led to the inclusion of the two studies that included any form of control treatment with which bosentan could be compared. These studies differed markedly in number of patients and length of follow-up. The RCT, based on a large patient population, found bosentan to significantly lower PVR, but failed to show a positive outcome for either 6MWD or WHO functional class. The nonrandomized trial, with a longer period of follow-up, did find a significant improvement of 6MWD and WHO class. A possible explanation is that the length of follow-up of the RCT was too short to determine a positive effect on functional parameters such as exercise capacity. Generally, the chosen outcome measures are suitable, as PVR has been established as prognostic factor in CTEPH, and with 6MWD, exercise capacity can be reliably assessed [, ]. Since it concerns a group of patients with a poor prognosis, improvement in functional parameters is important to ensure quality of life. This was formally assessed only in the RCT and not found to differ between treatment groups. The available evidence for other outcomes is weak, especially concerning time to clinical worsening and mortality, so that no sound conclusions can be drawn.

The studies included in this review have several limitations. The nonrandomized trial had a small study population of only 34 patients. Since CTEPH is a complication that occurs in about 4% of patients with pulmonary embolism, it is difficult to conduct a study that includes a sufficient number of patients. This difficulty was overcome in the RCT by setting up a large multicenter trial that led to the inclusion of 157 patients, thereby yielding sufficient power to detect clinically relevant outcomes. However, the RCT followed the patients for only 16 weeks, therefore giving no data on long-term outcomes such as, for example, 1-year survival. In addition, the article states only that the mean exposure was 16 weeks, with a range of 4 to 21 weeks, but not how these differences in follow-up were adjusted for. Limitations at review level include the possibility of publication bias, as no unpublished studies could be retrieved. This could bias the results in favour of bosentan if studies on its effectiveness that did not find significant results were not published.

In conclusion, the evidence for beneficial effects of bosentan in studies to date is insufficient to firmly establish its role as medical therapy for CTEPH. The RCT did not demonstrate a clear clinical benefit, and it is uncertain whether this is due to the relatively short period of follow-up. Bosentan is generally safe, though liver function has to be monitored to beware of transaminases elevation. A recent nonsystematic review comes to the same conclusion but also states that treatment with disease-modifying agents such as bosentan is preferable to no additional therapy []. Indeed, for this reason bosentan is often administered off-label, but this raises questions about the cost-effectiveness of this treatment. For future research, it is therefore essential to conduct randomized clinical trials that follow patients for a longer time in order to investigate the clinical outcome in the long run and whether functional parameters such as exercise capacity improve after a longer period of treatment. These trials could also yield data on long-term survival. Patients from the RCT were followed for several years as an extension to the original study, but bosentan was administered open-label, and the results are not available yet []. As several types of medication are available for PAH therapy, combination therapy also has to be considered for the treatment of CTEPH, as there are indications that it is more effective than single medication []. Interactions and synergistic effects of different agents for treating CTEPH will also have to be assessed in future studies.

Disclosure: No funding was received for this systematic review. The authors declare no conflict of interest.

References

1.   Pengo V, Lensing AW, Prins MH, et al. Incidence of chronic thromboembolic pulmonary hypertension after pulmonary embolism. N Engl J Med. 2004;350(22):2257-2264.
2.   Gabbay E, Fraser J, McNeil K. Review of bosentan in the management of pulmonary arterial hypertension. Vasc Health Risk Manag. 2007;3(6):887-900.
3.   Lang IM, Klepetko W. Chronic thromboembolic pulmonary hypertension: an updated review. Curr Opin Cardiol. 2008;23(6):555-559.
4.   Hill NS, Preston IR, Roberts KE. Inoperable chronic thromboembolic pulmonary hypertension: treatable with medical therapy. Chest. 2008;134(2):221-223.
5.   Frantz RP. Bosentan for pulmonary hypertension and other pulmonary diseases: emerging evidence. Future Cardiol. 2008;4(5):459-468.
6.   Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol. 2009;62(10):e1-e34.
7.   Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol. 2009;62(10):1006-1012.
8.   Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377-384.
9.   Segovia Cubero J, Ortiz Uribe JC, Gomez Bueno M, Monivas Palomero V, Gonzalez Gonzalez M, Alonso-Pulpon Rivera L. Role of bosentan in patients with chronic venous thromboembolic pulmonary hypertension. Med Clin (Barc). 2007;128(1):12-14.
10.   Bonderman D, Nowotny R, Skoro-Sajer N, et al. Bosentan therapy for inoperable chronic thromboembolic pulmonary hypertension. Chest. 2005;128(4):2599-2603.
11.   Hughes RJ, Jais X, Bonderman D, et al. The efficacy of bosentan in inoperable chronic thromboembolic pulmonary hypertension: a 1-year follow-up study. Eur Respir J. 2006;28(1):138-143.
12.   Post MC, Plokker HW, Kelder JC, Snijder RJ. Long-term efficacy of bosentan in inoperable chronic thromboembolic pulmonary hypertension. Neth Heart J. 2009;17(9):329-333.
13.   Seyfarth HJ, Hammerschmidt S, Pankau H, Winkler J, Wirtz H. Long-term bosentan in chronic thromboembolic pulmonary hypertension. Respiration. 2007;74(3):287-292.
14.   Ulrich S, Speich R, Domenighetti G, et al. Bosentan therapy for chronic thromboembolic pulmonary hypertension. A national open label study assessing the effect of Bosentan on haemodynamics, exercise capacity, quality of life, safety and tolerability in patients with chronic thromboembolic pulmonary hypertension (BOCTEPH-Study). Swiss Med Wkly. 2007;137(41-42):573-580.
15.   Hoeper MM, Kramm T, Wilkens H, et al. Bosentan therapy for inoperable chronic thromboembolic pulmonary hypertension. Chest. 2005;128(4):2363-2367.
16.   Jais X, D'Armini AM, Jansa P, et al. Bosentan for treatment of inoperable chronic thromboembolic pulmonary hypertension: BENEFiT (Bosentan Effects in iNopErable Forms of chronic Thromboembolic pulmonary hypertension), a randomized, placebo-controlled trial. J Am Coll Cardiol. 2008;52(25):2127-2134.
17.   Vassallo FG, Kodric M, Scarduelli C, et al. Bosentan for patients with chronic thromboembolic pulmonary hypertension. Eur J Intern Med. 2009;20(1):24-29.
18.   Miyamoto S, Nagaya N, Satoh T, et al. Clinical correlates and prognostic significance of six-minute walk test in patients with primary pulmonary hypertension. Comparison with cardiopulmonary exercise testing. Am J Respir Crit Care Med. 2000;161(2 Pt 1):487-492.
19.   Riedel M, Stanek V, Widimsky J, Prerovsky I. Longterm follow-up of patients with pulmonary thromboembolism. Late prognosis and evolution of hemodynamic and respiratory data. Chest. 1982;81(2):151-158.
20.   Confalonieri M, Kodric M, Longo C, Vassallo FG. Bosentan for chronic thromboembolic pulmonary hypertension. Expert Rev Cardiovasc Ther. 2009;7(12):1503-1512.
21.    Clinical Trials: Bosentan in Patients With Inoperable Chronic Thromboembolic Pulmonary Hypertension (CTEPH) (BENEFIT). U.S. National Institutes of Health [11 February 2010; cited 28 April 2010]. Available from: http://www.clinicaltrials.gov/ct2/show/NCT00319111?term=bosentan+AND+chronic+AND+pulmonary+hypertension&rank=222
22.   Bresser P, Pepke-Zaba J, Jais X, Humbert M, Hoeper MM. Medical therapies for chronic thromboembolic pulmonary hypertension: an evolving treatment paradigm. Proc Am Thorac Soc. 2006;3(7):594-600.

Can Mutations Identified in Congenital Fibrinogen Disorders Explain the Clinical Manifestations?

Thu, 24 Jun 2010 08:25:11 +0100

INTRODUCTION

This review aims to provide an update on the mutations of inherited disorders of fibrinogen and their possible relationship with clinical complications. Many detailed and thoroughly annotated reviews of identified causative mutations have been published [–], and a registry for hereditary fibrinogen abnormalities [] can be accessed at . A new list of mutations will be published soon in the Williams Hematology, 8th Edition [].

Fibrinogen plays a major role in hemostasis as the precursor molecule for the insoluble fibrin clot but, in addition, participates in numerous other biologic processes, such as inflammation, wound healing, and angiogenesis. Fibrinogen binds plasminogen, antiplasmin, fibronectin, and factor XIII, among others. It also binds to platelets and supports platelet aggregation. After fibrinogen is converted to fibrin by thrombin, it provides non-substrate binding sites for thrombin (thus, fibrinogen is sometimes known as antithrombin I []) and binds to vascular endothelial and other cells, plasma or tissue matrix components such as fibronectin and glycosaminoglycans, and peptide growth factors. Fibrin provides a template for assembly and activation of the fibrinolytic system and is the major substrate for the enzyme plasmin. Both fibrinogen and fibrin also serve as substrates for plasma factor XIII that catalyzes covalent cross-linking/ligation.

Diseases affecting fibrinogen may be acquired or inherited. Inherited disorders of fibrinogen are rare and can be subdivided into type I and type II disorders []. Type I disorders affect the quantity of fibrinogen in circulation: hypofibrinogenemia is characterized by fibrinogen levels lower than 1.5 g L; afibrinogenemia is characterized by the complete deficiency of fibrinogen. Type II disorders affect the quality of circulating fibrinogen: in dysfibrinogenemia fibrinogen antigen levels are normal, in hypodysfibrinogenemia levels are reduced.

FIBRINOGEN STRUCTURE AND SYNTHESIS

Fibrinogen is a 340-kDa glycoprotein synthesized in hepatocytes that circulates in plasma at a concentration of 1.5 to 3.5 g L. The core structure consists of two outer D regions (or D domains) and a central E region (or E domain) connected through coiled-coil connectors () []. The molecule exhibits a two-fold axis of symmetry perpendicular to the long axis, consisting of two sets of three polypeptide chains (A, B, ) that are joined in their amino-terminal regions by disulfide bridges to form the E region. The outer D regions contain the globular C terminal domains of the Bβ chain (βC) and γ chain (γC). Unlike the βC and γC domains the C-terminal domains of the A chain (C) are intrinsically unfolded and flexible and tend to be non-covalently tethered in the vicinity of the central E region.

The three genes encoding fibrinogen Bβ (), Aα (), and γ () are clustered in a region of approximately 50 kilobases on human chromosome 4. and are transcribed from the reverse strand, in the opposite direction to . Alternative splicing results in two isoforms for the fibrinogen alpha chain: the common Aα chain, encoded by exons 1-5, and an extended Aα-E isoform, encoded by exons 1-6, which represents only 1% to 2% transcripts. Alternative splicing for also produces two transcripts: the major mRNA species contains all 10 exons and encodes the common chain (or A), while the minor product (’) does not splice out intron 9 and the corresponding open reading frame replaces the four codons of exon 10 with 20 alternative codons. encodes a single 1.9-kb transcript with a 1.5-kb coding sequence. Each gene is separately transcribed and translated to produce nascent polypeptides of 644 amino acids (Aα), 491 amino acids (Bβ), and 437 amino acids ().

During translocation of the single chains into the lumen of the endoplasmic reticulum (ER), a signal peptide is co-translationally cleaved from each chain. A 15 amino-acid propeptide is also cleaved from the C-terminus of the Aα chain by furin and carboxypeptidase H. The resulting chains have 610 amino acids (Aα), 461 amino acids (Bβ), and 411 amino acids (). Assembly proceeds in the ER with the formation of an Aα-γ or Bβ-γ intermediate. The addition of either a Bβ or Aα chain gives rise to a [AαBβγ] half-molecule, which dimerizes to form the functional hexamer []. The protein undergoes several post-translational modifications in the Golgi complex, including maturation of N-linked oligosaccharides, phosphorylation, hydroxylation, and sulfation.

Following assembly, which is completed within minutes, the mature molecule is constitutively secreted into the circulation, where it exhibits a half-life of approximately 4 days and a fractional catabolic rate of 25% per day.

FIBRINOGEN CONVERSION TO FIBRIN AND POLYMERIZATION

Conversion of fibrinogen to a fibrin clot [, ] occurs in three distinct phases: (1) enzymatic cleavage by thrombin to produce fibrin monomers, (2) self-assembly of fibrin units to form an organized polymeric structure, and (3) covalent cross-linking of fibrin by factor XIIIa. In the first phase of conversion to fibrin, cleavage of fibrinogen at AR35/G36 (R16/G17 without the signal peptide) and later B R44/G45 (R14/G15 without the signal peptide) results in release of fibrinopeptides A (FpA) and B (FpB), respectively, thus exposing “A” knobs and “B” knobs (]. The “A” knob located at the new amino-terminal end of the fibrin chain starts with the GPRV amino acid sequence. The “A” knob in fibrin interacts with the constitutive complementary association site known as hole “a” in -chain of another molecule to initiate the fibrin assembly process. “A knob–a hole” interactions result in formation of double-stranded fibrils in which fibrin molecules become aligned in an end-to-middle, staggered overlapping arrangement [–]. Fibrils subsequently undergo branching by lateral fibril associations in which two fibrils converge to form a four-stranded “bilateral” fibril junction. Progressive lateral associations among fibrils result in larger fibril bundles or fibers. A second type of junction, known as equilateral branching, is formed by three fibrils converging to form a three-member junction. Together these two types of branch junctions provide scaffolding for the clot network, the ultimate structure of which is influenced by several variables, including salt concentration, pH, and thrombin concentration [, ].

FpB release occurs more slowly than FpA release and exposes another polymerization site, known as the “B” knob, beginning with the amino acid sequence GHRP. GHRP interacts with a constitutive “b” hole in the -chain. FpB cleavage is accelerated by fibrin polymerization, whereas FpA cleavage is independent of fibrin polymerization. “B knob–b hole” interactions are not required for lateral fibril associations, but they contribute to lateral association. Finally, additional self-associating sites in the D region participate in fibrin assembly and factor XIIIa cross-linking [].

THROMBIN BINDING TO FIBRINOGEN AND FIBRIN

Thrombin binds to its substrate, fibrinogen, through a fibrinogen recognition site in thrombin, known as exosite 1 []. The fibrin clot itself also exhibits significant thrombin-binding potential; this non-substrate binding potential of fibrin for thrombin is referred to as antithrombin I []. Antithrombin I activity is defined by two classes of non-substrate thrombin binding sites in fibrin, one of low-affinity in the E-region and the other of high-affinity in D regions of fibrin(ogen) molecules containing the variant ′ chain. Antithrombin I (fibrin) is an important inhibitor of thrombin generation that functions by sequestering thrombin in the forming fibrin clot and also by reducing the catalytic activity of fibrin-bound thrombin. Vascular thrombosis may result from absence of antithrombin I (as in afibrinogenemia, see below), reduced plasma ′ chain content [], or defective thrombin binding to fibrin as found in certain dysfibrinogenemias. On the contrary, increased susceptibility to arterial thrombosis has been reported when ′-chain levels are significantly elevated. Moreover, thrombin bound to /′-fibrin is less inhibited by antithrombin than thrombin bound to /-fibrin, and /′-fibrin serves as a reservoir of active thrombin, which may contribute to the prothrombotic nature of thrombi [].

AFIBRINOGENEMIA AND HYPOFIBRINOGENEMIA

We recently reviewed the main features of these two type I disorders (afibrinogenemia and hypofibrinogenemia) that affect the quantity of fibrinogen in circulation []. The first causative mutation for congenital afibrinogenemia was published in 1999 []. Since this date, more than 80 distinct mutations, the majority in have been identified in patients with afibrinogenemia (in homozygosity or in compound heterozygosity) or with hypofibrinogenemia, since a large number of these patients are in fact asymptomatic carriers of afibrinogenemia mutations. A complete list of mutations can be found at . Causative mutations can be divided into two main classes: null mutations (deletions, splice-site, frameshift, and early-truncating nonsense mutations), with no protein production at all, and mutations producing abnormal protein chains that impair fibrinogen assembly and/or secretion and are retained inside the cell (missense or late-truncating nonsense mutations).

Four large deletions (greater than 1 kb) have been identified to date, the most frequent being the 11 kb deletion [, ]. In afibrinogenemic patients of European origin, the most common mutation is a donor splice mutation in intron 4, c.510 + 1G > T (previously described as IVS4 + 1 G > T). Haplotype data suggest that this mutation, like the 11 kb deletion, is also recurrent or a very ancient mutation since the c.510 + 1G > T mutation is found on multiple distinct haplotypes. Several frameshift mutations account for complete fibrinogen deficiency (12 in , three in , and two in ). exon 5 has the largest number of frameshift mutations. Interestingly, seven single base pair deletions in exon 5 result in usage of the same new reading frame. Finally, more than 20 nonsense mutations accounting for afibrinogenemia and hypofibrinogenemia have been identified. Null mutations (ie, large deletions or frameshift, early-truncating nonsense, or splice-site mutations) account for the majority of afibrinogenemia alleles, as expected given the phenotype of the disorder (ie, complete deficiency of fibrinogen in circulation and bleeding). Of particular interest, therefore, are missense mutations leading to complete fibrinogen deficiency. These are clustered in the highly conserved C-terminal globular domains of the Bβ and γ chains. Expression studies in transfected cells for five missense mutations identified in homozygosity or compound heterozygosity in afibrinogenemic patients showed that these mutations allowed individual chain synthesis and intracellular assembly of the hexamer but impaired secretion, suggesting that an intact C-terminal domain is necessary for fibrinogen secretion into the circulation [–].

Ten missense mutations have also been identified in in heterozygosity in patients with hypofibrinogenemia. For the majority of these mutations, analysis of patient plasma fibrinogen by mass spectrometry has confirmed absence of the mutant γ-chain in the circulation. Others have been studied at the functional level in transfected cells: fibrinogen Matsumoto IV C179R (C153R) was found to impair intracellular hexamer assembly [], while fibrinogen Bratislava W253C (W227C) was found to impair fibrinogen secretion [].

In the majority of patients with afibrinogenemia or hypofibrinogenemia, there is no evidence of intracellular accumulation of the mutant fibrinogen chain. This implies the existence of an efficient degradation pathway for fibrinogen mutants that allow individual chain synthesis and assembly but not secretion. As shown in , three mutations, all in are known to cause hypofibrinogenemia accompanied by hepatic storage disease [–]. Two are missense mutations in (fibrinogen Brescia and Aguadilla), which cause fibrinogen deficiency in the heterozygous state due to the absence of the mutant γ chain in patient plasma, and also progressive liver disease associated with hepatocellular cytoplasmic inclusions. A third mutation was identified in (fibrinogen Angers) in a woman with chronic abnormal liver-function tests []. The molecular mechanism by which the fibrinogen Angers mutation, but also the fibrinogen Brescia and Aguadilla mutations, localized in the 5-stranded beta sheet of γC and the “a” hole, respectively, leads to impaired secretion, retention in the ER, and formation of aggregates remains to be determined []. In a simple model, the positions of these three mutations could delimit a region in the γC domain, which when mutated causes retention in the cell but escapes degradation. However, this is not the case, since there are not only numerous dysfibrinogenemia mutations (see below) occurring in this region that are efficiently assembled and secreted into the circulation but also hypofibrinogenemia mutations in this region that are not secreted but do not lead to ER inclusion bodies and liver disease.

CAN MUTATIONS IDENTIFIED IN AFIBRINOGENEMIA AND HYPOFIBRINOGENEMIA EXPLAIN THE CLINICAL MANIFESTATIONS?

The majority of afibrinogenemia patients have a bleeding diathesis, but no clear relationship exists between the various mutations and the clinical manifestations, especially for the cases with thrombotic complications. Current diagnostic tests are appropriate for establishing the diagnosis of quantitative fibrinogen disorders, but additional tests are necessary for a more accurate prediction of the clinical phenotype of a patient and, consequently, the appropriate treatment. Indeed, although in afibrinogenemia, all patients have non-measurable functional fibrinogen, the severity of bleeding is highly variable among patients, even among those with the same genotype. Similarly, there is no clear relationship between the molecular defect and the risk of thrombosis.

The most likely explanation for the observed variability of clinical manifestations is the existence of modifier genes/alleles: some variants may increase the severity of bleeding, while others may ameliorate the phenotype. Such modifiers have yet to be identified; however, common variants predisposing to thrombophilia (eg, factor V Leiden) most certainly play a role in decreasing the severity of bleeding, as in hemophilia []. The existence of modifying genes/polymorphisms is also strongly suspected in the previously discussed cases of hypofibrinogenemia associated with fibrinogen inclusion bodies in hepatocytes. Indeed, all individuals heterozygous for one of the three causative mutations identified in have hypofibrinogenemia, but not all have fibrinogen aggregates and associated liver disease.

Paradoxically, both arterial and venous thromboembolic complications are observed in afibrinogenemic patients []. These complications can occur in the presence of concomitant risk factors such as a coinherited thrombophilic risk factor or after replacement therapy. However, in many patients, no known risk factors are present. Several hypotheses have been put forward to explain this predisposition to thrombosis. One explanation is that even in the absence of fibrinogen, platelet aggregation is possible due to the action of von Willebrand factor [], and in contrast to hemophiliacs, afibrinogenemic patients are able to generate thrombin, both in the initial phase of limited production and also in the secondary burst of thrombin generation. In some patients, increase of prothrombin activation fragments or thrombin–antithrombin complexes has been observed, which may reflect an enhanced thrombin generation. These abnormal levels can be normalized by fibrinogen infusions [].

As previously mentioned, fibrin also acts as antithrombin I by both sequestering and down-regulating thrombin activity []. Thrombin that is not trapped by the clot is available for platelet activation and smooth muscle cell migration and proliferation, particularly in the arterial vessel wall. In fibrinogen-deficient mice, thrombus formation is maintained but the thrombus is unstable and has a tendency to embolize []. Similarly, the absence of fibrinogen in human plasma results in large but loosely packed thrombi under flow conditions [].

DYSFIBRINOGENEMIA AND HYPODYSFIBRINOGENEMIA

Dysfibrinogenemia is defined by the presence of normal levels of functionally abnormal plasma fibrinogen. Hypodysfibrinogenemia is defined by low levels of a dysfunctional protein. As in afibrinogenemia and hypofibrinogenemia, both are heterogeneous disorders caused by many different mutations in the three fibrinogen-encoding genes.

Most dysfibrinogenemia mutant molecules are found in plasma at normal antigenic levels; thus, they can be diagnosed by the combination of a prolonged thrombin time, normal levels of fibrinogen antigen, and low functional levels of fibrinogen. Over 400 cases of dysfibrinogenemia have been reported to date [, ], with more than 40 distinct mutations identified (more than 60 distinct mutations in dysfibrinogenemia and hypodysfibrinogenemia combined). Most cases are asymptomatic and are only identified as a result of routine coagulation screening. Although there is a publication bias tending to report cases with clinical manifestations, it has been published that approximately 25% of patients with dysfibrinogenemia have a history of bleeding, and in approximately 20%, a tendency toward thrombosis is observed [].

Dysfibrinogenemia mutations usually affect one or more phases of the fibrinogen-fibrin conversion and fibrin assembly process (ie, impaired release of fibrinopeptides, defects in fibrin polymerization, and defective factor XIIIa-mediated cross-linking). Other defects affect related aspects of fibrinogen/fibrin function or metabolism, such as catabolism, abnormal tissue deposition, defective assembly of the fibrinolytic system, abnormal interactions with platelets, endothelial cells, or calcium binding [–].

Mutations resulting in abnormal “A” knobs or deficient fibrinopeptide release

Fibrinogen Detroit was the first abnormal fibrinogen in which the specific mutation was identified at the protein level []. This R38S (R19S) mutation is located in the “A” knob (ie, GPRV), resulting in impaired fibrin polymerization and a bleeding tendency. Other substitutions involving residue R38 (R19) have been found to be associated with bleeding in some cases (eg, Munich I, R38N (R19N) and Mannheim I, R38G (R19G)) and with thrombosis in other cases (eg, Aarhus and Kumamoto also due to R38G (R19G)). The mechanism for thrombophilia remains unclear, but coexisting risk factors may contribute to the clinical manifestations. Furthermore, the inability of a mutant fibrin to effectively bind and sequester thrombin may play a role in such a clinical presentation. Bleeding that occurs under conditions involving defective fibrinopeptide release or production of a defective “A” knob is most likely related to the reduced polymerization potential of the mutant fibrins that are produced, with resulting defective clot formation.

Missense mutations at residue R35 (R16), which is part of the thrombin cleavage site in the fibrinogen α-chain, are the most common causative mutations accounting for dysfibrinogenemia, found in approximately 40% of cases []. The R35 (R16) residue can be mutated to either H (CGT > CAT) or C (CGT > TGT), leading to delayed or absent fibrinopeptide A release, respectively, and subsequent delayed polymerization. A prolonged reptilase time is observed for both variants. Most patients do not have a bleeding tendency. Some patients have been found to be homozygous for these mutations or phenotypically homozygous, due to compound heterozygosity for an R35 (R16) missense mutation and the large 11 kb deletion first characterized in afibrinogenemia []. In these cases, a mild bleeding tendency is observed. Missense mutations in affecting FpB release have been identified but are much less common than those affecting FpA release ().

Mutations leading to polymerization defects

Sites in the D region important for fibrin polymerization (ie, the “a” hole and D:D sites) are affected in many dysfibrinogenemias. Mutations affecting the “a” hole in the chain are numerous, while none specifically involving the “b” hole in the B chain have been described.

The interface for the end-to-end D:D site in the -chain lies between R301 (R275) and S326 (S300), with T306 (T280) contacting R301 (R275) at the D:D interface []. Mutations at the R301 (R275) residue to C (CGT > TGT) or H (CGT > CAT) are the second-most common cause of dysfibrinogenemia, accounting for more than 10% of fibrinogen variants [, ]. Impaired polymerization has been observed for all substitutions at this position. Most of these cases are asymptomatic, but some patients heterozygous for R301C (R275C) have thrombosis, sometimes in association with an additional thrombotic risk factor such as factor V Leiden [].

Mutations accounting for hypodysfibrinogenemia

Hypodysfibrinogenemia, which is defined by low levels of a dysfunctional protein, can be caused by different molecular mechanisms. One is heterozygosity for a single mutation (similar to most cases of dysfibrinogenemia), which leads to synthesis of an abnormal fibrinogen chain that is secreted less efficiently than normal fibrinogen (eg, fibrinogen Kyoto IV []). Another mechanism is the presence of two different mutations, with one mutation responsible for the fibrinogen deficiency (the “hypo phenotype”) and one mutation responsible for the abnormal function of the molecule (the “dys phenotype”). These mutations can be either in compound heterozyosity, as observed in fibrinogen Keokuk [], for example, which is caused by compound heterozygosity for the common afibrinogenemia splice-site mutation c.510G > T and a premature truncating nonsense mutation in Q347X (Q328X), or on the same allele, as observed in fibrinogen Leipzig II, which is caused by the common hypofibrinogenemia mutation A108G (A82G) and G377S (G351S) []. Finally, homozygosity for a single mutation that allows reduced secretion of a functionally impaired molecule (eg, fibrinogens Otago [] and Marburg [, ] can also account for hypodysfibrinogenemia.

CAN MUTATIONS IDENTIFIED IN DYSFIBRINOGENEMIA AND HYPODYSFIBRINOGENEMIA EXPLAIN THE CLINICAL MANIFESTATIONS?

Individuals with inherited dysfibrinogenemia are frequently asymptomatic; however, some patients suffer from bleeding, thromboembolic complications, or both (). Two mechanisms may explain most of the cases of thrombosis associated with dysfibrinogenemia: (1) the abnormal fibrinogen is defective in binding thrombin, which results in elevated levels of thrombin, and (2) the abnormal fibrinogen forms a fibrin clot that is resistant to plasmin degradation.

Some mutations are predictive of the clinical phenotype, for example, the R573C (R554C) substitution in the Aα chain (eg, fibrinogens Chapel Hill III, Paris V, and Dusart) predisposes patients to thrombosis. Here, the impaired fibrinolysis exhibited by this dysfibrinogen appears to be responsible for the thrombotic complications observed. Other examples associated with thrombosis include dysfibrinogens Barcelona III, Haifa I, or Bergamo II due to the common R301H (R275H) mutation in the γ chain. On the other hand, several dysfibrinogenemias, particularly those caused by mutations in the amino-terminal region of the Aα chain (eg, fibrinogen Detroit R38S(R19S) or Mannheim I R38G(R19G)) are associated with bleeding. These examples illustrate how determining the causative mutation can allow to take precautionary measures.

Other mutations in the Aα-chain of fibrinogen are associated with a particular form of hereditary amyloidosis [–). The E545V(E526V) amino acid substitution is the most common of these mutations. The abnormal fibrinogen fragments form amyloid fibrils, and the extracellular deposition of these fibrils leads to renal failure. Chronic renal dialysis is performed for managing renal failure. Renal transplantation should be considered as an alternative to chronic dialysis; however, renal transplantation is not a cure for hereditary renal amyloidosis-related renal failure, because continuous fibrinogen-related amyloid deposition ultimately results in allograft destruction. Combined liver and kidney transplantation prevents further amyloid deposition in the renal allograft and elsewhere but is associated with additional perioperative and subsequent risks.

CONCLUSIONS

The molecular analysis of congenital fibrinogen disorders provides several important pieces of information. In afibrinogenemia, no simple genotype–phenotype correlation exists, but genotyping is of value since it allows a better understanding of fibrinogen assembly and secretion and it also is of help for prenatal diagnosis. Further studies could explain why some of these patients develop thrombosis. In hypofibrinogenemia, some well-characterized mutations are linked with fibrinogen inclusion bodies in hepatocytes. The most convincing associations between clinical complications and genotype are found in dysfibrinogenemias. These mutations may explain bleeding, thrombotic complications, and renal amyloidosis in patients. Thus, the identification of the underlying causative mutation is valuable for patients with dysfibrinogenemia in order to provide an appropriate treatment.

Disclosure: The authors declare no conflict of interest.

Acknowledgments: The authors thank the Swiss National Science
Foundation for their continuous support for their research through
grants.

References

1. Mosesson MW. Hereditary fibrinogen abnormalities. In: Lichtman M, Beutler E, Kipps TJ, Seligsohn U, Kaushansky K, Prchal J, eds. Williams Hematology 7th Edition.2006 New York: McGraw-Hill. p. 1909-1927.[AQ4]
2. Maghzal GJ, Brennan SO, Homer VM, George PM. The molecular mechanisms of congenital hypofibrinogenemia. Cell Mol Life Sci. 2004;61:1427-1438.
3. Ebert RF. Index of Variant Human Fibrinogens. Boca Raton, FL: CRC Press; 1994.
4. Asselta R, Duga S, Tenchini ML. The molecular basis of quantitative fibrinogen disorders. J Thromb Haemost. 2006;4:2115-2129.
5. Neerman-Arbez M, de Moerloose P. Mutations in the fibrinogen gene cluster accounting for congenital afibrinogenemia: an update and report of 10 novel mutations. Hum Mutat. 2006;28:540-553.
6. Hanss M, Biot F. A database for human fibrinogen variants. Ann N Y Acad Sci. 2001;936:89-90.
7. Neerman-Arbez M, de Moerloose P. Hereditary fibrinogen abnormalities. In: Kaushansky K, Marshall A, Lichtman M, Beutler E, Kipps TJ, Seligsohn U, Prchal J, eds. Williams Hematology 8th Edition. New York: McGraw-Hill; in press.
8. Mosesson MW. Update on antithrombin I (fibrin). Thromb Haemost. 2007;98:105-108.
9. de Moerloose P, Neerman-Arbez M. Congenital fibrinogen disorders. Semin Thromb Hemost. 2009;35:356-366.
10. Medved L, Weisel JW. Recommendations for nomenclature on fibrinogen and fibrin. J Thromb Haemost. 2009;7:355-359.
11. Huang S, Mulvihill ER, Farrell DH, et al. Biosynthesis of human fibrinogen. Subunit interactions and potential intermediates in the assembly. J Biol Chem. 1993;268:8919-8926.
12. Mosesson MW. The structure and biological features of fibrinogen and fibrin. Ann N Y Acad Sci. 2001;936:11-30.
13. Weisel JW. Fibrinogen and fibrin. Adv Protein Chem. 2005;70:247-299.
14. Spraggon G, Everse SJ, Doolittle RF. Crystal structures of fragment D from human fibrinogen and its crosslinked counterpart from fibrin. Nature. 1997;389:455-462.
15. Lord ST. Fibrinogen and fibrin: scaffold proteins in hemostasis. Curr Opin Hematol. 2007;14:236-241.
16. Weisel JW. Structure of fibrin: impact on clot stability. J Thromb Haemost. 2007;1(5 Suppl):116-124.
17. Mosesson MW, Siebenlist KR, Hernandez I, et al. Evidence that alpha2-antiplasmin becomes covalently ligated to plasma fibrinogen in the circulation: a new role for plasma factor XIII in fibrinolysis regulation. J Thromb Haemost. 2008;6:1565-1570.
18. Fenton JW II, Olson TA, Zabinski MP, Wilner GD. Anion-binding exosite of human α-thrombin and fibrin(ogen) recognition. Biochemistry. 1988;27:7106-7112.
19. Uitte de Willige S, de Visser MC, Houwing-Duistermaat JJ, et al. Genetic variation in the fibrinogen gamma gene increases the risk for deep venous thrombosis by reducing plasma fibrinogen gamma' levels. Blood. 2005;106:4176-4183.
20. Fredenburgh JC, Stafford AR, Leslie BA, Weitz JI. Bivalent binding to to γA/γ′- fibrin engages both exosites of thrombin and protects it from inhibition by the antithrombin-heparin complex. J Biol Chem. 2008;283:2470-2477.
21. Neerman-Arbez M, Honsberger A, Antonarakis SE, Morris MA. Deletion of the fibrinogen alpha-chain gene (FGA) causes congenital afibrinogenemia. J Clin Invest. 1999;103:215-218.
22. Duga S, Asselta R, Santagostino E, et al. Missense mutations in the human beta fibrinogen gene cause congenital afibrinogenemia by impairing fibrinogen secretion. Blood. 2000;95:1336-1341.
23. Vu D, Bolton-Maggs PH, Parr JR, et al. Congenital afibrinogenemia: identification and expression of a missense mutation in FGB impairing fibrinogen secretion. Blood. 2003;102:4413-4415.
24. Spena S, Asselta R, Duga S, et al. Congenital afibrinogenemia: intracellular retention of fibrinogen due to a novel W437G mutation in the fibrinogen Bbeta-chain gene. Biochim Biophys Acta. 2003;1639:87-94.
25. Monaldini L, Asselta R, Duga S, et al. Fibrinogen Mumbai: intracellular retention due to a novel G434D mutation in the Bbeta-chain gene. Haematologica. 2006;91:628-633.
26. Terasawa F, Okumura N, Kitano K, et al. Hypofibrinogenemia associated with a heterozygous missense mutation gamma153Cys to Arg (Matsumoto IV): in vitro expression demonstrates defective secretion of the variant fibrinogen. Blood. 1999;94:4122-4131.
27. Vu D, de Moerloose P, Batorova A, et al. Hypofibrinogenaemia caused by a novel FGG missense mutation (W253C) in the gamma chain globular domain impairing fibrinogen secretion. J Med Genet. 2005;42:e57.
28. Brennan SO, Wyatt J, Medicina D, et al. Fibrinogen Brescia: hepatic endoplasmic reticulum storage and hypofibrinogenemia because of a gamma284 Gly→Arg mutation. Am J Pathol. 2000;157:189-196.
29. Brennan SO, Maghzal G, Shneider BL, et al. Novel fibrinogen gamma375 Arg→Trp mutation (fibrinogen Aguadilla) causes hepatic endoplasmic reticulum storage and hypofibrinogenemia. Hepatology. 2002;36:652-658.
30. Rubbia-Brandt L, Neerman-Arbez M, Rougemont A-L, et al. Fibrinogen gamma 375Arg→Trp mutation (fibrinogen Aguadilla) causes hereditary hypofibrinogenemia, hepatic endoplasmic reticulum storage disease and cirrhosis. Am J Surg Pathol. 2006;30:906-911.
31. Francalanci P, Santorelli FM, Talini I, et al. Severe liver disease in early childhood due to fibrinogen storage and de novo gamma375Arg→Trp gene mutation. J Pediatr. 2006;148:396-398.
32. Dib N, Quelin F, Ternisien C, et al. Fibrinogen Angers with a new deletion gamma GVYYQ 346-350 causes hypofibrinogenemia with hepatic storage. J Thromb Haemost. 2007;5:1999-2005.
33. Neerman-Arbez M. To aggregate or not to aggregate. J Thromb Haemost. 2007;5:1997-1998.
34. Franchini M, Lippi G. Factor V Leiden and hemophilia. Thromb Res. 2010;125:119-123.
35. De Marco L, Girolami A, Zimmerman TS, Ruggeri ZM. von Willebrand factor interaction with the glycoprotein IIb/IIa complex. Its role in platelet function as demonstrated in patients with congenital afibrinogenemia. J Clin Invest. 1986;77:1272-1277.
36. Korte W, Feldges A. Increased prothrombin activation in a patient with congenital afibrinogenemia is reversible by fibrinogen substitution. Clin Invest. 1994;72:396-398.
37. Ni H, Denis CV, Subbarao S, et al. Persistence of platelet thrombus formation in arterioles of mice lacking both von Willebrand factor and fibrinogen. J Clin Invest. 2000;106:385-392.
38. Remjin JA, Wu Y-P, Ijsseldijk W, et al: Absence of fibrinogen in afibrinogenemia results in large but loosely packed thrombi under flow conditions. Thromb Haemost. 2001;85:736-742.
39. Haverkate F, Samama M. Familial dysfibrinogenemia and thrombophilia. Report on a study of the SSC Subcommittee on Fibrinogen. Thromb Haemost. 1995;73:151-161.
40. Martinez J. Congenital dysfibrinogenemia. Curr Opin Hematol. 1997;4:357-365.
41. Roberts HR, Stinchcombe TE, Gabriel DA. The dysfibrinogenaemias. Br J Haematol. 2001;114:249-257.
42. Hill M, Dolan G. Diagnosis, clinical features and molecular assessment of the dysfibrinogenaemias. Haemophilia. 2008;14:889-897.
43. Blomback M, Blomback B, Mammen EF, Prasad AS. Fibrinogen Detroit-a molecular defect in the N-terminal disulphide knot of human fibrinogen? Nature. 1968;218:134-137.
44. Galanakis DK, Neerman-Arbez M, Scheiner T, et al. Homophenotypic A-alpha R16H fibrinogen (Kingsport): uniquely altered polymerization associated with slower fibrinopeptide A than fibrinopeptide B release. Blood Coagul Fibrinolysis. 2007;18:731-737.
45. Siebenlist KR, Mosesson MW, Meh DA, et al. Coexisting dysfibrinogenemia (γR275C) and factor V Leiden deficiency associated with thromboembolic disease (fibrinogen Cedar Rapids). Blood Coagul Fibrinolysis. 2000;11:293-304.
46. Okumura N, Terasawa F, Hirota-Kawadobora M et al. A novel variant fibrinogen, deletion of Bbeta111Ser in coiled-coil region, affecting fibrin lateral aggregation. Clin Chim Acta. 2006;365:160-167.
47. Lefebvre P, Velasco PT, Dear A, et al. Severe hypodysfibrinogenemia in compound heterozygotes of the fibrinogen AalphaIVS4 + 1G>T mutation and an AalphaGln328 truncation (fibrinogen Keokuk). Blood. 2004;103:2571-2576.
48. Meyer M, Dietzel H, Kaetzel R, et al. Fibrinogen Leipzig II (gamma351Gly→Ser and gamma82Ala→Gly): hypodysfibrinogenaemia due to two independent amino acid substitutions within the same polypeptide chain. Thromb Haemost. 2007;98:903-905.
49. Ridgway HJ, Brennan SO, Faed JM, George PM. Fibrinogen Otago: a major alpha chain truncation associated with severe hypofibrinogenaemia and recurrent miscarriage. Br J Haematol. 1997;98:632-639.
50. Koopman J, Haverkate F, Grimbergen J, et al. Fibrinogen Marburg: a homozygous case of dysfibrinogenemia, lacking amino acids A alpha 461-610 (Lys 461 AAA→stop TAA). Blood. 1992;80:1972-1979.
51. Sugo T, Nakamikawa C, Takebe M, et al. Factor XIIIa cross-linking of the Marburg fibrin: formation of alpha and gamma-heteromultimers and the alpha-chain-linked albumin gamma complex, and disturbed proto-fibril assembly resulting in acquisition of plasmin resistance relevant to thrombophilia. Blood. 1998;91:3282-3288.
52. Hamidi AL, Liepnieks JJ, Uemichi T, et al. Renal amyloidosis with a frame shift mutation in fibrinogen alpha-chain gene producing a novel amyloid protein. Blood. 1997;90:4799-4805.
53. Uemichi T, Liepnieks JJ, Yamada T, et al. A frame shift mutation in the fibrinogen A alpha chain gene in a kindred with renal amyloidosis. Blood 1996;87:4197-4203.
54. Uemichi T, Liepnieks JJ, Benson MD. Hereditary renal amyloidosis with a novel variant fibrinogen. J Clin Invest. 1994;93:731-736.
55. Benson MD, Liepnieks J, Uemichi T, et al. Hereditary renal amyloidosis associated with a mutant fibrinogen alpha-chain. Nat Genet. 1993;3:252-255.
56. Gillmore JD, Lachmann HJ, Rowczenio D, et al. Diagnosis, pathogenesis, treatment, and prognosis of hereditary fibrinogen A alpha-chain amyloidosis. J Am Soc Nephrol. 2009;20:444-451.
57.Yang Z, Kollman JM, Pandi L, Doolittle RF. Crystal structure of native chicken fibrinogen at 2.7A resolution. Biochemistry.2001;40:12515-12523