Haemophilia (2014), 20 (Suppl. 4), 137–144
DOI: 10.1111/hae.12415
REVIEW ARTICLE
Joint WFH-ISTH session: issues in clinical trial design F . P E Y V A N D I , * A . F A R R U G I A , † A . I O R I O , ‡ N . S . K E Y § and A . S R I V A S T A V A ¶ *Department of Pathophysiology and Transplantation, Angelo Bianchi Bonomi Haemophilia and Thrombosis Center, Fondazione IRCCS C a Granda Ospedale Maggiore Policlinico, Universit a degli Studi di Milano, Luigi Villa Foundation, Milan, Italy; †Faculty of Medicine and Surgery, School of Surgery, The University of Western Australia, Perth, WA, Australia; ‡Department of Clinical Epidemiology and Biostatistics, Hamilton Niagara Hemophilia Program, McMaster University, Hamilton, ON, Canada; §Department of Medicine, University of North Carolina, Chapel Hill, NC, USA; and ¶Department of Haematology, Christian Medical College, Vellore, India
Summary. Haemophilia therapy is experiencing an unprecedented expansion in the number and novelty of clotting factor concentrates. Every product must be licensed by regulatory authorities, primarily on the basis of its safety and efficacy profiles. The low prevalence of haemophilia, and other inherited bleeding disorders, presents a significant challenge to patient recruitment for preauthorization clinical trials, especially given the low frequency of inhibitory antibodies, the major adverse event related to clotting factor exposure. Other challenges include a lack of harmonization between the major regulatory
authorities in certain key areas, the selection of laboratory monitoring methodologies and the difficulty in obtaining high-quality phase IV safety data following authorization. These aspects will be reviewed in this session, which will also highlight the roles played by the World Federation of Hemophilia and International Society on Thrombosis and Haemostasis in the promotion of these discussions.
Clinical trials for market authorization of clotting concentrates – ISTH SSC recommendations – F. Peyvandi
applications for the authorization to market medicinal products. The market authorization of clotting factor concentrates for the treatment and prevention of bleeding in patients with haemophilia in Europe and the US follows the guidelines of the two principal regulatory authorities. Since these guidelines lack a uniform standard of recommendations for pre- and postregistration, a project group assembled by the Factor VIII/IX Subcommittee of the Scientific and Standardization Committee (SSC) of the International Society on Thrombosis and Haemostasis (ISTH) is developing a set of recommendations for the optimal design of clinical studies and trials for clotting factor concentrates for the treatment of haemophilia A and B. Clinical trial design recommendations promoted by the ISTH SSC project group are based on four priority considerations: (i) assessing the harmonized safety and efficacy data required by regulatory agencies for product registration, (ii) exploring the potential impact of alternative statistical approaches and innovative trial design on the preauthorization regulatory requirements for product safety and efficacy determination, (iii) examining the current scientific concepts of immunogenicity and neo-antigenicity and their potential influence on clinical trial design and novel approaches to antibody surveillance and (iv) preparing the
The development of medicinal products is widely seen as an important public health issue. The approval of new drugs requires clinical trials to ensure efficacy and safety. Every country has its own regulatory authority, which is responsible for enforcement of regulations and issuance of guidelines to regulate the marketing of drugs. The evaluation activity of the competent authority, prior to the authorization of a drug and its approval for marketing, aims to ensure that the product offers the necessary guarantees of quality, safety and efficacy. In the United States, the Food and Drug Administration (FDA) is responsible for protecting and promoting public health and supervising the drug approval process. European drug approvals are overseen by the European Medicines Agency (EMA). The EMA is responsible for the scientific evaluation of Correspondence: Nigel Key, 303 Mary Ellen Jones Bldg, CB #7035, Chapel Hill, NC 27599, USA. Tel.: +1-919-966-3311; fax: +1-919-966-7639; e-mail: [email protected] Accepted 24 February 2014 © 2014 John Wiley & Sons Ltd
Keywords: clinical trials, haemophilia, regulatory requirements
137
138
F. PEYVANDI et al.
assessment for the availability of innovative clinical trial design strategies and models that may be suitable for rare diseases such as haemophilia. In the debate on the harmonization between the FDA and the EMA, the ISTH SSC project group has two priority areas, namely, inhibitor assay methodology and the revision of the demand for paediatric trials for pre- and postregistration assessment. Currently, safety aspects in trials for clotting concentrates include viral safety while the most significant adverse event is that of immunogenicity. The development of neutralizing antibodies (inhibitors) against coagulation factor VIII (FVIII), occurring in up to 30% of patients with severe haemophilia A [1] and up to 3% of patients with haemophilia B [2], adds greatly to the difficulty and expense of preventing and treating bleeding episodes. Immunogenicity should be investigated prior to marketing authorization and substantiated with postmarketing studies. The detection of neutralizing antibodies is dependent on the method of measurement employed, and the standardization and optimization of assays used to quantify FVIII inhibitor levels is essential to the meaningful comparison of the results of inhibitor studies. The Nijmegen modification of the Bethesda assay has been recommended by the EMA as the gold standard in preauthorization studies and in postmarketing surveillance. A modification to the current Bethesda/Nijmegen method [3], which replaces FVIII deficient plasma with buffered normal plasma, promises to reduce the potential variability in the test. In addition, for the novel concentrates, additional assays such as an electrochemiluminescent immunoassay and a radioimmunoassay have been used with high sensitivity for neutralizing and non-neutralizing antibodies [4,5]. A Pediatric Investigational Plan (PIP) is required by the EMA in the assessment of new drugs to ensure that there is adequate information about how children fare on an experimental medication before it goes to market. The regulation states that the submission of the PIP should occur no later than at the completion of human pharmacokinetic studies, which is interpreted by the EMA as the end of phase I of the clinical trials. In contrast, the FDA recommends paediatric studies as a postmarketing phase IV commitment. The demands of the EMA, regarding paediatric trials, place an excessive requirement on manufacturers of new haemophilia products, and threaten to create a delay in access to these therapies among adults with this disorder in Europe. The number of children required for premarketing studies by the EMA amounts to at least 50 and 20 children for clinical trials in haemophilia A and B respectively. Given the rarity of haemophilia such paediatric trials will take years to complete. Therefore, these requirements need to be amended for rare disorders. A further proposal discussed by the ISTH SSC project group is to review alternative approaches to trial Haemophilia (2014), 20 (Suppl. 4), 137--144
design for preauthorization studies with respect to safety, particularly immunogenicity. The number of patients typically needed for preauthorization clinical trials is currently 100 for the EMA, and 80 for the FDA. The patient number required by the EMA has been selected by balancing the clinical data package needed to demonstrate efficacy and safety against the availability of patients suffering from a rare disease. The number of patients is expected to be adequate to provide relevant information on general safety aspects and to demonstrate the efficacy of a FVIII product in terms of its ability to restore FVIII levels and achieve haemostasis, i.e. arresting as well as preventing bleeding. In view of the limited number of patients in preauthorization trials, further information mainly focusing on safety aspects must be acquired through postmarketing investigations [6]. For the FDA, the sample size has been defined based on the evaluation of inhibitor development with the goal of showing one or fewer cases with the upper bound of the twosided 95% confidence interval (CI) for the product inhibitor incidence rate being below 6.8%, and the calculation based on an intent-to-treat (ITT) population. Of note, Bayesian and Adaptive Design approaches were considered as alternative statistical models to estimate the value and confidence interval around the inhibitor frequency, but were not determined to add to the efficiency of the espoused model [7]. Ultimately, in this proposed approach, subject requirement and trial duration are only moderately reduced from the current regulatory requirements, to meet the current standards of acceptable preauthorization product safety determination. The ISTH SSC project group attempts to delineate innovative approaches to the clinical design of new product safety (immunogenicity) trials based on the known epidemiology and immunology of FVIII inhibitor development in congenital haemophilia A. A biphasic ‘epidemic’ and ‘endemic’ pattern defines the post-exposure inhibitor incidence, and this parameter should be evaluated in the trial design. Specifically, after 20–50 exposure days (EDs) a high peak ‘epidemic’ rate (up 30%) is observed in previously untreated patients (PUPs) [8], followed by a lifelong low ‘endemic’ incidence of 0.1– 0.6% per patient-year, particularly after 150 EDs [9]. Therefore, the ISTH SSC project group discussed how methodology might best inform the traditional design of the single-arm prelicensure study with respect to study duration and subject number. The same considerations should be applied to trial designs in haemophilia B while taking into account two important differences in the study design, namely a lower prevalence and the more rare development of inhibitors. Clinical studies differ in key characteristics, such as in their definition of previously treated patients (PTPs). When preregistration studies are evaluated in © 2014 John Wiley & Sons Ltd
ISSUES IN CLINICAL TRIALS DESIGN
PTPs, they should be defined in a way that is most suitable to the study of product-related immunogenicity since the incidence of inhibitor formation declines with increasing numbers of infusions, but never disappears. Patients having approximately 150 EDs or more therefore provide an immunotolerant population in which an unusually high incidence of inhibitor formation, suggestive of neoantigenicity, would be unexpected and relatively easy to detect. Another reason for choosing PTPs is that in developed countries, PUPs are relatively few in number, making their recruitment an obstacle to conducting clinical studies. However, the ISTH SSC project group recommended the use of large-scale postmarketing pharmacovigilance studies in PTPs and PUPs for the continued epidemiological study of inhibitor risk in the interval between 50 and 150 EDs, to gain sufficient numbers of patients for statistically valid assessments. A further hurdle to the design of preauthorization clinical trials is the poor awareness of the precise nature of the interactions between therapeutic FVIII products and the recipient’s immune system. A systematic collection of clinical and biological data, e.g. information about genetics and immune status relative to therapeutic products, from subjects entering pre- and postauthorization new product studies will, therefore, be required to scientifically address these questions.
Design of pivotal clinical studies – what are the unresolved issues? – A. Farrugia Unresolved (and unresolvable) issues in haemophilia therapy assessment Given that technology has given treaters and patients potential access to coagulation factors which is unhampered by the limitations of naturally sourced factors, we ask: ‘what are the unresolved issues impeding the translation of this favourable technological landscape into optimal care?’ We suggest that these issues include the following: 1. Funding/reimbursement of treatment products; 2. Rapid assessment and approval of new products, whether by established or new technology; and 3. Understanding of residual hazards, particularly inhibitors.
The evidence-based medicine (EBM) era and haemophilia Increasing appreciation of the need for a societally accountable scrutiny of the process of medicinal market entry gradually led to the principles of EBM becoming the principal modus operandi of regulatory agencies worldwide. It should be noted that bloodderived therapies have been relatively late in their
© 2014 John Wiley & Sons Ltd
139
absorption into this paradigm [10], and that the first European directive for medicinal products in 1965 specifically excluded those from blood and plasma. Their gradual incorporation has tacitly reflected the challenges in establishing the efficacy of a range of products used by small patient populations, and little adherence to EBM is visible even in the current public documents detailing public processes. Hence, the first generations of haemophilia concentrates were either ‘grandfathered’ on the relevant markets, once regulations came into effect, or subjected to standards for quality and safety to the extent that were then appreciated. This minimalist approach was augmented, frequently in a somewhat ad hoc fashion, as the risks of viral infection came to be appreciated. A substantial and effective framework was in place by the mid1990s. The minimization, through this process and others, of the pathogen safety risk has shifted the focus of scrutiny and concern onto inhibitor risk (see below). In relation to efficacy, regulators continue to be responsive to the paucity of patients available for clinical trials, and to grant approval based on processes outside the demands of mainstream randomized controlled trials (RCTs) (see e.g. the process for congenital fibrinogen deficiency corrected by fibrinogen concentrate in [11]). Evidence-based medicine (EBM’s) impact on haemophilia therapy has been generally beneficial through the incorporation of necessary measures to ensure the safety and efficacy of products, although the influence of processes such as the Cochrane Collaboration has been less positive. As continuing issues and new products emerge, an uncertain effect on haemophilia treatment is starting to be evident.
From evidence-based to individualized medicine – the position of haemophilia de Leon [12] has pointed out that one aspect of EBM and the primacy of RCTs, is the assumption of statistical homogeneity in the patient population and in therapeutic effect. Hence, an average response to a treatment represents individual patients adequately [12]. This key presumption, which extends to the highest levels of the EBM hierarchy, is increasingly challengeable through developments in clinical pharmacology and the emergence of pharmacogenetics [13]. Demonstrable heterogeneity in the pharmacokinetics (PK) of drugs [12] through genetic, environmental and physical variables draws us to the new paradigm of ‘personalized medicine’ (PM). This paradigm recognizes heterogeneity and addresses outliers in therapeutic response, in contrast to EBM. The key component of PM is individual heterogeneity in PK, which is also demonstrated in the response
Haemophilia (2014), 20 (Suppl. 4), 137--144
140
F. PEYVANDI et al.
to factor concentrates in haemophilia [14]. In adult patients with haemophilia A, the half-life of FVIII varies between 8 and 24 h [14]. The implications of this for the structuring of treatment protocols have been developed and reviewed by Collins [15]. It may also have potential application in the personalized treatment of individual patients and in the design of pivotal studies for the assessment of effectiveness.
Personalized trials – essential for haemophilia The limitation imposed by rare patient populations has been proposed as the main reason for the abandonment of the standard statistical methodology reflected in RCTs and conformant to mainstream EBM [16]. These deficiencies in EBM lend themselves to the PM paradigm conformant to a patient-centric and personalized approach, and drawing on alternative methods for assessing effectiveness. These alternative methods have been reviewed [17] and the stated commitment of regulators to consider these approaches is encouraging [18,19]. In particular, the use of ‘N-of-1’ clinical trials merits serious consideration as the method of choice in a new paradigm of PM for rare disorders such as haemophilia [20]. The level of therapeutic precision possible from this approach exceeds the ‘one size fits all’ presumption of the conventional RCT considerably [21] and also decreases trial costs [22]. The application of N-of-1 trials to individual patients may be envisioned in the area of approval of current and emerging clotting factor concentrates such as longer acting FVIII and FIX [23]. Mannucci’s concern [24] that the shrinking pool of available patients needed for conforming to the increasing requirements of the European authorities may well be addressed through conducting N-of-1 trials of the product under review using the current standard of care as a control, e.g. comparing longer acting vs. unmodified products. This approach will enhance the quality of the evidence as a conventional blinded and randomized design is possible with subsequent use of Bayesian methodology to allow the projection of the data from individuals to a population of patients [25]. Improved thinking is also needed in specifying the end points of such studies. For example, we have mentioned the focus on longer acting products, but can a longer half-life be viewed as an automatic benefit? The actual clinical need is for the prolongation of the time during which the patients are at higher trough levels and, as Collins has shown [15], longer acting products can also result in the prolongation of suboptimal trough level periods. It should be noted that similar effects in prolonging optimal trough levels can be obtained by increasing the dosage of current established products. Such considerations should be factored into studies for novel molecules, keeping in mind that adverse effects, such as the possibility of Haemophilia (2014), 20 (Suppl. 4), 137--144
inhibitors, are unlikely to be detected in the preapproval phase of assessment, irrespective of the design.
Assessment of long-term safety and efficacy of clotting factor concentrates – A. Iorio The need for studies assessing long-term effectiveness After a drug has completed all the phases required for licensure, there is still much to learn about the real impact of the drug on the overall health of the target population for which it has been approved. Several factors come into play when routine use starts, among which the most relevant are the willingness of doctors to prescribe and of patients to be compliant to the prescribed regimen. Indeed, the effectiveness of a drug is the balance of the expected benefits, usually lower if the patient is not compliant, and the unwanted side effects, very often erratically related to exposure, so that the most likely consequence of non-adherence is a reduction in the net clinical benefit [26]. Furthermore, the role of anticipated and unanticipated drug interactions, and of variability in efficacy with concomitant comorbidities such as renal impairment or in specific populations like children or the elderly, is usually unknown at the time of initial approval. In most cases, new drugs are found to be beneficial in these populations, and long-term assessment of their use generates important knowledge. Finally, side effects uncommon enough to escape the detection power of registration trials may be recognized only when large populations are exposed for a sufficient length of time. This has sometimes led to drug withdrawals, as happened with thalidomide in the 1950s or rosiglitazone in 2011; or the issuing of drug warnings even for commonly prescribed drugs, as was the case for beta-agonists in children [27] or for certain opiates [28,29]. Factor concentrates are no exception to the need for long-term assessment of efficacy and safety, and especially so as we consider new therapeutics with enhanced characteristics, which no longer fit into the category of simple replacement therapy. This important concept cannot be stressed enough: haemophilia is a monogenic disease in which coagulation is disrupted by the absence of FVIII or FIX; the classical treatment for haemophilia is replacement therapy with concentrates of plasma-derived factors from plasma donors. Formerly, the only requirement for approval of such concentrates was to show their ability to restore a physiological factor concentration, stop bleeding and to allow bloodless surgery [30]. Unfortunately, these products were vehicles for the dissemination of blood-borne infections, and virus purification © 2014 John Wiley & Sons Ltd
ISSUES IN CLINICAL TRIALS DESIGN
processes were, therefore, introduced. As a result of this modification of concentrates, it immediately became evident that long-term postmarketing surveillance was needed to confirm both the efficacy of the purification steps [31] and the absence of antigenic modification of the molecule that might induce a higher than expected rate of inhibitors, as was indeed shown for one specific pasteurized concentrate in Belgium and the Netherlands [32]. The introduction of recombinant products, the manipulation of the production process (e.g. B-domain deletion or the introduction of filtration steps) and the more advanced enhancement of the new long-acting molecules have all increased the need for long-term surveillance.
The methodological requirements for an effective assessment As for any clinical research goal, a specific question has to be defined to identify the optimal study design. Broadly speaking, the long-term assessment of safety and efficacy answers the following question: In a broadly defined population of haemophilia patients, what is the net clinical benefit (the balance of efficacy and safety) of the use of a given factor concentrate? Of course, given that the population is a composite one (previously untreated patients, previously treated patients, patients with severe, moderate and mild haemophilia, etc.) and that the treatment goals also vary (on demand, prophylaxis, surgical use) the answer might require different specifications for different cases. Furthermore, given that patients need some form of treatment, long-term assessments are usually comparative in nature: the net clinical benefit of a drug has an intrinsic value, but this is very limited in its practical impact if it does not allow a comparison to the net clinical benefit of alternative treatments. The study design to answer this specific question is a large inception cohort of patients with haemophilia receiving the treatment of interest or alternative treatments [33,34]. Two main strategies are usually employed to build similar inception cohorts. The first strategy is the use of administrative databases, which means using prescription data (e.g. records of FVIII or FIX reimbursement) to identify patients, and diagnosis codes for the outcome (e.g. causes of death, hospital admissions, laboratory assessments of inhibitor levels, etc.). This method works well mostly in small countries with advanced healthcare systems (e.g. Denmark or Norway) or for large health insurance databases (e.g. Medicare or the Veteran’s Administration), and for commonly prescribed drugs and severe events. A more efficient model for a rare disease like haemophilia is to use a dedicated disease database, which is possible if the database is inclusive (covering most of the population) and has a standardized mechanism for reporting events. © 2014 John Wiley & Sons Ltd
141
The second strategy is the setting up of a research database or registry, i.e. a prospective data collection focused on collecting data from clinical practice to answer the above questions. The first category of studies is retrospective and the second is prospective, but this difference is not critical to the scope of long-term assessment. However, two aspects are of the utmost relevance. The first is comprehensiveness. ‘Inception’ is the key term: it means that virtually no eligible patient is excluded, and since the alternative treatment can be also no treatment, ideally every patient qualifies. The second is the set of measures adopted to reduce bias in the collection, analysis and interpretation of the results: ideally, those assessing or reporting the outcome of interest should be blinded to the treatment under study (e.g. the laboratory personnel testing for inhibitors should not know about the treatment of the patient and all laboratory results should automatically flow into the registry), and so should those performing the analysis. Finally, since these cohorts are not randomized, a multivariable or propensity score matching approach including all putative confounders must be adopted. The mandatory reporting of adverse events to health authorities and drug producers has not been mentioned as a research category. In fact, though a necessary activity, it generates a database of cases only, which makes any analysis dependent on linkage to other sources of data. A different model, in this perspective, has recently been proposed by the EMA, with authorization requiring a postmarketing supplementary provision of data to expand the evidence base in a theoretically more feasible way [35,36].
Examples of stakeholder barriers Examples of the first type of studies described above are the UK [37–40], Canadian [41–45] and Italian haemophilia databases [46–51] and the Centers for Disease Control and Prevention (CDC) Universal Data Collection (UDC) system [52–55]. Many other national and international registries are at a more advanced or initial stage of development, and will certainly contribute to the field. Performing long-term assessment of drugs is often not the main goal of these registries, but they have been shown to be able to provide important contributions. Examples of the second type are the EUHASS registry [56], the PedNet RODIN registry [57,58] and the Post Authorization Surveillance Studies (PASS) promoted by industry. They all have a predefined research question and a predefined protocol in common, but differ in other relevant aspects. The main stakeholders in each and every one of the above studies are the manufacturers, the patients, the haemophilia doctors and the regulators. The manufacturers are driven by the willingness to make profit by Haemophilia (2014), 20 (Suppl. 4), 137--144
142
F. PEYVANDI et al.
providing a safe and effective treatment, and thus they fund, in larger or smaller amount and more or less directly, all the above-mentioned studies. In this perspective the PASS studies have two peculiar characteristics. First, the manufacturer owns the data, which sometimes compromises trust in the results, even if ownership sits with industry for most registration trials as well. Second, PASS studies, independently of the promotional use of published reports, are sometimes seen more as promotional activities than research, which, if ever true, would have to be viewed as misconduct on the part of the investigators. The haemophilia patients, as individuals and through their patient organizations, are ideally the second most interested stakeholder, being the beneficiary of the evidence about long-term safety and efficacy, and contributing their own personal data and time. While self-evident, this concept does not fit with the misleading vision that treatment is meant to cancel the disease – thus implicitly leaving no room for tedious data collection activities: unfortunately, until treatment is required, the patient will remain such, and there will be no true progress without full support from the patient community. Haemophilia doctors and regulators, on different levels, play a critical part in the enrolment of every patient in long-term assessment programmes that produce the evidence that may inform future treatment decisions. Having set this framework, what are the barriers to performing long-term assessment studies in haemophilia? The most important is the absence of objective outcomes to measure both efficacy and safety. For efficacy, we have no objective way of assessing the initiation and cessation of joint bleeds [59], nor an effective statistical way of summarizing the number of bleeds over time, as the commonly used annualized bleeding rate (ABR) is far from optimal. Health-related quality of life measures, though available, are far from being routinely available in clinical practice. Regarding safety, the laboratory diagnosis of inhibitors is subject to important variability and the long-term relevance
References 1 Lusher JM, Arkin S, Abildgaard CF, Schwartz RS. Recombinant factor VIII for the treatment of previously untreated patients with hemophilia A. Safety, efficacy, and development of inhibitors. Kogenate Previously Untreated Patient Study Group. N Engl J Med 1993; 328: 453–9. 2 DiMichele D. Inhibitor development in haemophilia B: an orphan disease in need of attention. Br J Haematol 2007; 138: 305–15. 3 Raut S SMIA: A new approach in FVIII inhibitor measurement and standardization. 58th Annual SSC, 2012.
Haemophilia (2014), 20 (Suppl. 4), 137--144
of clinically relevant inhibitors is difficult to establish in the absence of agreed upon guidelines to proceed to immune tolerance therapy. Finally, we have no knowledge of the potential long-term effects of prolonged administration of modified molecules (e.g. conjugates with albumin, PEG, Fc receptor). The second barrier is the absence of standardization in surveillance schemes, which makes it challenging to a) gain power by pooling different datasets; and b) perform comparative assessments. In the latter perspective, the RODIN and EUHASS registries have proven the feasibility of comparative assessments. One final very important barrier is the multiple reporting of patients as both cases and exposed subjects in different studies, which again impairs the value of pooled analysis. Finally, it is important to acknowledge that proprietary data from PASS studies have been made available by companies for independent analysis [60], something which will ideally become routine, as suggested by the AllTrial initiative (http://www.alltrials.net/). In conclusion, there is a clear need for long-term assessment of safety and efficacy, and clear evidence of enormous progress in our capacity to perform such research. Sustained research efforts are needed to overcome existing barriers and to harmonize the various initiatives in the field. It is to be hoped that both the WFH and the ISTH will continue to support and facilitate these demanding efforts.
Disclosures AF was compensated for consultancy services to manufacturers of plasma protein therapies, including one mentioned in the paper. AI received research support from BioGen Idec and Novo Nordisk and honoraria as a consultant from Bayer and BioGen Idec. NSK received research support from Baxter Biosciences and honoraria as a consultant to Bayer, CSL Behring, Novo Nordisk and Baxter. FP has received honoraria for participating as a speaker at satellite symposia and educational meetings organized by Novo Nordisk, CSL Behring, LFB, Grifols, Bayer and Baxter and received research grant funding from Novo Nordisk, Kedrion and Biotest. No funds were received by any author in relation to the present work.
4 Powell JS, Josephson NC, Quon D et al. Safety and prolonged activity of recombinant factor VIII Fc fusion protein in hemophilia A patients. Blood 2012; 119: 3031–7. 5 Tiede A, Brand B, Fischer R et al. Enhancing the pharmacokinetic properties of recombinant factor VIII: first-in-human trial of glycoPEGylated recombinant factor VIII in patients with hemophilia A. J Thromb Haemost 2013; 11: 670–8. 6 Guideline on the clinical investigation of recombinant and human plasma-derived factor VIII products. EMA/CHMP/BPWP/ 144533/2009 Committee for medicinal products for human use (CHMP). Available at http://www.ema.europa.eu/docs/en_GB/doc-
ument_library/Scientific_guideline/2011/08/ WC500109692.pdf. Accessed July 21, 2011. 7 Lee ML, Roth DA. A Bayesian approach to the assessment of inhibitor risk in studies of factor VIII concentrates. Haemophilia 2005; 11: 5–12. 8 Gouw SC, van der Bom JG, Ljung R et al. PedNet and RODIN Study Group. Factor VIII products and inhibitor development in severe hemophilia A. N Engl J Med 2013; 368: 231–9. 9 McMillan CW, Shapiro SS, Whitehurst D, Hoyer LW, Rao AV, Lazerson J. The natural history of factor VIII: C inhibitors in patients with hemophilia A: a national cooperative study. II. Observations on the
© 2014 John Wiley & Sons Ltd
ISSUES IN CLINICAL TRIALS DESIGN
10
11
12
13
14
15 16
17
18
19
20
21
22
23
24
25
initial development of factor VIII: C inhibitors. Blood 1988; 71: 344–8. Farrugia A. The regulatory pendulum in transfusion medicine. Transfus Med Rev 2002; 16: 273–82. Food and Drug Administration. Summary Basis for Regulatory Action - Treatment of acute bleeding in patients with congenital fibrinogen deficiency, afibrinogenemia and hypofibrinogenemia [Internet]. 2009. Available at: http://www.fda.gov/downloads/BiologicsBloodVaccines/BloodBloodProducts/ ApprovedProducts/LicensedProductsBLAs/ FractionatedPlasmaProducts/UCM244507. pdf. Accessed November 25, 2013. De Leon J. Evidence-based medicine versus personalized medicine: are they enemies? J Clin Psychopharmacol 2012; 32: 153–64. Meyer UA. Pharmacogenetics - five decades of therapeutic lessons from genetic diversity. Nat Rev Genet 2004; 5: 669–76. Bj€ orkman S, Folkesson A, Berntorp E. In vivo recovery of factor VIII and factor IX: intra- and interindividual variance in a clinical setting. Haemophilia 2007; 13: 2–8. Collins PW. Personalized prophylaxis. Haemophilia 2012; 18(Suppl 4): 131–5. Gerss JWO, K€ opcke W. Clinical trials and rare diseases. Adv Exp Med Biol 2010; 686: 173–90. Farrugia A. Trialing plasma protein therapies for rare disorders: thinking outside the box. Pharm Policy Law 2009; 11: 345–52. Temple RJ. The regulatory pathway for rare diseases lessons learned from examples of clinical study designs for small populations [Internet]. 2013. Available at: http:// www.fda.gov/downloads/aboutfda/centersoffices/officeofmedicalproductsandtobacco/ cder/ucm352759.pdf. Accessed November 18, 2013. Committee for medicinal products for human USE, (CHMP). guideline on clinical trials in small populations [Internet]. 2006. Available at: http://www.ema.europa.eu/docs/ en_GB/document_library/Scientific_guideline/ 2009/09/WC500003615.pdf. Accessed November 18, 2013. Lillie EO, Patay B, Diamant J, Issell B, Topol EJ, Schork NJ. The n-of-1 clinical trial: the ultimate strategy for individualizing medicine? Per Med 2011; 8: 161–73. Kravitz RL, Duan N, Niedzinski EJ, Hay MC, Subramanian SK, Weisner TS. What ever happened to N-of-1 trials? Insiders’ perspectives and a look to the future. Milbank Q 2008; 86: 533–55. Scuffham PA, Nikles J, Mitchell GK et al. Using N-of-1 trials to improve patient management and save costs. J Gen Intern Med 2010; 25: 906–13. Valentino LA, Ismael Y, Grygotis M. Novel drugs to treat hemophilia. Expert Opin Emerg Drugs 2010; 15: 597–613. Mannucci PM. Evolution of the European guidelines for the clinical development of factor VIII products: little progress towards improved patient management. Haemophilia 2013; 19: 344–8. Zucker DR, Ruthazer R, Schmid CH. Individual (N-of-1) trials can be combined to
© 2014 John Wiley & Sons Ltd
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
give population comparative treatment effect estimates: methodologic considerations. J Clin Epidemiol 2010; 63: 1312–23. Haynes B, Ackloo E, Sahota N, McDonald HP, Yao X. Interventions for enhancing medication adherence. Cochrane Database Syst Rev 2008; CD000011. McMahon AW, Levenson MS, McEvoy BW, Mosholder AD, Murphy D. Age and risks of FDA-approved long-acting b₂adrenergic receptor agonists. Pediatrics 2011; 128: e1147–54. Buckeridge D, Huang A, Hanley J et al. Risk of injury associated with opioid use in older adults. J Am Geriatr Soc 2010; 58: 1664–70. Solomon DH, Rassen JA, Glynn RJ, Lee J, Levin R, Schneeweiss S. The comparative safety of analgesics in older adults with arthritis. Arch Intern Med 2010; 170: 1968–76. Glasziou P, Chalmers I, Rawlins M, McCulloch P. When are randomised trials unnecessary? Picking signal from noise BMJ 2007; 334: 349–51. Tabor E. The epidemiology of virus transmission by plasma derivatives: clinical studies verifying the lack of transmission of hepatitis B and C viruses and HIV type 1. Transfusion 1999; 39: 1160–8. Peerlink K, Arnout J, Gilles G, Saint-Remy J-M, Vermylen J. A Higher than Expected Incidence of Factor VIII Inhbitors in Multitransfused Haemophilia A Patients Treated with Intermediate Purity Pasteurized factor VIII Concentrate. Thromb Haemost 1993; 69: 115–18. Rothman KJ, Greenland S. Causation and causal inference in epidemiology. Am J Public Health 2005; 95(Suppl 1): S144–50. Tugwell P, Haynes B. Assessing claims of causation. In: Haynes RB, Sackett DL, Guyatt GH, Tugwell P eds. Clinical epidemiology: how to do clinical practice research. Philadelphia, PA: Lippincott, Williams and Wilkins, 2006: 356–387. EMA. Guideline on clinical investigation of recombinant and human plasma-derived factor VIII products. 2011. EMA/CHMP/ BPWP/144533/2009. EMA. Guideline on clinical investigation of recombinant and human plasma-derived factor IX products. 2011. Colvin BT, Hay CR, Hill FG, Preston FE. The incidence of factor VIII inhibitors in the United Kingdom, 1990–93. Inhibitor working party. United Kingdom haemophilia centre directors organization. Br J Haematol 1995; 89: 908–10. Darby SC, Keeling DM, Spooner RJD et al. The incidence of factor VIII and factor IX inhibitors in the hemophilia population of the UK and their effect on subsequent mortality, 1977–99. J Thromb Haemost 2004; 2: 1047–54. Hay CRM, Palmer B, Chalmers E et al. Incidence of factor VIII inhibitors throughout life in severe hemophilia A in the United Kingdom. Blood 2011; 117: 6367– 70. Darby SC, Kan SW, Spooner RJ et al. Mortality rates, life expectancy, and causes of death in people with hemophilia A or B in
41
42
43
44
45
46
47
48
49
50
51
52
53
143
the United Kingdom who were not infected with HIV. Blood 2007; 110: 815–25. Giles A, Rivard G, Teitel J, Walker I. Surveillance for factor VIII inhibitor development in the Canadian Hemophilia A population following the widespread introduction of recombinant factor VIII replacement therapy. Transfus Sci 1998; 19: 139–48. Nagel K, Walker I, Decker K, Chana KC, Pai MK. Comparing bleed frequency and factor concentrate use between haemophilia A and B patients. Haemophilia 2011; 17: 872–4. Webert KE, Rivard GE, Teitel J et al. Low prevalence of inhibitor antibodies in the Canadian haemophilia population. Haemophilia 2012; 18: e254–9. Walker I, Pai M, Akabutu J et al. The Canadian Hemophilia Registry as the basis for a national system for monitoring the use of factor concentrates. Transfusion 1995; 35: 548–51. Rubinger M, Lillicrap D, Rivard GE et al. A prospective surveillance study of factor VIII inhibitor development in the Canadian haemophilia A population following the switch to a recombinant factor VIII product formulated with sucrose. Haemophilia 2008; 14: 281–6. Iorio A, Oliovecchio E, Morfini M, Mannucci PM, On PMM, Of B. Italian registry of haemophilia and allied disorders. objectives, methodology and data analysis. Haemophilia 2008; 14: 444–53. Tagliaferri A, Rivolta GF, Iorio A et al. Mortality and causes of death in Italian persons with haemophilia, 1990–2007. Haemophilia 2010; 16: 437–46. Rocca A, Pizzinelli S, Oliovecchio E, Santagostino E, Rocino A, Iorio A. Replacement therapy with recombinant factor IX. A multicentre evaluation of current dosing practices in Italy. Blood Transfus 2010; 9: 60–9. Giampaolo A, Abbonizio F, Puopolo M et al. Consumption of clotting factors in severe haemophilia patients undergoing prophylaxis and on-demand treatment in Italy. Transfus Med 2011; 21: 280–4. Tagariello G, Iorio A, Santagostino E et al. Comparison of the rates of joint arthroplasty in patients with severe factor VIII and IX deficiency : an index of different clinical severity of the 2 coagulation disorders Comparison of the rates of joint arthroplasty in patients with severe factor VIII. Blood 2009; 114: 779–84. Tagariello G, Iorio A, Matino D et al. High rate of spontaneous inhibitor clearance during the long term observation study of a single cohort of 524 haemophilia a patients not undergoing immunotolerance. J Hematol Oncol 2013; 6: 63. Kempton CL, Soucie JM, Abshire TC. Incidence of inhibitors in a cohort of 838 males with hemophilia A previously treated with factor VIII concentrates. J Thromb Haemost 2006; 4: 2576–81. Guh S, Grosse SD, McAlister S, Kessler CM, Soucie JM. Health care expenditures for Medicaid-covered males with haemo-
Haemophilia (2014), 20 (Suppl. 4), 137--144
144
F. PEYVANDI et al.
philia in the United States, 2008. Haemophilia 2012; 18: 276–83. 54 Soucie JM, McAlister S, McClellan A, Oakley M, Su Y. The universal data collection surveillance system for rare bleeding disorders. Am J Prev Med 2010; 38: S475–81. 55 Carpenter SL, Michael Soucie J, Sterner S, Presley R. Increased prevalence of inhibitors in Hispanic patients with severe haemophilia A enrolled in the Universal Data Collection database. Haemophilia 2012; 18: e260–5.
Haemophilia (2014), 20 (Suppl. 4), 137--144
56 Makris M, Calizzani G, Fischer K et al. EUHASS: the european haemophilia safety Surveillance system. Thromb Res 2011; 127: S22–5. 57 Gouw SC, van den Berg HM, Fischer K et al. Intensity of factor VIII treatment and inhibitor development in children with severe hemophilia A: the RODIN study. Blood 2013; 121: 4046–55. 58 Gouw S, van den Berg M, van der Bom JG. Factor VIII prophylaxis and inhibitor development in previously untreated patients
with severe hemophilia A: the RODIN study. Haemophilia 2012; 18: 1–208. 59 Dimichele DM, Blanchette V, Berntorp E. Clinical trial design in haemophilia. Haemophilia 2012; 18(Suppl 4): 18–23. 60 Marcucci M, Cheng J, Oldenburg J et al. Meta-analysis of Post Authorization Safety Studies: worldwide postmarking surveillance of hemophilia A patients treated with antihemophilic factor recombinant plasma/ albumin-free method rAHF-PFM. J Thromb Haemost 2013; 11: 1075.
© 2014 John Wiley & Sons Ltd
DOI: 10.1111/hae.12415
REVIEW ARTICLE
Joint WFH-ISTH session: issues in clinical trial design F . P E Y V A N D I , * A . F A R R U G I A , † A . I O R I O , ‡ N . S . K E Y § and A . S R I V A S T A V A ¶ *Department of Pathophysiology and Transplantation, Angelo Bianchi Bonomi Haemophilia and Thrombosis Center, Fondazione IRCCS C a Granda Ospedale Maggiore Policlinico, Universit a degli Studi di Milano, Luigi Villa Foundation, Milan, Italy; †Faculty of Medicine and Surgery, School of Surgery, The University of Western Australia, Perth, WA, Australia; ‡Department of Clinical Epidemiology and Biostatistics, Hamilton Niagara Hemophilia Program, McMaster University, Hamilton, ON, Canada; §Department of Medicine, University of North Carolina, Chapel Hill, NC, USA; and ¶Department of Haematology, Christian Medical College, Vellore, India
Summary. Haemophilia therapy is experiencing an unprecedented expansion in the number and novelty of clotting factor concentrates. Every product must be licensed by regulatory authorities, primarily on the basis of its safety and efficacy profiles. The low prevalence of haemophilia, and other inherited bleeding disorders, presents a significant challenge to patient recruitment for preauthorization clinical trials, especially given the low frequency of inhibitory antibodies, the major adverse event related to clotting factor exposure. Other challenges include a lack of harmonization between the major regulatory
authorities in certain key areas, the selection of laboratory monitoring methodologies and the difficulty in obtaining high-quality phase IV safety data following authorization. These aspects will be reviewed in this session, which will also highlight the roles played by the World Federation of Hemophilia and International Society on Thrombosis and Haemostasis in the promotion of these discussions.
Clinical trials for market authorization of clotting concentrates – ISTH SSC recommendations – F. Peyvandi
applications for the authorization to market medicinal products. The market authorization of clotting factor concentrates for the treatment and prevention of bleeding in patients with haemophilia in Europe and the US follows the guidelines of the two principal regulatory authorities. Since these guidelines lack a uniform standard of recommendations for pre- and postregistration, a project group assembled by the Factor VIII/IX Subcommittee of the Scientific and Standardization Committee (SSC) of the International Society on Thrombosis and Haemostasis (ISTH) is developing a set of recommendations for the optimal design of clinical studies and trials for clotting factor concentrates for the treatment of haemophilia A and B. Clinical trial design recommendations promoted by the ISTH SSC project group are based on four priority considerations: (i) assessing the harmonized safety and efficacy data required by regulatory agencies for product registration, (ii) exploring the potential impact of alternative statistical approaches and innovative trial design on the preauthorization regulatory requirements for product safety and efficacy determination, (iii) examining the current scientific concepts of immunogenicity and neo-antigenicity and their potential influence on clinical trial design and novel approaches to antibody surveillance and (iv) preparing the
The development of medicinal products is widely seen as an important public health issue. The approval of new drugs requires clinical trials to ensure efficacy and safety. Every country has its own regulatory authority, which is responsible for enforcement of regulations and issuance of guidelines to regulate the marketing of drugs. The evaluation activity of the competent authority, prior to the authorization of a drug and its approval for marketing, aims to ensure that the product offers the necessary guarantees of quality, safety and efficacy. In the United States, the Food and Drug Administration (FDA) is responsible for protecting and promoting public health and supervising the drug approval process. European drug approvals are overseen by the European Medicines Agency (EMA). The EMA is responsible for the scientific evaluation of Correspondence: Nigel Key, 303 Mary Ellen Jones Bldg, CB #7035, Chapel Hill, NC 27599, USA. Tel.: +1-919-966-3311; fax: +1-919-966-7639; e-mail: [email protected] Accepted 24 February 2014 © 2014 John Wiley & Sons Ltd
Keywords: clinical trials, haemophilia, regulatory requirements
137
138
F. PEYVANDI et al.
assessment for the availability of innovative clinical trial design strategies and models that may be suitable for rare diseases such as haemophilia. In the debate on the harmonization between the FDA and the EMA, the ISTH SSC project group has two priority areas, namely, inhibitor assay methodology and the revision of the demand for paediatric trials for pre- and postregistration assessment. Currently, safety aspects in trials for clotting concentrates include viral safety while the most significant adverse event is that of immunogenicity. The development of neutralizing antibodies (inhibitors) against coagulation factor VIII (FVIII), occurring in up to 30% of patients with severe haemophilia A [1] and up to 3% of patients with haemophilia B [2], adds greatly to the difficulty and expense of preventing and treating bleeding episodes. Immunogenicity should be investigated prior to marketing authorization and substantiated with postmarketing studies. The detection of neutralizing antibodies is dependent on the method of measurement employed, and the standardization and optimization of assays used to quantify FVIII inhibitor levels is essential to the meaningful comparison of the results of inhibitor studies. The Nijmegen modification of the Bethesda assay has been recommended by the EMA as the gold standard in preauthorization studies and in postmarketing surveillance. A modification to the current Bethesda/Nijmegen method [3], which replaces FVIII deficient plasma with buffered normal plasma, promises to reduce the potential variability in the test. In addition, for the novel concentrates, additional assays such as an electrochemiluminescent immunoassay and a radioimmunoassay have been used with high sensitivity for neutralizing and non-neutralizing antibodies [4,5]. A Pediatric Investigational Plan (PIP) is required by the EMA in the assessment of new drugs to ensure that there is adequate information about how children fare on an experimental medication before it goes to market. The regulation states that the submission of the PIP should occur no later than at the completion of human pharmacokinetic studies, which is interpreted by the EMA as the end of phase I of the clinical trials. In contrast, the FDA recommends paediatric studies as a postmarketing phase IV commitment. The demands of the EMA, regarding paediatric trials, place an excessive requirement on manufacturers of new haemophilia products, and threaten to create a delay in access to these therapies among adults with this disorder in Europe. The number of children required for premarketing studies by the EMA amounts to at least 50 and 20 children for clinical trials in haemophilia A and B respectively. Given the rarity of haemophilia such paediatric trials will take years to complete. Therefore, these requirements need to be amended for rare disorders. A further proposal discussed by the ISTH SSC project group is to review alternative approaches to trial Haemophilia (2014), 20 (Suppl. 4), 137--144
design for preauthorization studies with respect to safety, particularly immunogenicity. The number of patients typically needed for preauthorization clinical trials is currently 100 for the EMA, and 80 for the FDA. The patient number required by the EMA has been selected by balancing the clinical data package needed to demonstrate efficacy and safety against the availability of patients suffering from a rare disease. The number of patients is expected to be adequate to provide relevant information on general safety aspects and to demonstrate the efficacy of a FVIII product in terms of its ability to restore FVIII levels and achieve haemostasis, i.e. arresting as well as preventing bleeding. In view of the limited number of patients in preauthorization trials, further information mainly focusing on safety aspects must be acquired through postmarketing investigations [6]. For the FDA, the sample size has been defined based on the evaluation of inhibitor development with the goal of showing one or fewer cases with the upper bound of the twosided 95% confidence interval (CI) for the product inhibitor incidence rate being below 6.8%, and the calculation based on an intent-to-treat (ITT) population. Of note, Bayesian and Adaptive Design approaches were considered as alternative statistical models to estimate the value and confidence interval around the inhibitor frequency, but were not determined to add to the efficiency of the espoused model [7]. Ultimately, in this proposed approach, subject requirement and trial duration are only moderately reduced from the current regulatory requirements, to meet the current standards of acceptable preauthorization product safety determination. The ISTH SSC project group attempts to delineate innovative approaches to the clinical design of new product safety (immunogenicity) trials based on the known epidemiology and immunology of FVIII inhibitor development in congenital haemophilia A. A biphasic ‘epidemic’ and ‘endemic’ pattern defines the post-exposure inhibitor incidence, and this parameter should be evaluated in the trial design. Specifically, after 20–50 exposure days (EDs) a high peak ‘epidemic’ rate (up 30%) is observed in previously untreated patients (PUPs) [8], followed by a lifelong low ‘endemic’ incidence of 0.1– 0.6% per patient-year, particularly after 150 EDs [9]. Therefore, the ISTH SSC project group discussed how methodology might best inform the traditional design of the single-arm prelicensure study with respect to study duration and subject number. The same considerations should be applied to trial designs in haemophilia B while taking into account two important differences in the study design, namely a lower prevalence and the more rare development of inhibitors. Clinical studies differ in key characteristics, such as in their definition of previously treated patients (PTPs). When preregistration studies are evaluated in © 2014 John Wiley & Sons Ltd
ISSUES IN CLINICAL TRIALS DESIGN
PTPs, they should be defined in a way that is most suitable to the study of product-related immunogenicity since the incidence of inhibitor formation declines with increasing numbers of infusions, but never disappears. Patients having approximately 150 EDs or more therefore provide an immunotolerant population in which an unusually high incidence of inhibitor formation, suggestive of neoantigenicity, would be unexpected and relatively easy to detect. Another reason for choosing PTPs is that in developed countries, PUPs are relatively few in number, making their recruitment an obstacle to conducting clinical studies. However, the ISTH SSC project group recommended the use of large-scale postmarketing pharmacovigilance studies in PTPs and PUPs for the continued epidemiological study of inhibitor risk in the interval between 50 and 150 EDs, to gain sufficient numbers of patients for statistically valid assessments. A further hurdle to the design of preauthorization clinical trials is the poor awareness of the precise nature of the interactions between therapeutic FVIII products and the recipient’s immune system. A systematic collection of clinical and biological data, e.g. information about genetics and immune status relative to therapeutic products, from subjects entering pre- and postauthorization new product studies will, therefore, be required to scientifically address these questions.
Design of pivotal clinical studies – what are the unresolved issues? – A. Farrugia Unresolved (and unresolvable) issues in haemophilia therapy assessment Given that technology has given treaters and patients potential access to coagulation factors which is unhampered by the limitations of naturally sourced factors, we ask: ‘what are the unresolved issues impeding the translation of this favourable technological landscape into optimal care?’ We suggest that these issues include the following: 1. Funding/reimbursement of treatment products; 2. Rapid assessment and approval of new products, whether by established or new technology; and 3. Understanding of residual hazards, particularly inhibitors.
The evidence-based medicine (EBM) era and haemophilia Increasing appreciation of the need for a societally accountable scrutiny of the process of medicinal market entry gradually led to the principles of EBM becoming the principal modus operandi of regulatory agencies worldwide. It should be noted that bloodderived therapies have been relatively late in their
© 2014 John Wiley & Sons Ltd
139
absorption into this paradigm [10], and that the first European directive for medicinal products in 1965 specifically excluded those from blood and plasma. Their gradual incorporation has tacitly reflected the challenges in establishing the efficacy of a range of products used by small patient populations, and little adherence to EBM is visible even in the current public documents detailing public processes. Hence, the first generations of haemophilia concentrates were either ‘grandfathered’ on the relevant markets, once regulations came into effect, or subjected to standards for quality and safety to the extent that were then appreciated. This minimalist approach was augmented, frequently in a somewhat ad hoc fashion, as the risks of viral infection came to be appreciated. A substantial and effective framework was in place by the mid1990s. The minimization, through this process and others, of the pathogen safety risk has shifted the focus of scrutiny and concern onto inhibitor risk (see below). In relation to efficacy, regulators continue to be responsive to the paucity of patients available for clinical trials, and to grant approval based on processes outside the demands of mainstream randomized controlled trials (RCTs) (see e.g. the process for congenital fibrinogen deficiency corrected by fibrinogen concentrate in [11]). Evidence-based medicine (EBM’s) impact on haemophilia therapy has been generally beneficial through the incorporation of necessary measures to ensure the safety and efficacy of products, although the influence of processes such as the Cochrane Collaboration has been less positive. As continuing issues and new products emerge, an uncertain effect on haemophilia treatment is starting to be evident.
From evidence-based to individualized medicine – the position of haemophilia de Leon [12] has pointed out that one aspect of EBM and the primacy of RCTs, is the assumption of statistical homogeneity in the patient population and in therapeutic effect. Hence, an average response to a treatment represents individual patients adequately [12]. This key presumption, which extends to the highest levels of the EBM hierarchy, is increasingly challengeable through developments in clinical pharmacology and the emergence of pharmacogenetics [13]. Demonstrable heterogeneity in the pharmacokinetics (PK) of drugs [12] through genetic, environmental and physical variables draws us to the new paradigm of ‘personalized medicine’ (PM). This paradigm recognizes heterogeneity and addresses outliers in therapeutic response, in contrast to EBM. The key component of PM is individual heterogeneity in PK, which is also demonstrated in the response
Haemophilia (2014), 20 (Suppl. 4), 137--144
140
F. PEYVANDI et al.
to factor concentrates in haemophilia [14]. In adult patients with haemophilia A, the half-life of FVIII varies between 8 and 24 h [14]. The implications of this for the structuring of treatment protocols have been developed and reviewed by Collins [15]. It may also have potential application in the personalized treatment of individual patients and in the design of pivotal studies for the assessment of effectiveness.
Personalized trials – essential for haemophilia The limitation imposed by rare patient populations has been proposed as the main reason for the abandonment of the standard statistical methodology reflected in RCTs and conformant to mainstream EBM [16]. These deficiencies in EBM lend themselves to the PM paradigm conformant to a patient-centric and personalized approach, and drawing on alternative methods for assessing effectiveness. These alternative methods have been reviewed [17] and the stated commitment of regulators to consider these approaches is encouraging [18,19]. In particular, the use of ‘N-of-1’ clinical trials merits serious consideration as the method of choice in a new paradigm of PM for rare disorders such as haemophilia [20]. The level of therapeutic precision possible from this approach exceeds the ‘one size fits all’ presumption of the conventional RCT considerably [21] and also decreases trial costs [22]. The application of N-of-1 trials to individual patients may be envisioned in the area of approval of current and emerging clotting factor concentrates such as longer acting FVIII and FIX [23]. Mannucci’s concern [24] that the shrinking pool of available patients needed for conforming to the increasing requirements of the European authorities may well be addressed through conducting N-of-1 trials of the product under review using the current standard of care as a control, e.g. comparing longer acting vs. unmodified products. This approach will enhance the quality of the evidence as a conventional blinded and randomized design is possible with subsequent use of Bayesian methodology to allow the projection of the data from individuals to a population of patients [25]. Improved thinking is also needed in specifying the end points of such studies. For example, we have mentioned the focus on longer acting products, but can a longer half-life be viewed as an automatic benefit? The actual clinical need is for the prolongation of the time during which the patients are at higher trough levels and, as Collins has shown [15], longer acting products can also result in the prolongation of suboptimal trough level periods. It should be noted that similar effects in prolonging optimal trough levels can be obtained by increasing the dosage of current established products. Such considerations should be factored into studies for novel molecules, keeping in mind that adverse effects, such as the possibility of Haemophilia (2014), 20 (Suppl. 4), 137--144
inhibitors, are unlikely to be detected in the preapproval phase of assessment, irrespective of the design.
Assessment of long-term safety and efficacy of clotting factor concentrates – A. Iorio The need for studies assessing long-term effectiveness After a drug has completed all the phases required for licensure, there is still much to learn about the real impact of the drug on the overall health of the target population for which it has been approved. Several factors come into play when routine use starts, among which the most relevant are the willingness of doctors to prescribe and of patients to be compliant to the prescribed regimen. Indeed, the effectiveness of a drug is the balance of the expected benefits, usually lower if the patient is not compliant, and the unwanted side effects, very often erratically related to exposure, so that the most likely consequence of non-adherence is a reduction in the net clinical benefit [26]. Furthermore, the role of anticipated and unanticipated drug interactions, and of variability in efficacy with concomitant comorbidities such as renal impairment or in specific populations like children or the elderly, is usually unknown at the time of initial approval. In most cases, new drugs are found to be beneficial in these populations, and long-term assessment of their use generates important knowledge. Finally, side effects uncommon enough to escape the detection power of registration trials may be recognized only when large populations are exposed for a sufficient length of time. This has sometimes led to drug withdrawals, as happened with thalidomide in the 1950s or rosiglitazone in 2011; or the issuing of drug warnings even for commonly prescribed drugs, as was the case for beta-agonists in children [27] or for certain opiates [28,29]. Factor concentrates are no exception to the need for long-term assessment of efficacy and safety, and especially so as we consider new therapeutics with enhanced characteristics, which no longer fit into the category of simple replacement therapy. This important concept cannot be stressed enough: haemophilia is a monogenic disease in which coagulation is disrupted by the absence of FVIII or FIX; the classical treatment for haemophilia is replacement therapy with concentrates of plasma-derived factors from plasma donors. Formerly, the only requirement for approval of such concentrates was to show their ability to restore a physiological factor concentration, stop bleeding and to allow bloodless surgery [30]. Unfortunately, these products were vehicles for the dissemination of blood-borne infections, and virus purification © 2014 John Wiley & Sons Ltd
ISSUES IN CLINICAL TRIALS DESIGN
processes were, therefore, introduced. As a result of this modification of concentrates, it immediately became evident that long-term postmarketing surveillance was needed to confirm both the efficacy of the purification steps [31] and the absence of antigenic modification of the molecule that might induce a higher than expected rate of inhibitors, as was indeed shown for one specific pasteurized concentrate in Belgium and the Netherlands [32]. The introduction of recombinant products, the manipulation of the production process (e.g. B-domain deletion or the introduction of filtration steps) and the more advanced enhancement of the new long-acting molecules have all increased the need for long-term surveillance.
The methodological requirements for an effective assessment As for any clinical research goal, a specific question has to be defined to identify the optimal study design. Broadly speaking, the long-term assessment of safety and efficacy answers the following question: In a broadly defined population of haemophilia patients, what is the net clinical benefit (the balance of efficacy and safety) of the use of a given factor concentrate? Of course, given that the population is a composite one (previously untreated patients, previously treated patients, patients with severe, moderate and mild haemophilia, etc.) and that the treatment goals also vary (on demand, prophylaxis, surgical use) the answer might require different specifications for different cases. Furthermore, given that patients need some form of treatment, long-term assessments are usually comparative in nature: the net clinical benefit of a drug has an intrinsic value, but this is very limited in its practical impact if it does not allow a comparison to the net clinical benefit of alternative treatments. The study design to answer this specific question is a large inception cohort of patients with haemophilia receiving the treatment of interest or alternative treatments [33,34]. Two main strategies are usually employed to build similar inception cohorts. The first strategy is the use of administrative databases, which means using prescription data (e.g. records of FVIII or FIX reimbursement) to identify patients, and diagnosis codes for the outcome (e.g. causes of death, hospital admissions, laboratory assessments of inhibitor levels, etc.). This method works well mostly in small countries with advanced healthcare systems (e.g. Denmark or Norway) or for large health insurance databases (e.g. Medicare or the Veteran’s Administration), and for commonly prescribed drugs and severe events. A more efficient model for a rare disease like haemophilia is to use a dedicated disease database, which is possible if the database is inclusive (covering most of the population) and has a standardized mechanism for reporting events. © 2014 John Wiley & Sons Ltd
141
The second strategy is the setting up of a research database or registry, i.e. a prospective data collection focused on collecting data from clinical practice to answer the above questions. The first category of studies is retrospective and the second is prospective, but this difference is not critical to the scope of long-term assessment. However, two aspects are of the utmost relevance. The first is comprehensiveness. ‘Inception’ is the key term: it means that virtually no eligible patient is excluded, and since the alternative treatment can be also no treatment, ideally every patient qualifies. The second is the set of measures adopted to reduce bias in the collection, analysis and interpretation of the results: ideally, those assessing or reporting the outcome of interest should be blinded to the treatment under study (e.g. the laboratory personnel testing for inhibitors should not know about the treatment of the patient and all laboratory results should automatically flow into the registry), and so should those performing the analysis. Finally, since these cohorts are not randomized, a multivariable or propensity score matching approach including all putative confounders must be adopted. The mandatory reporting of adverse events to health authorities and drug producers has not been mentioned as a research category. In fact, though a necessary activity, it generates a database of cases only, which makes any analysis dependent on linkage to other sources of data. A different model, in this perspective, has recently been proposed by the EMA, with authorization requiring a postmarketing supplementary provision of data to expand the evidence base in a theoretically more feasible way [35,36].
Examples of stakeholder barriers Examples of the first type of studies described above are the UK [37–40], Canadian [41–45] and Italian haemophilia databases [46–51] and the Centers for Disease Control and Prevention (CDC) Universal Data Collection (UDC) system [52–55]. Many other national and international registries are at a more advanced or initial stage of development, and will certainly contribute to the field. Performing long-term assessment of drugs is often not the main goal of these registries, but they have been shown to be able to provide important contributions. Examples of the second type are the EUHASS registry [56], the PedNet RODIN registry [57,58] and the Post Authorization Surveillance Studies (PASS) promoted by industry. They all have a predefined research question and a predefined protocol in common, but differ in other relevant aspects. The main stakeholders in each and every one of the above studies are the manufacturers, the patients, the haemophilia doctors and the regulators. The manufacturers are driven by the willingness to make profit by Haemophilia (2014), 20 (Suppl. 4), 137--144
142
F. PEYVANDI et al.
providing a safe and effective treatment, and thus they fund, in larger or smaller amount and more or less directly, all the above-mentioned studies. In this perspective the PASS studies have two peculiar characteristics. First, the manufacturer owns the data, which sometimes compromises trust in the results, even if ownership sits with industry for most registration trials as well. Second, PASS studies, independently of the promotional use of published reports, are sometimes seen more as promotional activities than research, which, if ever true, would have to be viewed as misconduct on the part of the investigators. The haemophilia patients, as individuals and through their patient organizations, are ideally the second most interested stakeholder, being the beneficiary of the evidence about long-term safety and efficacy, and contributing their own personal data and time. While self-evident, this concept does not fit with the misleading vision that treatment is meant to cancel the disease – thus implicitly leaving no room for tedious data collection activities: unfortunately, until treatment is required, the patient will remain such, and there will be no true progress without full support from the patient community. Haemophilia doctors and regulators, on different levels, play a critical part in the enrolment of every patient in long-term assessment programmes that produce the evidence that may inform future treatment decisions. Having set this framework, what are the barriers to performing long-term assessment studies in haemophilia? The most important is the absence of objective outcomes to measure both efficacy and safety. For efficacy, we have no objective way of assessing the initiation and cessation of joint bleeds [59], nor an effective statistical way of summarizing the number of bleeds over time, as the commonly used annualized bleeding rate (ABR) is far from optimal. Health-related quality of life measures, though available, are far from being routinely available in clinical practice. Regarding safety, the laboratory diagnosis of inhibitors is subject to important variability and the long-term relevance
References 1 Lusher JM, Arkin S, Abildgaard CF, Schwartz RS. Recombinant factor VIII for the treatment of previously untreated patients with hemophilia A. Safety, efficacy, and development of inhibitors. Kogenate Previously Untreated Patient Study Group. N Engl J Med 1993; 328: 453–9. 2 DiMichele D. Inhibitor development in haemophilia B: an orphan disease in need of attention. Br J Haematol 2007; 138: 305–15. 3 Raut S SMIA: A new approach in FVIII inhibitor measurement and standardization. 58th Annual SSC, 2012.
Haemophilia (2014), 20 (Suppl. 4), 137--144
of clinically relevant inhibitors is difficult to establish in the absence of agreed upon guidelines to proceed to immune tolerance therapy. Finally, we have no knowledge of the potential long-term effects of prolonged administration of modified molecules (e.g. conjugates with albumin, PEG, Fc receptor). The second barrier is the absence of standardization in surveillance schemes, which makes it challenging to a) gain power by pooling different datasets; and b) perform comparative assessments. In the latter perspective, the RODIN and EUHASS registries have proven the feasibility of comparative assessments. One final very important barrier is the multiple reporting of patients as both cases and exposed subjects in different studies, which again impairs the value of pooled analysis. Finally, it is important to acknowledge that proprietary data from PASS studies have been made available by companies for independent analysis [60], something which will ideally become routine, as suggested by the AllTrial initiative (http://www.alltrials.net/). In conclusion, there is a clear need for long-term assessment of safety and efficacy, and clear evidence of enormous progress in our capacity to perform such research. Sustained research efforts are needed to overcome existing barriers and to harmonize the various initiatives in the field. It is to be hoped that both the WFH and the ISTH will continue to support and facilitate these demanding efforts.
Disclosures AF was compensated for consultancy services to manufacturers of plasma protein therapies, including one mentioned in the paper. AI received research support from BioGen Idec and Novo Nordisk and honoraria as a consultant from Bayer and BioGen Idec. NSK received research support from Baxter Biosciences and honoraria as a consultant to Bayer, CSL Behring, Novo Nordisk and Baxter. FP has received honoraria for participating as a speaker at satellite symposia and educational meetings organized by Novo Nordisk, CSL Behring, LFB, Grifols, Bayer and Baxter and received research grant funding from Novo Nordisk, Kedrion and Biotest. No funds were received by any author in relation to the present work.
4 Powell JS, Josephson NC, Quon D et al. Safety and prolonged activity of recombinant factor VIII Fc fusion protein in hemophilia A patients. Blood 2012; 119: 3031–7. 5 Tiede A, Brand B, Fischer R et al. Enhancing the pharmacokinetic properties of recombinant factor VIII: first-in-human trial of glycoPEGylated recombinant factor VIII in patients with hemophilia A. J Thromb Haemost 2013; 11: 670–8. 6 Guideline on the clinical investigation of recombinant and human plasma-derived factor VIII products. EMA/CHMP/BPWP/ 144533/2009 Committee for medicinal products for human use (CHMP). Available at http://www.ema.europa.eu/docs/en_GB/doc-
ument_library/Scientific_guideline/2011/08/ WC500109692.pdf. Accessed July 21, 2011. 7 Lee ML, Roth DA. A Bayesian approach to the assessment of inhibitor risk in studies of factor VIII concentrates. Haemophilia 2005; 11: 5–12. 8 Gouw SC, van der Bom JG, Ljung R et al. PedNet and RODIN Study Group. Factor VIII products and inhibitor development in severe hemophilia A. N Engl J Med 2013; 368: 231–9. 9 McMillan CW, Shapiro SS, Whitehurst D, Hoyer LW, Rao AV, Lazerson J. The natural history of factor VIII: C inhibitors in patients with hemophilia A: a national cooperative study. II. Observations on the
© 2014 John Wiley & Sons Ltd
ISSUES IN CLINICAL TRIALS DESIGN
10
11
12
13
14
15 16
17
18
19
20
21
22
23
24
25
initial development of factor VIII: C inhibitors. Blood 1988; 71: 344–8. Farrugia A. The regulatory pendulum in transfusion medicine. Transfus Med Rev 2002; 16: 273–82. Food and Drug Administration. Summary Basis for Regulatory Action - Treatment of acute bleeding in patients with congenital fibrinogen deficiency, afibrinogenemia and hypofibrinogenemia [Internet]. 2009. Available at: http://www.fda.gov/downloads/BiologicsBloodVaccines/BloodBloodProducts/ ApprovedProducts/LicensedProductsBLAs/ FractionatedPlasmaProducts/UCM244507. pdf. Accessed November 25, 2013. De Leon J. Evidence-based medicine versus personalized medicine: are they enemies? J Clin Psychopharmacol 2012; 32: 153–64. Meyer UA. Pharmacogenetics - five decades of therapeutic lessons from genetic diversity. Nat Rev Genet 2004; 5: 669–76. Bj€ orkman S, Folkesson A, Berntorp E. In vivo recovery of factor VIII and factor IX: intra- and interindividual variance in a clinical setting. Haemophilia 2007; 13: 2–8. Collins PW. Personalized prophylaxis. Haemophilia 2012; 18(Suppl 4): 131–5. Gerss JWO, K€ opcke W. Clinical trials and rare diseases. Adv Exp Med Biol 2010; 686: 173–90. Farrugia A. Trialing plasma protein therapies for rare disorders: thinking outside the box. Pharm Policy Law 2009; 11: 345–52. Temple RJ. The regulatory pathway for rare diseases lessons learned from examples of clinical study designs for small populations [Internet]. 2013. Available at: http:// www.fda.gov/downloads/aboutfda/centersoffices/officeofmedicalproductsandtobacco/ cder/ucm352759.pdf. Accessed November 18, 2013. Committee for medicinal products for human USE, (CHMP). guideline on clinical trials in small populations [Internet]. 2006. Available at: http://www.ema.europa.eu/docs/ en_GB/document_library/Scientific_guideline/ 2009/09/WC500003615.pdf. Accessed November 18, 2013. Lillie EO, Patay B, Diamant J, Issell B, Topol EJ, Schork NJ. The n-of-1 clinical trial: the ultimate strategy for individualizing medicine? Per Med 2011; 8: 161–73. Kravitz RL, Duan N, Niedzinski EJ, Hay MC, Subramanian SK, Weisner TS. What ever happened to N-of-1 trials? Insiders’ perspectives and a look to the future. Milbank Q 2008; 86: 533–55. Scuffham PA, Nikles J, Mitchell GK et al. Using N-of-1 trials to improve patient management and save costs. J Gen Intern Med 2010; 25: 906–13. Valentino LA, Ismael Y, Grygotis M. Novel drugs to treat hemophilia. Expert Opin Emerg Drugs 2010; 15: 597–613. Mannucci PM. Evolution of the European guidelines for the clinical development of factor VIII products: little progress towards improved patient management. Haemophilia 2013; 19: 344–8. Zucker DR, Ruthazer R, Schmid CH. Individual (N-of-1) trials can be combined to
© 2014 John Wiley & Sons Ltd
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
give population comparative treatment effect estimates: methodologic considerations. J Clin Epidemiol 2010; 63: 1312–23. Haynes B, Ackloo E, Sahota N, McDonald HP, Yao X. Interventions for enhancing medication adherence. Cochrane Database Syst Rev 2008; CD000011. McMahon AW, Levenson MS, McEvoy BW, Mosholder AD, Murphy D. Age and risks of FDA-approved long-acting b₂adrenergic receptor agonists. Pediatrics 2011; 128: e1147–54. Buckeridge D, Huang A, Hanley J et al. Risk of injury associated with opioid use in older adults. J Am Geriatr Soc 2010; 58: 1664–70. Solomon DH, Rassen JA, Glynn RJ, Lee J, Levin R, Schneeweiss S. The comparative safety of analgesics in older adults with arthritis. Arch Intern Med 2010; 170: 1968–76. Glasziou P, Chalmers I, Rawlins M, McCulloch P. When are randomised trials unnecessary? Picking signal from noise BMJ 2007; 334: 349–51. Tabor E. The epidemiology of virus transmission by plasma derivatives: clinical studies verifying the lack of transmission of hepatitis B and C viruses and HIV type 1. Transfusion 1999; 39: 1160–8. Peerlink K, Arnout J, Gilles G, Saint-Remy J-M, Vermylen J. A Higher than Expected Incidence of Factor VIII Inhbitors in Multitransfused Haemophilia A Patients Treated with Intermediate Purity Pasteurized factor VIII Concentrate. Thromb Haemost 1993; 69: 115–18. Rothman KJ, Greenland S. Causation and causal inference in epidemiology. Am J Public Health 2005; 95(Suppl 1): S144–50. Tugwell P, Haynes B. Assessing claims of causation. In: Haynes RB, Sackett DL, Guyatt GH, Tugwell P eds. Clinical epidemiology: how to do clinical practice research. Philadelphia, PA: Lippincott, Williams and Wilkins, 2006: 356–387. EMA. Guideline on clinical investigation of recombinant and human plasma-derived factor VIII products. 2011. EMA/CHMP/ BPWP/144533/2009. EMA. Guideline on clinical investigation of recombinant and human plasma-derived factor IX products. 2011. Colvin BT, Hay CR, Hill FG, Preston FE. The incidence of factor VIII inhibitors in the United Kingdom, 1990–93. Inhibitor working party. United Kingdom haemophilia centre directors organization. Br J Haematol 1995; 89: 908–10. Darby SC, Keeling DM, Spooner RJD et al. The incidence of factor VIII and factor IX inhibitors in the hemophilia population of the UK and their effect on subsequent mortality, 1977–99. J Thromb Haemost 2004; 2: 1047–54. Hay CRM, Palmer B, Chalmers E et al. Incidence of factor VIII inhibitors throughout life in severe hemophilia A in the United Kingdom. Blood 2011; 117: 6367– 70. Darby SC, Kan SW, Spooner RJ et al. Mortality rates, life expectancy, and causes of death in people with hemophilia A or B in
41
42
43
44
45
46
47
48
49
50
51
52
53
143
the United Kingdom who were not infected with HIV. Blood 2007; 110: 815–25. Giles A, Rivard G, Teitel J, Walker I. Surveillance for factor VIII inhibitor development in the Canadian Hemophilia A population following the widespread introduction of recombinant factor VIII replacement therapy. Transfus Sci 1998; 19: 139–48. Nagel K, Walker I, Decker K, Chana KC, Pai MK. Comparing bleed frequency and factor concentrate use between haemophilia A and B patients. Haemophilia 2011; 17: 872–4. Webert KE, Rivard GE, Teitel J et al. Low prevalence of inhibitor antibodies in the Canadian haemophilia population. Haemophilia 2012; 18: e254–9. Walker I, Pai M, Akabutu J et al. The Canadian Hemophilia Registry as the basis for a national system for monitoring the use of factor concentrates. Transfusion 1995; 35: 548–51. Rubinger M, Lillicrap D, Rivard GE et al. A prospective surveillance study of factor VIII inhibitor development in the Canadian haemophilia A population following the switch to a recombinant factor VIII product formulated with sucrose. Haemophilia 2008; 14: 281–6. Iorio A, Oliovecchio E, Morfini M, Mannucci PM, On PMM, Of B. Italian registry of haemophilia and allied disorders. objectives, methodology and data analysis. Haemophilia 2008; 14: 444–53. Tagliaferri A, Rivolta GF, Iorio A et al. Mortality and causes of death in Italian persons with haemophilia, 1990–2007. Haemophilia 2010; 16: 437–46. Rocca A, Pizzinelli S, Oliovecchio E, Santagostino E, Rocino A, Iorio A. Replacement therapy with recombinant factor IX. A multicentre evaluation of current dosing practices in Italy. Blood Transfus 2010; 9: 60–9. Giampaolo A, Abbonizio F, Puopolo M et al. Consumption of clotting factors in severe haemophilia patients undergoing prophylaxis and on-demand treatment in Italy. Transfus Med 2011; 21: 280–4. Tagariello G, Iorio A, Santagostino E et al. Comparison of the rates of joint arthroplasty in patients with severe factor VIII and IX deficiency : an index of different clinical severity of the 2 coagulation disorders Comparison of the rates of joint arthroplasty in patients with severe factor VIII. Blood 2009; 114: 779–84. Tagariello G, Iorio A, Matino D et al. High rate of spontaneous inhibitor clearance during the long term observation study of a single cohort of 524 haemophilia a patients not undergoing immunotolerance. J Hematol Oncol 2013; 6: 63. Kempton CL, Soucie JM, Abshire TC. Incidence of inhibitors in a cohort of 838 males with hemophilia A previously treated with factor VIII concentrates. J Thromb Haemost 2006; 4: 2576–81. Guh S, Grosse SD, McAlister S, Kessler CM, Soucie JM. Health care expenditures for Medicaid-covered males with haemo-
Haemophilia (2014), 20 (Suppl. 4), 137--144
144
F. PEYVANDI et al.
philia in the United States, 2008. Haemophilia 2012; 18: 276–83. 54 Soucie JM, McAlister S, McClellan A, Oakley M, Su Y. The universal data collection surveillance system for rare bleeding disorders. Am J Prev Med 2010; 38: S475–81. 55 Carpenter SL, Michael Soucie J, Sterner S, Presley R. Increased prevalence of inhibitors in Hispanic patients with severe haemophilia A enrolled in the Universal Data Collection database. Haemophilia 2012; 18: e260–5.
Haemophilia (2014), 20 (Suppl. 4), 137--144
56 Makris M, Calizzani G, Fischer K et al. EUHASS: the european haemophilia safety Surveillance system. Thromb Res 2011; 127: S22–5. 57 Gouw SC, van den Berg HM, Fischer K et al. Intensity of factor VIII treatment and inhibitor development in children with severe hemophilia A: the RODIN study. Blood 2013; 121: 4046–55. 58 Gouw S, van den Berg M, van der Bom JG. Factor VIII prophylaxis and inhibitor development in previously untreated patients
with severe hemophilia A: the RODIN study. Haemophilia 2012; 18: 1–208. 59 Dimichele DM, Blanchette V, Berntorp E. Clinical trial design in haemophilia. Haemophilia 2012; 18(Suppl 4): 18–23. 60 Marcucci M, Cheng J, Oldenburg J et al. Meta-analysis of Post Authorization Safety Studies: worldwide postmarking surveillance of hemophilia A patients treated with antihemophilic factor recombinant plasma/ albumin-free method rAHF-PFM. J Thromb Haemost 2013; 11: 1075.
© 2014 John Wiley & Sons Ltd
