Challenges and progress in gene therapy for hemophilia A

In this issue, Powell and colleagues (page 2038) report the results of a phase 1 gene therapy clinical trial for hemophilia A, based on intravenous injection of a retroviral vector encoding B-domain–deleted factor VIII (FVIIIΔB). This phase 1 trial was based on encouraging preclinical studies, mostly in rabbits,¹  and essentially confirms the safety of this approach in patients. The vectors could be detected in the peripheral blood mononuclear cells for at least a year. Although some participants had detectable circulating FVIII levels (>1%) on repeated occasions, experienced fewer bleeding episodes, and required fewer FVIII protein infusions compared with historic rates, the clinical benefits were overall rather limited and a dose response was lacking. It appears therefore that the preclinical studies that constituted the basis of this trial may not have accurately predicted the vector doses required to achieve therapeutic FVIII levels. However, the limited efficacy of this particular gene therapy approach in adult patients was not entirely unexpected, due to the inability of retroviral vectors to transduce nondividing cells.

Previous studies had shown that stable therapeutic levels of FVIII or FIX could only be obtained in neonatal hemophilic mouse and dog models or in adult mice that received hepatocyte growth factor to stimulate hepatocyte cell division.^2,3  This inherent limitation of retroviral vectors justifies the development of vectors that can also transduce nondividing cells, such as lentiviral⁴  adeno-associated viral vectors (AAV). In this same issue, Scallan and colleagues (page 2031) report stable expression (> 14 months) of therapeutic levels of FVIII (2%-4%) in 2 dogs with hemophilia A following liverdirected gene therapy using an AAV-based vector encoding FVIIIΔB. Although AAV has successfully been used for gene therapy in hemophilia B dogs⁵  and results from clinical trials for hemophilia B are encouraging, progress in hemophilia A gene therapy has been hampered by the inherent, limited packaging capacity of AAV and the relatively large size of the B-domain–deleted FVIIIΔB cDNA. Scallan and colleagues showed that this limitation could be overcome by using small regulatory elements to drive FVIII expression, in accordance with previous reports.⁶  Although recent studies had shown that therapeutic levels of FVIII could be achieved in hemophilia A dogs with no apparent toxicity following gene therapy,⁷  the work by Scallan and colleagues is an important step forward since it is the first demonstration that long-term phenotypic correction of the bleeding diathesis could be achieved, albeit partial, in a clinically relevant, large animal model of hemophilia A. However, the reason for the lack of a dose response is not clear and warrants further studies in larger cohorts. Additional improvements in vector design and increased gene transfer efficiencies will be required to further increase FVIII expression levels, potentially allowing the use of lower, clinically acceptable vector doses. The simultaneous development of different gene therapy approaches is justified to bring a cure for hemophilia A one step closer to reality.