Hyperactive Factor IX Padua: A Game-Changer for Hemophilia Gene Therapy

Main Text

For the past 25 years, the development of gene therapy for hemophilia has fueled technological innovations and led to emerging insights that benefited the field at large.1, 2 It was particularly encouraging that sustained expression of coagulation factor IX (FIX) was achieved after liver-directed gene therapy with adeno-associated viral vectors (AAVs) in patients with severe hemophilia B.³ Neverthless, the circulating FIX levels in that study—2%–5% range of normal—fell short of preventing spontaneous and trauma-induced bleeds. Moreover, high vector doses resulted in liver inflammation and hepatotoxicity, requiring transient immunosuppressive treatment with corticosteroids. Therefore, it was important to obtain higher FIX activity levels, while reducing the overall vector load, and minimize the risk of liver inflammation. In a report published in the New England Journal of Medicine,⁴ George and colleagues (Children’s Hospital of Philadelphia) and Spark Therapeutics have now achieved this goal and demonstrated sustained FIX coagulant activity of 33.7 ± 18.5% (range, 14% to 81%) after AAV-based gene therapy in all trial participants (9 out of 9) receiving a vector dose of 5 × 10¹¹ vector genomes (vg)/kg. Moreover, there was no elevation in liver enzyme levels in a majority of the subjects (7 out of 9), suggesting a reduced risk of liver inflammation. The increased FIX activity after gene therapy enabled the discontinuation of baseline prophylaxis and nearly eliminated spontaneous bleeds and the need for exogenous recombinant FIX. This is an important advance for the field and brings a cure for hemophilia one step closer to reality.

The improvement in vector performance in this recent clinical trial can be ascribed mainly to the use of a modified FIX transgene encoding a hyperactive mutant FIX protein containing just a single point-mutation (i.e., R338L). This hyperactivating mutation, designated as FIX-R338L-Padua, was initially identified by Dr. Simioni and colleagues⁵ (University of Padua) in thrombophilic patients that expressed a FIX protein with an 8-fold increase of specific activity compared to wild-type FIX. Presumably, this R338L mutation results in more efficient thrombin generation. In 2012, we initially demonstrated that the incorporation of this hyperactivating R338L mutation in the FIX gene could augment the efficacy of gene therapy for hemophilia B up to 5- to 10-fold after liver-directed gene therapy (in collaboration with Naldini and colleagues).⁶ This allowed for higher FIX expression levels at lower, and thus potentially safer, vector doses. This was independently confirmed by Dr. Arruda and colleagues,⁷ following muscle-directed delivery with AAVs encoding the FIX-R338L-Padua. Although muscle-directed gene delivery had previously been attempted in clinical trials for hemophila B,⁸ obtaining robust levels of circulating FIX had been particularly challenging. To overcome these limitations, we and others therefore delivered the FIX-R338L-Padua transgene to hepatocytes with AAVs yielding robust FIX procoagulant activity, consistent with our initial lentiviral studies.9, 10, 11, 12 Thus, the “Padua-effect” was independent of the vector used or the target cells.6, 7, 9, 10, 11, 12, 13, 14 Large animal studies in the canine hemophilia B model confirmed the enhanced functionality of the FIX-R338L-Padua transgene after AAV or lentiviral transduction.7, 10, 15 Therefore, these encouraging preclinical studies in hemophilic animal models justified using the FIX-R338L-Padua for gene therapy in patients suffering from hemophilia B.

The first clinical trial based on the use of FIX-R338L-Padua in hemophilia B (BAX335, Shire; ClinicalTrials.gov NCT 01687608) showed sustained FIX activity for 12 months in the 20%–25% range in one of the patients receiving a self-complementary AAV8- FIX-R338L-Padua vector at a dose of 10¹² vg/kg, consistent with the lack of bleeds, obviating the need for regular FIX infusions (https://www.baxter.com/news-media/newsroom/press-releases/2015/06_24_15_bax335.page). Although FIX activity levels peaking above 50% could be attained in the high-dose cohort (i.e., 3 × 10¹² vg/kg), these levels were not sustained and declined to basal levels. The recent clinical trial results described by George et al.⁴ not only confirm the earlier preclinical findings, but also go beyond the BAX335 trial by demonstrating a more consistent response and sustained FIX expression in all of the trial participants. The exact reason for the different outcome between these two AAV trials based on the FIX-R338L-Padua is not entirely clear but may be due, at least in part, to differences in immune reactions, although the exact mechanisms remain elusive. It is noteworthy that differences in vector configuration (single-stranded versus double-stranded), expression cassette design, and promotor used (ApoE-HCR/AAT versus TTR) and the capsid itself (AAV-Spark100 versus AAV8) may have influenced the eventual outcome. As is often the case in gene therapy, the devil is in the details, and an apparent subtle change in vector design and/or manufacturing could potentially yield divergent results in patients. However, even within the same trial, including the recent Spark Pfizer study, significant differences in expression kinetics and/or clotting factor activity levels are apparent among different subjects.⁴ Moreover, in a recent AAV-based gene therapy trial for hemophilia A (BioMarin), factor VIII levels varied greatly among the trial subjects, even reaching sustained supra-physiologic levels corresponding to 200% of normal.¹⁶ Since epidemiological studies indicate that supra-physiologic FVIII levels are associated with a higher thrombotic risk,¹⁷ this causes some specific safety concerns. Hence, the impact of this inter-patient variability should not be underestimated because it has important consequences for patients’ quality of life.

From the patient’s perspective, it still looks a bit like a lucky draw because, ultimately, steady-state clotting factor levels cannot be predicted at this point. Steady-state FIX or FVIII levels will dictate whether the treated patients will still need to depend on factor substitution therapy, despite the gene therapy intervention. Although levels in the 30%–40% range can significantly minimize the risk of spontaneous bleeds, the risk for trauma-induced bleeds remains and it does not yet represent a bona fide cure. It is known that patients react differently to factor replacement therapy using recombinant or plasma-derived factors which may, at least in part, explain some of the variation in clincal responses. As gene therapy is picking up momentum in the clinic, it is likely that we will gain a better understanding of the molecular and cellular mechanisms that influence inter-patient variation in clinical responses. It is already becoming increasingly clear that the presence of pre-existing anti-AAV antibodies, even at low-titers, could significantly impede gene transfer and result in lower FIX levels, as confirmed in the current study.⁵

The trial data suggest that administration of lower vector doses (i.e., 5 × 10¹¹ vg/kg as opposed to 2 × 10¹² vg/kg or higher) may have contributed to a decreased risk of liver inflammation and hepatotoxicity, since only 2 patients out of 9 developed these unwanted side effects. Though this is encouraging, ideally we would want to eliminate this risk altogether, obviating the need for transient immune suppression with glucocorticoids. It is particularly intriguing that the participants in the trial that had an AAV capsid-directed immune response and liver inflammation also had the most rapid initial rise in FIX activity. Therefore, it cannot be ruled out that the immune response may somehow have facilitated gene expression, as suggested by George and colleagues.⁵ It seems more likely, however, that AAV transduction may have been more efficient in those patients, resulting not only in a more rapid onset of FIX expression levels, but a higher load of AAV capsids in the transduced hepatocytes. Consequently, this would have increased the likelihood of developing an AAV capsid-specific T cell response that contributes to the immune rejection of the transduced hepatocytes and the ensuing transaminitis.

Skeptics initially argued that shifting to the FIX-R338L-Padua might trigger thrombosis and increase the likelihood of developing neutralizing antibodies to the mutated FIX protein. The preclinical data did not support this, and the trial data have now confirmed the lack of thrombotic complications and absence of inhibitory antibodies to the FIX-R338L-Padua variant.⁵ Therefore, the aforementioned preclinical studies5, 6, 7, 9, 10, 11, 12, 13, 15 and the recent clinical trials make a strong case for FIX-R338L-Padua, which can be considered a gamechanger for the field. Indeed, other gene therapy programs (including UniQure’s AMT-061) are now replacing wild-type FIX with the hyperactive FIX-R338L-Padua version for their pivotal trials.

The city of Padua boasts some of the nicest examples of renaissance architecture and also houses the University of Padua, one of the oldest in Europe. It played an important role during the renaissance period, with Galileo Galilei, Nicolaas Copernicus, William Harvey, and Andreas Vesalius as some of their most famous alumni. The field of gene therapy is currently experiencing its own renaissance and, at least for hemophilia, Padua may again have had no small part to play.

Conflicts of Interest

The authors are not involved in the Spark/Pfizer study. The have obtained research grants from Shire, Pfizer, and Bayer and have been consultants and/or paid speakers for Shire, Pfizer, Bayer, and Biotest. They are inventors on patent applications and granted patents in the field of gene therapy, including for hemophilia.