Spk-9001: Adeno-Associated Virus Mediated Gene Transfer for Hemophilia B Achieves Sustained Factor IX with Minimal Immune Response

Background: Data from earlier hemophilia B (HB) AAV-mediated liver gene transfer trials demonstrated a dose-dependent, capsid-specific immune response that may result in clearance of transduced hepatocytes and loss of transgene expression (Manno et al. 2006, Mingozzi et al. 2007). This has not posed major safety concerns, but may limit efficacy. Prior work incorporated the use of steroids to abort this immune response and maintain factor IX (FIX) expression (Nathwani et al 2014). The percent of transgene expression lost increased with number of elapsed days from transaminase elevation to steroid initiation. Once vector responsiveness to steroids is established, the need for steroids is not a limitation in and of itself but highlights the requirement to reliably and expeditiously recognize an immune response and initiate steroids. Here we present our immunomonitoring data following infusion of SPK-9001, an AAV vector designed to achieve therapeutic FIX:C at a low vector dose to minimize capsid-specific T cell responses.

Method: Ten HB subjects were infused with 5 x 10¹¹ vg/kg SPK-9001, containing a transgene cassette encoding FIX-Padua that confers ~8-12 - fold higher specific activity than wild-type factor IX (Simioni et al. 2009, Crudele et al. 2016). T cell responses to the SPK-9001 AAV capsid and transgene product were monitored post-infusion using a validated interferon-γ enzyme-linked immunospot (ELISpot) assay, with a threshold for positivity of > 50 spot forming units (SFU) per 10⁶ PBMC.

Results: As of 7/26/17, 10 subjects are 26 -78 weeks post SPK-9001 infusion with mean sustained FIX:C of 33.8±18.1% (mean±SD). There were no vector or procedure-related adverse events. All subjects discontinued prophylaxis and had a significant reduction in the annualized bleeding rate (p<0.0001). No patients exhibited a significant T cell response to the transgene product, including the Padua mutation site. T cell responses against the viral capsid generally fell into 3 categories: 5 subjects had no detectable response to capsid; 3 subjects had a transient T cell response that peaked at week 2 that quickly resolved without intervention; 2 subjects demonstrated a T cell response that persisted for longer than 3 weeks. The persistent T cell responses observed in the 2 subjects correlated with an asymptomatic increase of liver transaminases. One subject was administered a course of prednisone starting on day 56, after which FIX:C levels stabilized and T cell reactivity quickly returned to baseline levels. An orthogonal peptide matrix approach identified 1 candidate AAV epitope, SEYQLPYVL, which is a MHC Class I predicted binder for the subject's HLA-B*18 haplotype according to bioinformatics prediction algorithms. This epitope is highly conserved across AAV serotypes. In response to grade I transaminase toxicity, the second subject was administered steroids starting on day 36. Concurrent ELISPOT values were elevated (1260 SFU/10⁶ PBMC), but not associated with a decline in FIX:C and the subject maintained FIX:C in the normal range. Like the subject's LFTs, the ELISPOT response markedly decreased with steroids while his FIX:C remained stable at ~80%. This subject continues to demonstrate a low-positivity to AAV capsid peptides through 32 weeks and off steroids, but the persistent T cell response has not been associated with a decline in FIX:C or impact on overall efficacy. Further ELISPOT analysis on this subject identified 4 AAV T cell epitopes that are predicted binders for 2 of this subject's alleles, HLA-A*03 and HLA-B*07.

Conclusion:

We report consistent levels of sustained plasma FIX:C of 33.8±18.1% (mean±SD) in 10 subjects to date. Observed T cell responses to the AAV capsid occurred in 2/10 subjects that required steroid intervention. The early administration of immunosuppression and subsequent decrease in T cell response in the absence of a corresponding loss in FIX:C suggests that prompt recognition of a capsid immune response and early steroid intervention is critical. Furthermore, persistence of detectable T cell responses late in the time-course, with no association with a diminution in FIX:C, may suggest that circulating T cells may be present long after transduced hepatocytes cease presenting capsid peptides. Preliminary data suggest that our validated method for assessing T cell responses can detect responses that require management with steroids to ensure continuous FIX expression.