Long-Term Persistence of Antibodies to Adeno-Associated Viral Vectors Following Gene Therapy with scAAV8-LP1-Fixco

Background

Gene therapy for hemophilia B (HB) consisting of a one-time, systemic, administration of adeno-associated viral (AAV) vectors encoding a functional Factor-IX gene is now approved and represents an important and long-awaited milestone. A rise in anti-AAV antibody titers follows systemic administration of AAV vectors. This does not have direct clinical consequences; however, persistence of neutralising anti-AAV antibodies (NAB) at high titers will preclude subsequent successful gene transfer with vector of the same serotype. This would be problematic, in the event that transgene expression should fall below therapeutic levels and repeat dosing might be needed. Furthermore, high titer NAB response to one serotype can cross-react with other serotypes thereby limiting the option of using alternative serotype AAV vectors. We report a longitudinal analysis of anti-AAV antibody response over a 10-year period in severe HB participants following systemic administration of scAAV2/8-LP1-hFIXco from the St Jude-UCL Phase I/II, gene therapy trial (AGT4HB ClinicalTrials.gov:NCT00979238).

Methods

Anti-AAV antibody levels were evaluated in the 10 trial participants following a bolus intravenous infusion of scAAV2/8-LP1-hFIXco at a dose of 2x10 ¹¹vg/kg (N=2), 6x10 ¹¹vg/kg (N=2), or 2x10 ¹²vg/kg (N=6) between 2010 and 2012. Total anti-AAV antibodies (TAB) were measured by ELISA, and levels quantitated using a pooled IgG reference. NAB were quantitated by an in-vitro transduction inhibition assay (TIA) in which plasma samples were preincubated with AAV CMV NanoLuc (MOI: AAV3 1, AAV5 100 and 8 100) prior to evaluation of transduction of HEK 293T cells. Results were normalised to virus control and the 50% inhibition dilution (IC50) was calculated. For comparison, we assessed anti-AAV antibody titers in 38 normal human plasma samples using the same methodology.

Results

At the time of data cut-off (31 ^st Dec 2022) the median follow-up for the cohort was 10.7 years (range: 4-12 years) with follow-up in two participants (one mid dose and the other high dose) being limited to 4 years. A dose dependent AAV8 TAB response was observed with low dose participants having a mean of 84±101mg/ml at 1-year, mid dose 507 mg/ml and high dose 1650±603 mg/ml. This represented a >1400-fold increase from baseline, pre-AAV administration values. These TAB levels were significantly higher than levels following natural infection in 38 seropositive individuals not exposed to recombinant vector (normal range 0.31-34.02mg/ml). In the high dose cohort, TAB titer against AAV3 and AAV5 at 1 year after gene transfer were 145±167 mg/ml (natural infection reference range 0.31-29.61mg/ml) and 96±55 mg/ml (normal range 0.31-36.13mg/ml) respectively, consistent with cross-reactivity between AAV serotypes. There was a decline in TAB levels against all serotypes prior to reaching a plateau. At year 5 the high dose cohort was 928.9 ±556.9mg/ml for AAV8, 90.4±95.6 mg/ml for AAV3 and, 72.8 ±49.1 mg/ml for AAV5.

AAV8 NAB titers had a similar kinetic with IC50 at 1 year 7261.5 ±9812 (low-dose), 15000 (mid-dose) and, 40204± 37823 (high dose). For comparison, AAV8 IC50 were significantly higher than levels following natural infection with wild-type AAV in seropositive individuals (range 5-747) (Mann Whitney p<0.0001). NABs levels declined to an IC50 level of 12907±8246 at 5 years for the high dose cohort, which was still 3000-fold higher than their baseline value (range 5-70.1) (Mann Whitney p<0.0001), and above the threshold for efficient gene transfer. The development of cross-reactive NAB levels was relatively modest and within the range observed in natural seropositive individuals. For the high dose cohort NAB to AAV3 at year 1 was 6536± 1869 (normal range 5- 13502), and for AAV5 at year 1 852± 596 (normal range 5-2143).

Conclusion

Following AAV gene therapy there is a robust dose dependent rise in anti-AAV 8 antibody levels that decline over time, but at 10 years are still significantly above the threshold for re-administration of vector of the same serotype. There was clear evidence of cross reactivity against AAV3 and 5. AAV5 NAB titres were lower than other serotypes declining to levels below a threshold required for efficient gene transfer in the low and mid dose cohort within 10 years. This suggests that, should FIX levels fall below therapeutic levels in this cohort, successful re-administration of an alternative vector may be possible.