Multiyear Factor VIII Expression after AAV Gene Transfer for Hemophilia A

Abstract

Background: The goal of gene therapy for patients with hemophilia A is to safely impart long-term stable factor VIII expression that predictably ameliorates bleeding with the use of the lowest possible vector dose.

Methods: In this phase 1-2 trial, we infused an investigational adeno-associated viral (AAV) vector (SPK-8011) for hepatocyte expression of factor VIII in 18 men with hemophilia A. Four dose cohorts were enrolled; the lowest-dose cohort received a dose of 5 × 10¹¹ vector genomes (vg) per kilogram of body weight, and the highest-dose cohort received 2 × 10¹² vg per kilogram. Some participants received glucocorticoids within 52 weeks after vector administration either to prevent or to treat a presumed AAV capsid immune response. Trial objectives included evaluation of the safety and preliminary efficacy of SPK-8011 and of the expression and durability of factor VIII.

Results: The median safety observation period was 36.6 months (range, 5.5 to 50.3). A total of 33 treatment-related adverse events occurred in 8 participants; 17 events were vector-related, including 1 serious adverse event, and 16 were glucocorticoid-related. Two participants lost all factor VIII expression because of an anti-AAV capsid cellular immune response that was not sensitive to immune suppression. In the remaining 16 participants, factor VIII expression was maintained; 12 of these participants were followed for more than 2 years, and a one-stage factor VIII assay showed no apparent decrease in factor VIII activity over time (mean [±SD] factor VIII activity, 12.9±6.9% of the normal value at 26 to 52 weeks when the participants were not receiving glucocorticoids vs. 12.0±7.1% of the normal value at >52 weeks after vector administration; 95% confidence interval [CI], -2.4 to 0.6 for the difference between matched pairs). The participants had a 91.5% reduction (95% CI, 88.8 to 94.1) in the annualized bleeding rate (median rate, 8.5 events per year [range, 0 to 43.0] before vector administration vs. 0.3 events per year [range, 0 to 6.5] after vector administration).

Conclusions: Sustained factor VIII expression in 16 of 18 participants who received SPK-8011 permitted discontinuation of prophylaxis and a reduction in bleeding episodes. No major safety concerns were reported. (Funded by Spark Therapeutics and the National Heart, Lung, and Blood Institute; ClinicalTrials.gov numbers, NCT03003533 and NCT03432520.).

Figure 1.. Factor VIII Activity after Infusion of SPK-8011 in 18 Participants.

The SPK-8011 vector was administered in four doses: 5 × 10¹¹ vector genomes (vg) per kilogram of body weight, 1 × 10¹² vg per kilogram, 1.5 × 10¹² vg per kilogram, and 2 × 10¹² vg per kilogram. On the y axis, factor VIII activity is shown on a log scale. The number to the right of each line is the participant number. Factor VIII activity was determined with the use of a one-stage factor VIII assay. The gray shaded area indicates the range of factor VIII activity in mild hemophilia A. Participants 5 and 12 lost transgene expression.

Figure 2 (facing page).. Preliminary Efficacy of SPK-8011.

Panel A shows the annualized rate of bleeding events before and after vector infusion. Panel B shows the annualized number of exogenous factor VIII infusions before and after vector infusion. Participants 5 and 12 lost all transgene expression within 1 year after vector administration, and the data shown are for the period of transgene expression only. Panel C shows the percentage of participants with no bleeding events after SPK-8011 administration. The number of participants who completed each follow-up interval is indicated below the x-axis.

Figure 3.. Correlation between Factor VIII Activity and the Rate of Spontaneous Bleeding Events.

Panel A shows the factor VIII activity relative to the annualized rate of spontaneous bleeding among Participants 1 through 18. Factor VIII activity was determined with the use of a one-stage factor VIII assay. A Bayesian negative binomial regression analysis with noninformative priors was used to analyze the relationship between the annualized rate of spontaneous bleeding after vector administration and the proportion of time participants had factor VIII activity that was greater than 10% of the normal value. The blue curve indicates the regression line, and the shaded area the 95% credible interval. Each X denotes one participant, and the participant numbers are shown. Data shown for Participants 5 and 12 are based on observations before the loss of transgene expression. Panel B shows the posterior density of the mean annualized rate of spontaneous bleeding among participants with factor VIII activity of more than 10% of the normal value at least half the time; among these participants, there was a 99% probability of an annualized rate of spontaneous bleeding of less than 1 event. The gray vertical line indicates an annualized rate of spontaneous bleeding of 1 event.