OBJECTIVES : Prophylaxis with recombinant factor VIII (rFVIII) is the current standard of care for haemophilia A (HA). Several approaches have been used to extend the half-life of rFVIII to improve outcomes with prophylaxis in HA. In the absence of comparative studies to establish the relative efficacy of these extended half-life (EHL) products, indirect comparison of clinical trial data provides an alternative approach. This analysis was performed to evaluate the relative efficacy of two EHL therapies approved for the prophylactic treatment of HA: recombinant FVIII Fc (rFVIIIFc) and pegylated rFVIII (BAY 94-9027).

METHODS : rFVIIIFc and BAY 94-9027 were compared using matching-adjusted indirect comparison. Data for rFVIIIFc were extracted from the individualised prophylaxis arm of the A-LONG phase 3 clinical trial (n=117) and from the pooled prophylaxis population of the PROTECT VIII phase 2/3 study (n=110), corresponding to the approved treatment regimens of each product. Following matching for baseline characteristics of age, body mass index, prior bleeds, target joints and ethnicity, mean annualised bleeding rate (ABR) and the proportion of patients with zero bleeds were determined for rFVIIIFc and BAY 94-9027.

RESULTS : After matching, the effective sample size for A-LONG was 81 patients. Mean ABR was lower in the rFVIIIFc individualised prophylaxis group versus the BAY 94-9027 pooled prophylaxis population (2.9 vs 4.9). This difference was clinically relevant and statistically significant (mean difference ‒1.8; 95% CI ‒3.4, ‒0.2). The difference in the proportion of patients with zero bleeds between rFVIIIFc (45.5%) and BAY 94-9027 (38.2%) was not statistically significant (odds ratio: 1.35; 95% CI 0.77, 2.36).

CONCLUSIONS : The results of this indirect treatment comparison show that individualised prophylaxis with rFVIIIFc results in lower ABR compared to BAY 94-9027, a difference that is both statistically significant and clinically relevant. Our findings provide important data to inform treatment decisions and will be incorporated into further cost/utility research.