Effect on Joint Health of Routine Prophylaxis with Bayer’s Sucrose-Formulated Recombinant Factor VIII (rFVIII-FS) in Adolescents and Adults Previously Treated on Demand: MRI Analyses from the 3-Year Spinart Study

Introduction: The benefits of primary prophylaxis with a factor VIII (FVIII) product in pediatric patients with severe hemophilia A are well established. Fewer data are available on the benefits of secondary prophylaxis (started after ≥2 joint bleeds but before the onset of documented joint disease). The 3-year SPINART study compared the efficacy and safety of routine prophylaxis vs on-demand treatment in adolescents and adults with severe hemophilia A, all of whom were treated with Bayer's sucrose-formulated recombinant FVIII (rFVIII-FS). Primary 3-year data on magnetic resonance imaging (MRI) joint assessments in SPINART have been recently reported. Here we present additional analyses of the SPINART 3-year MRI data.

Methods: SPINART was a 3-year, randomized, controlled, parallel-group, open-label study conducted at 31 centers in the United States, Bulgaria, Romania, and Argentina. Male patients aged 12–50 years were eligible for SPINART if they had severe hemophilia A (FVIII:C <1%), ≥150 exposure days to any FVIII product, no current evidence or history of FVIII inhibitors, no prophylaxis for >12 consecutive months in the past 5 years, and 6–24 documented bleeding events or treatments in the previous 6 months. Eligible patients were randomly assigned 1:1 to on-demand treatment or prophylaxis. Patients assigned to prophylaxis received rFVIII-FS 25 IU/kg 3 times weekly; in patients with ≥12 bleeding episodes per year, dose increases of 5 IU/kg were permitted at years 1 and 2. All patients underwent MRI assessments at baseline and year 3 to evaluate the structure of 6 index joints (knees, ankles, elbows). Each MRI was read by 3 radiologists blinded to treatment assignment who independently completed the Extended MRI (eMRI) scale. The eMRI scale has 2 domains (soft tissue, osteochondral), and total eMRI scores range from 0 to 45 based on soft-tissue domain scores of 0 to 9 and osteochondral domain scores of 0 to 36; higher eMRI scores indicate greater joint structural damage. Change from baseline to year 3 in eMRI total score based on all 6 index joints was analyzed for the following baseline characteristics: region (US vs non-US), age (≤29 vs >29 years), and number of bleeding episodes in the previous 6 months (<8 vs ≥8). For patients with target joints, change from baseline to year 3 in eMRI scores in the worst target joint was analyzed using analysis of covariance adjusted for bleeding frequency during the prior 6 months.

Results: Eighty-four patients (42 per treatment group) were enrolled in the SPINART study. Target joint analysis data for patients with target joints who completed the study were available for 28 on-demand and 20 prophylaxis patients. Least squares (LS) mean change from baseline to year 3 in eMRI total score in the analyzed target joint was 0.91 (95% CI, –0.06 to 1.88) and 1.09 (95% CI, 0.12–2.07) for the on-demand and prophylaxis groups, respectively; the difference was not statistically significant (LS mean difference, 0.18; 95% CI, –1.05 to 0.70; P=0.68). Results for the subgroup analyses are shown in the Table.

Conclusions: Over 3 years of treatment, change in eMRI total score for target joints was similar for the on-demand and prophylaxis groups in SPINART. In the prophylaxis group, progression of joint structural damage after 3 years of treatment, as indicated by changes in eMRI total scores based on all 6 index joints, did not differ by number of bleeding episodes in the preceding 6 months but appeared to be less pronounced among younger patients compared with older patients and among those in the non-US group compared with the US group; results by age and region in the on-demand group were opposite of those seen in the prophylaxis group. These results must be interpreted with caution given the small patient numbers, the possibility that the study duration was not sufficient to show changes on MRI, and the fact that target joints were assessed. These results may underscore the importance of preventing target joint development and show that once a target joint has developed, MRI may not show reversal of pre-existing damage despite prophylaxis.