Type of Factor VIII Product as Inhibitor Risk Factor in Patients with Severe Hemophilia A and Null Mutations.

This retrospective study evaluates the interactions of known determinants of the inhibitor risk in severe and moderate hemophiliacs followed-up at 3 Italian Hemophilia Centers.

clinical data up to inhibitor development or 150 exposure days (EDs) were collected in 393 patients with Factor VIII <5% (FVIII <1%: 331, 84%) and median age of 27 years (interquartile range, IQR:15-36) first treated with plasma-derived (pdFVIII, 70%) or recombinant products (rFVIII, 30%). Exposure to blood components for more than 4 days was an exclusion criterion.

the F8 genotype was known in 334 patients (85%), 57% with null mutations (nonsense mutations, inversions and large deletions). Inhibitors (titer ≥ 0.6 BU/ml on at least two consecutive occasions and FVIII recovery <66% of expected) were detected in 112 patients (28%) after a median of 17 EDs (IQR:10-29), being high responding (HR, peak titer ≥ 5 BU/ml) in 23% and transient in 7%. Table 1 shows the crude and adjusted hazard ratios of potential determinants of inhibitor development in this cohort. Intensive treatment was defined as treatment duration ≥ 3 consecutive days at a dose ≥ 50 IU/kg/day. Sub-analyses were performed in 272 patients who never switched class of product (pdFVIII and rFVIII) and in 314 patients who did not started prophylaxis (≥ 1 regular infusion per week): in both cases the use of rFVIII at first exposure was associated with an increased inhibitor risk (adjusted hazard ratio 5.1, 95%CI 1.5-16.9 and 4.1, 95%CI 1.8-9.4, respectively). Similar results were obtained also excluding 103 patients (26%) who were first exposed to blood components (adjusted hazard ratio 3.5, 95%CI 1.5-7.9). Interestingly, patients with null mutations treated with rFVIII had a higher inhibitor risk than those first treated with pdFVIII (adjusted hazard ratio 3.3, 95%CI 1.4-7.6), while no difference was found in patients with non null mutations treated with rFVIII (hazard ratio 1.4, 95%CI 0.5-3.7). These results were confirmed including only high-responding inhibitors in the analysis.

our results show the influence of the interaction between type of FVIII product and F8 genotype on inhibitor risk, being the use of rFVIII associated with an higher risk in patients with null mutations. In our cohort this association was confirmed also after adjustment for the other known genetic and environmental risk factors of inhibitor development.

No relevant conflicts of interest to declare.