Model of the Impact of Delayed Inhibitor Development on Cumulative Breakthrough Bleeds and Costs in Persons with Hemophilia A Receiving Emicizumab Prophylaxis

Introduction: Approximately 30% of persons with hemophilia A (PwHA) develop neutralizing inhibitors to factor VIII (FVIII) treatment, with peak inhibitor development estimated to occur at 20-50 FVIII exposure days. Emicizumab is a recombinant factor IXa- and factor X-directed antibody, which is Food and Drug Administration approved for prophylaxis in PwHA with inhibitors and is also being evaluated for the treatment of PwHA without inhibitors. We developed a simulation model to estimate the time to inhibitor development in PwHA receiving emicizumab prophylaxis compared with those receiving FVIII prophylaxis.

Methods: We developed a decision tree model using Microsoft Excel to estimate the frequency of breakthrough bleeds (BTBs) and development of inhibitors in severe hemophilia A patients over a 20-year period (starting age of patients: 12 months). Patients in the model were assumed to receive either FVIII or emicizumab as prophylaxis. All patients received FVIII for the treatment of BTBs. BTB rates were obtained from HAVEN trials and published sources. Each day of FVIII infusion, regardless of indication (prophylaxis or bleed treatment) was counted as 1 FVIII exposure day. Patients in either treatment arm could develop inhibitors after 20 exposure days, with the probability of developing inhibitors and the number of exposure days required to develop inhibitors assumed to be the same in both arms. However, the risk of inhibitor development was assumed to vary by FVIII treatment i.e. plasma-derived or recombinant. Inhibitors could be transient (resolving spontaneously within 6 months) or persistent. Patients who developed persistent inhibitors could undergo immune tolerance induction (ITI) and, if unsuccessful, switch to prophylaxis and BTB treatment with bypassing agents (BPAs; activated prothrombin complex concentrate [aPCC] for prophylaxis and recombinant activated factor VII [rFVIIa] or aPCC for bleed treatment) in the FVIII arm, and prophylaxis with emicizumab and BTB treatment with BPAs (rFVIIa or aPCC) in the emicizumab arm, without additional ITI attempts. Patients who resolved their inhibitors were assumed not to regain them during the model time horizon. Treatment costs were calculated using drug unit list price (FVIII short acting, emicizumab, rFVIIa and aPCC), dosing was obtained from package inserts and other published sources, and body weight by age was obtained from US growth charts. Key model outcomes included the time to inhibitor development, the total number of BTBs, and total drug costs (in 2018 USD) over the model time horizon. Sensitivity analyses were performed to identify model inputs that were influential on study outcomes, by varying model inputs (±20%).

Results: Patients receiving FVIII prophylaxis were estimated to reach 20 FVIII exposure days in 4 months, compared with 162 months in patients receiving emicizumab prophylaxis. The total number of estimated BTBs per patient over 1 year was 1.5 and 5.0; over 5 years, 7.5 and 26.8; and over 20 years, 32.3 and 108.9 in the emicizumab and FVIII arms, respectively. The associated cost savings were estimated at $1.7 million over 20 years for each severe hemophilia A patient initiating emicizumab prophylaxis compared with FVIII prophylaxis. The incremental number of bleeds with FVIII compared with emicizumab prophylaxis was most sensitive to the annual number of BTBs with each prophylaxis regimen; however, results consistently showed fewer bleeds with emicizumab prophylaxis.

Conclusions: Assuming the same risk of inhibitor development, emicizumab prophylaxis delayed the development of inhibitors by more than 10 years compared with FVIII prophylaxis. The model suggests that prophylaxis with emicizumab compared with FVIII leads to an avoidance of bleeds and considerable healthcare cost savings.