Model of Short and Long-Term Outcomes of Emicizumab Prophylaxis Treatment for Persons with Hemophilia Α

Introduction: Persons with hemophilia A (PwHA) can experience bleeds that result in pain, disability, and a decrease in quality of life over time. Emicizumab is a Food and Drug Administration-approved bispecific antibody to factors IXa/X for prophylaxis of PwHA with inhibitors. Emicizumab was evaluated for the treatment of PwHA without inhibitors in the HAVEN 3 trial. However, its long-term impact is still unknown. This model aimed to estimate the lifetime clinical and economic outcomes of prophylaxis with emicizumab versus factor VIII (FVIII) replacement in persons with severe hemophilia A (HA).

Methods: An Excel-based Markov model was developed to predict long-term clinical and economic outcomes associated with HA treatment options from the US societal perspective. In the model base-case, a hypothetical cohort of patients aged 1 year with severe HA without inhibitors were assigned to prophylaxis with emicizumab or FVIII. Patients in both arms could receive FVIII episodic treatment for breakthrough bleeds. In the model, patients could experience serious adverse events, develop FVIII inhibitors (based on FVIII exposure days), undergo immune therapy induction, develop arthropathy (conditional on number of joint bleeds and Pettersson score), and undergo orthopedic surgery (based on achieving a Pettersson score of 28). Indirect costs associated with productivity loss due to missed work or disabilities were also estimated and applied when patients would be in working age (18-65 years). Patients were assumed to receive life-long treatment. Model inputs were obtained from the HAVEN trials, published literature, and expert opinion. One-way sensitivity analysis was performed to assess inputs that were influential on the model outcomes. We report number of treated bleeds, mean age at inhibitor development, mean age at arthropathy onset, number of orthopedic surgeries, and total direct medical and indirect productivity costs. Outcomes are reported over 1 year, 5 years and lifetime time horizons. Cost results were inflation-adjusted to 2017 USD. Scenario analysis of moderate and severe PwHA was also performed.

Results: The number of treated bleeds and joint bleeds was consistently lower for patients in the emicizumab arm across all time horizons (emicizumab vs. FVIII treated bleeds 1 year: 1.5 vs. 5.8; 5 years: 7.4 vs. 29.1; lifetime: 125.7 vs. 444.0; Joint bleeds 1 year: 1.1 vs. 3.6; 5 years: 5.4 vs. 17.8; lifetime: 92.2 vs. 273.3). Consequently, arthropathy onset (defined as Pettersson score ≥1) was postponed for patients receiving emicizumab versus FVIII prophylaxis (mean age at arthropathy onset: 12.9 vs. 5.4 years). Orthopedic surgeries were rare events in both emicizumab and FVIII arms (0.0 vs. 0.2). FVIII inhibitor development was also delayed in the emicizumab arm (mean age at inhibitor development: 13.9 vs. 1.1 years). Total cumulative direct and indirect costs were lower for emicizumab prophylaxis compared with FVIII prophylaxis for all time horizons (1 year: $97,127 vs. $312,020; 5 years: $603,057 vs. $1,281,059; lifetime: $15,373,443 vs. $20,134,246). Sensitivity analysis indicated that results were driven by drug list price, dosing schedule, and differences in efficacy rate. Results for moderate and severe PwHA were consistent with findings in the severe HA-only population.

Conclusions: The model suggests that prophylaxis treatment of HA with emicizumab can result in a lower number of bleeding events as well as delayed onset of arthropathy and inhibitor development. Findings indicate that prophylaxis with emicizumab leads to better patient outcomes and has a lower financial impact from the US societal perspective.