Deciding between Multiple Curative Options in Sickle Cell Disease: Cost-Effectiveness of Non-Myeloablative/Reduced Intensity Conditioning Haploidentical Allo-HSCT Vs Gene Therapy Vs Standard of Care in Adult Patients with Sickle Cell Disease

Introduction: The recent approval and high prices of curative gene therapy for sickle cell disease (SCD), a disease known to substantially decrease quality of life, has also invigorated efforts within the allotransplantation (allo-HSCT) space. As another curative option, allo-HSCT for SCD was substantially limited in prior years, especially in adults, due to considerable morbidity and mortality from both chronic graft-versus-host disease (cGVHD) and the myeloablative conditioning (MAC) process. Moreover, the process of identifying a well-matched stem cell donor (MD) was often difficult and expensive. In recent years though, advancements and improvements in non-myeloablative (NMAC) and reduced intensity conditioning (RIC), the use of post-transplant cyclophosphamide (PTCy), and the expansion of donor pools using haploidentical donors (HID) has led to large improvements in mortality, morbidity, cost, and availability. While these curative options continue to refine and evolve to enable long-term quality-of-life improvements in patients with SCD, the cost-effectiveness of this latest wave of allo-HSCT has yet to be established in the United States. Accordingly, we performed the first cost-effectiveness analysis of non-myeloablative/reduced intensity haploidentical allo-HSCT (NMAC/RIC-HID) for young adults with SCD in the United States.

Methods: In this independent analysis free of industry influence, we built a 4-arm Markov model of adult patients with SCD to determine the cost-effectiveness of 1) NMAC/RIC-HID allo-HCST vs 2) gene therapy vs 3) standard-of-care (SOC). To further assess NMAC-HID allo-HSCT against a traditional, historical reference, 4) a MAC-MD allo-HCST arm was also tested. Our model assumed the best-reported performance for each transplant strategy, 100% effectiveness for gene therapy using exa-cel ($2.1 million), and real-world health resource utilization for SOC; conducting the analysis over a lifetime time-horizon from the US health system perspective. Allo-HSCT-specific mortality and cGVHD transition probabilities were derived from a recent international NMAC-HID trial (NCT01850108) and, for MAC-MD recipients, from 154 SCD HLA-matched donor recipients reported across the European Society for Blood and Marrow Transplantation, Center for International Blood and Marrow Transplant Research, and Eurocord. The SOC arm employed disease-severity-specific transition probabilities and costs using data from US commercially insured patients with SCD, consisting of utilization of hydroxyurea, opioid therapy, antibiotics, vaccinations, and blood transfusions across 2007-2017 (Optum). SCD severity was informed by the annual number of vaso-occlusive crises requiring hospitalization. For all arms, age-, sex- and disease-specific background mortality was used. Effectiveness was measured using quality-adjusted life-years (QALYs) and was either derived from EORTC QLQ C-30 and EQ-5D data or was informed from published health state utility values. The primary outcome was the incremental net monetary benefit (iNMB) at a willingness-to-pay threshold of $100,000/QALY; the decision rule in a cost-effectiveness analysis is to choose the strategy with the highest iNMB. We further conducted a probabilistic sensitivity analysis to capture uncertainty in all parameters simultaneously over 10,000 Monte Carlo simulations.

Results: In the base-case, NMAC/RIC-HID allo-HSCT, gene therapy, and SOC accrued 22.6, 24.3, and 14.3 discounted lifetime QALYs at discounted costs of $1.0, $2.5, and $1.0 million, respectively. When assessed against historical MAC-MD allo-HSCT, NMAC-HID allo-HSCT accumulated 0.9 additional QALYs and saved $240,000. In probabilistic sensitivity analysis, NMAC/RIC-HID allo-HSCT was the cost-effective strategy in 100% of 10,000 iterations, with an incremental net monetary benefit of $810,000 [95% credible interval $500,000-1.1 million] vs SOC and $1.3 million [$840,000-$1.5 million] vs gene therapy.

Conclusion: With advancements in NMAC/RIC-HID allo-HSCT outcomes in adults with SCD, allotransplantation offers a cost-effective therapeutic option for lifelong disease amelioration for people living with SCD. The population health benefits of NMAC/RIC-HID allo HSCT versus myeloablative matched donor allo-HSCT are particularly important in the context of an increased donor eligibility pool with the former.