Cost-Effectiveness Analysis of Gilteritinib Versus Best Supportive Care (BSC) for the Treatment of Relapsed or Refractory (R/R) FLT3 Mutation-Positive (FLT3mut+) Acute Myeloid Leukemia (AML)

Abstract

Introduction: AML is an aggressive hematopoietic malignancy, and ~30% of patients have mutations in the FLT3 gene. Historically, patients with R/R FLT3mut+ AML experienced dismal survival outcomes. The approval of FLT3-targeted agents has resulted in a paradigm change in the management of these patients. In 2018, gilteritinib was approved as the first targeted agent for R/R FLT3mut+ AML based on results of the phase 3 ADMIRAL trial, showing significantly improved median overall survival (OS) of 9.3 months. Historically, due to limited effective options and tolerability concerns, 11-40% of patients with R/R FLT3mut+ AML received BSC (defined as no active leukemia-directed therapy). Patients managed by BSC still incur substantial resource use and have poor outcomes. The estimated median OS was 2-3 months, with the majority of patients (> 90%) dying within one year. To quantify the incremental benefit and economic value of gilteritinib over BSC, a cost-effectiveness analysis (CEA) model was constructed to model disease trajectory, quality of life impact, and economic impact. Methods: A CEA was developed to compare the costs and effectiveness of gilteritinib and BSC from a US third-party payer's perspective over a lifetime horizon. The model used a 1-month cycle and 3% discount rate and comprised a decision tree followed by a three-stage partitioned survival model. In the decision tree stage, patients were first classified based on whether they received allogeneic hematopoietic stem cell transplantation (HSCT). Following stratification, patients were distributed among three health states: event-free survival (EFS), alive and post-event, and death. Due to the palliative nature of BSC, all patients in the BSC arm were assumed to start in the alive and post-event state and were not eligible to receive subsequent HSCT based on clinical inputs. For patients in the gilteritinib arm, the efficacy inputs were based on the ADMIRAL trial for patients without HSCT or literature (Evers 2018) for those with HSCT. For patients in the BSC arm, the efficacy inputs were based on available literature (Sarkozy 2013). Parametric survival models or hazard ratios were used to predict probabilities in different health states until year 3. After year 3, all patients who remained alive were considered cured with a two-fold increase in mortality risk compared to the general population. Treatment costs (drug, administration, and hospitalization), adverse event (AE) costs, subsequent HSCT costs, medical costs for health states, FLT3 testing, post-progression treatment, and terminal care costs were obtained from public databases and literature. Patients managed by BSC were assumed not to incur any costs related to treatment, AEs, or HSCT. All costs were inflated to 2018 US dollar (USD). Utilities for each health state were derived from the ADMIRAL trial and disutilities for AEs were derived from the literature. The model calculated total costs, and total effectiveness measured by life years (LYs) and quality-adjusted life years (QALYs) for gilteritinib and BSC, respectively. Incremental cost-effectiveness ratios (ICERs), defined as the incremental cost per additional gain in health effect (i.e., LY and QALY), were estimated comparing gilteritinib to BSC. To test the robustness of the results, deterministic sensitivity analyses (DSA) and probabilistic sensitivity analyses (PSA) were performed. Results: Over a lifetime horizon, treatment with gilteritinib, compared to BSC, was associated with an incremental LY gain of 2.82 and QALY gain of 2.32, respectively. With an additional cost of $239,623 relative to BSC, the ICER per one LY gained was $84,922, and ICER per one QALY gained was $103,110. The results were most sensitive to discount rate, cure assumptions, and gilteritinib cost. PSA showed 100% probability of being cost-effective at an acceptable willingness-to-pay threshold of $150,000/QALY in the US. Conclusions: Gilteritinib led to substantial clinical benefit compared to BSC for treatment of R/R FLT3mut+ AML. Despite the large cost difference, the ICER is still within a reasonable range of less than $150,000/QALY, as established by the US Institute for Clinical and Economic Review. Gilteritinib fulfills an important unmet need in the treatment landscape and represents a potential paradigm shift in the current management of R/R FLT3mut+ AML for patients who otherwise would not have received active therapy. Disclosures Zeidan: BeyondSpring: Honoraria; Acceleron Pharma: Consultancy, Honoraria, Research Funding; Celgene Corporation: Consultancy, Honoraria, Research Funding; Boehringer-Ingelheim: Consultancy, Honoraria, Research Funding; Trovagene: Consultancy, Honoraria, Research Funding; Incyte: Consultancy, Honoraria, Research Funding; Takeda: Consultancy, Honoraria, Research Funding; ADC Therapeutics: Research Funding; Abbvie: Consultancy, Honoraria, Research Funding; Cardinal Health: Honoraria; Otsuka: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria, Research Funding; Jazz: Honoraria; Ariad: Honoraria; Agios: Honoraria; Novartis: Honoraria; Astellas: Honoraria; Daiichi Sankyo: Honoraria; Medimmune/AstraZeneca: Research Funding; Seattle Genetics: Honoraria. Pandya:Astellas Pharmaceuticals: Employment. Qi:Astellas: Other: I am an employee of Analysis Group, Inc., which provided paid consulting services to Astellas for the conduct of this study; Analysis Group, Inc.: Employment. Garnham:Astellas: Employment. Yang:Analysis Group, Inc.: Employment; Astellas: Other: I am an employee of Analysis Group, Inc., which provided paid consulting services to Astellas for the conduct of this study. Shah:Astellas: Employment.