Overall survival of glasdegib in combination with low-dose cytarabine, azacitidine, and decitabine among adult patients with previously untreated AML: comparative effectiveness using simulated treatment comparisons.

Tremblay, Gabriel; Westley, Tracy; Cappelleri, Joseph C; Arondekar, Bhakti; Chan, Geoffrey; Bell, Timothy J; Briggs, Andrew; (2019) Overall survival of glasdegib in combination with low-dose cytarabine, azacitidine, and decitabine among adult patients with previously untreated AML: comparative effectiveness using simulated treatment comparisons. CLINICOECONOMICS AND OUTCOMES RESEARCH, 11. pp. 551-565. ISSN 1178-6981 DOI: https://doi.org/10.2147/CEOR.S203482

Abstract

BACKGROUND: Until recently, treatments for older patients with AML ineligible to receive intensive chemotherapies were limited to hypomethylating agents, low-dose cytarabine (LDAC), or clinical trials. In 2018, the FDA approved combination glasdegib (GLAS) plus LDAC based on Phase II results demonstrating improved overall survival (OS) versus LDAC alone in previously untreated AML. However, no randomized clinical trials have directly compared GLAS + LDAC with other AML treatments. OBJECTIVE: Using both indirect treatment comparison (ITC) and simulated treatment comparison (STC), which adjusts for baseline differences between trials, the comparative effectiveness of GLAS + LDAC was compared with hypomethylating agent azacitidine (AZA) or decitabine (DEC). METHODS: A systematic literature review identified published trials of AZA or DEC versus LDAC among older AML patients ineligible for high-intensity chemotherapy. In addition to standard and covariate-adjusted ITC, STC was performed following guidance from the NICE Decision Support Unit (DSU). Using individual patient data from the Phase II GLAS + LDAC study, population-specific OS hazard ratios (HR) for GLAS + LDAC versus AZA or DEC were compared. Furthermore, covariate-adjusted ITC (Cox multivariate models) and STC were repeated using GLAS + LDAC versus LDAC data propensity-weighted for within-trial mean cytogenetic risk. As this initial step was not specified in the DSU, results from this second method were compared to the first STC following DSU guidance only. RESULTS: Standard ITC and STC both demonstrated significantly improved OS for GLAS + LDAC versus either AZA or DEC. Adjusting for key covariates, STC stepwise exponential models demonstrated GLAS + LDAC superiority to both AZA (HR=0.424; 95% CI: 0.228, 0.789) and DEC (HR=0.505; 95% CI: 0.269, 0.949). These significant results held using full or step-wise approaches, following DSU guidance only or the weighted STC approach. CONCLUSION: Using ITC and STC, GLAS + LDAC demonstrated superior OS to AZA or DEC in an adult population with previously untreated AML for whom intensive chemotherapy is not an option.