Older patients with acute myeloid leukemia (AML) unfit for intensive chemotherapy have a poor prognosis, which is further hindered by the shortage of well-tolerated standard therapies. Nonintensive chemotherapy options recommended by the National Comprehensive Cancer Network Guidelines include combinations of glasdegib with low-dose cytarabine (GLAS + LDAC) and venetoclax with LDAC (VEN + LDAC); however, these regimens have not been compared in a head-to-head trial. Therefore, indirect treatment comparisons (ITCs) to assess their comparative effectiveness are of clinical relevance. Unlike landmark measures of overall survival (OS), which may overvalue or undervalue survival at a single time point, hazard ratios (HRs) for OS estimate relative survival valued equally throughout follow-up. Therefore, a simulated treatment comparison (STC) study was designed to compare OS HRs between GLAS + LDAC and VEN + LDAC in older patients with AML who are unfit for intensive chemotherapy.
Individual AML patient data from the BRIGHT AML 1003 study, which compared in a randomized, blinded fashion LDAC with or without glasdegib, were extracted for the active arm (GLAS + LDAC) population (n = 116) and published summary data were extracted from the VIALE-C phase 3 VEN + LDAC trial (n = 211); these 2 data sets were used in this analysis. STC was employed based on Decision Support Unit guidelines from the United Kingdom National Institute for Health and Care Excellence. Parametric time-to-event regression models for OS were generated using the GLAS + LDAC individual AML patient data, and optimal exponential models are presented based on fit statistics. To match the VEN + LDAC trial, OS HRs were estimated after adjusting for baseline patient characteristics of the GLAS + LDAC trial. Full and backwards stepwise models (retention P value <.2) were constructed, taking into account all mutually available patient covariates, including age, sex, de novo versus secondary AML, bone marrow blast >50%, Eastern Cooperative Oncology Group performance status, and cytogenetic risk. OS HRs were compared indirectly by ITC (unadjusted for covariates) and STC (adjusted) with 95% confidence intervals (CIs). Both ITC and STC utilized post-hoc analysis results for VEN + LDAC, and a sensitivity analysis of prespecified analysis results for VEN + LDAC was performed.
As suggested by overlapping Kaplan-Meier curves for GLAS + LDAC and VEN + LDAC, the survival probability appeared similar throughout follow-up. However, differences in patient characteristics between studies were not accounted for in comparisons of crude survival rates. When compared with the VEN + LDAC trial, patients in the GLAS + LDAC trial were more likely to be male and have secondary AML, as well as have worse cytogenetic risk profiles.
Unadjusted ITC estimated that GLAS + LDAC was favored over VEN + LDAC numerically (not statistically significant) with a reduction in mortality (HR, 0.66; 95% CI, 0.38-1.15). The estimated comparative advantage of GLAS + LDAC over VEN + LDAC increased following STC with full covariate adjustment (HR, 0.56; 95% CI, 0.24-1.31). Sensitivity analysis results were consistent with these findings.
In conclusion, this study indirectly compared OS HRs of GLAS + LDAC and VEN + LDAC among older patients with AML who were unfit for intensive chemotherapy. A well-designed and conducted STC provides the best adjusted estimate of comparative effectiveness between treatments in the absence of a head-to-head trial. Both unadjusted ITC and adjusted STC revealed that the OS HR numerically, but not statistically, favored GLAS + LDAC over VEN + LDAC. These findings suggest safety profile, burden of administration, and patient preference must be factored into clinical treatment decisions between these 2 recommended nonintensive chemotherapy regimens, rather than being based on differences in survival outcomes alone.
Tremblay G, Daniele P, Dolph M, et al. Comparative Effectiveness of Glasdegib or Venetoclax in Combination with Low-Dose Cytarabine Using Simulated Treatment Comparisons. Presented at: 62nd American Society of Hematology Annual Meeting & Exposition, December 5-8, 2020. Abstract 624.