Based on the results of the open-label AG120-C-001 study, ivosidenib (IVO), an oral, potent, targeted inhibitor of mutant isocitrate dehydrogenase 1 (IDH1mut), was approved by the US Food and Drug Administration for the treatment of IDH1mut relapsed/refractory (R/R) acute myeloid leukemia (AML), and in newly diagnosed AML in adults aged ≥75 years or patients ineligible for intensive chemotherapy (IC) in 2019.
In this study, analyses of matched patient data were conducted, using outcomes from IDH1mut R/R AML patients from the German AML Study Group (AMLSG) registry and a real-world chart review study (RWD) from France, Germany, United Kingdom, and Spain, to evaluate the comparative benefit of IVO within the proposed European Union indication. Patient data from Arm 1+ of the AG120-C-001 study (N = 159) was compared with a historical control (HC) by combining patient data from the AMLSG registry (N = 127) and the RWD (N = 148). A medical review was performed to identify Arm 1+ IVO patients in the AG120-C-001 study and HC patients who fell within the proposed European Union indication. Treatment with IVO was compared with the most recent therapy received by HC patients. Based on AG120-C-001 study eligibility criteria, IVO patients were not considered candidates for IC; therefore, HC patients treated with IC as their most recent therapy were excluded. Propensity score–based matching/weighting methods were used to adjust for imbalances in baseline prognostic factors between the 2 cohorts (optimal full matching and inverse probability of treatment weighting [IPTW]). Six baseline prognostic factors for estimation of propensity scores (age, history of hematopoietic stem-cell transplantation, number of prior regimens for AML, nature of AML, cytogenetic risk, and primary refractory status) were included based on a review of the literature and available data. Balance between populations was assessed pre- and postmatch via comparison of weighted standardized differences (WSDs) for each covariate. Time-to-event data were summarized via Kaplan-Meier (KM) estimators with 2-sided 95% confidence intervals (CIs). Cox regression analysis was employed to estimate hazard ratios (HRs) of overall survival (OS) using the key prognostic factors as covariates and corresponding 95% CIs were estimated using the sandwich estimator. In addition, complete remission (CR) rates were compared between IVO patients and RWD non-IC HC patients. AMLSG patients were excluded from the CR analysis, as the response data for these patients did not allow for the distinction between CRs and other response types.
A total of 109 IVO patients and 60 HC patients fell within the proposed European Union indication. As it more strongly minimized the absolute WSDs between cohorts compared with optimal full matching, the IPTW-matched data set was selected for this analysis, with all WSDs <0.05. The median OS with IVO was 8.1 months (95% CI, 5.7-9.8) compared with 2.9 months (95% CI, 1.9-4.5) in HC patients. The HR for OS was strongly in favor of IVO at 0.396 (95% CI, 0.279-0.562; P <.0001). Clear and early separation was seen between the KM curves for IVO and HC, suggesting an early and sustained benefit of IVO treatment in this setting. The landmark 6-month OS rate was 57.7% (95% CI, 48.2-67.2) in the IVO cohort versus 29.1% (95% CI, 17.4-40.8) in the HC cohort. The landmark 12-month OS rate was 35.0% (95% CI, 25.7-44.3) in the IVO cohort versus 10.8% (95% CI, 2.7-18.9) in the HC cohort. Higher CR rates were also observed in the IVO cohort, at 18.3% (95% CI, 11.6-26.9) compared with the HC cohort, at 7.0% (95% CI, 1.5-19.1).
In conclusion, monotherapy with IVO demonstrated extended OS and the potential to increase CR rates compared with standard-of-care therapies in an HC population.
Reference
Paschka P, Dombret H, Thomas X, et al. Ivosidenib Improves Overall Survival Relative to Standard Therapies in Relapsed or Refractory Mutant IDH1 AML: Results from Matched Comparisons to Historical Controls. Presented at: 62nd American Society of Hematology Annual Meeting & Exposition, December 5-8, 2020. Abstract 625.