A multicenter real-world study on response and survival in older patients with acute myeloid leukemia treated with intensive chemotherapy versus VEN/HMA by 2022 and 2024 ELN risk stratifications: An analysis from the MARROW consortium

Abstract INTRODUCTION Median age at diagnosis for acute myeloid leukemia (AML) in the U.S. is 69 years (yrs), with a dismal 5-yr overall survival (OS) rate of approximately 5% in those ≥75 yrs. Intensive chemotherapy (IC), often with a “7+3” regimen, is considered standard for medically fit patients (pts) 60 yrs and a subset of pts 60 yrs. In contrast, treatment with venetoclax (VEN) plus hypomethylating agents (HMAs) has led to a paradigm shift in the management of older pts 75 yrs and those unfit for IC and is also being investigated in younger pts. To investigate other factors besides medical fitness, we analyzed treatment response and survival with IC versus VEN/HMA in older AML pts per the prognostic molecular signatures identified in the 2022 and 2024 European LeukemiaNet (ELN) guidelines. METHODS Eleven institutions participating in the Myeloid Malignancy Association on Rapid Research Outcomes Working Group (MARROW) consortium abstracted clinical and molecular data on all newly diagnosed pts with AML aged 60-80 yrs who received upfront VEN/HMA or IC. Composite complete remission (CRc) was defined per 2022 ELN criteria. Categorical and continuous variables were compared using Chi-square/Fisher's exact test and two sample T-test/Wilcoxon rank-sum test, respectively. OS was estimated by the Kaplan-Meier method and groups were compared by log-rank tests. RESULTS A total of 711 pts across 11 institutions were identified. Seventy percent (n=501) received IC and 30% (n=210) received VEN/HMA. Median age at diagnosis was 67 yrs (range: 60-80) for the whole cohort. As expected, median age was less in those treated with IC vs VEN/HMA (66 yrs vs 71 yrs, p0.001). Most pts (78%, n=554) were White, 9% (n=65) were Black, and 7% (n=51) were Hispanic. More pts treated with IC had ECOG 0 (29% vs 12%, p=0.001) and more pts treated with VEN/HMA had ECOG 1 (71% vs 56%, p=0.001). There was no difference in baseline hemoglobin (median 8.8 g/dL; range: 3.1-15.4), platelet count (55 x 109/L; 3-1163), white blood cell count (6.1 x 109/L; 0.1-426), or percent bone marrow blasts (45%; 0-100) between the two groups. All pts with core-binding factor AML (5%, n=32) received IC. Of the 40% (n=284) of pts with normal cytogenetics, most (72%, n=205) received IC. The most common mutations were in DNMT3A (18%, n=128), IDH2 (17%, n=124), NPM1 (16%, n=116), ASXL1 (16%, n=111), TET2 (15%, n=104), RUNX1 (14%, n=98), SRSF2 (14%, n=96), TP53 (12%, n=86), IDH1 (12%, n=82), NRAS (10%, n=72), and FLT3-ITD (10%, n=71). KRAS mutations were present in 5% (n=33). Per 2022 ELN, 19% (n=110) were classified as having favorable risk disease, 18% (n=105) as intermediate, and 64% (n=376) as adverse. More pts with favorable risk received IC (21% vs 12%, p0.001) whereas more pts with adverse risk disease received VEN/HMA (79% vs 58%, p0.001). One-third of the cohort (n=225) is alive with a median follow-up time of 3.3 yrs (range: 0.05-18). CRc was significantly higher in pts treated with IC (66% vs 54%, p=0.003) and more pts proceeded to allogeneic stem cell transplant (alloSCT) after IC (39% vs 23%, p0.001). A similar proportion of pts with CRc after IC and VEN/HMA had relapsed disease (36% and 37% respectively, p=0.91). Median OS for IC was 18 months (mo) vs 15.6 mo for VEN/HMA (p=0.054). When censored at alloSCT, median OS for both groups was 14.4 mo (p=0.15). We then analyzed OS by 2022 ELN risk classification and censored at alloSCT. For favorable risk disease treated with IC vs VEN/HMA, median OS was not reached vs 16.8 mo, respectively (p=0.07). For intermediate risk disease, median OS was similar regardless of treatment type (16.8 mo for IC vs 18 mo for HMA/VEN, p=0.60). For adverse risk disease, median OS was also similar by treatment type (9.6 mo for IC vs 10.8 mo for VEN/HMA, p=0.19). We also analyzed OS by 2024 ELN risk stratification censored for alloSCT. Median OS for favorable, intermediate, and adverse risk was 19.2, 14.4, and 6 mo respectively for IC (p0.0001) vs 19.2, 10.8, and 6 mo respectively for VEN/HMA (p0.0001). CONCLUSIONS This multi-center real-world analysis suggests that AML pts 60 yrs of age with intermediate and adverse risk disease by 2022 and 2024 ELN have similar OS regardless of treatment approach. Therefore, a VEN-based treatment—even for “medically fit” pts with intermediate and adverse risk disease—may provide comparable OS while sparing the toxicity and logistics of IC.