Survival outcomes with the use of gemtuzumab-ozogamycin in patients with acute myeloid leukemia receiving intensive chemotherapy stratified by molecular risk category

Abstract Introduction Gemtuzumab-ozogamicin (GO) is a CD33 antibody-drug conjugate that is approved for use in acute myeloid leukemia (AML) patients receiving induction chemotherapy (IC). A meta-analysis of randomized studies showed improved relapse-free survival (RFS) with GO use in patients with favorable (fav) or intermediate (int) risk cytogenetics by older Medical Research Council (MRC) cytogenetic criteria. The benefit of GO in patients with European LeukemiaNet (ELN) 2022 defined fav, int, or adverse (adv) risk disease, which incorporates molecular features, is not well defined. We examined outcomes of patients receiving GO in a retrospective cohort of IC-treated AML patients stratified by the molecularly informed ELN 2022 risk criteria. Methods Sequential patients with newly diagnosed AML receiving IC were retrospectively identified at a single institution from 2016-2023. GO was used from 2021 onward with fractionated dosing (3g/m2 day 1, 4, and 7) in combination with 7+3 IC or cytarabine consolidation. Patients with adv-risk karyotypes did not receive GO or had it discontinued. Cytogenetics and next-generation sequencing were performed in a clinical lab and patients excluded if these were absent. Risk stratification and outcomes were retrospectively assessed by ELN 2022 criteria. Morphologic or molecular relapse was considered a RFS event. Survival analysis was by Kaplan-Meier. Propensity score matching (PSM) within ELN risk cohort and by NPM1 status was by K-nearest neighbors matching. Factors utilized included: age >60, receipt of allogeneic hematopoietic cell transplantation (allo-HCT), TP53-mutated (mt), DNMT3A-mt, FLT3-mt, and sex. Results We identified 496 patients (150 fav, 146 int, and 200 adv) with a median age of 62 years (range 19-78). Among fav-risk patients 49 received any GO, 41 received 3 IC doses, and 36 received GO during consolidation. 32 int-risk patients received any GO of which 16 had had 3 IC doses, while 2 had GO during consolidation. There were 33 adv-risk patients that received GO, of which 17 received 3 IC doses, and 1 also during consolidation. Median RFS for fav-risk patients was significantly prolonged for GO vs no-GO (not reached (NR) vs 18.3 months (mo),p=0.019) with a trend for OS (NR vs 73.4 mo, p=0.1); 24-mo RFS was 64% vs 41%. Amongst int-risk patients, there was no difference in median RFS for GO vs no-GO (31.9 vs 13.1 mo, p=0.12) or OS (NR vs 73.4 mo, p=0.22) with a 24-mo RFS of 52% vs 36%. Amongst adv-risk patients, there was a trend towards improved RFS for GO vs no-GO (36.9 vs 13.5 mo, p=0.059) and OS (44.3 vs 24.4 mo, p=0.022) with a 24-mo RFS of 57% vs 44%. However, upon excluding patients with complex karyotypes, who are enriched in the no-GO group, the observed differences in RFS for GO vs no-GO (NR vs 19.7 mo, p=0.08) and OS (44.3 vs 31.2 mo, p=0.13) are largely eliminated. We then examined outcomes in patients with NPM1-mt irrespective of ELN classification with a PSM analysis. Among 128 patients with NPM1-mt, 44 received GO and 25 received 3 doses of GO. Unadjusted, patients with NPM1-mt had an improved RFS with GO versus no-GO (NR vs 15.4 mo, p=0.031) but not OS (NR vs 45.9 mo, p=0.091). PSM matching achieved a standardized mean difference (SMD) of <0.1 for each covariate except DNMT3A-mt (SMD=0.109). By PSM, there was a non-significant trend for improved RFS in GO vs no-GO patients (NR vs 16.4 mo, p=0.1) and OS (NR vs. 31.9 mo, p=0.14:) with a 24-mo RFS of 58% vs 37%. When performing PSM for NPM1-mt patients who received 3 doses of GO versus no GO doses, we observe no significant difference in RFS (NR vs 25.9 mo, p=0.2) or OS (NR vs 75.1 mo, p=0.36). For GO vs no-GO, the 24-mo RFS was 72% vs 51% and OS 76% vs 76%. Amongst patients receiving allo-HCT (56 ELNfav, 98 ELNint, and 128 ELNadv) non-relapse mortality for patients with GO versus no-GO was 8% vs 11.8% with 2 vs 0 fatal veno-occlusive disease events. Discussion In our real-world cohort, GO use during IC was associated with improved RFS in ELN 2022fav AML. No statistically significant improvement in RFS or OS was observed in ELN 2022 int- or adv-risk groups. In AML patients with NPM1-mt, there was a significant improvement in RFS with GO; however, this difference was not observed after matching for known baseline modifiers of outcomes in NPM1-mt patients. These results support GO use in the ELN 2022 fav-risk group. More work is needed to more precisely define which patients may benefit from GO beyond risk group.