Venetoclax-Based Therapy Versus Intensive Chemotherapy Followed By Allogeneic-Stem Cell Transplantation for High-Risk Elderly Acute Myeloid Leukemia

Introduction. Acute myeloid leukemia (AML) with ELN2022 adverse risk cytogenetics have poor outcomes. The current standard of treatment for “fit” elderly patients aged 60-74 years is intensive chemotherapy (IC) with anthracycline plus cytarabine (7+3) or liposomal daunorubicin plus cytarabine (CPX-351) followed by allogeneic stem-cell transplantation (allo-SCT), when feasible. However, response rate and median overall survival (OS) remain low due to persistent disease or relapse. Non-intensive approaches based on the BCL2 inhibitor venetoclax (VEN) are commonly used for IC-ineligible AML patients based on the VIALE-A trial. However, limited data is available on the utility of VEN-based non-intensive regimens as a first-line therapy followed by allo-SCT for “fit” elderly patients with cytogenetically defined high-risk AML.

Method. We retrospectively reviewed patients with newly diagnosed AML age 60-74 with chromosomal (chr.) abnormalities defining adverse risk in ELN 2022 classification, treated upfront with non-intensive VEN-based therapies compared with IC followed by allo-SCT in an international academic consortium (3 US centers [DFCI, MSKCC, Yale] and 3 Europe centers [IPC Marseille, France, CHU Nice, France, and Vilnius University Hospital, Lithuania]). Overall response rate (ORR) was defined as complete response (CR)+CR with incomplete count recovery (CRi)+Morphologic leukemia-free state (MLFS) and composite CR (cCR) as CR + CRi. OS was defined from treatment initiation to death and relapse-free survival (RFS) from the date of remission to relapse or death whichever comes first. Log-rank test was used to compare OS and RFS between groups. Cumulative incidence of relapse (CIR) and non-relapse mortality (NRM) were measured with Gray’s test.

Results. We included 86 patients (20 patients treated with VEN [HMA-VEN, n=16, Cladribine-Fludarabine-VEN, n=4] and 66 patients treated with IC [CPX-351, n=12, 7+3, n=44]). The median age was 66 (60-74). Pts treated with VEN-based therapies were older than IC-treated pts (69 [ranges, 61-74], vs 64 [ranges, 60-74], p=0.0001, respectively). Thirteen pts had AML with prior hematological malignancy (15.1%), and 18 pts a therapy-related AML (20.9%). Except for age, all other baseline characteristics did not differ between treatment groups.

The median number of chr. abnormalities was 3 (range1-13), with most frequent cytogenetic alterations being complex karyotypes (70.0% vs. 57.6%) followed by chr. 5/7 abnormalities (45.0% vs. 50.0%), monosomal karyotype (10.0% vs. 22.7%), chr. 17 abnormalities (5% vs. 18.2%), MLL alteration (5.0% vs. 16.7%) and chr. 3 abnormalities (20.0% vs. 9.1%), in VEN vs IC groups, respectively. Mutations in TP53 were found in 15 out of the 60 tested pts (25.0%) and secondary-like mutations (Lindsley et al. Blood 2015) in 23 of the 59 tested pts (38.9%). The ORR and cCR were 90% and 90% in VEN-treated pts treated vs. 78.7% and 80.3%, in pts treated with IC, p= 0.338 and p=0.505, respectively. Allo-SCT was performed in CR in 10 (50.0%) pts vs. 48 (72.7%, p=0.101), CRi in 8 (40.0%) vs. 4 (6.0%, p=0.0001) and MLFS in 0 vs. 1 (1.5%, p=0.999), in VEN and IC groups, respectively. Two pts (10%) in the VEN group vs. 13 (19.7%) went to allo-SCT while not in remission (p=0.503). The median time from treatment initiation to allo-SCT was 159 days (interquartile ranges [IQR], 129-178) vs. 119 (IQR, 93-155, p=0.663). Median follow up was 17 mo in the VEN group (IQR, 13-24) vs. 66 (IQR, 39-77, p=<0.001). Median OS and RFS were 20 mo and 29 mo (95%CI, 0-61) vs. 19 (95%CI, 11-26) and 30 (95%CI, 22-38, p=0.508 and 0.765, respectively).The median OS for pts who underwent allo-SCT in remission was NR vs. 21 mo (95% CI, 11-79, p=0.311), in the VEN vs IC groups, respectively. Seven pts (35%) vs. 21 (31.8%) relapsed after allo-SCT, median time to relapse was 166 days (IQR, 98-210) vs. 242 (73-355) in the VEN and IC group, respectively (p=0.559). One-year NRM incidences were 22% (95%CI, 6-43) vs. 28% (95%CI, 16-42) in the VEN and IC group, respectively (p = 0.465) and 1-year CIR were 32% (95%CI, 12-54) and 21% (95%CI, 10-35) in the Ven and IC groups, respectively (p = 0.319).

Conclusion. VEN-based non intensive approaches allowed successful allo-SCT with long-term disease control. This strategy may be preferred to IC for AML fit pts with adverse risk cytogenetics; however, this will have to be examined in prospective validation trials.