The combination of venetoclax and hypomethylating agents (HMA) has demonstrated potent activity in acute myeloid leukemia (AML), both in newly diagnosed patients (pts) and those with relapsed/refractory (r/r) disease. We analyzed the association between response to therapy and leukemic somatic mutations, cytogenetics, and other pertinent patient- and leukemia-related features in a large series of newly diagnosed and r/r AML in adults treated with venetoclax in combination with HMA at City of Hope between October 2016 and May 2018.

We identified 107 evaluable adults with AML treated with the combination of venetoclax and HMA. Sixty-one (57%) pts had r/r AML at the time of initiating treatment (median prior lines of therapy: 2; range: 1-10), while 46 (43%) were treated in the frontline setting. The median age of pts was 68 years (range: 19-86). AML was de novo in 57 (53%), therapy-related in 23 (21%) and secondary in 27 (25%) pts. Thirty-six (34%) pts had prior exposure to HMA, and 21 (20%) pts had relapsed following prior allogeneic hematopoietic cell transplantation (HCT). The majority of treated pts had unfavorable (52%) or intermediate-risk (39%) AML based on combined cytogenetics and molecular profiles. The most common detected somatic mutations (majority by next generation sequencing) were FLT3 (17%), followed by DNMT3A (15%), RAS and TET2 (each 14%), RUNX1 (13%), TP53 (12%), and IDH1/2 (11%). Most pts received decitabine in combination with venetoclax (N=97, 91%); only 10 (9%) pts received 5-azacitidine together with venetoclax.

Complete remission (CR)/CR with incomplete hematologic recovery (CRi) was achieved in 57 (53%) pts after a median of 2 (range 1-4) cycles. For 36 pts who achieved CR/CRi and had available minimal residual disease (MRD) assessment by multicolor flow cytometry (MFC), 23 (64%) became MRD-. CR/CRi was higher in pts carrying favorable- or intermediate-risk AML compared to poor-risk AML (100% vs. 60% vs. 45%, P=0.029). CR/CRi was 48% in those with complex cytogenetics (N = 31), 45% in monosomal karyotype (N = 22), 36% in KMT2A gene rearrangement (N = 11), 74% in normal karyotype (N = 19), and 25% in inversion 3 (N =4). The CR/CRi rate was not significantly different between newly diagnosed or r/r AML (61% vs. 48%, P = 0.17), nor was there a difference associated with AML type (de novo vs. therapy-related vs. secondary, P= 0.26), patient age (> or ≤ 65 years) at time of therapy (P = 0.13), prior allogeneic HCT (P = 0.29), prior administration of HMA (P = 0.37) and the type or schedule (5- or 10-day decitabine) of HMA (P = 0.52). In multivariate analysis, only favorable- or intermediate-risk cytogenetics was associated with better CR/CRi (P = 0.036). CR/CRi was also comparable regardless of the presence or absence of various analyzed somatic AML mutations. However, in recursive partitioning analysis of detectable somatic mutations and response to therapy, the combined lack of RAS, TP53 and RUNX1 mutations was linked to an improved rate of CR/CRi. When AML cases were stratified into functional gene alteration subgroups (according to the TCGA data set), there was no significant difference in CR/CRi according to the presence or absence of certain functional genes/fusions.

Median overall survival (OS) for all pts was 12.5 months and was 14.6 months for pts who achieved CR/CRi, in contrast to 4.6 months for non-responders (P <0.001). Only AML subtype (de novo vs. therapy-related vs. secondary) (P <0.001) and AML genetic risk (favorable/intermediate vs. high) (P = 0.042) independently impacted OS in multivariate analysis. None of the AML individual somatic mutations influenced OS for this cohort, however, in recursive partitioning analysis of detectable mutations, the presence of any of SRSF2, IDH1/2 or RUNX1 were associated with improved OS. Furthermore, the presence of myeloid transcription factor (P = 0.033) and spliceosome complex mutations (P = 0.004) predicted superior OS, whereas the presence of a chromatin modifying mutation predicted inferior OS (P = 0.004). Thirteen (23%) responders subsequently underwent allogeneic HCT.

We report remarkable activity with venetoclax and HMA across various high-risk genetics and clinical features in AML patients. Prospective studies are warranted to compare this combination directly with chemotherapy in all AML subsets. This is particularly true for high risk AML where response to conventional chemotherapy is poor.


Ali:Incyte Corporation: Membership on an entity's Board of Directors or advisory committees. Salhotra:Kadmon Corporation, LLC: Consultancy. Khaled:Alexion: Consultancy, Speakers Bureau; Juno: Other: Travel Funding; Daiichi: Consultancy. Stein:Celgene: Speakers Bureau; Amgen Inc.: Speakers Bureau. Forman:Mustang Therapeutics: Other: Licensing Agreement, Patents & Royalties, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract