Outcomes of De Novo Acute Myeloid Leukemia with Monocytic Differentiation (FAB M4/5) Treated with Venetoclax and Decitabine

Background: Venetoclax (VEN) with hypomethylating agent (HMA) is a new standard for older/unfit patients (pt) with acute myeloid leukemia (AML). Approximately 20-40% pts do not respond to VEN+HMA regimens and subsequently have poor outcomes. Hence, mechanisms of resistance to VEN is an area of active investigation. Pre-clinical data suggests that monocytic differentiation (FAB AML M4/5) may be associated with intrinsic resistance to VEN in AML (Pei et al. Cancer Discovery 2020). We conducted a subgroup analysis of pts with de novo M4/M5 AML treated on a prospective trial with VEN+HMA and compared outcomes with pts with non-M4/5 AML.

Methods: In this phase 2 trial of decitabine (DEC) and VEN, the frontline cohort enrolled pts >60 yrs. Only pts with newly diagnosed de novo AML were included in this analysis. Pts received DEC 20 mg/m² for 10 days every 4-6 wks for induction followed by DEC for 5 days after CR/CRi (DEC10-VEN, NCT03404193). VEN dose was 400 mg daily or equivalent. VEN duration was reduced in cases of myelosuppression. Responses were determined per European LeukemiaNet 2017 Criteria. Full protocol of the study has been published previously (DiNardo et al. Lancet Haematol. 2020. In press).

Results: Among de novo AML pts treated on this prospective trial, 16 pts with FAB AML M4/5 and 30 pts with FAB non-M4/5 AML. The median age of all pts was 72 yrs (interquartile range [IQR] 69-77, Table 1). In the M4/5 group, 81% (n=13) had NPM1^mut and 69% pts (n=11) had ELN favorable risk AML compared to the non-M4/5 group where 23% pts with NPM1^mut and 23% (n=7) had ELN favorable risk AML. In M4/5 group, 19% pts (n=3) had ELN adverse risk AML compared to 60% (n=18) in non-M4/5 group. Response rates were similar in M4/5 AML compared to non-M4/5 AML (Table 2). CR/CRi rate in pts with M4/5 AML was 88% (n=14) and in non-M4/5 AML was 80% (n=24, p=.52). No measurable residual disease at the level of 0.01% sensitivity was detected by flow cytometry in 85% (11/13) AML M4/5 patients and in 58% (14/24) non-M4/5 AML patients, (p=.10). Rate of relapse in M4/5 AML was 27% (4/15) and in non-M4/5 AML was 36% (9/25, p=.54). Median no. of cycles to response was 1 (IQR 1-2) in both M4/5 and non-M4/5 AML. Overall survival (OS) was NR in M4/5 AML and was 19.4 months (mo) in non-M4/5 AML (hazard ratio [HR] 0.68, 95% CI 0.27, 1.70, p=.4, Fig. 1). Duration of CR/CRi in M4/5 AML was NR and in non-M4/5 AML was 20.0 mo (HR 0.68, 95% CI 0.23, 2.08, p=.5, Fig. 2). OS and CR/CRi DOR were similar among pts with FAB M4 and M5 AML (Fig. 3, 4). Overall, 2 pts with M4 AML had MRD at the time of CR and MLFS, respectively. The pt who achieved CR MRD+ had TP53^mut and BRINP3^mut and relapsed after 8.3 mo. The pt with MLFS MRD+ had NPM1^mut and SRSF2^mut, developed febrile neutropenia and died from renal failure 1.5 mo after starting therapy. In total, 4 pts with M4/5 AML relapsed out of whom 3 pts had M4 AML and 1 pt had M5A AML. 3 pts had achieved CR, 1 pt achieved CRi and 2/3 pts had achieved MRD negativity. The median DOR for these pts was 5.4 mo. 3 pts had N/RAS^mut at screening and 1 pt had TP53^mut. Additional analysis with inclusion of secondary AML and RAEB-T pts with monocytic markers will be presented.

Conclusion: In our cohort of pts treated with DEC10-VEN, de novo AML with monocytic differentiation (FAB M4/5) had favorable risk features and excellent outcomes. Rate, depth, durability of response, and overall survival were comparable to AML without monocytic differentiation (non-M4/5).