Paper Details

ABSTRACT

Background:

Azacitidine (AZA) combined with venetoclax (VEN) is standard of care (SOC) for newly diagnosed (ND) acute myeloid leukemia (AML) in older patients (pts) or those ineligible for intensive chemotherapy (IC). However, response and overall survival (OS) remain suboptimal in pts with monocytic and/or RAS/RTK pathway (RASp) mutated AML who derive inferior benefit from AZA/VEN, with complete remission/complete remission with incomplete hematologic recovery (CR/CRi) rates of ~60% and median OS of 8-12 months (Lachowiez et al. Blood Cancer Discovery 2025; Döhner et. al. Blood 2024).

Cladribine/cytarabine-based regimens may overcome VEN resistance in these AML subsets through increased dependence on purine/pyrimidine metabolism (Pei et al. Cancer Discovery 2023). CLAD/LDAC/VEN alternating with AZA/VEN was safe and effective in older patients with AML (Kadia et al. AJH 2026) and may be particularly active in monocytic and RAS pathway-mutated disease. We report interim results from a prospective, multicenter phase 2 study of CLAD/LDAC/VEN in newly diagnosed monocytic and/or RAS pathway-mutated AML (NCT06504459).

Methods:

Pts age ≥ 75 or considered unfit for IC with ND monocytic AML (by WHO 5^th ed.) or with RASp mutations (N/KRAS, FLT3-ITD/TKD, NF1, PTPN11, or CBL) were eligible with a planned accrual of up to 40 pts. Treatment consisted of 28-day cycles (C) of CLAD (5 mg/m² IV days [D] 1-5), LDAC (20 mg SubQ twice daily D1-10), and VEN (400 mg PO D1-21) administered in an alternating 2-cycle pattern with SOC AZA/VEN. Pts were hospitalized for induction and re-induction, per protocol. For pts attaining a disease response of at least morphologic leukemia-free state (MLFS) after C1, the CLAD duration was reduced to D1-3 in C2 onward and administered in the outpatient setting.

The primary endpoint was composite complete remission (CRc: CR, CR with partial hematologic recovery [CRh], or CRi) at the end of induction. Secondary endpoints included overall response rate (ORR: CR+CRh+CRi+MLFS+partial response), event-free survival (EFS), and Grade 3 treatment-emergent adverse events (TEAEs).

Results:

Of 18 pts enrolled to date, all were safety-evaluable and 17 were efficacy-evaluable. Median age was 69 years (range, 54–76). Sixteen pts (89%) harbored a RASp mutation, most commonly NRAS (n=9), PTPN11 (n=5), FLT3-ITD (n=4), and NF1 (n=3). Fourteen pts (78%) had de novo AML and 4 (22%) had secondary AML arising from a prior myeloproliferative neoplasm. Among the nine pts (50%) with monocytic AML, 7 had RASp mutations and 6 had NPM1mutations (5 with concurrent RASp and NPM1 mutations). Per ELN 2024 Less-Intensive classification, 5 (28%), 11 (61%), and 2 (11%) pts had favorable-, intermediate-, and adverse-risk disease, respectively.

ORR was 94% (16/17) and CRc rate was 82% (14/17). Among the 5 efficacy-evaluable pts with NPM1-mutated AML, all achieved CR/CRi by end of C2 and four (80%) achieved MRD-negative CRc (via qPCR with 10?5 sensitivity) by end of C3. One pt with monocytic AML and mutated NRAS, KRAS, ASXL1, SRSF2, and IDH2 did not respond following C1 and is undergoing re-induction.

After a median follow up of 8.8 months (range, 1.3-17.8), median EFS was not reached (estimated 12-month EFS: 92%).  One pt with monocytic AML at diagnosis relapsed with a primitive AML phenotype and emergent RUNX1 mutations after prolonged cytopenias precluded continuation to AZA/VEN maintenance. The other 15 pts with response have not relapsed or had disease progression at the time of analysis. Six pts (33%) underwent allogeneic hematopoietic cell transplantation.

The most common grade  3 TEAEs were febrile neutropenia (35%, n=6), neutrophil count decreased (35%, n=6), platelet count decreased (18%, n=3), anemia (12%, n=2), and hypokalemia (12%, n=2).

Correlative analyses of BCL2 gene expression by qPCR were available for 16 pts. Relative to a reference population of 100 ND-AML pts treated with HMA/VEN, mean BCL2 expression was lower in monocytic (32^nd percentile) vs. non-monocytic (47^th) and in RASp-mutated (43^rd percentile) vs. wild-type (88^th) cases, consistent with less BCL2 dependency in these subgroups.

Conclusions:

CLAD/LDAC/VEN alternating with AZA/VEN demonstrates encouraging activity in monocytic and RASp-mutated AML and compares favorably with historical experience using AZA/VEN in these high-risk AML subgroups with lower BCL2 dependency. Accrual is ongoing and additional follow-up and translational data will be presented.