Phase 2a Study of SLS009, a Highly Selective CDK9 Inhibitor, in Combination with Azacitidine and Venetoclax for Relapsed/Refractory Acute Myeloid Leukemia after Prior Venetoclax Treatment

Background

Although hypomethylating agents (HMAs) plus venetoclax (HMA/VEN) is a standard frontline treatment regimen for older/unfit patients (pts) with acute myeloid leukemia (AML), long-term outcomes are poor as the majority of pts who achieve a complete remission (CR) will ultimately relapse. In the absence of targetable mutations (ie, IDH1/2 or FLT3), there are no known effective agents for relapsed/refractory (R/R) AML after HMA/VEN and median overall survival (mOS) in those pts is <2.5 months (Maiti et al, Haematologica 2020). Resistance mechanisms to VEN-based therapies include up-regulation and dependence on MCL-1 activity. We hypothesized that the addition of SLS009, a highly selective CDK9 inhibitor, to Azacitidine/VEN (AZA/VEN) may abrogate resistance to venetoclax and result in clinical activity in R/R AML after VEN-based combinations.

Methods

This Phase 2a trial was a dose ranging, open label, single arm, study of AZA/VEN + SLS009 in R/R AML after prior Ven treatment conducted across 5 US institutions. The primary objectives of the trial were to assess efficacy, safety, tolerability, maximum tolerated dose (MTD) and optimal dosing regimen of SLS009 in combination with AZA/VEN. There were three different dose levels (DLs) of SLS009: DL1: 45 mg IV QW, DL2: 60 mg IV QW and DL3: 30 mg IV BIW (on day 1 and 2 of the week). AZA/VEN was administered as standard doses for 28-day cycles. Clinical activity was assessed based on standard ELN 2022 criteria.

Results:

Between July, 2023 and June, 2024, 30 pts were enrolled across all three dosing cohort (DL1: n=10; DL2: n=9, DL3: n=11 ). Overall, median age was 71 years (range: 41-84 years) while 67% of patients were >70 years and 17 (57%) patients were male. By race/ethnicity: 22 patients (73%) were non-Hispanic White , 4 (13%) were Black, 3 (10%) were Hawaiian/Pacific Islander and 1 (3%) was Asian. All but one patient (97%) had ELN adverse-risk AML and 77% were classified as AML-Myelodysplasia Related (AML-MR) per WHO 2022 criteria. Most frequent mutations seen overall were RUNX1 (33%), ASXL1 (30%), TP53 (30%), TET2 (20%), and SRSF2 (17%). Median number of prior treatments was 2 (range: 1-7).

Across all 3 dose levels, no dose limiting toxicities (DLTs) were observed; no treatment-related mortality and no serious adverse reactions were observed. Drug related toxicities of any grade occurring in more than one patient across all cohorts were nausea (23%), diarrhea (13%), hyperphosphatemia (10%), pyrexia (7%) and white blood cell count decreased (7%). There were no significant differences in AEs between DLs. Grade ≥3 drug-related AEs occurring in more than one patient per cohort were: 45 mg QW: White blood cell count decreased (20%); 60 mg QW: Neutropenia (11%); 30 mg BIW: none.

Among 29 evaluable pts, 16 (55%) had ≥50% reduction in bone marrow (BM) blasts compared to baseline (DL1: 60%; DL2: 33%; DL3: 80%). Nine (31%) pts achieved an overall response (i.e., CR+CRi+MLFS), including 5 (17%) who achieved CR/CRi. Response rates per dose level: DL1: 10%; DL2: 33%; DL3: 50%. Responses were observed within 1^st cycle of treatment in 8/9 overall responders. All 9 responders had AML-MR (9/23 of AML MR pts responded) and 8/15 pts (53%) with somatic MR mutations responded. Among those with ASXL1 mutations, 5/9 (56%) achieved an overall response. 2/9 (22%) with TP53 mutations achieved a response including one pt with concomitant TP53 and ASXL1 mutation who had an ongoing response at data cut-off. Among 9 responders, 1 proceeded to stem cell transplant (SCT), and 3 additional patients are under evaluation for SCT.

Median follow-up was 3.6 months at data cutoff. Median duration of responses was not reached though 7/9 responses are ongoing (Range: 1-6 months). Fifteen patients were still alive at the time of the data cutoff and the median OS for the trial has not been reached. At the first DL in which 8/10 pts died, mOS was 5.5 months.

Discussion:

Addition of SLS009 to AZA/VEN was found to be safe and feasible without DLT’s. SLS009 30 mg IV BIW (DL3) was chosen as the optimal dose. Clinical activity was seen particularly in pts with ASXL1 mutation which may be a subpopulation of patients with preferential sensitivity to SLS009 + AZA/VEN. Further development will be focused on AML-MR patients with and without ASXL1 mutations.