Abstract
Background: QHRD107, a potent CDK9 inhibitor, has demonstrated promising activity in relapsed/refractory acute myeloid leukemia (R/R AML). The 107VA regimen, a triple combination of QHRD107, venetoclax (VEN) and azacitidine (AZA), has showed favorable safety profile. With enrollment now completed, we report updated efficacy and long-term survival data.
Methods: This phase 2a, open-label, multicenter study (NCT06532058) consisted of a dose escalation phase and expansion phase. Patients with R/R AML were enrolled and received QHRD107 at one of three dose levels (40 mg, 60 mg, or 80 mg, PO, Q12 hours, days 1-28 in combination with AZA (75 mg/m2, days 1-7) and VEN (400 mg with dose ramp-up, days 1-28). The primary objective of the dose escalation phase was to determine the maximum tolerated dose (MTD) and recommended phase 2 dose (RP2D), while the expansion phase focused on assessing composite complete remission rate (cCR) and duration of response (DOR).
Results: Between August 2023 and February 2025, a total of 61 patients underwent screening, with 53 included in the intention-to-treat population. No dose-limiting toxicities (DLTs) were observed during the dose escalation phase, establishing 80mg of QHRD107 twice daily as the MTD. Due to gastrointestinal toxicity, patients were randomized 1:1 to receive QHRD107 at 60mg or 80mg during the expansion phase.
The median age of the cohort was 60 years (range: 19-80), with 24 (45.3%) patients being male. Cytogenetic information was available in 49 (92.5%) patients, and next-generation sequencing results were available in all patients. At baseline, 7 (13.2%) patients had a RUNX1::RUNX1T1 fusion gene, 8 (15.1%) had an NPM1 mutation, 10 (18.9%) had a CEBPA bZIP mutation, 5 (9.4%) had a FLT3-ITD mutation and 6 (11.3%) had a TP53 mutation. Seven patients (13.2%) were classified as AML with myelodysplasia-related gene mutations/cytogenetic abnormalities (AML-MR) according to 2022 ELN recommendation.
A total of 39 (73.6%) patients had been previously exposed to BCL-2 inhibitors (venetoclax or sonrotoclax), with 35 (66.0%) being refractory or relapsed from VEN-AZA regimen. Three patients in the expansion phase terminated treatment before cycle 1 day 28. Response rates were similar between the 60 and 80mg groups, with no obvious dose-dependent trend. Among 50 efficacy-evaluable patients, the overall response rate (ORR) was 54.0% (CR: 4, CRh: 7, CRi: 9, MLFS: 4, PR: 3), with a cCR rate of 40.0%. Measurable residual disease (MRD) negativity by flow cytometry was achieved in 9 responding patients.
Subgroup analysis revealed response differences among patients with different genotypes. Patients with TP53 mutation and AML-MR showed higher response rates, with ORR of 66.7% and 100%, respectively. In contrast, patients with solo NPM1 mutation or NPM1 and FLT3-ITD co-mutations showed lower response rates.
Notably, patients who were refractory or relapsed from VEN-AZA treatment achieved high response rates, with cCR rate of 34.3% (12 of 35) and ORR of 54.3% (19 of 35). Among these patients, all 5 AML-MR patients and 4 TP53-mutated patients achieved treatment responses.
Three (6%) patients proceeded to allogeneic stem cell transplant and 2 patients received autologous stem cell transplant. The median overall survival (OS) was 12.9 months (95% CI: 5.4 to not reached) vs 13.5 months (95% CI: 3.5 to not reached) in the QHRD107 60mg and 80mg groups, respectively.
Conclusions: The 107VA regimen demonstrated high response rates in VEN-AZA relapse or refractory AML, particularly in patients with TP53 mutations and myelodysplasia-related features. Based on these data, a phase 2b randomized controlled study comparing 107VA with other salvage regimens is currently being conducted for patients who relapse after VEN-AZA treatment.
