Addition of Sorafenib to Cladribine, High-Dose Cytarabine, G-CSF, and Mitoxantrone (CLAG-M) in Adults with Newly-Diagnosed Acute Myeloid Leukemia (AML) and High-Grade Myeloid Neoplasms Independent of FLT3-Mutation Status: Final Results of a Phase 1/2 Study

BACKGROUND: We previously showed that cladribine, cytarabine, G-CSF and dose-escalated mitoxantrone (CLAG-M) led to a higher rate of complete remissions without measurable residual disease (MRD^neg CR) than 7+3. Recently, a randomized trial demonstrated addition of the multikinase inhibitor sorafenib to 7+3 improved event free survival (EFS) irrespective of FLT3 mutation status. Here, we report the results of a phase 1/2 study (NCT02728050) evaluating addition of sorafenib to CLAG-M (CLAGM+S) in adults with newly-diagnosed AML or other high-grade myeloid neoplasms (≥10% blasts). Phase 1 results were reported previously.

METHODS: Patients aged 18-60 years were eligible if they had a treatment-related mortality score (TRM) of ≤13.1 (corresponding to a predicted <13.1% risk of death by day 28 following intensive induction chemotherapy) and adequate organ function. Patients with uncontrolled infection or concomitant illness with expected survival <1 year were excluded. Cohorts of 6-12 patients were assigned to 1 of 6 dose levels of mitoxantrone (10, 12, 15, or 18 mg/m²/day, days(d) 1-3) and sorafenib (200 to 400 mg twice daily [BID] on d10-20 with induction and 8-27 with consolidation). G-CSF dose was 300 or 480 μg/day (weight-based, d0-5), cladribine 5 mg/m²/day (d1-5), and cytarabine 2 g/m²/day (d1-5). Up to 4 cycles of post-remission therapy with CLAG+S (mitoxantrone omitted) were allowed, followed by 12 months of maintenance sorafenib. Patients were taken off trial if they underwent allogeneic transplant. As primary endpoint in Phase 2, the rate of MRD^neg CR following CLAGM+S was compared to a matched historical control population who received CLAG-M alone. Secondary endpoints included rate of CR, overall response rate (ORR defined as CR+CR with incomplete hematologic recovery [CRi]), overall survival (OS), EFS and the toxicity profile. EFS and OS were estimated using the Kaplan-Meier method and compared using Cox regression models.

RESULTS: 84 patients were enrolled from 12/8/2016-10/7/2021. In Phase 1, CLAG-M with mitoxantrone at 18 mg/m²/day and sorafenib 400 mg BID was established as the recommended phase 2 dose (RP2D). Sixty-one patients with a median age of 50 [range: 21-60] years were treated at the RP2D. ELN2017 risk category was favorable in 28%, intermediate in 36%, and adverse in 34%. FLT3-ITD and TKD mutations were present in 28% and 7% respectively. Mortality within 28 days was 2%. At the RP2D, a response of MRD^neg CR occurred in 49/61 patients (80%, 95% CI: 68-89%). Additional responses of MRD^pos CR and CRi (3 patients each) led to an ORR of 90%. The MRD^neg CR rate in patients with and without FLT3-mutations was similar (90% vs. 83%, p=0.30). Median times to an absolute neutrophil count ≥500/μL and a platelet count of ≥50,000/μL were 29 and 25 days. Besides infections/neutropenic fever, rash was the most common grade ≥3 adverse event. With a median follow up of 35 months, median OS was not reached; 12-month OS was 82%. There was no difference in OS and EFS in the FLT3-mutated and FLT3-wild type patients.

We next compared CLAGM-S to a historical cohort of 71 newly diagnosed patients treated with CLAG-M alone, matched for mitoxantrone dose, age and TRM score (Table 1). In univariate analysis, rates of MRD^neg CR, OS and EFS were similar (p=0.83, 0.16, and 0.37 respectively). Likewise, there was no difference in OS (p=0.54) or EFS (p=0.63) in FLT3-mutated patients who received CLAGM+S vs CLAG-M. In subgroup analysis, we observed a benefit favoring CLAGM+S for both OS (p<0.01) and EFS (p=0.03) among ELN2017 intermediate risk patients (Figure 1). No benefit of CLAGM+S vs CLAG-M alone in either OS or EFS was observed among adverse risk patients. Analysis of the favorable-risk cohort was limited by small sample size. Multivariate models comparing CLAGM-S to CLAG-M alone (reference) and controlling for age, secondary disease, FLT3 status and cytogenetic risk showed improved OS (HR=0.38, 95%CI 0.17-0.84, p= 0.02) and EFS (HR= 0.43, 95% CI 0.22-0.85 p=0.02) following CLAGM-S; MRD^neg CR was also more likely with CLAGM-S though not statistically significant (OR= 2.56, 95% CI 0.68-9.61, p= 0.16).

CONCLUSION: Addition of sorafenib to CLAG-M produces high rates of MRD^neg CR in newly diagnosed AML patients ≤60 and prolongs OS and EFS in multivariate analysis compared to historical, matched control cohort receiving GLAG-M alone, a benefit that was particularly evident in patients with intermediate-risk disease.