G-CSF, Cladribine, Cytarabine, and Dose-Escalated Mitoxantrone (GCLAM) in Adults with Relapsed/Refractory Acute Myeloid Leukemia (AML) or High-Grade Myeloid Neoplasms: Final Results from a Phase 1/2 Trial

Background: Outcomes with standard chemotherapies for relapsed/refractory myeloid neoplasms are poor, with complete remission (CR) rates often not exceeding 15-20%. Results from a previous 2 study in poor-risk relapsed/refractory acute myeloid leukemia (AML) suggested encouraging activity of a combination of G-CSF, cladribine, cytarabine, and mitoxantrone (GCLAM). Given data suggesting benefit of escalated doses of anthracyclines in AML, we conducted a phase 1/2 study (NCT02044796) of GCLAM using escalated doses of mitoxantrone for patients with relapsed/refractory AML or high-risk myeloid neoplasms (≥10% blasts).

Patients and Methods: Patients≥18 years were eligible if they had a treatment-related mortality (TRM) score of ≤6.9 (corresponding to a predicted 28 day mortality of ≤6.9% with standard induction chemotherapy) and adequate organ function (LVEF ≥45%, creatinine ≤2.0 mg/dL, bilirubin ≤3.25 mg/dL). Excluded were patients with uncontrolled infection or concomitant illness with expected survival <1 year. In phase 1, cohorts of 6-12 patients were assigned to 1 of 4 dose levels of mitoxantrone (12, 14, 16, or 18 mg/m²/day, days 1-3, compared to 10 mg/m²/day used in previous trials). Other drugs doses were G-CSF (days 0-5), cladribine 5 mg/m²/day (days 1-5), and cytarabine 2 g/m²/day (days 1-5). In phase 2, patients were treated at the maximum tolerated dose (MTD) of mitoxantrone. A second course of GCLAM was given if CR was not achieved in cycle 1. Up to 4 cycles of consolidation with GCLA (mitoxantrone omitted) were allowed if CR or CR with incomplete blood count recovery (CRi) was achieved with 1-2 cycles of induction therapy. Dose-limiting toxicities (DLTs) were: 1) grade 3 non-hematologic toxicity lasting >48 hours that resulted in >7 day delay of the next treatment cycle; 2) grade ≥4 non-hematologic toxicity if recovery to grade ≤2 within 14 days, both excluding infection or constitutional symptoms.

Results: Among 26 patients treated in phase 1, 2 DLTs occurred at dose level 4 (encephalopathy and cardiogenic shock), establishing GCLAM with mitoxantrone at 16 mg/m²/day as the MTD. Forty patients, median age 63 [range: 33-77] years, median TRM score 2.0 [0.2-6.39], with AML (n=34) or high-grade myelodysplastic syndrome (n=6) received GCLAM at MTD; 28% had secondary disease. Nineteen were primary refractory and 21 had relapsed disease (median duration of CR1: 12 months). Cytogenetics were favorable in 1, intermediate in 21, and adverse in 18; 6 patients had NPM1 and 1 had FLT3 mutations. Ten patients achieved CR (25%, 95% confidence interval [CI]: 13-41%) and 13 a CRi (32%, 95% CI: 19-49%) with 1-2 induction cycles, for an overall response rate (ORR) of 58% (95% CI: 40-73%). Nine of the 10 CR patients (90%) were negative for minimal residual disease by flow cytometry. One further patient had a morphologic leukemia free state, 11 had resistant disease, 2 were not evaluable, and 2 died in aplasia within 28 days of therapy for a TRM rate of 5%. The median overall survival (OS) was 11.5 months. Besides infections/fever, the most common grade ≥3 adverse events were rash and hypoxia (fluid overload/infection-related). Median times to an absolute neutrophil count ≥500/μL and platelet count ≥50,000/μL were 29 days each. In multivariable analysis controlling for baseline prognostic features, compared to 41 patients treated with GCLAM outside this study with mitoxantrone at 10mg/m²/day, GCLAM with mitoxantrone at 16mg/m² was associated with better OS (hazard ratio [HR]=0.45, 95% CI: 0.24-0.85, p=.01) and a nonsignificantly higher ORR (odds ratio [OR]=1.87; 95% CI: 0.69-5.10, p=.22). Further, compared to patient treated with other salvage regimens used at our institution (GCLAC [G-CSF, clofarabine and high-dose cytarabine], n=61; d/MEC [decitabine priming, mitoxantrone, etoposide, and cytarabine], n=52), GCLAM with mitoxantrone at 16mg/m² was associated with better ORR and OS than d/MEC (OR=3.33, p=.012; HR for death= 0.50, p=.01) and a similar ORR compared to GCLAC (OR=1.69, p=.25) after multivariable adjustment with a HR for death of 0.68 (95% CI: 0.39-1.18, p=.17).

Conclusion GCLAM with mitoxantrone at 16 mg/m²/day is well tolerated with a low TRM rate, and has potent anti-leukemia activity in relapsed/refractory myeloid neoplasms. Survival for this regimen appears better than when the regimen is used with mitoxantrone at 10 mg/m²/day and compared to other salvage regimens such as d/MEC.