Bemcentinib as monotherapy and in combination with low-dose cytarabine in acute myeloid leukemia patients unfit for intensive chemotherapy: a phase 1b/2a trial

Abstract

Beyond first line, the prognosis of relapsed/refractory (R/R) acute myeloid leukemia (AML) patients is poor with limited treatment options. Bemcentinib is an orally bioavailable, potent, highly selective inhibitor of AXL, a receptor tyrosine kinase associated with poor prognosis, chemotherapy resistance and decreased antitumor immune response. We report bemcentinib monotherapy and bemcentinib+low-dose cytarabine combination therapy arms from the completed BerGenBio-funded open-label Phase 1/2b trial NCT02488408 ( www.clinicaltrials.gov ), in patients unsuitable for intensive chemotherapy. The primary objective in the monotherapy arm was identification of maximum tolerated dose with secondary objectives to identify dose-limiting toxicities, safety and efficacy, and bemcentinib pharmacokinetic profile. In the combination arm, the primary objective was safety and tolerability, with efficacy and pharmacokinetics as secondary objectives. Safety and tolerability were based on standard clinical laboratory safety tests and Common Terminology Criteria for Adverse Events version 4. Bemcentinib monotherapy (32 R/R, 2 treatment-naïve AML and 2 myelodysplasia patients) was well-tolerated and a loading/maintenance dose of 400/200 mg was selected for combination treatment, comprising 30 R/R and 6 treatment-naïve AML patients. The most common grade 3/4 treatment-related adverse events were cytopenia, febrile neutropenia and asymptomatic QTcF prolongation, with no grade 5 events reported. In conclusion, bemcentinib+low-dose cytarabine was safe and well tolerated.

Fig. 1. Flowchart of patient disposition in the DE and LDAC cohorts.

AML, acute myeloid leukemia; LDAC, low-dose cytarabine; MDS, myelodysplastic syndrome; No., Number; R/R, relapsed/refractory.

Fig. 2. Treatment-emergent adverse events of grade ≥ 3 observed in ≥ 10% of patients.

a Dose escalation cohort. b LDAC cohort. Grade 3, severe (green); Grade 4, life-threatening (blue); Grade 5, death (red). Source data are provided as a Source Data file.

Fig. 3. Kaplan-Meier survival curves.

a OS, b EFS and c RFS for treatment-naive and R/R patients. 95% CI, 95% confidence interval of the median; NA, confidence limit could not be determined. p-values from one-tailed log-rank test. Source data are provided as a Source Data file.

Fig. 4. Dose response curve of maximal pAXL inhibition in AML blasts.

AML blasts were identified within the peripheral blood mononuclear cell population by CyTOF and the maximum inhibition observed for each patient was plotted against the bemcentinib serum concentration at the time of maximum inhibition. A robust log-logistic dose response model with lower limit set to 0 was fitted. The calculated EC₅₀ (+/- standard error) is indicated (text and dotted line). The black line indicates the fitted model, with the 95% confidence interval indicated by grey shading. n = 13 individual patients. Patients for whom maximum inhibition occurred before treatment onset are not shown because the corresponding serum concentration (0 ng/ml) cannot be accommodated on the logarithmic concentration scale. However, they were included for calculation of the dose response curve. Source data are provided as a Source Data file.