A Phase II Study of Decitabine with or without Carboplatin and Arsenic Trioxide in MDS and AML Patients

• Decitabine/arsenic trioxide was more clinically effective, against MDS/AML, than decitabine/carboplatin, absent adverse toxicity.

Although decitabine (DAC) shows activity against myelodysplastic syndrome (MDS) and acute myelocytic leukemia (AML), patient responses are limited, with poor prognosis. Following preclinical studies showing arsenic trioxide (ATO) and carboplatin (Carbo) to enhance DAC’s gene derepression, we initiated a randomized phase 2 clinical trial comparing DAC safety and efficacy, alone or combined with Carbo or ATO, in MDS and AML patients. Following randomization of thirty patients to one of three regimens: DAC only (20 mg/m², days 1 -5), DAC 20 mg/m² days 1-5 plus Carbo AUC 5 on day 8, or DAC with ATO 0.15 mg/kg days 1-5, we used adaptive patient randomization (based on response rates), with a minimum 3 cycles of 28 days each. The primary endpoint was composite response rate, while secondary endpoints were 1-year median survival, safety, and epigenetic effects. Of 91 total patients, 44 had relapsed/refractory disease but no significant Grade 3 or 4 toxicities. 4/15 patients on DAC alone (26.7%), 2/14 on DAC/Carbo (14.3%), and 20/62 on DAC/ATO (32.3%), responded. DAC/ATO had the most stable disease, or better, responses (p=0.018), and a response rate statistically higher than DAC alone (p=0.041). In particular, MDS diagnosis and previous treatment status were the best predictors of response, with MDS patients receiving DAC/ATO surviving the longest (16.5 months), versus DAC/Carbo (4.6 months) and DAC (9.3 months) (p=0.039). Epigenetic effects were similar across all groups. In conclusion, combined DAC and ATO was well tolerated, with clinical responses better than DAC or DAC/Carbo, improving survival for MDS/CMML patients. NCT02190695.