Paper Details

ABSTRACT

Background: The outcomes of patients with myelodysplastic syndromes (MDS) after hypomethylating agent (HMA) failure remain dismal, with an overall survival of 4 to 6 months for higher-risk MDS. A previous randomized phase 3 trial of rigosertib, a small molecule RAS mimetic, versus best supportive care in patients with higher-risk HMA failure MDS suggested a benefit of rigosertib in specific subsets of patients, including those with Revised International Prognostic Scoring System (IPSS-R) very high risk (VHR) disease (Garcia-Manero G et al, Lancet Oncology 2016). Therefore, we designed and now report the final results of a phase III trial with rigosertib versus physician choice of treatment in patients with MDS after HMA failure.

Methods: Adult patients with refractory anemia with excess blasts (RAEB) with at least 1 cytopenia after HMA failure were eligible. HMA failure was defined as progression on, relapse after a response with, or intolerance to HMA, as well as failure to achieve response after at least 6 cycles of azacitidine or 4 cycles of decitabine. Patients were randomly assigned in a 2:1 ratio to rigosertib 1800 mg per 24 hrs as a continuous infusion on days 1-3 of a 2 week cycle for the first 8 cycles and on days 1-3 of a 4 week cycle thereafter or physician choice of treatment. The primary endpoint was overall survival from the time of randomization to death from any cause, with a subgroup analysis of patients with IPSS-R VHR disease. Clinical trial information: NCT02562443.

Results: From December 2015 through April 2020, 499 patients were screened and 372 patients enrolled, with 248 randomized to receive rigosertib and 124 physician choice of treatment. Of these, 237 patients received at least 1 dose of rigosertib and 109 patients assigned to physician choice cohort did not discontinue before week 1 visit. Patient characteristics were well-matched between the 2 arms, including median age of 72.5 (range: 47, 85) vs 73 (range: 40, 85) and duration of prior HMA of 5.9 months (range: 0.2, 15.2) vs 6.2 (range: 0.2, 13.5) in the rigosertib and physician choice arms, respectively, as well as number of cycles of prior HMA of 6 (range: 1, 10) in both.

At the July 26, 2020 cut-off date for survival analysis, 290 (78%) of the 372 enrolled patients had died. Treatment was ongoing in 18 patients receiving rigosertib and 9 patients receiving physician choice of therapy. The median follow-up time was 23.5 months (95% confidence interval [CI]: 21.1, 36.6). The median overall survival was 6.37 months (95% CI: 5.78, 7.73) and 6.28 months (95% CI: 5.39, 10.05) in the rigosertib and physician choice of treatment arms, respectively (p=0.26). Specifically in the IPSS-R VHR patients, the median overall survival was 5.19 months (95% CI: 4.04, 6.01) for those who received rigosertib and 5.19 months (95% CI: 4.24, 6.11) for those who received physician choice of treatment (p=0.40). When evaluating the impact of various factors, including gender, age, treatment region, therapy-related MDS, MDS classification, prior HMA, type of HMA failure, IPSS-R risk group, and cytogenetic abnormalities, on overall survival, only those with RAEB in transformation had improved overall survival with rigosertib than physician choice of therapy (hazard ratio 0.51 [95% CI: 0.29, 0.89], p=0.02).

In the safety population, 235 patients (99%) in the rigosertib and 104 patients 95%) in the physician choice of treatment experienced treatment-emergent adverse events (AEs), with the most common grade ≥ 3 AEs being thrombocytopenia, anemia, neutropenia, febrile neutropenia, and pneumonia. Serious AEs occurred in 156 patients (66%) receiving rigosertib and 66 patients (61%) receiving physician choice of treatment. A total of 47 patients (20%) on the rigosertib arm and 26 patients (24%) on the physician choice of treatment arm died due to adverse events. Four of these deaths were attributed to rigosertib, with cause of acute respiratory distress syndrome, acute respiratory failure, cardiac failure, and metastatic neoplasm.

Conclusion: Treatment with single-agent rigosertib did not improve overall survival compared to physician’s choice of treatment in patients with MDS after HMA failure, including those with IPSS-R VHR disease. These findings highlight the ongoing challenge of developing effective therapeutics once standard-of-care HMA therapy has been exhausted.