Primary Results of Stimulus-MDS1: A Randomized, Double-Blind, Placebo-Controlled Phase II Study of TIM-3 Inhibition with Sabatolimab Added to Hypomethylating Agents (HMAs) in Adult Patients with Higher-Risk Myelodysplastic Syndromes (MDS)

Background Patients (pts) with higher-risk MDS need tolerable treatment options with durable clinical benefits. Sabatolimab (MBG453) is a novel immunotherapy targeting T-cell immunoglobulin domain and mucin domain-3 (TIM-3), an immuno-myeloid regulator expressed on both immune cells and leukemic stem cells. Sabatolimab+HMAs delivered durable responses in a Phase (Ph) Ib study in pts with high-risk/very high risk (HR/vHR)-MDS (Brunner AM, ASH 2021). Here, we report the primary results from the ongoing STIMULUS-MDS1 (NCT03946670), a randomized, double-blind, placebo-controlled, Ph II study of sabatolimab+HMA in pts with intermediate risk (IR; and ≥5% bone marrow [BM] blasts), HR, or vHR-MDS who were ineligible for intensive chemotherapy or hematopoietic stem cell transplant at screening.

Methods Treatment-naive pts aged ≥18 years with IR/HR/vHR-MDS as defined by the Revised International Prognostic Scoring System (IPSS-R) were eligible. Pts were randomized 1:1 to sabatolimab+HMA or placebo+HMA and stratified by risk category (per investigator assessment) and type of HMA (azacitidine [AZA] or decitabine [DEC]; per investigator discretion). Intravenous (IV) sabatolimab (400 mg) or placebo were administered on Day (D)8 and D22 every 2 weeks, AZA (75 mg/m² IV or subcutaneous) on D1-7 or D1-5 + D8-9, or DEC (20 mg/m² IV) on D1-5. The 2 primary endpoints were complete remission (CR, per modified International Working Group-MDS criteria) rate and progression-free survival (PFS: time from randomization to relapse from CR, progression [including acute myeloid leukemia], or death). Secondary endpoints included response rates, overall survival (OS), duration of response, and safety.

Results A total of 127 pts were randomized (sabatolimab+HMA, N=65; placebo+HMA, N=62) between 29 JUL 2019 and 10 AUG 2020; 89% received AZA and 11% DEC. Reported results are from the primary analysis, when the study was unblinded (data cutoff 1 MAR 2022), unless noted otherwise. Median follow-up was 24 months (mo; randomization to cutoff). Baseline characteristics were similar between arms; median age was 73 years; 16.5%, 37.8%, and 45.7% of pts had IR, HR, and vHR-MDS, respectively. Median PFS was (sabatolimab+HMA vs placebo+HMA) 11.1 vs 8.5 mo (P=0.102, 1-sided; not significant [ns]; Figure A). Primary CR rate based on data up to 7 mo after the last pt’s first visit was evaluated earlier by an independent data monitoring committee; CR rate was (sabatolimab+HMA vs placebo+HMA) 21.5% (14/65) vs 17.7% (11/62) (P=0.769, 1-sided; ns). Updated CR rate at primary analysis was (sabatolimab+HMA vs placebo+HMA) 23.1% vs 21.0%, CR+partial remission (PR)+hematologic improvement (HI; includes marrow CR with HI and stable disease with HI) was 49.2% vs 37.1%, median duration of CR was 18.0 vs 9.2 mo, median duration of CR+PR+HI was 13.4 vs 9.2 mo, and OS was 19.0 vs 18.0 mo (hazard ratio 0.905 [95% CI: 0.565, 1.450]). In an exploratory subgroup analysis of pts with <10% BM blasts at baseline (sabatolimab+HMA, n=32; placebo+HMA; n=29), CR was 28.1% vs 17.2%; CR+PR+HI was 53.1% vs 34.5%; and median PFS was 11.3 vs 8.3 mo (Figure B). Similar beneficial results were seen in IR/HR IPSS-R pts. The grade ≥3 adverse events (AEs) with >20% in either arm were (sabatolimab+HMA vs placebo+HMA) neutropenia (53.2% vs 63.5%), thrombocytopenia (37.1% vs 42.9%), febrile neutropenia (35.5% vs 23.8%), anemia (22.6% vs 42.9%), and leukopenia (22.6% vs 28.6%). The most common non-hematological all grade AEs were constipation (46.8% vs 38.1%) and diarrhea (43.5% vs 22.2%). One pt developed a serious potential immune-mediated AE (pneumonitis) on study treatment (sabatolimab+DEC; 6 doses of 400 mg of sabatolimab) and was treated with immunosuppressants including steroids. This AE was suspected as study treatment related and resulted in fatality.

Conclusion Sabatolimab+HMA is associated with a favorable safety profile in pts with higher-risk MDS. Although improvements in CR and PFS were not statistically significant, when assessed with the better duration of response, the data may suggest a delayed-onset benefit in the sabatolimab arm, as well as a preferential effect in pts with lower disease burden. Additional exploratory and biomarker analyses will be reported in the meeting. The ongoing Ph III STIMULUS-MDS2 with a primary endpoint of OS (NCT04266301) has completed accrual and will definitively inform on the impact of sabatolimab in higher-risk MDS.