Abstract Sabatolimab is a novel immunotherapy with immuno‐myeloid activity that targets T‐cell immunoglobulin domain and mucin domain‐3 (TIM‐3) on immune cells and leukemic blasts. It is being evaluated for the treatment of myeloid malignancies in the STIMULUS clinical trial program. The objective of this analysis was to support the sabatolimab dose‐regimen selection in hematologic malignancies. A population pharmacokinetic (PopPK) model was fit to patients with solid tumors and hematologic malignancies, which included acute myeloid leukemia, myelodysplastic syndrome (including intermediate‐, high‐, and very high‐risk per Revised International Prognostic Scoring System), and chronic myelomonocytic leukemia. The PopPK model, together with a predictive model of sabatolimab distribution to the bone marrow and binding to TIM‐3 was used to predict membrane‐bound TIM‐3 bone marrow occupancy. In addition, the total soluble TIM‐3 (sTIM‐3) kinetics and the pharmacokinetic (PK) exposure‐response relationship in patients with hematologic malignancies were examined. At intravenous doses above 240 mg Q2w and 800 mg Q4w, we observed linear PK, a plateau in the accumulation of total sTIM‐3, and a flat exposure‐response relationship for both safety and efficacy. In addition, the model predicted membrane‐bound TIM‐3 occupancy in the bone marrow was above 95% in over 95% of patients. Therefore, these results support the selection of the 400 mg Q2w and 800 mg Q4w dosing regimens for the STIMULUS clinical trial program.