PK/PD model‐informed dose selection for oncology phase I expansion: Case study based on PF‐06939999, a PRMT5 inhibitor

The optimal dose for targeted oncology therapeutics is often not the maximum tolerated dose. Pharmacokinetic/pharmacodynamic (PK/PD) modeling can be an effective tool to integrate clinical data to help identify the optimal dose. This case study shows the utility of population PK/PD modeling in selecting the recommended dose for expansion (RDE) for the first‐in‐patient (FIP) study of PF‐06939999, a small‐molecule inhibitor of protein arginine methyltransferase 5. In the dose escalation part of the FIP trial (NCT03854227), 28 patients with solid tumors were administered PF‐06939999 at 0.5 mg, 4 mg, 6 mg, or 8 mg once daily (q.d.) or 0.5 mg, 1 mg, 2 mg, 4 mg, or 6 mg twice daily (b.i.d.). Tolerability, safety, PK, PD biomarkers (plasma symmetrical dimethyl‐arginine [SDMA]), and antitumor response were assessed. Semimechanistic population PK/PD modeling analyses were performed to characterize the time‐courses of plasma PF‐06939999 concentrations, plasma SDMA, and platelet counts collected from 28 patients. Platelet counts were evaluated because thrombocytopenia was the treatment‐related adverse event with clinical safety concern. The models adequately described the PK, SDMA, and platelet count profiles both at individual and population levels. Simulations suggested that among a range of dose levels, 6 mg q.d. would yield the optimal balance between achieving the PD target (i.e., 78% reduction in plasma SDMA) and staying below an acceptable probability of developing grade ≥3 thrombocytopenia. As a result, 6 mg q.d. was selected as the RDE. The model‐informed drug development approach informed the rational dose selection for the early clinical development of PF‐06939999.