First-in-human phase Ia study of the PI3Kα inhibitor CYH33 in patients with solid tumors

Abstract

PIK3CA mutations are highly prevalent in solid tumors. Targeting phosphatidylinositol 3-kinase α is therefore an attractive strategy for treating cancers harboring PIK3CA mutations. Here, we report the results from a phase Ia, open label, dose-escalation and -expansion study (NCT03544905) of CYH33, a highly selective PI3Kα inhibitor, in advanced solid tumors. The primary outcomes were the safety, tolerability, maximum tolerated dose (MTD) and recommended phase 2 dose (RP2D) of CYH33. The secondary outcomes included evaluation of pharmacokinetics, preliminary efficacy and changes in pharmacodynamic biomarkers in response to CYH33 treatment. The exploratory outcome was the relationship between the efficacy of CYH33 treatment and tumor biomarker status, including PIK3CA mutations. A total of 51 patients (19 in the dose escalation stage and 32 in the dose expansion stage) including 36 (70.6%) patients (4 in the dose escalation stage and 32 in the dose expansion stage) with PIK3CA mutations received CYH33 1-60 mg. The MTD of CYH33 was 40 mg once daily, which was also selected as the RP2D. The most common grade 3/4 treatment-related adverse events were hyperglycemia, rash, platelet count decreased, peripheral edema, and fatigue. Forty-two out of 51 patients were evaluable for response, the confirmed objective response rate was 11.9% (5/42). Among 36 patients harboring PIK3CA mutations, 28 patients were evaluable for response, the confirmed objective response rate was 14.3% (4/28). In conclusion, CYH33 exhibits a manageable safety profile and preliminary anti-tumor efficacy in solid tumors harboring PIK3CA mutations.

Fig. 1. Trial profile.

Diagram indicating participant numbers and disposition through the course of the trial. QD, once daily.

Fig. 2. Pharmacokinetic results.

a CYH33 and b its active metabolite I27 Concentration-Time Profiles (Mean ± SD) on Cycle 1 Day 1 (1 mg, n = 1; 5 mg, n = 1; 10 mg, n = 2; 20 mg, n = 5; 30 mg, n = 12; 40 mg, n = 3; 60 mg, n = 4); c CYH33 and d its active metabolite I27 Concentration-Time Profiles (Mean ± SD) on Cycle 1 Day 28 (1 mg, n = 1; 5 mg, n = 1; 10 mg, n = 2; 20 mg, n = 6; 30 mg, n = 9; 40 mg, n = 7; 60 mg, n = 2); e Pharmacodynamic results showing changes of FBG levels from baseline on Cycle 1 Day 8 (1 mg, n = 1; 5 mg, n = 1; 10 mg, n = 2; 20 mg, n = 6; 30 mg, n = 12; 40 mg, n = 20; 60 mg, n = 4). The box represents 25th, 50th, and 75th percentiles of observed values; the whiskers represent the minimum and maximum values; green dots represent individual values. hr hour, FBG fasting blood glucose, SD standard deviation.

Fig. 3. Representative CT images and HE stained biopsy sections from a patient in the 40 mg CYH33 treatment group with advanced breast cancer (Luminal B, HER2-, PIK3CA E545K mutation) who achieved a rapid and robust treatment response.

a Baseline CT image (September 18, 2019): a mass (red circle) was detected in the upper outer quadrant of the right breast on enhanced CT. b Cycle 2 Day 15 CT image (November 25, 2019): the mass (red circle) had reduced on enhanced CT compared with baseline. c End of study visit CT image (April 24, 2020): the mass (red circle) had increased on enhanced CT. HE stained biopsy sections from pre-baseline (d scale bar = 200 μm; g scale bar = 100 μm), at partial response (e scale bar = 200 μm; h scale bar = 100 μm) and at disease progression (f scale bar = 200 μm; i and j scale bar = 100 μm). PR partial response, PD progressive disease, HER2- human epidermal growth factor receptor 2 negative, CT computed tomography, HE hematoxylin-eosin.