Abstract CT015: Phase I dose-escalation study of the allosteric AKT inhibitor BAY 1125976 in advanced solid cancer

This open-label, phase I first-in-human study (NCT01915576) of BAY 1125976, a highly specific and potent allosteric inhibitor of AKT1/2, was conducted to evaluate the safety, pharmacokinetics, and maximum tolerated dose (MTD) of BAY 1125976 in patients with advanced solid tumors. A dose expansion in hormone receptor (HR) positive metastatic breast cancer (MBC) patients, enriched for patients harboring the AKT1E17K mutation, was included to evaluate the clinical benefit at the recommended Phase II dose (R2D).79 patients (including 39 MBC patients) were enrolled. Mean age was 56.7 years and 61 (77.2%) patients were female. All patients, except one, had received ≥1 prior line of systemic anticancer therapy. Response was measured using RECIST v.1.1.Oral dose escalation was initiated with a continuous once daily (OD) dosing (21 days/cycle), starting with a liquid formulation for the first two dose steps (10 mg and 20 mg OD) and bridging to a tablet formulation for further dose steps (40 mg, 80mg and 120mg OD). Based on toxicity assessment during cycle 1, and additional pharmacokinetic and pharmacodynamic (p-AKT, p-PRAS40 in platelet rich plasma) data, a separate dose escalation using twice daily (BID) dosing tested 40 mg, 60mg and 80mg BID, respectively. The following dose limiting toxicities (DLT) occurred: At 120 mg OD, grade 3 and 4 liver enzyme elevation in two out of 6 patients (2/6) and grade 3 alkaline phosphatase elevation in one patient; at 80mg BID, grade 3 liver enzyme elevation in two out of 4 patients and grade 3 hyperglycemia in one patient. After dose de-escalation according to the continuous reassessment method based study design at 60 mg BID, two out of 6 patients experienced grade 3 liver enzyme elevation. The MTD of BAY 1125976 was determined as 80 mg OD and 60 mg BID, respectively. 28 patients with HR+ MBC were enrolled in the expansion cohort using BAY 1125976 60 mg BID, including nine patients with the AKT1E17K mutation.Pharmacological inhibition of AKT1/2 as shown by inhibition of p-AKT and p-PRAS40 was seen in platelet rich plasma samples. However, among the 78 patients evaluable for response, only 1 (1%) patient with HR+ MBC (AKT1 wild-type) had a partial response (PR), 30 (38%) patients had stable disease (SD) and 38 (49%) patients had progressive disease as best response (data missing from 9 patients). Among 43 patients treated at the R2D, CBR was 27.9%. AKT1E17K mutation status was not associated with tumor response. Molecular charac