A pharmacokinetically (PK) and pharmacodynamically (PD) driven phase I trial of the pan-AKT inhibitor AZD5363 with expansion cohorts in PIK3CA mutant breast and gynecological cancers.

Background: AZD5363 is a novel potent pan-AKT inhibitor (IC50of AKT1, AKT2 and AKT3 of 3, 7 and 7nM respectively) with preclinical activity across a range of models. Methods: The trial had an adaptive design that allowed changes in schedule based on toxicity, PK, and PD findings. AZD5363 was administered orally (PO) twice a day (BID). Three schedules were explored: continuous dosing (7/7), four days a week, (4/7) and two days a week (2/7). PD biomarkers including pAKT, pGSK3?, and pPRAS40 were measured by IHC in pre- and post-treatment tumor biopsies. Once a RP2D was established, two expansion cohorts of PIK3CA-mutant ER+ve breast (B) and gynecological (G) cancers were explored. Results: 47, 21 and 22 patients were treated on the 7/7, 4/7 and 2/7 schedules respectively, with a further 27 and 18 patients recruited to the B and G cohorts to date. The MTDs of 7/7, 4/7 and 2/7 were 320mg BID, 480mg BID and 640mg BID respectively. The dose limiting toxicities (DLTs) were rash and diarrhea for 7/7, and hyperglycemia for 2/7. No DLTs were identified for 4/7. The most common causally-related adverse events CTC Grade 3 were hyperglycemia (20%), diarrhea (10%), rash (10%), nausea (3%) and fatigue (1%). PK profiles at the RP2D of 480mg BID (4/7) showed a multi-dose Css,max of 1426ng/mL and AUCssof 7952ng.hr/mL, which were consistent with exposures that gave tumor regression in preclinical models. Pre- and post-treatment biopsies confirmed target engagement in tumor tissue, with an increase in pAKT and reductions in pGSK3? and pPRAS40. Based on toxicity, PK and PD profiles 480mg BID (4/7) was chosen as the R2PD for single agent AZD5363, with the option of using 640mg BID (2/7) as a pharmacologically active dose for future combination studies. Target lesion shrinkage was observed in 7/15 and 4/14 in the B and G cohorts respectively to date, and with RECIST responses in evaluable patients of 3/15 (20%) and 1/14 (7%). Conclusions: Based on toxicity, PK and PD data two intermittent schedules of AZD5363 have been identified for further exploration. Promising single agent activity has been seen in PIK3CA-mutant breast cancer providing support for ongoing combination studies. Clinical trial information: NCT01226316.