Abstract Background: GSK2636771 is a potent, orally bioavailable and selective inhibitor of PI3Kβ (apparent Ki = 0.89 nM), with >900-fold selectivity over PI3Kα/PI3Kγ and 10-fold over PI3Kδ. It inhibits AKT phosphorylation and downstream signaling measured as decrease of PRAS40-, GSK3β-, and RPS6-phosphorylation in PTEN deficient cell lines. Methods: An ongoing phase I/IIa FTIH, open label dose escalation study of GSK2636771, once daily (QD), is being conducted to evaluate safety, pharmacokinetics (PK), pharmacodynamics (PD) and efficacy in pts with advanced tumors deficient in PTEN. The study comprised 3 parts: starting-dose selection, 3+3 dose escalation, and phase II expansion. PD was assessed in tumor and platelet rich plasma samples. Archival tumor samples were analyzed for alterations in a custom panel of select oncogenes and tumor suppressors by next-generation sequencing and copy number analyses. Results: As of Nov 2014, 62 pts (38 m: 24 f, mean 60 yrs) were enrolled and dosed at 7 dose levels (25 - 500 mg). In total 5 DLTs (3x hypophosphatemia/ 2x hypocalcemia G3) were observed in 3 pts at 500 mg, defining MTD. AEs ≥20% (all grades) included diarrhoea, nausea, vomiting, fatigue, abdominal pain, anemia, decreased appetite and headache. Cmax was reached 4-6 h after single and repeat dosing, with a mean t1/2 of 17.1-38.6 h across cohorts. Increases in mean Cmax and AUC(0-24 h) were dose-proportional up to 350 mg, but less than proportional at 400 and 500 mg. Phospho/total AKT ratio was decreased by >50% in platelet-rich plasma in 21 of 23 pts across various doses. Similarly, phospho/total AKT ratio was decreased (median 35%) in post-treatment tumor biopsies compared to pre-treatment tumors in 4 of 5 pts treated at 400 mg dose. One pt (prostate) had a RECIST PR and 17 pts had SD, 8 of whom were on study for ≥6 months. Exploratory genomic analyses of tumor samples from 10 prostate cancer pts revealed a PIK3CB L1049R mutation in the tumor of one of the pts who remained on study for 33 weeks and an increased PIK3CB gene copy number in the tumor of a pt who achieved PR. Preclinical characterization of PIK3CB L1049R mutation is ongoing, as it is homologous to the oncogenic PIK3CA H1047R mutation. These data suggest potential clinical benefit with GSK2636771 in pts with tumors harboring genetic alterations in PIK3CB. Tumors from other histologies are being analyzed for alterations in key cancer-relevant genes, including PIK3CB. Conclusions: MTD and recommended phase II dose of GSK2636771 is 400 mg QD. Pathway inhibition was observed with limited anti-tumor activity in pts with PTEN deficient tumors. Pts with tumors harboring concomitant genetic alterations in PIK3CB may benefit from GSK2636771 treatment. Recruitment into phase II expansion is ongoing. Citation Format: Johann de Bono, Hendrik-Tobias Arkenau, Joaquin Mateo, Jeffrey R. Infante, Howard A. Burris, Yung-Jue Bang, Joseph Eder, Sunil Sharma, Hyun C. Chung, Shaun Decordova, Karen E. Swales, Michelle D. Garrett, Desamaparados Roda-Perez, Meichun Ding, Mark Russo, Li Yan, Ben Suttle, Jerry M. Tolson, Wendy S. Halsey, Ganji Gopi, Harjeet K. Van Der Keyl, Shanker Kalyana-Sundaram, Ganesh M. Sathe, Monica Motwani, Rakesh Kumar. Exploratory genetic analysis of tumors from a phase I/II dose escalation study of GSK2636771 in patients (pts) with PTEN deficient advanced tumors. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr CT328. doi:10.1158/1538-7445.AM2015-CT328