RESUMO
BACKGROUND: In PTEN-loss models, the phosphatidylinositol 3-kinase (PI3K)/AKT and androgen receptor signaling pathways cross-regulate by reciprocal feedback whereby inhibition of one activates the other, creating a rationale for co-targeting. We studied the irreversible, pan-isoform inhibitor of Class I PI-3K PX-866 singly (part A) and with abiraterone acetate (AA) in patients on AA with rising prostate-specific antigen (PSA) (part B). PATIENTS AND METHODS: The primary endpoint was lack of progression at 12 weeks. Exploratory endpoints included changes in circulating tumor cells (CTC), pharmacodynamic studies on platelets (part A), and archival tumor exploration of PTEN as predictor of response (part B). RESULTS: A total of 43 and 25 patients accrued to parts A and B, respectively. In part A, 14 (33%) patients were progression-free at 12 weeks, with 2 partial objective responses and 1 confirmed PSA response. Favorable CTC conversion (< 5 CTC/7.5 mL) occurred in 6 (24%) of 25 evaluable patients. In part B, 11 of 25 patients had measurable disease. Six (24%) patients were progression-free at 12 weeks. No objective or PSA responses were observed. For all 68 patients, the most common toxicities were diarrhea (53 patients), nausea (36), anorexia (24), fatigue (22), and vomiting (20). Among 17 patients for whom PTEN testing was possible, 3 had PTEN homozygous deletion and 14 had no change. No correlation between PTEN status and response was seen. CONCLUSIONS: PX-866 had modest single agent activity. Adding AA to PX-866 showed no evidence of resistance reversal. Strategies to combine PI3K inhibition with androgen receptor-targeted therapies could consider initiation earlier, combination with other agents, and/or recruiting a selected population.
