Personalizing Cancer Treatment of Combined Targeting of PI3K/AKT and RAS/MEK Pathways in Advanced Cancer Patients: From the Start Phase I Clinical Trial Program

ABSTRACT Background Frequent coactivation of both PI3K/AKT and RAS/MEK pathways has been seen in a number of different tumor types. Preclinical studies also provide a clear rationale for the co-inhibition of these frequently-activated, semi-parallel pathways. Methods We reviewed the records of 1672 consecutive advanced cancer patients in our phase I clinical trials program and investigated the clinical impact of simultaneous blockade of both PI3K/AKT and RAS/MEK pathways in patients with oncogenic alterations in both signaling pathways from 317 patients who received PI3K pathway inhibitor and/or MAPK pathway inhibitor. Results Of 317 patients, 163 (51.4%) were tested for tumor genomic analysis using DNA array-based CGH/PCR-based DNA sequencing. PI3K pathway genetic alterations were detected in 29 of the 163 (17.8%) patients. RAS/RAF pathway genetic alterations were detected in 43 of the 163 (26.4%) patients. Simultaneous oncogenic alterations in both PI3K/AKT and RAS/MEK pathways were detected in 12 patients (colorectal cancer, n = 7; pancreatic cancer, n = 2; melanoma, n = 2; non-seminomatous germ cell tumor, n = 1). Six of the eight (75%) patients treated with personalized treatment based on dual pathways inhibition had tumor regression with prolonged duration of time to treatment failure compared with that of patients treated with single pathway inhibition. A patient with pancreatic cancer harboring simultaneous KRAS mutation and AKT2 amplification treated with combination MEK inhibitor and AKT inhibitor showed central cavitations of numerous lung metastatic lesions along with a decrease in CA19-9, which may reflect treatment effect. Conclusions These data could serve as a platform for therapeutic decision-making. The strategy of targeting parallel pathways may be especially important in patients with coexisting PI3K/AKT and MAPK pathway genetic alterations. The data also suggest the need to further investigate the role of molecular profiling and matching patients with targeted drugs for personalized treatment.