Molecular profiling of patients with colorectal cancer and matched targeted therapy in phase I clinical trials.

Clinical experience increasingly suggests that molecular prescreening and biomarker enrichment strategies in phase I trials with targeted therapies will improve the outcomes of patients with cancer. In keeping with the exigencies of a personalized oncology program, tumors from patients with advanced chemorefractory colorectal cancer were analyzed for specific aberrations (KRAS/BRAF/PIK3CA mutations, PTEN and pMET expression). Patients were subsequently offered phase I trials with matched targeted agents (MTA) directed at the identified anomalies. During 2010 and 2011, tumor molecular analysis was conducted in 254 patients: KRAS mutations (80 of 254, 31.5%), BRAF mutations (24 of 196, 12.2%), PIK3CA mutations (15 of 114, 13.2%), KRAS and PIK3CA mutations (9 of 114, 7.9%), low PTEN expression (97 of 183, 53.0%), and high pMET expression (38 of 64, 59.4%). In total, 68 patients received 82 different MTAs: phosphoinositide 3-kinase (PI3K) pathway inhibitor (if PIK3CA mutation, n = 10; or low PTEN, n = 32), PI3K pathway inhibitor plus MEK inhibitor (if KRAS mutation, n = 10; or BRAF mutation, n = 1), second-generation anti-EGF receptor monoclonal antibodies (if wild-type KRAS, n = 11), anti-hepatocyte growth factor monoclonal antibody (if high pMET, n = 10), mTOR inhibitor plus anti-insulin-like growth factor-1 receptor monoclonal antibody (if low PTEN, n = 5), and BRAF inhibitor (if BRAF mutation, n = 3). Median time-to-treatment failure on MTA was 7.9 versus 16.3 weeks for their prior systemic antitumor therapy (P < 0.001). Partial response was seen in 1 patient [1.2%, PI3K inhibitor with PIK3CA mutation] and stable disease >16 weeks in 10 cases (12.2%). These results suggest that matching chemorefractory patients with colorectal cancer with targeted agents in phase I trials based on the current molecular profile does not confer a significant clinical benefit.