Multiplexed protein signal pathway mapping identifies patients with rectal cancer that responds to neoadjuvant treatment

Background: Currently there is no reliable technique for predicting clinical or pathologic complete tumor response after radiochemotherapy (RCT) in patients with rectal cancer. We applied reverse phase protein microarray (RPMA) technology to find a signal pathway that may predict the response to preoperative treatment. Patients and Methods: Fifteen rectal cancer samples were collected during preoperative RCT. Seven patients had a good response to preoperative therapy (Mandard grade I-II) and 8 patients had a poor response (Mandard grade III-V). Using laser capture microdissection (LCM) and RPMA analysis, we measured the phosphorylation level of nearly 80 end points and analyzed the signaling pathways. Results: We identified 4 signaling proteins whose phosphorylation levels were significantly different (P <.05) between the good vs. poor responders; CHK2 and β-catenin were more highly phosphorylated in poor responders, whereas PDK1 and glycogen synthase kinase (GSK)-3α/β had lower phosphorylation levels in poor responders. Interestingly GSK-3α/β, β-catenin, and PDK1 are all present in the phosphatidylinositol-3-kinase (PI3K)-AKT signaling pathway. Conclusions: Based on our results, we hypothesize that the activating state of the PI3K-AKT pathway can stratify patients who could benefit most from neoadjuvant treatment. Moreover, identification of theranostic targets has the potential to pinpoint new therapeutic strategies for the nonresponsive population.