Predicting risk of chemotherapy-induced side effects in patients with colon cancer with single-nucleotide polymorphism (SNP) Bayesian networks (BNs).

Abstract

344 Background: Patients receiving standard chemotherapy for colon cancer are at significant, but unequal, risk for development of treatment-induced side effects. Currently, there is no accurate way to identify these risks for individual patients. This study adopted a novel approach to creating a risk prediction model for chemotherapy-induced side effects through the development of clusters of SNPs defined by BNs. It was designed to assess the feasibility of identifying SNP-BNs that could accurately predict the risk for 6 common, and often debilitating, side effects: chemotherapy-induced nausea/vomiting (CINV), diarrhea, oral mucositis (OM), cognitive dysfunction (CD), peripheral neuropathy (PN), and fatigue. Methods: Patients (n=57) with colon cancer who received at least 3 cycles of FOLFOX6 +/- bevacizumab (along with standard supportive care strategies) were enrolled. Saliva samples were collected, DNA was isolated, and SNPs were analyzed on Illumina Omni microarrays (2.5 x 106 SNPs). Side effects under consideration were observed using Patient Care Monitor, a validated patient-reported symptom assessment instrument. BNs were developed for each of the 6 side effects and cross-validated using robust statistical analyses. Results: The percentage of patients who experienced moderate-to-severe side effects was notable despite supportive care measures. These included the following: CINV (32%), diarrhea (16%), OM (26%), CD (21%), PN (26%), and fatigue (56%). SNP-BNs were defined for each of the 6 side effects and were found to predict risk with a high degree of accuracy (>90%) and receiver operating characteristic (ROC) curves (>0.920). Conclusions: This study calls attention to the fact that treatment-related side effects persist in patients receiving FOLFOX6 +/- bevacizumab for colon cancer. SNP-BNs were identified that could accurately predict 6 of the most common treatment-related side effects. This genomic-based tool has the potential to offer clinicians accurate and actionable information for individualizing a patient’s cancer care plan.