Identification of Novel Biomarkers of Response to Irinotecan in Patients with Colorectal Cancer

Jing Li¹; Águeda Martínez-Barriocanal¹,²; Diego Arango¹,²

¹Group of Biomedical Research in Digestive Tract Tumors, CIBBIM- Nanomedicine, Vall d’Hebron University Hospital, Research Institute (VHIR), Universitat Autònoma de Barcelona, Passeig Vall d’Hebron, 119- 129, 08035 Barcelona, Spain. ²Group of Molecular Oncology, IRBLleida, 25198, Lleida, Spain.

Colorectal cancer (CRC) is the third most common type of cancer worldwide, causing approximately 935,000 deaths annually. The efficacy of chemotherapeutic compounds as single agents is not greater than 30%. Treatment failure is caused by both, primary and acquired resistance to drug exposure. Due to significant side effects and high medical costs, it is crucial to identify prediction biomarkers to chemotherapy in patients with CRC. The in vitro sensitivity was assessed in 90 CRC cell lines. The methodology was as follows: treatment of cell lines for 72h with the three agents using 11 different concentrations in monotherapy, followed by sulforhodamine B (SRB) staining as a surrogate marker of cell number. To identify gene signatures associated with drug response, we conducted bioinformatic analysis using sensitivity and OMICS data. We used both gene/protein expression, specifically RNA sequencing, mRNA expression microarrays, shotgun proteomics, and mutational profile of the CRC cell lines, namely whole exome-sequencing for the identification of predictive biomarkers. We summarized drug sensitivity both, as a continuous variable, i.e., GI50 (drug concentration able to reduce cell growth 50%), and as a discrete variable stratifying cell lines into resistant and sensitive according to the growth in the presence of a single drug concentration. In the analysis with CPT-11, we observed that the sensitivity to the chemotherapeutic agent (GI50) is higher in cell lines with microsatellite instability (MSI) compared to MSS, and KRAS wild type compared with KRAS mutant. We have identified one gene and four proteins whose independent expression show a significant difference (p<0.05) between resistant and sensitive cell lines upon multiple test correction using Benjamini-Hochberg and a false discovery rate (FDR) adjusted to 0.2. We will confirm the predictive power of the selected biomarkers in a panel of CRC cell lines by immunohistochemistry and test the consistency in additional OMICS data. Subsequently, we will conduct a validation in a large cohort of formalin- fixed, paraffin-embedded tumor samples from CRC patients treated with irinotecan. Finally, the functional role of the predictive biomarkers will be investigated in vitro by upregulation or downregulation in the appropriate cell line models. We determined the sensitivity to 5-FU, CPT-11 and Oxaliplatin in a panel of 90 CRC cell lines. MSS/MSI show significantly different sensitivity to CPT-11. We have identified one gene and four proteins whose independent expression show a significant difference (p<0.05) between resistant and sensitive cell lines.