Abstract

689 Background: Pancreatic ductal adenocarcinoma (PDAC) is associated with poor overall survival (OS), however there are significant outliers. Understanding the biological factors driving PDAC cancer may reveal new biomarkers and treatment strategies. Mutational signatures, imprinted in the cancer genome, encode the life history of the cancer and may be used to stratify tumors into clinically-relevant subgroups. Methods: In this study we performed mutational signature analysis (MutationalPatterns, COSMIC v3.3) on whole genomes from 434 PDAC patients, comprising 167 patients with resected primary tumors, and 267 advanced PDACs from patients enrolled in the COMPASS trial. We evaluated whether the mutational signatures detected associate with OS using the log-rank test in the overall cohort and stratified by disease stage, adjusting for multiple hypotheses with 5% false discovery rate (FDR). We further characterized these associations in exploratory analyses. Results: Among the 55 COSMIC signatures detected in the overall cohort, the presence of three signatures significantly associated with OS after FDR correction. SBS85, a signature associated with indirect effects of activation-induced cytidine deaminase, was present in 18/434 (4.1%) samples, and was significantly associated with improved OS in the overall study population (log-rank p=6.5e-4, FDR-adjusted-p=0.026). SBS93, a signature of unknown etiology previously seen in other gastrointestinal cancers, was present in 4/434 (0.9%) samples, and was significantly associated with worse OS in the overall study population (log-rank, p=7e-6; FDR-adjusted-p=1.1e-3). SBS33, also of unknown etiology and previously seen in other gastrointestinal cancers, was present in 73/434 (16.8%) samples. It was seen more frequently in advanced disease (Fisher’s-exact p=1.9e-4), but was significantly associated with worse OS only in the resected cohort (log-rank, p=2.3e-4, FDR-adjusted-p=0.012). Conclusions: We identified three mutational signatures—SBS85, SBS93, and SBS33—that have not previously been described in PDAC and associate with OS. Further characterization of these novel signatures in PDAC and their underlying biology may uncover new treatment avenues.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call