Abstract 3381: Determining the relationship between a measure of stress-induced mutagenesis and patient survival in cancer

Abstract

Abstract Background: Clustering of mutations observed in cancer cells is reminiscent of the clustering caused by the stress-induced mutagenesis (SIM) response in bacteria. SIM occurs when double-strand breaks (DSB) in DNA are accompanied by cellular stress, leading to mutations concentrated around the DSB with a characteristic decay in abundance with distance from the DSB. We hypothesized that clustered mutations in cancer are caused by SIM and measures of SIM correlate with patient survival. Methods: Data from whole genome sequencing of 1939 tumors from the International Cancer Genomics Consortium portion of the Pan-Cancer Analysis Working Groups (PCAWG) were analyzed. Clusters were defined as 3 or more single nucleotide variants (SNVs) with inter-variant distances of less than 25kb and a < 1% probability of occurrence based on the negative binomial regression given by the total rate of observed mutations in the genome. Cluster shape was measured by computing a normalized cumulative probability distribution of SNVs within 250kb of cluster centroids. A stress-introduced heterogeneity (SItH) score was calculated to quantify the difference between the observed distribution and a uniform, random distribution (the null hypothesis). This score can be computed on a per cluster basis as well as a single overall score on a per patient basis. Results: Overall SItH scores ranged from 0.145 to 0.999 and varied significantly by organ site (ANOVA, F=141.9, p<2.2x10-16). After controlling for organ site, we found that in primary tumors, more positive SItH scores predicted increased patient survival (Cox Proportional Hazard Regression [CPHR]; Hazard Ratio (HR) = 0.431, 95% CI: 0.2033-0.9138, p=0.0282). However, when recurrent and metastatic tumors were considered as a group, higher SItH scores predicted decreased patient survival (CPHR; HR = 7.87, 95% CI: 1.229-50.49, p= 0.0295). When we looked at the diversity of SItH scores computed on a cluster basis, the type of tumor sample was no longer relevant. The inner quartile range (IQR) of the per interval SItH score associated with worse survival, with a HR of 4.59 (Cox Proportional Hazard Regression, 95% CI: 1.261-16.7, p= 0.0208). Conclusions: The heterogeneity represented by the IQR is likely derived from both the strength of SIM as mutational process within a tumor and the clonal diversity of the tumor, both of which would be expected to impact disease outcome. The relationship between overall SItH and the IQR with respect to survival suggests that cluster heterogeneity predominantly represents the amount of time SIM has been active during carcinogenesis while overall SItH represents the ratio of SIM relative to other mutational processes. Citation Format: Luis Cisneros, Charles Vaske, Kimberly J. Bussey. Determining the relationship between a measure of stress-induced mutagenesis and patient survival in cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3381.