Abstract
One of the benefits of neoadjuvant chemotherapy and radiation therapy (CRT) for the management of rectal cancer is tumor downstaging that can ultimately lead to complete tumor regression (known as complete pathological response - pCR). In a previous study using PET/CT imaging to assess tumor response to CRT, 50% of rectal cancers showed a continuous decrease in metabolic activity (estimated by standard uptake value measurements) between 6 and 12 weeks from CRT completion. However the remaining 50% of the cases showed increased metabolic activity within that period. We reasoned that the increased metabolic activity observed in these later patients could be determined by the clonal expansion of a genetically distinct subpopulation of tumor cells resistant to CRT. To address this question we performed exome sequencing and mutation detection analysis on normal tissue, primary tumor and residual tumor from 7 patients with rectal cancer that exhibited an increase in metabolic activity after CRT. Overall, about 20Gb of unambiguously mapped sequences were generated for each sample resulting in an average fold-coverage of 30X. Captured sequences mapped to the reference human genome were then used for the detection of SNPs and non-synonymous somatic point mutations in all three samples from each patient. Approximately 30,000 single nucleotide variants (SNVs) were identified in each sample and most of these SNVs were common to all samples. As expected, the majority of these common inherited variants (96%) have already been described in dbSNP. To identify non-synonymous somatic mutations occurring in primary and residual tumor samples, we excluded those present in the normal sample and those already described as a known SNP in dbSNP resulting in a mean of 243 SNVs per patient. Noteworthy, a significant number of non-synonymous somatic mutations were exclusively found in the primary and residual tumor samples of each patient, 30 and 32 SNVs on average respectively, revealing a high degree of tumor genetic heterogeneity. Furthermore, we were able to identify non-synonymous somatic mutations that were presented in both samples (mean of 107 SNVs), and for these mutations we determined the mutant allele frequency (number of reads representing the mutation/total number of reads covering the mutated base) in the primary and the residual tumor. We then searched for mutations with significantly different allele frequencies in the two samples, as these mutations would likely represent genetically distinct subpopulations of tumor cells selected during CRT. We were able to identify, on average, 4 SNVs enriched in primary tumor and 37 SNVs enriched in residual tumor per patient. In conclusion, based on exome sequencing of rectal tumors that exhibit incomplete response to CRT and have increased tumor metabolism we were able to identify non-synonymous mutations that may be associated with specific mechanisms of resistance or sensitivity to CRT. Functional analysis of these mutations and mutated genes will be presented, providing new insights into the molecular events influencing response to neoadjuvant CRT in rectal cancer.
Citation Format: Fabiana Bettoni, Elisa R. Donnard, Bruna R. S. Correa, Paula F. Asprino, Fernanda C. Koyama, Natália M. Felício, Bruna H. Hessel, Pedro A. F. Galante, Anamaria A. Camargo, Angelita Habr-Gama, Rodrigo O. Perez. Genetic heterogeneity in rectal cancer - Identification of subpopulations of tumor cells resistant to neoadjuvant CRT. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Integrating Clinical Genomics and Cancer Therapy; Jun 13-16, 2015; Salt Lake City, UT. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(1_Suppl):Abstract nr 05.