Parallel mutation screening and methylation quantification improves the molecular diagnostic yield for colorectal cancer.

e13003 Background: Recurrent somatic mutations and methylation patterns have previously been identified as biomarkers for non-invasive screening of colorectal cancer (CRC). However, most clinical screening assays have focused on a limited number of genomic loci, potentially reducing sensitivity and specificity. Here, we examined the potential utility of using an expanded set of mutation and methylation markers to improve screening accuracy of CRC. Methods: Pairs of colorectal tumor and tumor-adjacent tissues were resected from 41 CRC patients in Nanfang Hospital (NFH), China. The DNA was extracted, processed, and sequenced with Singlera Genomics’ OncoAim assay, which detects mutations across 59 genes implicated in cancer and quantifies the methylation level for key CRC genes. Matched pre- or post-surgery plasma samples were also collected and processed in parallel. Results: Thirty-eight tumor samples (93%) contained somatic mutations with greater than 5% allele frequency, while twenty-nine tumor samples (71%) contained aberrant methylation patterns. Among the tumor-adjacent samples, six contained somatic mutations and eight contained epigenetic abnormalities. Out of three tumor samples without any detectable somatic mutations, one showed an aberrant methylation pattern. Importantly, aberrant methylation patterns in tumor samples were also detected in matched plasma samples. Conclusions: Parallel mutation screening and methylation analysis not only improves the diagnostic yield, but also paves the way for non-invasive monitoring on patients with no common mutations.