Abstract 4252: Detecting microsatellite instability in ffpe tissue from crc subjects using next generation sequencing

Abstract

Abstract Background Microsatellite instability (MSI) is a hypermutable phenotype resulting from DNA mismatch repair deficiency and is observed in up to 10-15% of early-stage colorectal cancers (CRC). It manifests as abnormal lengthening or shortening of DNA repeats at specific genomic loci. MSI-High (MSI-H) CRC patients have been shown to respond favorably towards both chemotherapy and immunotherapy and have different prognosis than MSS patients and the FDA has approved Pembrolizumab as the first cancer therapy based on a class of biomarkers rather than a cancer type. With increasing options for targeted therapy in solid tumors, being able to combine MSI testing with other cancer biomarkers for approved therapies is beneficial. We show proof of concept of a pan-cancer gene panel using an NGS-based assay and bioinformatics workflow that enables classification of MSI status along with detection of SNVs, fusions, indels and CNVs from tumor tissue samples. Unlike the gold standard PCR-based test, our method does not require a matched normal sample, thus enabling MSI detection based on tumor tissue alone. Methods Our workflow combines whole genome library preparation, hybrid capture target enrichment, high-throughput sequencing and a proprietary analysis algorithm. We expanded the number of genomic loci to improve the sensitivity of MSI detection. The analysis algorithm was trained using DNA from pure and mixed cell lines, tumor tissue and normal samples with known MSI status to classify the molecular alterations between MSI-H and MSS genotypes from sequencing reads. The algorithm leverages repeat lengths in homopolymeric MSI loci in order to predict instability of individual loci followed by aggregation using a statistical framework to make the final call on MSI status. Results The above algorithm was evaluated on a cohort of 134 stage II and stage III CRC subjects who underwent curative intent surgery. MSI status for these samples was orthogonally tested using a PCR-based MSI Analysis System (31 positives, 103 negatives) and a dMMR system. Using a pre-determined threshold, the algorithm yielded 100% sensitivity and 100% specificity on the above cohort without using matched normal tissue. A second cohort evaluated includes 47 CRC patients that are enriched for MSI positive status by virtue of their selection criteria (BRAF positive). Very high concordance on MSI status with an orthogonal method was observed. Data demonstrating high concordance with a PCR_based MSI system will be available at the presentation. Conclusions Here we present an NGS-based assay and bioinformatics workflow with robust analytical performance for MSI detection in FFPE tissue samples without a paired normal. Citation Format: Amrita Pati, Hao Wang, Hamid Mirebrahim, Seng Saelee, Joshua Lefkowitz, Sean Chien, Ashla Singh, Fergal Casey, Vera Rapoport, Xiaoju Max Ma, John Lee, Alex Lovejoy, Daniel Klass, Hans-Peter Adams. Detecting microsatellite instability in ffpe tissue from crc subjects using next generation sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4252.