Pan-cancer panel testing in individuals with abnormal immunohistochemistry (IHC) staining of the mismatch-repair (MMR) genes.

Abstract

e13014 Background: IHC staining of the MMR genes is often used to screen for Lynch syndrome (LS) in patients with newly diagnosed colorectal cancer (CRC) and endometrial cancer (EC). Genetic testing of MMR genes guided by IHC has historically been used to confirm a diagnosis of LS. Given the known genetic heterogeneity of LS-related cancers, we assessed the extent to which IHC results reliably predicted findings with pan-cancer panel testing. Methods: We analyzed results from pan-cancer panel testing of 750 individuals with abnormal IHC results from 2013-2016. The panel included the MMR genes and other genes associated with CRC and/or other cancers. All clinical data, including IHC results, were obtained from provider-completed test request forms. Results: CRC (66.0%) was the most common cancer among tested individuals with abnormal IHC results, followed by EC (35.1%). 236 pathogenic variants (PVs) were identified in 228 (30.4%) individuals, 8 of whom carried two distinct mutations. Overall, 191 (80.9%) PVs occurred in MMR genes that were concordant with IHC results. The remaining 45 (19.1%) PVs were not predicted by IHC results, and included two PVs in MMR genes ( MSH6 PV with MLH1/PMS1 absent in IHC, MSH2 PV with PMS2 absent in IHC). The majority of PVs that were not predicted by IHC results occurred in non-MMR genes associated with CRC and EC (n=27, 60.0%; see Table). Conclusions: In this cohort, PVs were identified in 30% of individuals with abnormal IHC who were tested with a pan-cancer panel. Nearly 20% of PVs occurred in non-MMR genes that would be missed by MMR gene-specific testing based on IHC results. Collectively, this suggests pan-cancer panel testing may increase diagnostic yield among individuals with abnormal IHC. [Table: see text]