A single institutional analysis of missed opportunities in Lynch syndrome diagnosis.

Abstract

95 Background: The National Comprehensive Cancer Network (NCCN) recommended germline testing for all patients with early onset colorectal cancer (EOCRC) in 2017. Older adults continued to rely on other pathways such as family history and universal immunohistochemistry (IHC) screening to identify Lynch syndrome (LS) carriers. Here, we describe cascade completion after abnormal IHC in the era of universal IHC screening for CRC. Methods: We reviewed deidentified data from the Emory Lynch Syndrome Screening Network (LSSN) database. This database tracks patients with cancers that are considered Lynch syndrome associated. We evaluated IHC results with deficient mismatch repair (dMMR); defined as loss of ≥1 protein on complete IHC. LS suspicious cases were those with MSH2/MSH6 loss or MLH1/PMS2 loss without BRAF mutation or MLH1 promoter methylation (obtained on reflex testing). Outcomes included referral for germline testing as well as completed genetic counseling/testing. Results: Among 565 patients with CRC (450 colon, 115 rectal) in the LSSN database diagnosed between 2013 to 2023, IHC was completed in 99.5% (562/565). dMMR was identified in 17.8% (100/562). Among dMMR tumors, MLH1/PMS2 loss (n=73), MSH2/MSH6 loss (n=17), and rare single protein losses (MSH6, PMS2, or MSH2 only) (n=10). Reflex testing was not completed in 20% of MLH1/PMS2 cases, leaving 41 patients suspicious for LS from the MLH1/PMS2 cohort. In total, 62 patients were LS suspicious. Among LS suspicious cases, 40% (25/62) were referred for germline testing and 29% (18/62) completed germline testing. 61% (11/18) of those tested carried a pathogenic variant. In age stratified analysis, 100 patients had EOCRC. Germline testing was completed in 27% (27/100) and 19% (5/27) carried a pathogenic variant. All 5 pathogenic variants were confined to dMMR tumors, while no variants were detected in pMMR tumors (0/19). In patients ≥50 years (n=461), 83 had tumor with dMMR, and 11% (9/83) were tested for germline mutations. Pathogenic mutation was identified in 44% (4/9) of those ≥50 years. Conclusions: In this cohort, the greatest missed opportunity for Lynch syndrome diagnosis was in patients ≥50 years. Despite a 44% pathogenic yield, most did not complete germline testing. Improving completion in older adults with abnormal IHC and ensuring universal uptake in younger patients are both essential to maximize Lynch carrier detection and enable cascade testing.