YIELD OF MULTIGENE PANEL GERMLINE GENETIC TESTING AMONG THOSE WITH ADVANCED COLORECTAL ADENOMAS

Background & Aim: Advanced adenomas (AAs) [≥1cm, villous, or high-grade dysplasia (HGD)] are precursors to colorectal cancer (CRC) and although guidelines support multi-gene panel testing (MGPT) in those with CRC, it is unclear if MGPT should be performed in individuals with CRC precursors. Our aim was to assess the yield of MGPT among those with AAs and asess factors associated with pathogenic/likely pathogenic variants (PV/LPV).

Methods: Individuals age ≥18 with at least one adenoma ≥10mm, or with HGD or villous architecture were identified between January 2011 to June 2023 from a statewide healthcare system including 13 endoscopy centers. Individuals with an established hereditary GI cancer syndrome, known PV/LPV in CRC gene, personal history of CRC, inflammatory bowel disease, or prior MGPT were excluded. For those who provided informed consent, telehealth genetic counseling and commercial 84-91 MGPT was offered. Demographics, medical/family history, and colonoscopy/pathology data were gathered from the electronic health record and participant surveys. Wilcoxon rank sum test, Pearson's Chi-squared test, and Fisher's exact test were used to assess factors associated with PV/LPV. NCT04160832

Results: Of 842 individuals with an AA over the study period, 167 consented and completed MGPT (Table 1). Overall, median age was 50, and majority were female (52.7%) and White (92.8%). Indications for colonoscopy were screening (39.5%; 21.5% for family history of CRC & 15.4% average risk), polyp surveillance (15.0%) and diagnostic (34.1%). AAs were based on size only (29.3%), histology only (9.0%) and both size and histology (34.7%). Anatomic distribution of AAs was predominantly rectum (26.1%) and left colon (40.0%), compared with right colon (33.9%). At least one PV/LPV was found in 24 patients (14.4%); however, when excluding MUTYH carriers and the CHEK2 c.470T>C variant, 17/167 (10.1%) had a clinically actionable PV/LPV, with three (1.8%) in a mismatch repair (MMR) gene (Table 2). Having a first degree relative with CRC was significantly associated with PV/LPV (72.7% vs 27.1%, p=0.011). Only 11.8% (2/17) with a clinically actionable PV/LPV had a PREMM5 score ≥2.5. At least one variant of uncertain significance (VUS) was found in 45.5% (76/167) and 15.0% (25/167) had more than one VUS. Of the 108 total VUS’ found, 10 were in MMR genes and 13 in other CRC genes (2 APC, 3 CHEK2, 3 POLD1, 2 POLE, 1 PTEN, 1 SMAD4, 1 STK11).

Conclusions: In this study, 10% of patients with AAs had a clinically actionable PV/LPV in a cancer predisposition gene, with 1.8% in mismatch repair genes, and the majority in genes not typically associated with CRC risk. Only 12% of those with a PV/LPV met current criteria for MGPT. Yield was higher among individuals with a family history of CRC. These data can guide clinicians on expected yield of MGPT in those with AAs.