Genetic testing of an unselected gastric cancer cohort: Uncovering germline mutations and their clinical implications.

Abstract

e22602 Background: Gastric cancer (GC) is the fifth most common cancer in Portugal and the third leading cause of cancer death worldwide. While most GC cases are sporadic, 10% show familial aggregation (FA), with 1-3% being hereditary. Advances in genomics may further improve treatment outcomes, particularly in locally advanced or metastatic GC. This study aimed to assess the incidence of pathogenic variants (PV) in cancer risk-associated genes in a prospective cohort of GC patients (pts). Methods: This prospective study included 69 consecutive pts out of 147 potential candidates with histologically confirmed locally advanced or metastatic gastric adenocarcinoma (ADC), referred to the Medical Oncology department between October 2020 and October 2022. The multigene panel, covering cancer risk-associated genes (including BRCA1, BRCA2, PALB2, BARD1, CHEK2, MLH1, MSH2, MSH6, PMS2), was used for genetic testing. For pts meeting hereditary GC criteria, CTNNA1 was also analysed. Clinical and pathological data were extracted from medical reports and clinical interviews. Results: Of the 69 pts (median age 63 years [27-83]; 28 F, 41 M), 42 (60.9%) had stage II/III disease, and 26 (37.7%) had stage IV. Histological subtypes included tubular ADC (42%), poorly cohesive cell ADC (18.8%), mixed ADC (31.9%) and other ADC types (7.2%). Primary treatments included perioperative chemotherapy (ChT) with surgery (66.7%), surgery with adjuvant ChT (14.5%), and palliative ChT (11.6%). Germline PVs were identified in 6 pts (8.7%): 2 BRCA2 (2.9%), 1 BARD1 (1.4%), 1 MSH2 (1.4%), and 2 MSH6 (2.9%). All pts with PVs in MSH2, MSH6 and BARD1 genes had tubular ADC, while the 2 BRCA2 pts had a diagnosis of poorly cohesive cell ADC. Only 2 with PVs (1 BRCA2, 1 MSH2) had family histories suggestive of hereditary syndrome (HS). No PVs were found in CTNNA1 or CDH1. The median follow-up was 29.95 months, with 39 pts alive at the time of analysis. Notably, both pts with BRCA2 PVs (stage IIB and III) were alive and disease-free after 42.5 months, while the other pts with PVs had died (stage IIB, III and IVA). Conclusions: A surprisingly detection rate of 8.7% was observed in this GC cohort unselected for FA. PVs were predominantly found in genes related to homologous recombination repair and DNA mismatch repair. The unexpected identification of a PV in BARD1, a gene less commonly associated with GC, warrants further investigation into its prognostic and therapeutic implications. The presence of BRCA2 PVs, probably associated with better responses to platinum-based ChT, could explain a more favourable prognosis in poorly cohesive cell ADC, a subtype typically resistant to ChT. These findings underline the importance of genetic testing in GC for improving pt management. Further studies are needed to confirm these results and evaluate the potential for identifying HS, which could enhance prevention, early detection and treatment strategies.