Genetic Predisposition to Gastric Cancer: CDH1 and Beyond

Abstract

Gastric cancer is a complex disease influenced by strong genetic and environmental factors. Hereditary gastric cancer syndromes are thought to account for between 1-3% of all cases. The most common hereditary gastric cancer syndrome is Hereditary Diffuse Gastric Cancer (HDGC), an autosomal dominant cancer syndrome that is primarily characterised by an extreme risk of developing diffuse-type gastric cancer. Approximately 40% of families that fit the clinical criteria for HDGC carry a pathogenic variant in germline CDH1. An explanation for the remaining 60% of cases remains largely elusive.

While New Zealand as a whole is a country with a low-incidence of gastric cancer, Māori have an age-standardised incidence of gastric cancer more than three times that of non-Māori. Additionally, Māori have an average age of diagnosis approximately 10 years younger than non-Māori, and are one of the few populations worldwide with a higher incidence of the diffuse-type disease. To assess the contribution of HDGC to the high-incidence of diffuse gastric cancer for Māori, we analysed the CDH1 sequence from an unselected cohort of 94 Māori gastric cancer patients and 200 healthy matched controls using next-generation amplicon sequencing, multiplex ligation-dependent probe amplification, and Sanger sequencing. Pathogenic CDH1 variants were identified in 18% of all cases, 34% of diffuse gastric cancers, and 66% of early-onset cases (< 45 years of age). After adjusting for the effect of clinical genetic testing for known Māori HDGC families, we estimate 6% of all Māori gastric cancer patients and 13% of diffuse gastric cancer patients carry pathogenic germline CDH1 variants.

Chile is a country with a high-incidence of gastric cancer and no formal genetic screening programme for gastric cancer patients. To explore pathogenic germline CDH1 variants as a cause of gastric cancer in Chile, next-generation amplicon sequencing and Sanger sequencing were used to screen a cohort of 51 Chilean gastric cancer patients that presented with a striking family history or early-onset disease. Overall, one clear pathogenic CDH1 variant was identified, representing 2.0% of all probands and 3.6% of probands who met the clinical criteria for HDGC. Although pathogenic CDH1 variants were rare in this Chilean cohort, we were able to screen the extended family of the one proband with a confirmed mutation and identify five further carriers. These carriers will now benefit from surveillance and early intervention.

Finally, whole-exome sequencing was used to examine 14 diffuse gastric cancer patients that fit the clinical criteria for HDGC and did not carry a pathogenic variant in their germline CDH1. Variants in these patients were filtered and prioritised for further evaluation and validation using Sanger sequencing. Single probands were found to carry pathogenic variants in ATM and TP53, genes which are not associated with HDGC, but are known to increase gastric cancer risk. Additional mutations of interest were identified in FARP2, FGD4, and LMO7, genes that are important in the coordination of the actin cytoskeleton and/or cell adhesion, pathways which are dysregulated in diffuse-type gastric tumours. Until now, FARP2, FGD4, and LMO7 were not linked with diffuse gastric cancer risk. It is clear from the current study and other HDGC studies, that there is no other common gene for HDGC, however families may carry private variants in genes rarely associated with disease.

Taken together, these studies demonstrate the variable frequency of pathogenic variants in germline CDH1 in different populations, the absence of other commonly mutated genes in familial diffuse gastric cancers, and the importance of genetic screening and targeted interventions for those that carry pathogenic variants.