Somatic Alteration Characteristics of Early-Onset Gastric Cancer

Abstract

Gastric cancer is one of the most common and deadly cancer types worldwide, which brings millions of dollars of economic loss each year. Patients diagnosed with early-onset gastric cancer were reported to have a worse prognosis compared to other gastric cancer patients, while the mechanisms behind such phenomenon are unknown. To identify age-dependent somatic alternations in gastric cancer, next-generation sequencing targeting 425 genes was performed on 1688 gastric tumor tissues and corresponding plasma samples. In our study, the microsatellite instability (MSI) and chromosomal instability score (CIS) values increased along with the age of patients, which indicates that older patients display a less genomic stability pattern. The differences of somatic alternations between young and old groups were compared. Somatic mutations CDH1 and copy number gains of FGFR2 were identified to enrich in the younger gastric cancer patients, which may contribute to the worse prognosis of early-onset gastric cancer patients.

Figure 1

Clinical details at each age group in gastric cancer patients. (a) Number of patients in different age groups. The red vertical line denotes the age cutoff of young and old gastric cancer patients. (b) Sex distribution in young and old gastric cancer patient groups. (c) Subtype distribution in young and old gastric cancer patient groups. (d) EBV and MSI subtype distribution in young and old gastric cancer patient groups.

Figure 2

Age-dependent trends of TMB and CIS values in gastric cancer patients. p values represent the trend in changes across different age groups. (a) Boxplot of TMB values of young and old gastric cancer patients. (b) Boxplot of CIS values of young and old gastric cancer patients. (c) Boxplot of TMB values in different age groups. (d) Boxplot of CIS values in different age groups.

Figure 3

Logistic regression analysis of different somatic mutations and frequencies of different somatic mutations in multiple age groups. (a) Logistic regression analysis of different somatic mutations in gastric cancer patients. Odds ratios (ORs) represent the risk of detecting the somatic mutation when diagnosed at a young age. (b) Bar plot of somatic TP53 mutation frequencies in different age groups. (c) Bar plot of somatic ARID1A mutation frequencies in different age groups. (d) Bar plot of somatic CDH1 mutation frequencies in different age groups.

Figure 4

Logistic regression analysis of different somatic copy number variations (CNVs) and arm copy number variations (arm CNVs), and somatic mutation signature analysis of young and age gastric cancer groups. (a) Logistic regression analysis of different somatic CNVs in gastric cancer patients. (b) Logistic regression analysis of different arm CNVs in gastric cancer patients. (c) Somatic mutation signature patterns of young and old gastric cancer patients.

Figure 5

Oncoprint of most frequent somatic mutations (TP53, ARID1A) in gastric cancer patients and significant somatic alterations in logistic regression analysis.