Gastric cancer is a malignant tumor arising from the stomach lining and is the 5^th most common cancer worldwide (1). Chronic inflammation of the stomach has several different etiologies leading to gastric cancer, many of which are contributed to by social and geographic factors. There has been a known difference in prevalence and survival between Asians and other ethnicities specifically regarding gastric cancer likely as a result of these factors (2,3). Increased prevalence among Asian peoples has long been attributed to infections by Helicobacter Pylori and high nitrate and salt diets (4-6). With efforts to screen the public in Asian countries (Korea and Japan) for H. Pylori (7,8), the incidence of infection has significantly decreased. Due to this change in etiology, the molecular characterization of gastric cancer among Asians is likely to have likewise changed and requires reexamination.

Targeted gene immunotherapy in gastric cancer has been shown, among a variety of molecular differences, to improve survival and cancer-related sequelae (9). While it is known to some degree that there are differences in genetic mutations and copy number variations between Asians and Whites (10), the data, especially after preventative screening, is sparse. In a study by Jia et al. that examined somatic mutation differences in gastric cancer between Asians and Whites, multiple gene mutations were shown to have significant differences in prevalence between the populations-including APC, ARIDIA, KMT2A, PIK3CA, and PTEN. The study, however, did not examine the survival differences that derived from the variants of these genes. Survival curves for the same genes mutated in gastric cancer in Asians and Whites may shed light on differences in clinical practice, environmental risk factors, possible racial bias or other unknown molecular interactions.

Improvements in H. pylori screening as well as improved gene mapping, and identification of molecular differences in gastric cancers between Asians and Whites may reveal new gene targets for therapy.

The study aimed to elucidate the molecular differences in gastric cancer between Asian and White populations. We analyzed patient data from The Cancer Genome Atlas (TCGA) Genomic Data Commons (GDC) 2024 data set, which included 89 Asian patients and 278 White patients. It’s important to emphasize that this analysis encompasses individuals from various regions, not just Asian Americans or White Americans, as data contributions included samples from locations outside the U.S., particularly from East Asia.

We assessed the most prevalent genetic mutations across the two groups. We investigated significant differences in mutation prevalence and their impact on disease-free survival (DFS), disease-specific survival (DSS), and overall survival (OS). Additionally, we analyzed mRNA expression, methylation, and copy number variation (CNV) differences across the two subgroups. Notable differences emerged, and we generated OncoPrints and survival curves to illustrate these findings. Statistical analyses were performed, including p-values (p), hazard ratios (HR), and 95% confidence intervals (CIs). Downstream analysis of differentially expressed genes was conducted using NIH’s DAVID Bioinformatics, with the goal of identifying functional annotation clusters between the subgroups (11). Cox regression was used to depict any potential confounders within this experiment, specifically analyzing age, sex and tumor stage. Multivariate Analysis of Variance (MANOVA) was also used to test for significant differences between groups across multiple dependent variables.

Data for the TCGA studies were sourced from patients at leading academic institutions and compiled for accessibility through the cBioPortal platform. cBioPortal is an open-access resource that consolidates relevant data points from selected studies and provides statistical analyses. The platform enables filtering by specific genes, racial demographics, reported habits, survival curves, and other pertinent factors, facilitating comprehensive exploration of the data.

For a sample size of 100, a cutoff of approximately 5-10% is generally considered acceptable, meaning at least 5-10 cases with the alteration would provide a reasonable basis for initial analysis. While there are few explicit citations for strict cutoff percentages, these thresholds are typically guided by statistical power and the prevalence of alterations. In this specific gene study, the Asian cohort consisted of 89 patients, and we selected genes with alterations present in at least 10% of this population.

A total of 367 patients were included in this study: 89 Asians and 278 Whites. As seen in Figure 1, American Joint Committee on Cancer (AJCC) pathologic staging for Asians was I: n=9 (10.3%); II: n=41 (47.1%); III: n=34 (39.1%); IV: n=3 (3.4%). AJCC Pathologic staging for Whites was I: n=33 (14.7%); II: n=76 (33.8%); III: n=102 (45.3%); IV: n=14 (6.2%). The median age at diagnosis was 66 years for both Asian and White patients, with age distributions further divided into quartiles as shown in Figure 2. Cox-regression elicited several confounders in the data. Age >65 (p=0.039), stage 1 and stage 3 (p=0.007), stage 1 and stage 4 (p=0.008), stage 2 and stage 3 (p=0.043), and stage 2 vs. stage 4 (p=0.034). However, no significant difference was observed between stages 1 and 2 (p=0.09), stages 3 and 4 (p=0.394), and between men and women (p=0.669).

This study aimed to compare the molecular landscape of Asian and White gastric cancer patients, focusing on genes previously associated with DFS. Specifically, we investigated alterations in TP53, SYNE1, LRP1B, FAT4, and FLG, as well as sex and age, to determine if these factors are significantly associated with race. To assess the overall relationship between race and these variables, we employed multivariate analysis using Pillai’s Trace.

Multivariate analysis of variance (MANOVA) using Pillai’s trace was conducted on subsets of genes exhibiting different survival trends between Asians and Whites, with separate analyses for DFS and OS. For both DFS and OS, race was not significantly associated with survival outcomes (DFS: Pillai’s Trace=0.002656, F=0.1443, p=0.990; OS: Pillai’s Trace=0.004516, F=0.3708, p=0.829), suggesting that the observed differences between racial groups may not be statistically significant in this dataset. Similarly, sex did not show significant effects on survival in either analysis (DFS: Pillai’s Trace=0.008608, F=0.4703, p=0.830; OS: Pillai’s Trace=0.004299, F=0.3530, p=0.842). However, diagnosis age was found to be a statistically significant factor influencing both DFS (Pillai’s Trace=0.067615, F=3.9281, p=0.0008) and OS (Pillai’s Trace=0.044879, F=3.8413, p=0.0046), reinforcing its important role in survival outcomes. AJCC pathologic stage showed a marginal association with both DFS (Pillai’s Trace=0.057233, F=1.6006, p=0.087) and OS (Pillai’s Trace=0.038192, F=1.5964, p=0.122), suggesting that disease stage may contribute to survival variability. While race was not a significant factor in this analysis, further research with larger sample sizes or alternative methodologies may be needed to fully understand its potential role in survival outcomes.

Overall, as seen in Figure 3, Asians have lower disease-free rates of gastric cancer compared to whites after 36 months (p=0.014). There is no significant difference in OS between Asians and Whites (p=0.203). When compared against Whites, certain gene mutations (minimum Asians n >13) were associated with significantly DFS of Asians-TP53 (p=0.036), LRP1B (p=0.001), SYNE1 (p=0.018), FAT4 (p=0.009), FLG (p=0.024), HMCN1 (p=0.029), and ACVR2A (p=0.013).

Some genes were found to be associated with improved DFS of Whites when comparing altered versus unaltered groups such as FAT4 shown in Figure 4, TP53 (p=0.025), ACVR2A (p=0.043), and SYNE2 (p=0.046). However, LRP1B (p=0.014), FAT4 (p=0.021), DNAH7 (p=0.032), and KRAS (p=0.033) were associated with significantly worse DFS of affected Asians as compared to the unaffected group.

FLG (p=0.028) and DNAH7 (p=0.042) affected groups had significantly increased overall survival as compared to the unaffected. MUC16 mutation in Asians was also significantly associated with improved OS (p=0.040). CDH1 (p=0.00005) was significantly associated with worse OS in Asians in the altered group versus unaltered.

Certain copy number variations (CNV) were also significantly different between Asians and Whites. CCNE1 (p=0.0172) (Figure 5) was associated with significantly worse rates OS in Whites.

In the analysis of gastric cancer, significant differences emerged between White and Asian populations. For Whites, a notable annotation cluster revealed a strong association with immunological functions, including Immunoglobulin, Ig-like domains, V-type Immunoglobulins, antigen binding, V-set Immunoglobulins, Immunoglobulin complexes, and Immunoglobulin-mediated immune responses. This cluster had an enrichment score of 2.88, highlighting the importance of immune-related pathways in gastric cancer within this group.

Conversely, the Asian population exhibited three distinct annotation clusters. The first cluster demonstrated a strong involvement in the assembly of spliceosomal tri-snRNP complexes, including the quadruple SL/U4/U5/U6 snRNP and U4/U6 and U5 tri-snRNP complexes, achieving a high enrichment score of 5.07. The second cluster focused on pre-mRNA 5’-splice site binding, mRNA 5’-splice site recognition, and U1 snRNP, with an enrichment score of 4.41. Lastly, the third cluster related to mRNA splicing, RNA-binding, and the catalytic step 2 spliceosome, with an enrichment score of 1.54.

These findings suggest that while immune-related processes play a critical role in gastric cancer among Whites, splicing mechanisms are particularly enriched in the Asian population, indicating potentially different biological pathways influencing gastric cancer in these groups.

Cox regression analysis confirmed that advanced cancer staging and age greater than 65 are both associated with decreased survival in patients with gastric cancer, which aligns with expected outcomes. Notably, White patients were more frequently diagnosed at later stages but exhibited improved survival outcomes for several gene mutations. Additionally, sex was not identified as a confounding variable, enhancing the overall validity of the study findings.

MANOVA using Pillai’s Trace showed that the most important factor by far was age and that race had a much more minimal impact on survival. This can be interpreted in a number of ways. Asians and Whites comparably had similar staging and age at diagnosis; with Asians generally being diagnosed at earlier stages with later ages. While this study was not limited to North America, discrepancies may be related to patient-doctor interactions and perceived and real biases in diagnosis. Genetic discrepancies may also play a role in how fast or how late cancer manifests in Asians and Whites.

Overall, survival of Asians is comparable to that of Whites with respect to gene mutations in gastric cancer. Some CNVs surprisingly were associated with improved DFS of Asians compared to whites. Considering the markedly different gene clusters present between Asians and Whites, there is likely a different etiology and pathology for gastric cancer in the two populations. It is possible that a factor or combination of factors between social norms, access to healthcare, healthcare practices, diet, or provider racial bias may account for these gene discrepancies. It’s also possible that biological differences such as genetic makeup, gut flora, or infectious agents such as H. pylori are responsible. Differences in gene mutations and CNVs show promise for future prospective studies as gene targets in gastric cancer chemotherapy.

The primary limitation of this study was the sample size, particularly for Asian participants. Although numerous genes with established roles in gastric cancer were included in the analysis, the small number of Asian patients with the relevant genetic mutations restricted the scope of the study. Additionally, the limited sample size for specific mutations overall hindered more detailed analyses, such as cluster analysis and the exploration of variables like age, cancer subtype, and sex after accounting for ethnicity. Variations in donor site documentation also posed challenges for cross-sample analyses, particularly regarding gene panels and data collection. For instance, a substantial portion of patient data lacked assigned ethnicity and was excluded from the study.

Future research with larger and more diverse sample sizes is essential to generate more robust and precise insights into gene mutations and CNVs across ethnic groups, particularly between Asians and Whites. Such findings could pave the way for more targeted and individualized therapies for gastric cancer.

Considering the relatively low population of Asians in this study (n=89), a larger sample size may yet reveal more significant differences and help better characterize the nature of gastric cancer for future therapy.

Our analysis of The Cancer Genome Atlas reveals distinct molecular profiles of gastric cancer between Asian and White populations, highlighting differences in mutational burden, pathway activation, and gene expression patterns. These findings reveal the importance of incorporating racial and ethnic diversity in cancer genomics research, as population-specific molecular features may influence tumor behavior, therapeutic responses, and clinical outcomes. Further studies integrating genomic, environmental, and clinical data are warranted to deepen our understanding and support the development of personalized treatment strategies that address disparities in gastric cancer incidence and prognosis.

The Authors declare no conflicts of interest in relation to this study.

All Authors contributed equally to this work. Their roles, in accordance with the CRediT taxonomy, include: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; Writing - original draft; and Writing - review & editing.

The Authors would like to express their sincere gratitude to the University of Central Florida College of Medicine for their invaluable assistance and support in the completion of this study. Their resources, guidance, and encouragement were instrumental in facilitating this research.

This study was supported by University of Central Florida College of Medicine, FL, USA.

During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning-based image enhancement tools.