Abstract / Summary
Background: Whether early-onset gastric cancer (EOGC) represents a distinct clinical entity requiring age-specific management remains uncertain. We synthesized molecular profiles, treatment-associated survival, and prognosis to determine what current evidence supports in practice and requires prospective validation. Methods: This PROSPERO-registered, PRISMA-compliant review searched four databases (PubMed, Web of Science Core Collection, Embase, and the Cochrane Library) from 1990 to 2025. Random-effects models compared EOGC with late-onset gastric cancer (LOGC) using each study’s reported age definition, provided that the upper age cutoff for EOGC did not exceed 50 years, and evaluated treatment–survival associations within EOGC. Robustness was evaluated by restricted maximum likelihood (REML) estimation with Knapp–Hartung adjustment, subgroup analyses, meta-regression, and prediction intervals; certainty was assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE). Results: Fifty-eight studies involving 194,671 patients were included. EOGC showed lower observed odds of human epidermal growth factor receptor 2 (HER2) positivity (odds ratio [OR], 0.66; 95% confidence interval [CI], 0.53–0.84) and microsatellite instability (MSI)/deficient mismatch repair (dMMR) positivity (OR, 0.33; 95% CI, 0.12–0.91); p53 protein overexpression was less frequent, whereas the lower tumor protein p53 (TP53) mutation frequency represented only a non-significant numerical trend. E-cadherin loss was more frequent only under uniform immunohistochemical criteria (OR, 1.82; 95% CI, 1.16–2.85). These largely unadjusted molecular estimates indicate prevalence patterns rather than independent age effects. Radical resection was associated with lower mortality (hazard ratio [HR], 0.37; 95% CI, 0.22–0.61), but confounding by indication and a prediction interval crossing unity precluded causal inference. The chemotherapy association was not retained under REML–Knapp–Hartung analysis. Evidence was insufficient to establish differences in overall, disease-free, or cancer-specific survival. Certainty was low to very low. Conclusions: Age alone does not justify altered biomarker testing, systemic treatment, or surveillance. This review distinguishes actionable practice from unsupported extrapolation: established biomarker testing should be applied irrespective of age, and young age should not deter assessment for curative-intent resection. It also defines the evidence standards required before any EOGC-specific stratification can be adopted.