Introduction
Gastric cancer (GC) remains one of the most consequential malignancies worldwide, ranking as the fifth most commonly diagnosed cancer and the fifth leading cause of cancer-related death globally. In 2022, an estimated 969,000 new cases were diagnosed and approximately 660,000 deaths recorded, representing 4.8% of all incident cancers and 6.8% of all cancer mortality 10. Over 95% of GC cases are adenocarcinomas. Despite declining age-standardized rates in many high-resource countries, the absolute global burden is projected to increase 87.5% by 2050—reaching approximately 1.82 million cases—driven by demographic aging and population growth, particularly in low- and middle-income countries 20. This review synthesizes current epidemiological data, molecular classification frameworks, and emerging therapeutic strategies to inform clinical practice across oncology, gastroenterology, and translational medicine.
Global Epidemiology and Geographic Distribution
GC exhibits striking geographic heterogeneity (Table 1). The Western Pacific Region bears the highest incidence globally (age-standardized rate [ASR] 15.2 per 100,000 person-years; 543,757 cases), followed by the European Region (ASR 8.4; 161,553 cases) and the Americas (ASR 6.4; 103,924 cases) 10. Within East Asia, Mongolia records the highest national rates (male ASR 53.0; female ASR 21.9 per 100,000 person-years), followed by Japan (male ASR 40.9) and the Republic of Korea (male ASR 38.4) 10. China alone accounts for approximately 37% of all global GC cases (359,000 in 2022), reflecting both elevated incidence rates and its large population 9.
Table 1. Global Burden of Gastric Cancer (2022)
| Metric | Value |
|---|---|
| Global new cases | ~969,000 |
| Global deaths | ~660,000 |
| % of all cancers (incidence) | 4.8% |
| % of all cancer deaths | 6.8% |
| Global age-standardized incidence rate | 9.2 per 100,000 person-years |
| Highest-incidence WHO region | Western Pacific (ASR 15.2; 543,757 cases) |
| Highest-incidence country (male) | Mongolia (ASR 53.0) |
| Non-cardia subtype prevalence | 82% of cases (~795,000) |
| Cardia subtype prevalence | 18% of cases (~174,000) |
| Projected cases by 2050 | ~1.82 million (87.5% increase) |
Sex differences are clinically significant: the age-standardized incidence rate in males is 12.8 per 100,000 person-years compared with 6.0 per 100,000 in females, with a cumulative lifetime risk of 1.53% versus 0.67% by age 74 years 1. These differences reflect both biological susceptibility and differential exposure to risk factors including tobacco and alcohol.
Helicobacter pylori (H. pylori) infection is the dominant modifiable risk factor, responsible for approximately 76–87% of non-cardia GC (NCGC) cases in high-risk settings 10. Global H. pylori prevalence is estimated at 44% in adults, with substantial regional variation (Africa 70%; Oceania 24%) 10. A systematic review and meta-analysis of randomized controlled trials demonstrated that H. pylori eradication reduces GC incidence in healthy individuals (relative risk [RR] 0.54; 95% CI 0.40–0.72; number needed to treat [NNT] 72) and GC-related mortality (RR 0.61; 95% CI 0.40–0.92; NNT 135) 4. Additional risk factors include high dietary salt and preserved food intake, smoking, alcohol, obesity (particularly for cardia GC), and socioeconomic deprivation 3.
Incidence Trends, Screening, and Prevention
During 2008–2017, GC incidence declined in most countries, with annual decreases exceeding 3% in nations including South Korea, Norway, and the Netherlands 20. However, incidence is rising in younger populations (<50 years) in 15 of 34 studied countries, particularly among women in high-income settings 20. The projected 87.5% increase in absolute case burden by 2050 is most pronounced in the African Region (162% increase in new cases) and low Human Development Index (HDI) nations, underscoring the urgency of global prevention strategies 20.
Organized screening programs have markedly improved stage at diagnosis and survival. Japan's nationwide radiographic and endoscopic screening program (established 1983) and South Korea's national endoscopic screening program (initiated 2000) have shifted diagnoses toward earlier stages, yielding mortality rates approximately one-quarter of incidence rates in these countries—an outcome not observed in regions lacking systematic screening 10. A meta-analysis of 67 GC screening studies identified a pooled endoscopic uptake rate of 46% and an early gastric cancer detection rate of 59%, with developed countries significantly outperforming developing nations in both metrics (78% vs. 44% early detection rate; P<0.001) 5.
Molecular Classification and Biomarker Landscape
Modern GC management requires integration of histopathological and molecular classification. The Cancer Genome Atlas (TCGA) project defined four molecular subtypes: Epstein-Barr virus (EBV)-positive (~10% of cases), microsatellite instability-high (MSI-H; ~14–20%), genomically stable (GS; ~20%), and chromosomal instability (CIN; ~50%) 69. The Asian Cancer Research Group (ACRG) further classifies tumors as MSI, MSS/EMT (epithelial-mesenchymal transition), MSS/TP53+, and MSS/TP53−, providing complementary prognostic resolution 6.
Table 2. Major Molecular Biomarkers and Subtypes in Gastric Adenocarcinoma
| Biomarker / Subtype | Prevalence | Key Features | Prognostic/Predictive Significance |
|---|---|---|---|
| EBV-positive | ~10% | CpG methylation, PIK3CA/JAK2 alterations, immune infiltration | Favorable prognosis; enhanced immunotherapy responsiveness |
| MSI-H / dMMR | 14–20% (global) | High tumor mutational burden (TMB), mismatch repair (MMR) deficiency | 68% reduction in death risk with immune checkpoint inhibitors (ICI) vs. chemotherapy (HR 0.32, 95% CI 0.19–0.54) 6 |
| Genomically stable (GS) | ~20% | CDH1 mutations, diffuse/signet-ring histology, low TMB | Poor prognosis; limited targeted options |
| Chromosomal instability (CIN) | ~50% | TP53 mutations, aneuploidy, HER2 amplification | Variable; HER2+ subset (~15%) benefits from trastuzumab-based therapy |
| HER2 (human epidermal growth factor receptor 2) | ~15–20% | HER2 amplification/overexpression, enriched in CIN | Predictive for anti-HER2 therapy; emerging antibody-drug conjugate (ADC) relevance |
| PD-L1 CPS (combined positive score) | Variable; ~40–50% at CPS ≥10 | PD-L1 expression on tumor and immune cells | Predicts benefit from PD-1/PD-L1 inhibition; thresholds ≥1, ≥5, ≥10 used across trials |
| CLDN18.2 (claudin 18.2) | ~60–80% (any expression); ~38–57% (high expression) | Tight-junction protein selectively expressed in GC | FDA-approved target for zolbetuximab; correlates with ADC response |
| FGFR2b (fibroblast growth factor receptor 2b) | ~16% (overexpression ≥10%) | FGFR2 amplification/overexpression, HER2-negative enriched | Emerging target for bemarituzumab and FGFR inhibitors |
| MET | ~5–10% (amplification) | MET proto-oncogene amplification/overexpression | Investigational; ongoing trials with MET-directed agents |
| NTRK | <1% | NTRK1/2/3 oncogenic fusions | Tumor-agnostic FDA approval (larotrectinib, entrectinib); rare in GC |
Biomarker standardization remains a critical unmet need. PD-L1 evaluation varies across trials using CPS, tumor proportion score (TPS), and different thresholds; MSI testing methods and TMB definitions similarly lack harmonization, complicating cross-trial comparisons and clinical implementation 6.
Current and Emerging Therapeutic Strategies
Treatment paradigms have evolved from histology-driven cytotoxic chemotherapy to molecularly stratified, biomarker-directed precision oncology (Table 3). For resectable disease, the perioperative FLOT regimen (fluorouracil, leucovorin, oxaliplatin, docetaxel—4 cycles pre- and post-surgery) represents the Western standard of care, endorsed by NCCN 2025 guidelines 9. Biomarker testing at diagnosis should guide subsequent systemic therapy selection.
For advanced/metastatic disease, key milestones include: the ToGA trial establishing trastuzumab plus chemotherapy as first-line standard in HER2-positive GC (overall survival [OS] 13.8 vs. 11.1 months) 6; the CheckMate-649 trial demonstrating nivolumab plus chemotherapy superiority in PD-L1 CPS ≥5 patients (median OS 14.4 vs. 11.1 months) 6; and the DESTINY-Gastric01 trial confirming trastuzumab deruxtecan (T-DXd) efficacy in trastuzumab-refractory HER2-positive GC (median OS 12.5 vs. 8.4 months; objective response rate [ORR] 51% vs. 14%) 6.
The most consequential recent regulatory advance is the FDA approval of zolbetuximab-clzb (Vyloy) on October 18, 2024, the first CLDN18.2-directed therapy approved for first-line treatment of locally advanced or metastatic HER2-negative GC whose tumors are CLDN18.2-positive 7. In the SPOTLIGHT trial (n=565), zolbetuximab plus mFOLFOX6 achieved median progression-free survival (PFS) of 10.6 months versus 8.7 months (hazard ratio [HR] 0.751; P=0.0066) and median OS of 18.2 months versus 15.5 months (HR 0.750; P=0.0053) 7. The GLOW trial (n=507) similarly demonstrated improved median PFS (8.2 vs. 6.8 months; HR 0.687; P=0.0007) and median OS (14.4 vs. 12.2 months; HR 0.771; P=0.0118) with zolbetuximab plus CAPOX 7. The VENTANA CLDN18 (43-14A) RxDx Assay was concurrently approved as a companion diagnostic 7.
In the HER2 landscape, zanidatamab, a bispecific antibody binding HER2 extracellular domains 2 and 4, demonstrated superiority over trastuzumab-based chemotherapy in the phase 3 HERIZON-GEA-01 trial (data cutoff October 2025; median follow-up 26 months): median PFS 12.4 months versus 8.1 months, and median OS 26.4 months (zanidatamab plus tislelizumab plus chemotherapy) versus 19.2 months with trastuzumab plus chemotherapy—the first phase 3 GC trial to report median PFS >12 months and median OS >24 months 19.
For FGFR2b-overexpressing (~16%) HER2-negative GC, bemarituzumab plus mFOLFOX6 demonstrated median OS of 17.9 versus 12.5 months (HR 0.61; P=0.005) at primary analysis of the FORTITUDE-101 phase 3 trial, though longer follow-up showed OS convergence (14.5 vs. 13.2 months; HR 0.82) 18. Corneal adverse events were the predominant toxicity (largely reversible).
Next-generation CLDN18.2-directed ADCs are advancing rapidly. IBI343, an anti-CLDN18.2 antibody conjugated to exatecan via site-specific glycan conjugation (drug-to-antibody ratio 4), demonstrated confirmed ORR of 29–47% in CLDN18.2-high (≥75% expression) GC at the recommended phase 2 dose of 6–8 mg/kg every 3 weeks in a phase 1 trial (n=127), with a disease control rate of 88–90% and no interstitial lung disease events observed 17.
Table 3. Emerging and Established Therapies in Gastric Cancer by Target
| Drug / Agent | Target | Drug Class | Status | Key Efficacy (GC/GEJ) | Setting |
|---|---|---|---|---|---|
| Trastuzumab | HER2 | Monoclonal antibody (mAb) | Standard of care | ToGA: OS 13.8 vs. 11.1 mo 6 | 1st-line metastatic, HER2+ |
| Pembrolizumab | PD-1 | ICI | Standard of care | KEYNOTE-062: OS 17.4 vs. 10.8 mo (CPS ≥10) 6 | 1st-line metastatic |
| Nivolumab | PD-1 | ICI | Standard of care | CheckMate-649: OS 14.4 vs. 11.1 mo (CPS ≥5) 6 | 1st-line metastatic |
| Ramucirumab | VEGFR2 | mAb | Standard of care | RAINBOW: OS 9.6 vs. 7.4 mo 6 | 2nd-line metastatic |
| Trastuzumab deruxtecan (T-DXd) | HER2 | ADC | Approved (2021; tumor-agnostic expansion 2024) | DESTINY-Gastric01: OS 12.5 vs. 8.4 mo; ORR 51% 6 | 2nd-line+, HER2+ |
| Zolbetuximab-clzb (Vyloy) | CLDN18.2 | mAb | FDA approved Oct 2024 7 | SPOTLIGHT: OS 18.2 vs. 15.5 mo; GLOW: OS 14.4 vs. 12.2 mo | 1st-line metastatic, CLDN18.2+ HER2− |
| Zanidatamab | HER2 (domains 2+4) | Bispecific mAb | Phase 3 (HERIZON-GEA-01) | OS 26.4 vs. 19.2 mo (with tis + chemo) 19 | 1st-line metastatic, HER2+ |
| Bemarituzumab | FGFR2b | mAb | Phase 3 (FORTITUDE-101) | OS 17.9 vs. 12.5 mo (primary; HR 0.61) 18 | 1st-line, FGFR2b+ |
| IBI343 | CLDN18.2 | ADC (exatecan) | Phase 1/2 | ORR 29–47% (CLDN18.2-high) 17 | Refractory GC |
| Durvalumab + FLOT | PD-L1 | ICI + chemo | Phase 3 (MATTERHORN) | pCR 19% vs. 7% (ESMO 2024); EFS data maturing | Perioperative |
| Savolitinib / Crizotinib | MET | Small molecule | Investigational | Phase 1/2 data; limited GC-specific efficacy confirmed | MET-amplified GC |
Clinical Implications and Future Directions
The integration of TCGA-informed molecular classification and companion diagnostics into clinical practice mandates a comprehensive biomarker work-up for all patients with advanced GC: at minimum, HER2 status, PD-L1 CPS, MMR/MSI status, and CLDN18.2 expression should be assessed, with FGFR2b, MET, EBV, and TMB evaluated where trial access permits 97. Biomarker-first patient selection is now the organizing principle of GC drug development, as demonstrated by the companion diagnostic co-approval with zolbetuximab 7.
Substantial unmet needs persist. Late-stage diagnosis remains prevalent outside organized screening regions, limiting curative surgical candidacy 10. Treatment resistance across HER2-directed, checkpoint-directed, and kinase-directed axes demands combinatorial and sequencing strategies—reflected in the pipeline of dual-checkpoint constructs (LAG-3/PD-1, TIGIT/PD-1, CTLA-4/PD-L1), CLDN18.2-directed ADCs, and novel bispecific formats 16. Geographic disparities in access to molecular testing and novel agents—particularly across sub-Saharan Africa and parts of Asia—represent a critical equity challenge; sub-Saharan Africa faces a projected sixfold increase in GC burden relative to 2022 estimates 20. Perioperative immunotherapy trials (MATTERHORN with durvalumab, KEYNOTE-585 with pembrolizumab, DANTE with nivolumab) have begun to reshape neoadjuvant and adjuvant paradigms, with overall survival data still maturing 17. Finally, an H. pylori vaccine—potentially preventable for 76% of GC cases—remains an unfulfilled priority with only one candidate having completed phase 3 testing to date 9.
Conclusion
Gastric cancer imposes a profound and geographically heterogeneous global burden, with incidence concentrated in East Asia and an absolute case load projected to nearly double by 2050. Molecular classification—anchored in the TCGA framework—has transformed GC from a clinically uniform entity into a collection of biologically distinct subtypes with defined therapeutic vulnerabilities. The past 12 months have seen landmark advances, including FDA approval of the first CLDN18.2-directed agent, the emergence of zanidatamab as a potential new HER2 standard, and advancing CLDN18.2 ADC and FGFR2b-directed data. Realizing the full clinical benefit of these advances requires coordinated investment in prevention programs, molecular diagnostics infrastructure, biomarker standardization, and equitable access to precision oncology across all geographic and socioeconomic settings.