Duke-NUS Medical School Unveils Groundbreaking Insights into Intestinal Metaplasia and Stomach Cancer Risk

In a significant advancement for gastric cancer research, scientists at Duke-NUS Medical School in Singapore have led an international study revealing critical genetic and molecular mechanisms underlying intestinal metaplasia (IM), a key precancerous condition that heightens the risk of stomach cancer. Published in Cancer Discovery on January 14, 2026, the research analyzed over 1,500 IM samples from six countries, identifying novel mutational signatures, driver genes, and the role of clonal hematopoiesis (CH) in disease progression. This work not only sheds light on why stomach cancer rates vary globally but also offers potential biomarkers for early detection and intervention, particularly relevant in Singapore where gastric cancer claims around 300 lives annually and ranks among the top 10 cancer-related deaths.

The study, spearheaded by researchers including Kie Kyon Huang and senior investigator Patrick Tan, Duke-NUS Dean and Professor of Cancer Science, highlights how factors like ageing, smoking, oral bacteria overgrowth, and specific genetic mutations converge to drive IM towards malignancy. By decoding these processes, the findings pave the way for precision medicine strategies in Singapore's robust biomedical ecosystem.

What is Intestinal Metaplasia and Why Does It Matter?

Intestinal metaplasia (IM) occurs when the normal cells lining the stomach transform into cells resembling those of the intestine, a change often triggered by chronic inflammation from Helicobacter pylori (H. pylori) infection—the primary cause in over 75% of gastric cancer cases. This metaplastic shift is part of the Correa cascade, a stepwise progression from gastritis to IM, dysplasia, and finally invasive gastric cancer (GC). While IM itself is reversible in some cases through H. pylori eradication, persistent IM significantly elevates GC risk, making it a crucial target for surveillance.

In Singapore, an intermediate-risk region for GC (incidence around 8.0 per 100,000 for men), IM prevalence is notable among endoscopy patients, underscoring the need for advanced risk stratification. The Duke-NUS study provides the largest genomic dataset on IM to date, demonstrating low-clonality somatic mutations (average 23 per sample, variant allele frequency [VAF] 2.9%) that accumulate differently across risk strata.

Deciphering Mutational Signatures: SBS17 as an IM Hallmark

Mutational signatures are characteristic patterns of DNA mutations (e.g., single base substitutions or SBS) imprinted by specific mutagenic processes, akin to a molecular fingerprint. The Duke-NUS team used whole-genome sequencing (WGS) on 20 IM samples and targeted sequencing on over 1,500, uncovering SBS17—a previously underrecognized signature comprising 27.1% of mutations in IM, absent in normal gastric tissue.

SBS17 is enriched 14.5-fold in late-replicating genomic regions, correlates with hypomethylation and oxidative phosphorylation (OXPHOS) activity, and shows ties to tobacco exposure. Prevalence reached 26.2% across the cohort, higher in high-GC-risk countries like Japan (51.5%) versus low-risk areas. Unlike age-related SBS1 (17.0%), SBS17 mutations had lower VAFs, suggesting ongoing activity in IM evolution. These insights explain regional GC disparities and highlight smoking cessation as a modifiable risk reducer.

Clonal Hematopoiesis: An Age-Related Immune Saboteur in IM

Clonal hematopoiesis (CH), or more specifically clonal hematopoiesis of indeterminate potential (CHIP when VAF >2%), involves somatic mutations in hematopoietic stem cells (e.g., DNMT3A, TET2, ASXL1, PPM1D), leading to their clonal dominance in blood. Age-associated, CH affects up to 30% of older adults and promotes systemic inflammation via altered immune cell function.

In the study, CH was detected in 21% of 1,067 IM patients (225 cases), rising sharply with age (P = 2.7 × 10⁻¹²) and smoking (ASXL1 mutations: 4.9% smokers vs. 0.9% never-smokers, P = 1.7 × 10⁻⁴). High-CH (VAF >5%) predicted progression to dysplasia (P = 0.045) and early gastric neoplasia (EGN; P = 6.4 × 10⁻³), co-occurring with PIGR truncations (OR=2.8, P=0.017)—a gene crucial for mucosal IgA secretion. This duo fostered oral bacteria infiltration (e.g., Streptococcus, Neisseria), amplifying inflammation. Models incorporating CH improved GC risk prediction (AUC 0.72 to 0.773).

Photo by Galen Crout on Unsplash

• CH elevates somatic mutations in IM epithelium (P=1.3×10⁻²).
• Increased IgA+ plasma cells and T-cells (P<0.01).
• Spatial transcriptomics confirmed bacterial co-localization with inflammation markers like CXCL8.

Driver Genes Fueling IM Progression

Forty-seven significantly mutated genes (q<0.005) were pinpointed, including 25 novel ones like SMAD3, PPP2R1A, and KRAS/MAPK pathway members (KRAS, BRAF, MAP2K1, MAP3K1, MAP2K4). High-risk populations showed ARID1A/ARID2 enrichment (worse prognosis, OR=6.2 for EGN), while Singapore/China featured SOX9 truncations (94/909 vs. 2/142 in Japan/Korea). PIGR mutations disrupted immunity, synergizing with CH.

Higher mutation burdens in high-risk areas (28.5 vs. 20 median, P=4.1×10⁻¹¹) underscore environmental-genetic interplay. Organoid models validated KRAS-ERK dependency, responsive to MEK inhibitors.Read the full study abstract on PubMed

Smoking, Ageing, and Microbial Factors Amplify Risks

Ageing drives SBS1/5/40/18 signatures and CH, explaining late-onset GC. Smoking links to SBS17 (P=6.2×10⁻³) and ASXL1 CH mutations. Notably, CH-high IM harbored elevated oral taxa (Streptococcus OR>1, P=0.034), validated by saliva metagenomics (173 Singapore samples) and FISH/Stereo-seq showing bacterial-tumor proximity.

H. pylori CagA variants differed by risk (e.g., E106D in Singapore vs. high-risk strains), modulating host ASPP2 binding. These 'dual-hits' (CH-impaired immunity + dysbiosis) propel Correa cascade.

Global Variations: Lessons from Six Countries

Samples spanned high-risk (Japan, Korea), intermediate (Singapore, Hong Kong, Taiwan), and low-risk (US) regions. High-risk IM had more mutations and ARID1A hits; SBS17 was universal but peaked in high-risk. CH/microbiome patterns were consistent, suggesting universal mechanisms modulated by local factors like H. pylori strains.

This multinational effort, involving Duke-NUS Genome Biology Facility, underscores Singapore's leadership in collaborative oncology.Access the open-access paper

Implications for Prevention and Screening in Singapore

Singapore's National University Health System (NUHS) and Duke-NUS partnership positions the city-state for IM-guided screening. Biomarkers (ARID1A, SBS17, CH, PIGR) could refine endoscopy triage, especially post-H. pylori eradication. Interventions: smoking cessation, CH monitoring in elderly, probiotics/microbiome modulation, KRAS-targeted therapies.

With GC incidence stable but mortality high, these tools align with Singapore's precision health initiatives. Researchers advocate longitudinal cohorts for validation.

Photo by GEE MENG WAH on Unsplash

Duke-NUS Medical School: Pioneering Cancer Genomics in Singapore

Duke-NUS, a graduate-level medical school under National University of Singapore (NUS), excels in translational research. Prof. Patrick Tan's lab at the Cancer Science Institute drives GC genomics, leveraging PRECISE and GCEP1000 cohorts. This study exemplifies Duke-NUS's global impact, fostering PhD/postdoc training in bioinformatics and oncology.Explore research positions in Singapore universities

Alumni often secure roles in biotech; platforms like Rate My Professor highlight mentorship excellence.

Future Outlook: From Bench to Bedside

Prospective trials testing CH/SBS17 for risk prediction, MEK inhibitors for KRAS-high IM, and anti-Streptococcus therapies are next. Singapore's A*STAR and NMRC funding will accelerate this. For aspiring academics, Duke-NUS offers career advice on CVs and scholarships.

• Biomarker panels for endoscopy triage.
• Personalized interventions based on CH status.
• Integration with Singapore's health records for population screening.

Visit faculty jobs or university jobs for openings in cancer research.