Short-term risk prediction model for pancreatic cancer in type 2 diabetes: development and validation across two independent Chinese cohorts.

The STOP-DM model predicts 6-month pancreatic cancer risk in type 2 diabetes with an AUC of up to 0.89 across two Chinese cohorts, outperforming existing tools.

• Why it matters: Early detection of pancreatic cancer in individuals with long-standing diabetes is crucial, yet current models mainly focus on new-onset diabetes, missing high-risk long-term cases.
• What they did: Using a retrospective cohort design, researchers developed and validated a random forest-based model with 11 clinical predictors across over 430,000 Chinese patients, assessing risk at multiple time points.
• The result: The model accurately identifies high-risk individuals, with the top 5% accounting for nearly 60% of incident cases, enabling targeted screening and improved early diagnosis strategies.