A tool to predict risk for gastric cancer in patients with peptic ulcer disease on the basis of a nationwide cohort.

ABSTRACT

BACKGROUND & AIMS: Patients with gastric ulcers have significantly higher risk of gastric cancer, especially within 2 years after diagnosis. We used data from a national database to develop a personalized risk prediction model for patients with peptic ulcer diseases. METHODS: We collected data from Taiwan's National Health Insurance Research Database on 278,898 patients admitted for the first time with a primary diagnosis of peptic ulcer disease. We used the data to develop a nomogram, which we validated by discrimination and calibration, and in a test cohort. Cumulative incidences of study subjects predicted by the nomogram were examined. RESULTS: In total, 1269 subjects developed gastric cancer. Age, sex, peptic ulcer sites, peptic ulcer complications, Helicobacter pylori eradication, nonsteroidal anti-inflammatory drug use, and surveillance endoscopy were independent factors associated with risk of gastric cancer (all P < .001). The concordance index for the nomogram developed on the basis of these factors was 0.78. Study subjects were divided into quartiles of predicted risk scores; from lowest score quartile to highest, cumulative incidences at 1 year were 7.4/10,000 people, 14.2/10,000 people, 25.5/10,000 people, and 86.6/10,000 people. The cumulative incidences at 2 years were 9.3/10,000 people, 20.9/10,000 people, 38.0/10,000 people, and 135.7/10,000 people for the same quartiles of risk scores. The nomogram was validated in an independent cohort, and similar incidence values were determined. CONCLUSIONS: We developed and validated a nomogram to predict risk for gastric cancer 1 and 2 years after diagnosis of peptic ulcer disease. The nomogram provides a prognostic tool that can be easily used for individuals and can help physicians explain risk levels to patients.