Background: Selection of high-risk subjects for endoscopic screening of esophageal squamous cell carcinoma (ESCC) lacks individual predictive tools based on environmental risk factors.
Methods: We performed a large population-based case-control study of 1418 ESCC cases and 1992 controls in a high-risk area of China. Information on potential risk factors was collected via face-to-face interview using an electronic structured questionnaire. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using unconditional logistic regression models, and predictive nomograms were established accordingly. A weighted analysis was further conducted to introduce age into predictive nomograms due to frequency matching study design.
Results: Most cases were usually exposed to 4 to 6 risk factors, but most controls were usually exposed to 3 to 5 risk factors. The AUCs of male and female predictive nomograms were 0.75 (95%CI: 0.72, 0.77) and 0.76 (95%CI: 0.73, 0.79), respectively. The weighted analysis adding age in the predictive model improved the AUC in both men and women (0.81 (95%CI: 0.79, 0.84) and 0.88 (95%CI: 0.85, 0.90), respectively).
Conclusions: An easy-to-use preclinical predictive tool is provided to select candidate population with high ESCC risk for endoscopic screening. Its usefulness needs to be further evaluated in future screening practice.
Keywords: Environmental risk factors; Esophageal squamous cell carcinoma; Nomogram; Predictive model; Prevention; Screening.