A model for projecting individuals' risk of esophageal squamous cell carcinoma in a high-risk Chinese population

Early detection of esophageal squamous cell carcinoma (ESCC) may improve survival, but universal screening is not feasible. This study aimed to develop and validate a model for identifying individuals at high absolute risk of ESCC in a high-risk Chinese population.

The model was developed by using data from a case-control study in Yanting County, Sichuan, China between 2011 and 2013, including 942 ESCC cases and 942 age- and sex-matched control participants. Conditional logistic regression and the Gail algorithm were used to construct the model. Model performance was assessed by using the area under the receiver-operating characteristic curve (AUC) with 10-fold cross-validation. External validation was performed in two independent populations, i.e. another Chinese case-control study and the UK Biobank cohort.

The model included six risk factors: education level, marital status, tobacco smoking, alcohol consumption, body mass index (overweight status), and family history of cancer. The model incorporated age- and sex-specific incidence rates in the population to estimate the 5-year absolute risk. The AUC was 0.72 (95% confidence interval [CI] 0.69-0.74) in the derivation dataset and 0.68 (95% CI, 0.67-0.69) after cross-validation. In external validation, the AUC was 0.65 (95% CI, 0.61-0.69) in the independent Chinese case-control study and 0.66 (95% CI, 0.56-0.75) in UK Biobank. The estimated 5-year absolute risk of ESCC in the population in Yanting ranged from 0.0005% to 18.2%. In the group at the highest predicted risk, six individuals would need to undergo screening to detect one ESCC case within 5 years.

The developed model demonstrated acceptable performance and has the potential to identify high-risk individuals for targeted prevention and early detection in the studied high-risk population.