Purpose: To assess the performance of our health risk appraisal (HRA) models for screening individuals at high risk of esophageal/pharyngeal squamous cell carcinoma (EPSCC).
Methods: Based on the results of our previous case-control study, we invented HRA models that enable screening for EPSCC cases in Japanese men with high sensitivity and specificity based on either their aldehyde dehydrogenase-2 genotype (HRA-G model) or alcohol flushing (HRA-F model) and drinking, smoking, and dietary habits. Follow-up endoscopy combined with esophageal iodine staining (median follow-up period: 5.0 years) was done on 404 Japanese men (50-78 years) who were registered as cancer-free controls in the previous study.
Results: The follow-up endoscopy resulted in a diagnosis of 6 esophageal SCC (T(is) in 5 and T(1) in 1), 1 hypopharyngeal SCC (T(2)), and 1 oropharyngeal SCC (T(2)). Seven and 6 of the 8 EPSCC cases were in the top 10% risk group at baseline according to the HRA-G and HRA-F models, respectively. The EPSCC detection rates per 100 person-years in the top 10% risk groups by the HRA-G and HRA-F models were 4.38 (95% confidence interval, 1.76-9.01) and 3.48 (95% confidence interval, 1.28-7.58), respectively. Their age-adjusted relative risk was 95.1- and 26.3-fold, respectively (P < 0.0001), higher than in the bottom 90% risk groups.
Conclusions: The high detection rates for EPSCC in the top 10% risk group of this preliminary follow-up study were in good agreement with those predicted by the HRA models and thus encouraged the screening based on our HRA models in larger populations of Japanese men.