BACKGROUND: Because early squamous cell carcinoma (SCC) of the esophagus is detectable by endoscopic esophageal iodine staining with high accuracy and is easily treated by endoscopic mucosectomy, it is important to develop efficient methods for screening candidates for the endoscopic examination. Inactive aldehyde dehydrogenase-2 (ALDH2) is a very strong risk factor for esophageal SCC in alcohol drinkers and thus may be suitable as a screening tool. PURPOSE: To assess the performance of health risk appraisal (HRA) models in screening for esophageal SCC in the Japanese male population. METHODS: Two types of HRA models were developed based on our previous case-control study, which included assessment of ALDH2 activity and selected risk factors (HRA-G and HRA-F: activities of ALDH2 assessed by genotype and questionnaire for alcohol flushing, respectively). Each individual's risk of esophageal SCC was calculated quantitatively as a risk score. The sensitivity and specificity of the HRA models at various cutoff values of risk score was estimated by a leave-one-out cross-validation. The positive predictive value was estimated assuming the prevalence of esophageal SCC in the whole population to be 0.17% or 0.39% according to literatures. RESULTS: When individuals ranked in the top 10% of the HRA-F risk score was screened, the sensitivity was 57.9% and positive predictive value was 0.93% or 2.12% according to the above assumptions, respectively. The sensitivity was slightly better by the HRA-G model than by the HRA-F model. CONCLUSION: The HRA models may provide an important approach to early intervention strategies to control esophageal SCC in Japanese men.
BACKGROUND: Because early squamous cell carcinoma (SCC) of the esophagus is detectable by endoscopic esophageal iodine staining with high accuracy and is easily treated by endoscopic mucosectomy, it is important to develop efficient methods for screening candidates for the endoscopic examination. Inactive aldehyde dehydrogenase-2 (ALDH2) is a very strong risk factor for esophageal SCC in alcohol drinkers and thus may be suitable as a screening tool. PURPOSE: To assess the performance of health risk appraisal (HRA) models in screening for esophageal SCC in the Japanese male population. METHODS: Two types of HRA models were developed based on our previous case-control study, which included assessment of ALDH2 activity and selected risk factors (HRA-G and HRA-F: activities of ALDH2 assessed by genotype and questionnaire for alcohol flushing, respectively). Each individual's risk of esophageal SCC was calculated quantitatively as a risk score. The sensitivity and specificity of the HRA models at various cutoff values of risk score was estimated by a leave-one-out cross-validation. The positive predictive value was estimated assuming the prevalence of esophageal SCC in the whole population to be 0.17% or 0.39% according to literatures. RESULTS: When individuals ranked in the top 10% of the HRA-F risk score was screened, the sensitivity was 57.9% and positive predictive value was 0.93% or 2.12% according to the above assumptions, respectively. The sensitivity was slightly better by the HRA-G model than by the HRA-F model. CONCLUSION: The HRA models may provide an important approach to early intervention strategies to control esophageal SCC in Japanese men.