Development and external validation of a nomogram for predicting the risk of developing esophageal cancer based on a questionnaire: a multicenter case-control study

Zhang Yaling¹Tao Sun¹Xiaoqi Chen¹Dongdong Li²Feifan Feng¹Li-han Zhang³Xiaoyu Tang¹Jian-Ping Liu⁴Yuling Zheng^5*

• ¹Department of Oncology, the First Affiliated Hospital Henan of University of Chinese Medicine, Zhengzhou, China
• ²Henan University of Chinese Medicine Henan Provincial Hospital of TCM, Zhengzhou, China
• ³Henan Cancer Hospital Affiliated Cancer Hospital of Zhengzhou University, Zhengzhou, China
• ⁴Beijing University of Chinese Medicine, Beijing, China
• ⁵Henan University of Chinese Medicine, Zhengzhou, China

Abstract: Background: Early detection of esophageal cancer (EC) is an effective strategy for reducing mortality. This study aims to construct a predictive nomogram based on a questionnaire survey to assess the risk of EC in the Chinese population and achieve risk stratification. Methods: To ensure and verify the performance of the nomogram, we have established partnerships with 45 research institutions in 10 provinces in China. Our study comprised a total of 4016 participants (3423 participants were considered for the development and internal validation of the nomogram, while 593 participants were considered for the independent external validation). Data were collected via standardized questionnaires, and relevant predictors were screened using a least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression analysis. Finally, 10 core predictive factors were selected based on clinical significance and predictive performance to construct the predictive nomogram. The performance of the nomogram was evaluated using the area under the receiver operating characteristic curve (AUC) and calibration curves, while its clinical benefit was assessed using decision curve analysis (DCA) and clinical impact curves (CIC). Results: In the training cohort, using LASSO regression technology, multivariate logistic regression analysis, and consideration of the clinical significance of variables, 10 predictive factors were ultimately selected to construct a nomogram, including nutritional status, preference for hot food, rate of eating, fruit intake, preference for pickled food, preference for hard food, sex, vegetable intake, age, and cooking oil intake. The nomogram performed well in both the internal validation (corrected AUC = 0.806) and external validation datasets (corrected AUC = 0.843). The calibration curve showed that the nomogram was in good agreement with the observed outcomes. The results of DCA and CIC show that our nomogram demonstrated favorable clinical consistency. Conclusion: This study developed a nomogram, which has been proven to be a convenient, economical, and accurate tool for effectively predicting the risk of developing EC in individual Chinese populations, enabling risk stratification, and serving as a pre-screening tool prior to endoscopic screening for EC. Keywords: population screening; esophageal cancer (EC); nomogram; questionnaire-based screening; risk stratification.