Response

Abstract

We thank Etemadi et al1Etemadi A. Dawsey S.M. Abnet C.C. et al.Predicting the risk of esophageal high-grade lesions in opportunistic screening.Gastrointest Endosc. 2020; 92: 1136-1137Abstract Full Text Full Text PDF Scopus (1) Google Scholar for their interest in our study,2Liu Z. Guo C. He Y. et al.A clinical model predicting the risk of esophageal high-grade lesions in opportunistic screening: a multicenter real-world study in China.Gastrointest Endosc. 2020; 91: 1253-1260.e3Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar and we would like to address their questions. We agree with the authors that the aim of screening is to detect malignant lesions in the preclinical stage. However, “preclinical” does not always equate with “asymptomatic.” In rural China, for example, many cancer patients do not seek medical attention until symptoms are quite obvious, because of socioeconomic limitations and poor health awareness.3Wang J. Liu F. Gao H. et al.The symptom-to-treatment delay and stage at the time of treatment in cancer of esophagus.Jpn J Clin Oncol. 2008; 38: 87-91Crossref PubMed Scopus (29) Google Scholar Gastroenterologists, on the other hand, often fail to refer esophageal cancer patients in a timely manner because symptoms of esophageal squamous cell carcinoma (ESCC), including, for example, dysphagia and retrosternal pain, are not always specific. Thus, “diagnosing patients with symptoms as early as possible (downstaging)” should be another important goal in clinical opportunistic screening, just as with breast cancer and cervical cancer.4Dos-Santos-Silva I. De Stavola B.L. Renna Jr., N.L. et al.Ethnoracial and social trends in breast cancer staging at diagnosis in Brazil, 2001-14: a case only analysis.Lancet Glob Health. 2019; 7: e784-e797Abstract Full Text Full Text PDF PubMed Scopus (33) Google Scholar, 5Ananth R. Downstaging of cervical cancer.J Indian Med Assoc. 2000; 98: 41-44PubMed Google Scholar, 6Yip C.H. Downstaging is more important than screening for asymptomatic breast cancer.Lancet Glob Health. 2019; 7: e690-e691Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar Our model will facilitate motivating the patient to seek endoscopic examination and assist clinicians in precision referral, ultimately contributing to the establishment of the backbone of cost-effective screening programs.7Perbtani Y. Qumseya B.J. A prediction model for detection of esophageal squamous cell cancer: a new beginning or more of the same?.Gastrointest Endosc. 2020; 91: 1261-1263Abstract Full Text Full Text PDF PubMed Scopus (2) Google Scholar Regarding the weighting of predictors, the area under the curve (AUC) should be the correct indicator rather than coefficients in the model because coefficients are largely determined by the dimension and variation of the variable per se. In single-variable models with age, smoking, body mass index, dysphagia, and retrosternal pain, the AUCs were 0.797, 0.594, 0.566, 0.671, and 0.581, respectively, and the AUC of the model containing only 2 symptom variables was also significantly lower than our 5-variable model (0.707 vs 0.871, P < .001), showing the important role nonsymptom predictors play. According to our population-based surveys in China, weight loss, rather than a specific symptom of advanced ESCC,8He Z. Liu Z. Liu M. et al.Efficacy of endoscopic screening for esophageal cancer in China (ESECC): design and preliminary results of a population-based randomised controlled trial.Gut. 2019; 68: 198-206Crossref PubMed Scopus (68) Google Scholar,9Liu M. Liu Z. Cai H. et al.A model to identify individuals at high risk for esophageal squamous cell carcinoma and precancerous lesions in regions of high prevalence in China.Clin Gastroenterol Hepatol. 2017; 15: 1538-1546.e7Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar could occur in early-stage malignant esophageal lesions. Finally, the discrimination power of our model was 87% in the development set and 84% in the validation set, not 89% as quoted in the letter. All authors disclosed no financial relationships. Predicting the risk of esophageal high-grade lesions in opportunistic screeningGastrointestinal EndoscopyVol. 92Issue 5PreviewWe read with interest the article by Liu et al1 and were particularly intrigued by the 89% discrimination power, which is much higher than those of previous models,2,3 even those reported by the same group,4 or well-known models such as the Gail model for breast cancer.5 For a disease with a very high mortality, screening by a simple noninvasive model can offer clear survival benefits. Full-Text PDF