Development and validation of a prognostic prediction model for patients with lung metastasis of esophageal cancer

Pulmonary metastasis is relatively rare among esophageal cancer patients, and there is a limited body of research in this regard. This study used clinical and pathological indicators from monitoring, epidemiological, and Surveillance, Epidemiology, and End Results (SEER) databases to look at the risk factors for patients who develop pulmonary metastasis. This study aims to explore the risk factors for lung metastasis in esophageal cancer patients and their impact on prognosis, and to construct a nomogram model for predicting the survival period of patients with lung metastasis. We obtained data on esophageal cancer patients from the Surveillance, SEER database between 2010 and 2015. Patients with esophageal cancer lung metastasis were divided into a training group and a testing group. Univariate and multivariate analyses were performed on the training set to identify independent risk factors for esophageal cancer lung metastasis, and a nomogram model was constructed to predict overall survival (OS). The model’s discriminatory ability and calibration were evaluated using the C-index, area under the receiver operating characteristic (ROC) curve (AUC), and calibration curve. A total of 10,035 esophageal cancer patients were included in this study, with 590 diagnosed with lung metastasis. Single-factor and multi-factor logistic regression analyses revealed several risk factors associated with esophageal cancer lung metastasis, including male gender, higher tumor grades (Grade II and Grade III), advanced T stage, and the presence of bone, brain, and liver metastases. Patients diagnosed with esophageal cancer who have pulmonary metastases have a median overall survival of six months. For patients with esophageal cancer with lung metastasis, significant prognostic factors identified by Cox regression analysis included tumor grade (grade II and III), T2 lymph node metastasis, T3 lymph node metastasis, and the presence of bone metastasis. The constructed nomogram was validated by ROC analysis and calibration curves, demonstrating good predictive accuracy and discriminatory power. There are certain differences in risk and prognostic factors between patients with esophageal cancer and lung metastasis. ROC curves and calibration curves validated that the dynamic nomogram has certain predictive performance.