A novel nomogram for predicting occurrence of bone metastasis in lung cancer: a retrospective study

Background: Lung cancer is the leading cause of cancer-related deaths and bone metastases (BM) are common in lung cancer. In this study, we aimed to identify the risk factors and to construct nomogram for predicting occurrence of BM in lung cancer. Methods: A retrospective study of all lung cancer patients at a single institution from January 2005 to January 2018 was performed. The following variables were compared between the BM group and non-BM (nBM) group, including age, gender, histopathology, Alkaline Phosphatase (ALP) levels, calcium and hemoglobin (Hb) levels. Variables with differences were further analyzed by the multivariate logistic regression analysis to identify the risk factors. For predicting the risk for occurrence in lung cancer patients with BM, we constructed nomogram and the predictive performance was estimated by the concordance index, the receiver operating characteristics cure and calibration curve. Results: Data of 4204 patients with lung cancer were collected and 888 eligible patients with BM were enrolled in the study. Younger age, adenocarcinoma, higher ALP levels and lower Hb were the risk factors of BM in lung cancer. Nomogram was constructed to predict the development of BM and was proved to have good calibration and discrimination. Conclusion: BM is highly prevalent in lung cancer patients. Nomogram based on the risk factors for BM in lung cancer was proved to have good performance for predicting the probability of BM occurrence. Keywords: Lung cancer; Bone metastasis; Risk factors; Nomogram