Establishment of predictive model for patients with kidney cancer bone metastasis: a study based on SEER database

Background: Bone is a common metastatic tissue of kidney cancer. Accurate prediction of the prognosis of patients with kidney cancer bone metastasis (KCBM) can help doctors and patients choose a further appropriate treatment. Methods: During the period from January 1, 2010 to December 31, 2015, screening patients with kidney cancer diagnosed with bone metastases from the SEER database. Summary of demographic, pathology, number of other metastatic organs, and treatment for KCBM patients. All prognostic factors were plotted for Kaplan-Meier survival curves and log-rank test. Prognostic factors of P<0.001 in the log-rank test were chosen and used to establish nomograms of OS and KCSS. We used C-index, ROC curve, and calibration plot to test the prediction accuracy of two nomograms. Results: A total of 4,234 KCBM patients were included in the study, and patients were diagnosed between January 1, 2010 and December 31, 2015. The model establishment group included 2,966 KCBM patients and the validation group included 1,268 KCBM patients. We have established nomograms for OS and KCSS respectively. These two nomograms included factors such as age, marital status, insurance status, histological type, grade, T stage, N stage, number of extra-bone metastatic organs, surgery, RT, and CT. The C-index of nomograms of OS and KCSS was 0.733 and 0.752, respectively. In all ROC curves, all AUC values were greater than 0.7, proving that the nomograms of both OS and KCSS have achieved medium prediction accuracy. The calibration plots of the model establishment group and the validation group showed good consistency between the predicted nomograms of OS and KCSS. Conclusions: In this study, nomograms of OS and KCSS were established based on the published data of KCBM patients in the SEER database, and the model was validated internally and externally. The prediction accuracy of nomograms of OS and KCSS achieved satisfactory results. At present, this model has the ability to predict the prognosis of KCBM patients and can be used in clinical work.