PURPOSE: Renal sinus invasion is an attributive factor affecting the prognosis of renal cell carcinoma (RCC). This study aimed to construct a risk prediction model that could stratify patients with RCC and predict renal sinus invasion with the help of a machine learning (ML) algorithm. PATIENTS AND METHODS: We retrospectively recruited 1229 patients diagnosed with T1 stage RCC at the Baotou Cancer Hospital between November 2013 and August 2021. Iterative analysis was used to screen out predictors related to renal sinus invasion, after which ML-based models were developed to predict renal sinus invasion in patients with T1 stage RCC. The receiver operating characteristic curve (ROC), decision curve analysis (DCA), and clinical impact curve (CIC) were performed to evaluate the robustness and clinical practicability of each model. RESULTS: A total of 21 candidate variables were shortlisted for model building. Iterative analysis screened that neutrophil to albumin ratio (NAR), hemoglobin level * albumin level * lymphocyte count/platelet count ratio (HALP), prognostic nutrition index (PNI), body mass index*serum albumin/neutrophil-lymphocyte ratio (AKI), NAR, and fibrinogen (FIB) concentration (NARFIB), platelet to lymphocyte ratio (PLR), and R.E.N.A.L score was related to renal sinus invasion and contributed significantly to ML-based algorithm. The areas under the ROC curve (AUCs) of the random forest classifier (RFC) model, support vector machine (SVM), eXtreme gradient boosting (XGBoost), artificial neural network (ANN), and decision tree (DT) ranged from 0.797 to 0.924. The optimal risk probability of renal sinus invasion predicted was RFC (AUC = 0.924, 95% confidence interval [CI]: 0.414-1.434), which showed robust discrimination for identifying high-risk patients. CONCLUSION: We successfully develop practical models for renal sinus invasion prediction, particularly the RFC, which could contribute to early detection via integrating systemic inflammatory factors and nutritional parameters.
PURPOSE: Renal sinus invasion is an attributive factor affecting the prognosis of renal cell carcinoma (RCC). This study aimed to construct a risk prediction model that could stratify patients with RCC and predict renal sinus invasion with the help of a machine learning (ML) algorithm. PATIENTS AND METHODS: We retrospectively recruited 1229 patients diagnosed with T1 stage RCC at the Baotou Cancer Hospital between November 2013 and August 2021. Iterative analysis was used to screen out predictors related to renal sinus invasion, after which ML-based models were developed to predict renal sinus invasion in patients with T1 stage RCC. The receiver operating characteristic curve (ROC), decision curve analysis (DCA), and clinical impact curve (CIC) were performed to evaluate the robustness and clinical practicability of each model. RESULTS: A total of 21 candidate variables were shortlisted for model building. Iterative analysis screened that neutrophil to albumin ratio (NAR), hemoglobin level * albumin level * lymphocyte count/platelet count ratio (HALP), prognostic nutrition index (PNI), body mass index*serum albumin/neutrophil-lymphocyte ratio (AKI), NAR, and fibrinogen (FIB) concentration (NARFIB), platelet to lymphocyte ratio (PLR), and R.E.N.A.L score was related to renal sinus invasion and contributed significantly to ML-based algorithm. The areas under the ROC curve (AUCs) of the random forest classifier (RFC) model, support vector machine (SVM), eXtreme gradient boosting (XGBoost), artificial neural network (ANN), and decision tree (DT) ranged from 0.797 to 0.924. The optimal risk probability of renal sinus invasion predicted was RFC (AUC = 0.924, 95% confidence interval [CI]: 0.414-1.434), which showed robust discrimination for identifying high-risk patients. CONCLUSION: We successfully develop practical models for renal sinus invasion prediction, particularly the RFC, which could contribute to early detection via integrating systemic inflammatory factors and nutritional parameters.
Authors: Umberto Capitanio; Karim Bensalah; Axel Bex; Stephen A Boorjian; Freddie Bray; Jonathan Coleman; John L Gore; Maxine Sun; Christopher Wood; Paul Russo Journal: Eur Urol Date: 2018-09-19 Impact factor: 20.096
Authors: Jasmir G Nayak; Premal Patel; Olli Saarela; Zhihui Liu; Anil Kapoor; Antonio Finelli; Simon Tanguay; Ricardo Rendon; Ron Moore; Peter C Black; Louis Lacombe; Rodney H Breau; Jun Kawakami; Darrel E Drachenberg Journal: Urology Date: 2016-03-31 Impact factor: 2.649
Authors: M Stotz; M Pichler; G Absenger; J Szkandera; F Arminger; R Schaberl-Moser; H Samonigg; T Stojakovic; A Gerger Journal: Br J Cancer Date: 2013-12-19 Impact factor: 7.640
Authors: Francesco Guzzi; Luigi Cirillo; Elisa Buti; Francesca Becherucci; Carmela Errichiello; Rosa Maria Roperto; James P Hunter; Paola Romagnani Journal: Int J Mol Sci Date: 2020-09-19 Impact factor: 5.923