Comparison And Development Of Preoperative Systemic Inflammation Markers-Based Models For The Prediction Of Unfavorable Pathology In Newly Diagnosed Clinical T1 Renal Cell Carcinoma

Background: We sought to investigate the preoperative risk factors associated with the unfavorable pathology (UP) of clinical T1 (cT1) renal lesions. The aims of this study were to develop and compare several novel models capable of accurately identifying those patients at high risk of harboring occult adverse histopathological characteristics. Methods: The clinical parameters and preoperative laboratory test results from 1281 cT1 renal cell carcinomas (RCCs) patients who underwent partial nephrectomy (PN) or radical nephrectomy (RN) were collected. The data was randomly split into training (70%) and testing (30%) datasets. We performed univariable and multivariable logistic regression analyses for significant predictors and, subsequently, constructed predictive models based on those significant risk factors. Receiver operating characteristic (ROC) analysis was used to determine the model with the highest discrimination power with corresponding area under the curve (AUC). Calibration curves were plotted and decision curve analyses (DCAs) were applied to explore clinical net benefit. Results: UP was identified in 21.1% (n = 270), 21.0% (n = 188) and 21.3% (n = 82) patients in the total population, training cohort and validation cohort, respectively. R.E.N.A.L. (radius, exophytic/endophytic properties, nearness of tumor to collecting system or sinus, anterior/posterior, location relative to the polar lines) nephrometry score, tumor size, neutrophil-to-lymphocyte ratio (NLR) and albumin-to-globulin ratio (AGR) were independent predictors of UP. Among those predictive models, the model that consisted of tumor size, hemoglobin, NLR and AGR performs best according to the highest AUC of 0.70 and the highest net benefit. When tumor histology was added to the biomarker-based model, including tumor size, hemoglobin, NLR and AGR, the AUC improved from 0.60 to 0.63 in the validation cohort. Conclusions: In this analytical model study, our findings verified that systemic inflammation response markers showed high potential for identifying UP. Our biomarker-based models well predicted occult aggressive histopathological characteristics among patients with cT1 renal lesions, and the use of models may be greatly beneficial to urologists in tailoring precise management and therapy for patients. Robust validation is warranted prior to adoption into clinical practice.