Development and validation of a diagnostic nomogram integrating anatomical scores and systemic immune-inflammatory biomarkers for

Background: At initial diagnosis, approximately 30% of renal cell carcinoma (RCC) patients have Methods: A retrospective analysis was conducted in accordance with the TRIPOD (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis) reporting guideline, involving patients with RCC treated at the First Hospital of Lanzhou University from January 2016 to August 2025. Patients were randomly assigned to a training cohort and a validation cohort in a 7:3 ratio. Variable selection was performed using three machine learning algorithms: LASSO, SVM-RFE, and Boruta. Independent predictors were identified through multivariate logistic regression, and a diagnostic nomogram was constructed. Model performance was evaluated using receiver operating characteristic (ROC) curves, area under the curve (AUC), calibration curves, and decision curve analysis (DCA). Results: Three independent predictors were identified: lymphocyte-to-monocyte ratio (LMR) (OR = 0.78, 95% CI: 0.62-0.98), serum albumin-to-alkaline phosphatase ratio (AAPR) (OR = 0.05, 95% CI: 0.01-0.36), and PADUA (Preoperative Aspects and Dimensions Used for an Anatomical) score (OR = 1.41, 95% CI: 1.18-1.69). The AUC of the nomogram was 0.771 in the training cohort and 0.747 in the validation cohort. Calibration curves demonstrated excellent agreement between predicted and actual probabilities, while decision curve analysis highlighted the nomogram's net clinical benefit across a wide range of threshold probabilities. Conclusion: The developed nomogram demonstrated moderate discriminatory ability and high clinical applicability in identifying cancer metastasis in patients with initially diagnosed RCC. However, further validation with larger sample sizes and multicenter external cohorts is essential to confirm its generalizability.