Development and validation of multidimensional nomograms for predicting prostate cancer risk: a retrospective study

PSA screening has limited specificity in the 10-20 ng/mL range, making biopsy decisions difficult. Metabolic disorders and inflammation may improve risk prediction. We aimed to develop a model integrating multidimensional indicators to predict prostate cancer and high-grade disease in men with tPSA >10 ng/mL.

This retrospective study included 461 men who underwent prostate biopsy (tPSA >10 ng/mL). Patients were randomly split into training (70%) and validation (30%) sets. Two logistic regression models were built: Model 1 for benign vs. malignant; Model 2 (in malignant cases) for high-grade (ISUP ≥3) vs. low-grade. Variables with VIF>5 were excluded. Backward stepwise (AIC) and univariate P < 0.05 guided variable selection. Performance was assessed by AUC, calibration, DCA, and compared with fPSA% alone using DeLong test.

Among 461 patients, 252 (54.7%) had prostate cancer, and 139 (55.2% of malignant cases) had high-grade disease. Malignant patients were older, had higher BMI, TyG index, NLR, and tPSA, and lower fPSA% (all P < 0.05). High-grade patients showed similar metabolic differences but no tPSA difference versus low-grade. Model 1 retained age, TyG index, NLR, fPSA%, smoking, hypertension, and lesion location. Model 2 retained age, TyG index, LDH, and smoking history. Validation AUC was 0.871 for Model 1 and 0.779 for Model 2. At optimal thresholds, sensitivity/specificity were 80.3%/74.6% (Model 1) and 71.7%/62.1% (Model 2). Both models showed good calibration (Hosmer-Lemeshow P>0.05). Decision curve analysis and DeLong test confirmed that each full model provided higher net benefit than fPSA% alone (P < 0.05).

Two nomograms using routine clinical and laboratory variables may assist in risk stratification for prostate cancer and high-grade disease in men with tPSA >10 ng/mL, outperforming fPSA% alone. These tools can help reduce unnecessary biopsies.