论文部分内容阅读
Objectives: To develop and externally validate a prostate health index (PHI)-based nomogram for predicting the presence of prostate cancer (PCa) at biopsy in Chinese men with prostate-specific antigen (PSA) 4-10 ng/mL and normal digital rectal examination (DRE).Patients and Methods: 347 men were recruited from two hospitals between 2012 and 2014 to develop a PHI-based nonogram to predict PCa.To validate these results, we used a separate cohort of 230 men recruited at another center between 2008 and 2013.Inclusion criteria for both cohorts were a PSA 4-10 ng/mL and a normal DRE.Serum total PSA, free PSA and [-2]proPSA (p2PSA) were determined, and various derivatives, such as free-to-tPSA ratio (%fPSA),percentage of p2PSA (%p2PSA), PSA density (PSAD), and PHI, were calculated.Receiver operator curves (ROC) were used to assess the ability to predict PCa.Results: Overall, men in the validation cohort were older (median age=66, interquartile range [IQR] 61-71) than men in the development cohort (median age=64, IQR 59-70;p=0.011).Median PHI and %p2PSA were statistically significant lower (p<0.001) and median %fPSA was higher in men in the validation cohort (p=0.034);however there were no differences in prostate volume, tPSA and p2PSA (all p≥0.190).PHI achieved the highest AUC of 0.839 in the development cohort compared to the other predictors (AUC=0.839, p<0.001).Including age and prostate volume, a PHI-based nomogram was constructed and rendered an AUC of 0.877 (95% confidence interval [CI] 0.813-0.938).The AUC of the nomogram in the validation cohort was 0.786 (95% CI 0.678-0.894).In clinical effectiveness analyses, the PHI-based nomogram reduced unnecessary biopsies from 42.6% to 27% using a 5% threshold risk of PCa to avoid biopsy with no increase in the number of missed cases relative to PSAD based biopsy decision.Conclusion: We developed and externally validated a PHI-nomogram to predict the presence of PCa at biopsy.