Age-specific PCA3 score reference values for diagnosis of prostate cancer

Abstract

Purpose: We evaluated the impact of age on PCA3 score and the utility of age-specific reference values in predicting initial prostate biopsy (pBx) outcomes.

Patients and methods: This single-center, retrospective study included 205 men who underwent an initial 14-core TRUS-guided pBx due to PSA > 3.0 ng/ml or suspicious digital-rectal examination (DRE). PCA3 scores were measured with the Progensa assay. Linear regression models were fit to identify factors that impact PCA3 score and to determine age-specific reference values. Predictive accuracies of logistic regression models predicting presence of prostate cancer (PCa) were analyzed.

Results: The positive biopsy rate was 37%. In multivariable linear regression, age (P < 0.001), presence of PCa (P < 0.001), and multifocal HG-PIN (P = 0.012) were independent predictors of PCA3 score. Age showed the strongest impact on PCA3 score (T = 4.77). The upper 95% confidence interval of PCA3 score in each age category was defined as the age-specific limit. A PCA3-score over the age-specific limit (PCA3-age) was associated with an 4.17-fold increased odds of being diagnosed with PCa (P < 0.001). In multivariable logistic regression models predicting the presence of PCa, predictive accuracy of a base model (age, DRE, PSA, volume) increased from 69.6 to 75.4% (P = 0.037) by adding the continuous PCA3 score, to 73.9% (P = 0.098) with the 35 cutoff (PCA3-35) and to 77.1% (P = 0.008) with PCA3-age.

Conclusions: PCA3 score increases with age, independent of PCa presence. Age-specific PCA3 score reference values are superior to PSA, continuous PCA3 score, and PCA3-35 in predicting initial pBx outcome. Therefore, an age-adjusted PCA3 score should be used for interpretation of the results.