Detection of prostate cancer in an ethnically diverse, disadvantaged population with multiple prostate specific antigen measurements before biopsy.

ABSTRACT

PURPOSE: We assessed the predictive effect of prostate specific antigen velocity for men with a minimum of 3 pre-biopsy prostate specific antigen measurements in a racially diverse population. MATERIALS AND METHODS: We identified 795 patients who underwent 3 or more prostate specific antigen tests before prostate biopsy. Prostate specific antigen velocity was calculated by linear regression and used to assess associations with the risk of prostate cancer overall and of high grade prostate cancer (Gleason score 7-10). We created ROC curves and determined the AUC for several models, including only prostate specific antigen velocity or the last prostate specific antigen measurement before biopsy, to predict prostate cancer and high grade prostate cancer. RESULTS: The risk of prostate cancer and high grade prostate cancer increased linearly with increasing prostate specific antigen velocity quartiles (each p trend<0.001). Older patients were more likely to be diagnosed with prostate cancer, given the same prostate specific antigen velocity. In black and Hispanic patients there were strong linear associations between increasing prostate specific antigen velocity and the risk of prostate cancer overall and high grade prostate cancer. ROC curves incorporating prostate specific antigen velocity to predict prostate cancer and high grade prostate cancer varied significantly by race. The AUC of models in black and Hispanic patients was significantly higher than in white patients (0.62 and 0.64, respectively, vs 0.47, p=0.03). CONCLUSIONS: Prostate specific antigen velocity is a significant predictor of prostate cancer and high grade prostate cancer in men with 3 or more prostate specific antigen tests before prostate biopsy. Black and Hispanic patients appear to be at increased risk for prostate cancer at higher prostate specific antigen velocity, as are men older than 60 years. Further studies should confirm these results and create age and race specific guidelines to assess prostate specific antigen velocity.