Prostate-specific antigen (PSA) density in the diagnostic algorithm of prostate cancer

Background Screening for prostate cancer using prostate-specific antigen (PSA) alone leads to un-necessary biopsying and overdiagnosis. PSA density is easily accessible, but early evidence on its use for biopsy decisions was conflicting and use of PSA density is not commonly recommended in guidelines. Methods We analyzed biopsy outcomes in 5291 men in the population-based STHLM3 study with PSA >= 3 ng/ml and ultrasound-guided prostate volume measurements by using percentages and regression models. PSA density was calculated as total PSA (ng/ml) divided by prostate volume (ml). Main endpoint was clinically significant cancer (csPCa) defined as Gleason Score >= 7. Results The median PSA-density was 0.10 ng/ml(2) (IQR 0.075-0.14). PSA-density was associated with the risk of finding csPCa both with and without adjusting for the additional clinical information age, family history, previous biopsies, total PSA and free/total PSA (OR 1.06; 95% CI: 1.05-1.07 and OR 1.07, 95% CI 1.06-1.08). Discrimination for csPCa was better when PSA density was added to a model with additional clinical information (AUC 0.75 vs. 0.73, P < 0.05). The proportion of men with Gleason Score 6 (ISUP 1) was similar across stratas of PSA-density. Omitting prostate biopsy for men with PSA-density <= 0.07 ng/ml(2) would save 19.7% of biopsy procedures, while missing 6.9% of csPCa. PSA-density cutoffs of 0.10 ng/ml(2) and 0.15 ng/ml(2) resulted in detection of 77% (729/947) and 49% (461/947) of Gleason Score >= 7 tumors. Conclusions PSA-density might inform biopsy decisions, and spare some men from the morbidity associated with a prostate biopsy and diagnosis of low-grade prostate cancer.