Prostate Health Index density improves detection of clinically significant prostate cancer

ObjectivesTo explore the utility of Prostate Health Index (PHI) density for the detection of clinically significant prostate cancer (PCa) in a contemporary cohort of men presenting for diagnostic evaluation of PCa. Patients and MethodsThe study cohort included patients with elevated prostate-specific antigen (PSA; >2 ng/mL) and negative digital rectal examination who underwent PHI testing and prostate biopsy at our institution in 2015. Serum markers were prospectively measured per standard clinical pathway. PHI was calculated as ([{-2}proPSA/free PSA] x [PSA](1/2)), and density calculations were performed using prostate volume as determined by transrectal ultrasonography. Logistic regression was used to assess the ability of serum markers to predict clinically significant PCa, defined as any Gleason score 7 cancer or Gleason score 6 cancer in >2 cores or >50% of any positive core. ResultsOf 118 men with PHI testing who underwent biopsy, 47 (39.8%) were found to have clinically significant PCa on biopsy. The median (interquartile range [IQR]) PHI density was 0.70 (0.43-1.21), and was 0.53 (0.36-0.75) in men with negative biopsy or clinically insignificant PCa and 1.21 (0.74-1.88) in men with clinically significant PCa (P < 0.001). Clinically significant PCa was detected in 3.6% of men in the first quartile of PHI density (<0.43), 36.7% of men in the IQR of PHI density (0.43-1.21), and 80.0% of men with PHI density >1.21 (P < 0.001). Using a threshold of 0.43, PHI density was 97.9% sensitive and 38.0% specific for clinically significant PCa, and 100% sensitive for Gleason score 7 disease. Compared with PSA (area under the curve [AUC] 0.52), PSA density (AUC 0.70), %free PSA (AUC 0.75), the product of %free PSA and prostate volume (AUC 0.79), and PHI (AUC 0.76), PHI density had the highest discriminative ability for clinically significant PCa (AUC 0.84). ConclusionsBased on the present prospective single-centre experience, PHI density could be used to avoid 38% of unnecessary biopsies, while failing to detect only 2% of clinically significant cancers.