Document Type: Systematic Review and Meta-analysis
Hospital Carabineros of Chile, Nunoa, Santiago, Chile
Hospital Carabineros of Chile, Nunoa, Santiago, Chile.
Faculty of Medicine, University Finis Terrae, Providencia Diego Portales University, Manuel Rodrıguez Sur, Santiago, Chile.
Faculty of Medicine, Diego Portales University, Manuel Rodrıguez Sur, Santiago, Chile.
Background: The limitations of total serum PSA values remain problematic, especially after an initial negative prostate biopsy. In this prospective study of Chilean men with a continued suspicion of prostate cancer due to a persistently elevated total serum PSA, abnormal digital rectal examination and initial negative prostate biopsy were compared with the use of the on-line Chun nomagram, detection of primary malignant circulating prostate cells (CPCs) and free percent PSA to predict a positive second prostate biopsy. We hypothesized that men negative for circulating prostate cells have a small risk of clinically significant prostate cancer and thus may be conservatively observed. Men positive for circulating prostate cells should undergo biopsy to confirm prostate cancer.
Materials and Methods: Consecutive men with a continued suspicion of prostate cancer underwent 12 core TRUS prostate biopsy; age, total serum PSA and percentage free PSA and Chun nomagram scores were registered. Immediately before biopsy an 8ml blood simple was taken to detect primary mCPCs. Mononuclear cells were obtained by differential gel centrifugation and identified using double immunostaining with anti-PSA and anti-P504S. Biopsies were classifed as cancer/no-cancer, mCPC detecton test as negative/positive and the total number of cells/8ml registered. Areas under the curve (AUC) for percentage free PSA, Chun score and CPCs were calculated and compared. Diagnostic yields were calculated with reference to the number of possible biopsies that could be avoided and the number of clinically significant cancers that would be missed.
Results: A total of 164 men underwent a second biopsy; 41 (25%) had cancer; the AUCs were 0.65 for free PSA, 0.76 for the Chun score and 0.87 for CPC detection, the last having a significantly superior prediction value (p=0.01). Using cut off values of free PSA <10%, Chun score >50% and ≥1 CPC detected, CPC detection had a higher diagnostic yield. Some 4/41 cancers complied with the criteria for active surveillance, free PSA and the Chun score missed a higher number of significant cancers when compared with CPC detection.
Conclusions: Primary CPC detection outperformed the use of free PSA and the Chun nomagram in predicting clinically significant prostate cancer at repeat prostate biopsy.