Prostate Cancer Screening in the Fit Chilean Elderly: a Head to Head Comparison of Total Serum PSA versus Age Adjusted PSA versus Primary Circulating Prostate Cells to Detect Prostate Cancer at Initial Biopsy

Abstract

Background: Prostate cancer is predominately a disease of older men, with a median age of diagnosis of 68years and 71% of cancer deaths occurring in those over 75 years of age. While prostate cancer screening is notrecommended for men >70 years, fit elderly men with controlled comorbidities may have a relatively long lifeexpectancy. We compare the use of age related PSA with the detection of primary malignant circulating prostatecells mCPCs to detect clinically significant PC in this population. Materials and
Methods: All men undergoing PCscreening with a PSA >4.0ng/ml underwent TRUS 12 core prostate biopsy (PB). Age, PSA, PB results defined ascancer/no-cancer, Gleason, number of positive cores and percentage infiltration were registered. Men had an 8mlblood sample taken for mCPC detection; mononuclear cells were obtained using differential gel centrifugationand mCPCs were identified using immunocytochemistry with anti-PSA and anti-P504S. A mCPC was definedas a cell expressing PSA and P504S; a positive test as at least one mCPC detected/sample. Diagnostic yieldsfor subgroups were calculated and the number of avoided PBs registered. Esptein criteria were used to definesmall grade tumours.
Results: A total of 610 men underwent PB, 398 of whom were aged <70yrs. Men over 70yrs had: a higher median PSA, 6.24ng/ml versus 5.59ng/ml (p=0.04); and a higher frequency of cancer detected90/212 (43%) versus 134/398 (34%) (p=0.032). Some 34/134 cancers in men <70yrs versus 22/90 (24%) of men>70yrs complied with criteria for active surveillance. CPC detection: 154/398 (39%) men <70yrs were CPC (+),specificity for cancer 86%, sensitivity 88%, 14/16 with a false (-) result had a small low grade PC. In men >70years, 88/212 (42%) were CPC (+); specificity 92%, sensitivity 87%, 10/12 with a false (-) had small low gradetumours. False (+) results were more common in younger men 36/154 versus 10/88 (p<0.02). With a PSA cutoffof 6.5ng/ml, in men <70yrs, 108 PB would be avoided, missing 56 cancers of which 48 were clinically significant.Using CPC detection, 124 biopsies would be avoided, missing only 2 clinically significant cancers. In men >70yrs using a PSA >6.5ng/ml would have resulted in 108 PB with 34 PC detected, of which 14(41%) were smalllow grade tumours.
Conclusions: The use of CPC detection in the fit elderly significantly decreases the numberof PBs without missing clinically significant cancers, indicating superiority to the use of age-related PSA.

Keywords