A prediction model for early biochemical failure after radical prostatectomy based on the CAPRA-S score and the presence of secondary circulating prostate cells.

To establish a prediction model for early biochemical failure based on the CAPRA-S score and secondary circulating prostate cells.

A prospective single center study of men who underwent radical prostatectomy as monotherapy for prostate cancer.

Clinical-pathological findings were used to calculate the CAPRA-S score. 90 days after surgery blood was taken for CPC detection, mononuclear cells were obtained using differential gel centrifugation, and CPCs identified using immunocytochemistry. A CPC was defined as a cell expressing PSA but not CD45. The CPC test was defined as positive or negative. Patients were followed up for up to 5 years, biochemical failure was defined as a PSA >0. 2ng/ml. The validity of the CAPRA-S score was calibrated using partial validation, and Cox proportional hazard regression to build three models, CAPRA-S, CPC and combined models.

321 men participated, mean age 65. 5 years, after 5 years of follow up the biochemcial free survival was 98. 55%. The model using CAPRA-S showed a HR of 7. 66, that of CPC 34. 52 and the combined model showed a HR of 2. 60 for CAPRA-S and 22. 5 for CPC. Using the combined model, 23% of men changed from low risk to high risk or vice versa.

The incorporation of CPC detection significantly increased the discrimination in establishing the probability of biochemcial failure, high risk CAPRA-S patients who are negative for CPCs have a much better prognosis. The addition of CPC detection gives clinically significant information of who may be eligible for adjuvant therapy.

A single center prospective observational study of men following radical prostatectomy for prostate cancer. CAPRA-S scores were obtained from the surgical specimen analysis; secondary CPCs were detected using inmunocytochemistry three months post surgery, a positive sample contained ≥1 PSA (+) CD45 (-) staining cell/blood sample and BF was defined as a serum total PSA >0. 20ng/ml. Five year BF was determined using Cox regression analysis for models using the CAPRA-S, CPC, and combined data, they were compared using a decision analysis curve (DAC), Harrell's C concordance test and predicted versus observed survival using Kaplan-Meier curves.

321 men, mean age 65. 5yrs participated, in whom 193 (60%) had secondary CPCs detected. After 5 years of follow up the predicted biochemical free survival was 98. 6%. For the DAC, the combined CAPRA-S/CPC model was superior to both single variable models with a Harrell's C score of 0. 86. Using the combined model 23. 7% of men changed risk group.

The incorporation of CPC detection into the CAPRA-S score improved significantly its prognostic value, it identified a low risk CAPRA-S sub-group with intermediate risk and a high risk CAPRA-S subgroup with low risk. The incorporation of CPC detection into the CAPRA-s score provides clinically important information on possible treatment decisions. This article is protected by copyright. All rights reserved.

BJU international. 2015 Oct 28 [Epub ahead of print]

Nigel P Murray, Socrates Aedo, Eduardo Reyes, Nelson Orellana, Cynthia Fuentealba, Omar Jacob

Go “Beyond the Abstract” - Read an article commentary written by the authors

Hospital Carabineros of Chile, Nuñoa, 7770199, Santiago, Chile. , Faculty of Medicine, University Finis Terrae, Providencia, 7501015, Santiago, Chile. , Faculty of Medicine, Diego Portales University, Manuel Rodrıguez Sur 415, 8370179, Santiago, Chile. , Hospital DIPRECA, Las Condes, 7601003, Santiago, Chile. , Hospital Carabineros of Chile, Nuñoa, 7770199, Santiago, Chile. , Hospital Carabineros of Chile, Nuñoa, 7770199, Santiago, Chile.

PubMed