Heparin br Given the importance of improving risk discrimination among

Given the importance of improving risk discrimination among AA patients, we evaluated the prognostic utility of the CCP score and PTEN Heparin in predicting clinical outcomes in a cohort of men with localized prostate cancer highly enriched in AA patients.
2. Patients and methods
This retrospective study included patients diagnosed with clinically localized adenocarcinoma of the prostate who were treated at The Ochsner Clinic (New Orleans, LA, USA) between January 1, 2006 and December 31, 2011 with available biopsy formalin-fixed, paraffin-embedded (FFPE) tumor block (IRB #2013.243.A). Biopsies had been prospectively collected in a curated tissue bank. Patients were excluded from the study if they had any prior prostate therapy, a presurgical PSA level >100 ng/ml, histology on diagnostic biopsy other than adenocarci-noma of prostate, or nonlocalized disease (clinical T4 or M1). We also excluded patients who were treated with transurethral resection of the prostate, cryosurgery, or laser vaporization of the prostate.
Clinical and demographic data at the time of diagnosis were collected for each patient via chart review and included all the variables required to calculate a CAPRA score (presurgical serum PSA, biopsy Gleason scores, clinical stage, percent needle cores positive, and age at diagnosis) [18]. Self-reported ancestry was categorized as either AA or non-AA. In this study population, almost all non-AA patients were non-Hispanic Caucasians. Initial treatment type, progression of prostate cancer, subsequent treatment, vital status, cause of death, and the dates associated with these events were collected from The Ochsner Clinic tumor registry and/or medical chart review.
2.2. Patient outcomes
The primary clinical outcome was progression to metastatic disease, as confirmed by imaging. Disease-specific mortality (DSM) was an exploratory clinical outcome. Date and cause of death were sought from the Louisiana State Cancer Registry and National Death Index, and were used if prostate cancer death occurred within 6 mo of last patient contact with The Ochsner Clinic Department of Urology.
Time to clinical outcomes was measured in days elapsed since date of diagnosis. All non-events were censored at the date of last follow-up with The Ochsner Clinic or 10 yr from the date of diagnosis, whichever occurred first. Patients without follow-up were censored at the date of last contact with the clinic.
2.3. Biomarker testing
All molecular testing was completed blinded to patient outcomes at Myriad Genetics. CCP testing was performed as previously described with the exception that quality criteria were modified to allow for lower RNA content and greater RNA degradation in the archival tissue [14,19]. A board-certified pathologist (Z.S.) identified carcinoma tissue for analysis from FFPE biopsy samples. Selected tissue regions were macrodissected and deparaffinized (Deparaffinization Solution; Qiagen, Valencia, CA, USA) and RNA extraction was performed using miRNeasy (Qiagen). The expression of 31 CCP genes and 15 housekeeper genes was quantified in triplicate using TaqMan Low Density Arrays (Applied Biosystems, Foster City, CA, USA).
The CCP score was calculated as the average expression of the CCP genes normalized by the expression of the housekeeper genes [14]. CCP scores were considered not passing if the expression for more than nine CCP genes was missing or if the standard deviation for the triplicate score was >0.5. The CCR score was calculated as a linear combination of the CCP and CAPRA scores (0.39 CAPRA + 0.57 CCP) [19].
PTEN immunohistochemistry was performed using rabbit monoclo-nal antibody 138G6 (Cell Signaling Technology, Danvers, MA, USA) as previously described [22]. Negative PTEN expression was defined as no staining in 90% of tumor cells, and positive PTEN expression as staining in >10% of tumor cells.