Molecular Factors Showing Multivariate Significance for Outcome in a Conservatively Treated Prostate Cancer Biopsy Cohort
Daniel Berney, Gabrielle Fisher, Zi Hua Yang, Henrik Moller, Sak Kudahetti, Christopher Foster, Victor Reuter, Peter Scardino, Jack Cuzick. Barts and the London School of Medicine, London, United Kingdom; Memorial Sloan-Kettering Cancer Center, New York; Kings College London, London, United Kingdom; University of Liverpool, Liverpool, United Kingdom
Background: Using standard clinical parameters it is still not possible to differentiate indolent from aggressive prostate cancer (PC). We have previously reported on a cohort of 808 clinically localised PCs diagnosed by TURP and treated conservatively with long term follow up, and shown in the TURP cohort that ki-67 and P53 predict outcome in a multivariate model independently of Gleason score (GS) and Serum PSA. A 31 gene cell cycle progression (CCP) RNA signature outperformed all immunochemical markers. We here present the results of translation of this work into the more clincally relevant biopsy cohort.
Design: Biopsy specimens from 1991-6 were collected from patients with conservatively treated PC. PSA and GS were available with an endpoint of death from PC. A biopsy tissue micro-array (TMA) was constructed, and immunohistochemically (IHC) stained for Ki-67 and P53. Immunochemical positivity was measured in each core in a semi-quantitative manner examining strength of signal from 0-3 and percentage positivity to give a composite score. Also, RNA levels of 46 cell cycle progression (CCP) genes were determined in this cohort and a mean CCP score calculated for each patient.
Results: Ki-67 (n=295) correlated with GS (p<0.001). Assessed in 3 categories (≤5%, >5-≤10%, >10%), it is significant in univariate (χ2 trend=10.08, p<0.002) and multivariate models including GSand PSA (χ2 trend=5.66, p<0.017). P53 was not significant in univariate or multivariate models. However, CCP score outperformed all IHC markers. In a multivariate analysis which included Gleason and PSA, CCP dominated (HR for change from 25th to 75th percentile = 2.07, 95% CI(1.50, 2.85) χ2 = 20.8, P = 5 x 10-6) with only GS providing a significant additional contribution (χ2 = 15, 1df, P = 0.0001).
Conclusions: IHC scoring on TMA prostate biopsies is practicable and yields significant results for Ki-67, but is currently considerably outperformed by CCP score. Ki67 results are less impressive than those on the TURP cohort; this may be due to the smaller samples analysed, where inadequate samplng may be an issue.
We suggest that despite the wealth of IHC data in radical prostatectomy and TURP cohorts, translating this into useful assays on small amounts of biopsy tissue may not be as informative as other techniques such as RNA expression data which showed its clear superiority in this set of material.
Category: Genitourinary (including renal tumors)
Tuesday, March 20, 2012 1:00 PM
Poster Session IV # 106, Tuesday Afternoon