RNA Yields and RT-PCR Gene Expression Profiles Obtained from Manual-Microdissected Fixed Paraffin-Embedded Prostate Cancer Needle Core Biopsies
C Magi-Galluzzi, CL Millward, T Maddala, DB Cherbavaz, A Chen, SM Falzarano, M Lee, FL Baehner, EA Klein. Cleveland Clinic, Cleveland, OH; Genomic Health, Inc, Redwood City, CA
Background: Clinical tools based on prostate needle core biopsies (Bx) are needed to guide treatment planning at diagnosis for men with localized prostate cancer (PCA). Limited tissue in Bx specimens poses significant challenges to the development of molecular diagnostic tests. We examined RNA extraction yields and gene expression profiles using an RT-PCR assay to characterize RNA from manually micro-dissected formalin-fixed paraffin embedded (FFPE) PCA biopsy cores. We also investigated the association of RNA yields and gene expression profiles with Gleason Score (GS).
Design: Forty-eight FFPE prostate Bx cases from localized PCA were selected to represent a range of percent tumor involvement (<33%, 33-66%, and ≥66%) and GS (≤6, GS7, and GS≥8). Enriched tumor tissue from up to12 unstained sections per case were subject to RNA extraction. A 24-gene panel of cancer-related (n=18) and reference (n=6) genes was run in triplicate PCR assay wells on specimens with sufficient RNA yield. Descriptive statistics and analysis of variance were performed stratifying on percent tumor involvement and GS. Ordinal logistic regression was used to evaluate the relationship between gene expression and GS category.
Results: The RNA yield ranged from 16 to 2406 ng. Higher RNA yield was observed in samples with higher percent tumor involvement (p=0.02) and higher GS (p=0.01). RNA yield was sufficient (>200ng) in 74% of cases to permit 96-well RT-PCR, with 87% of cases having >100ng RNA yield. The RT-PCR performance, as assessed by PCR signal strength and low well-to-well variability, was comparable to that seen with Oncotype DX® Breast Cancer Assay. A large dynamic range was observed for many cancer-related genes and a number of potentially important PCA biomarkers were identified. Increased expression levels of 6 genes were significantly associated with higher GS, while increased expression levels of 3 genes were significantly associated with lower GS (only 1 gene would be expected to be statistically significant by chance alone).
Conclusions: Sufficient quantity and quality of RNA can be extracted from FFPE prostate biopsy needle cores for cancer biomarker studies. The potentially important PCA biomarkers identified in this study merit further investigation.
Category: Genitourinary (including renal tumors)
Tuesday, March 23, 2010 1:00 PM
Poster Session IV # 124, Tuesday Afternoon