TMPRSS2-ERG GENE FUSIONS IN MINIMAL PROSTATIC CARCINOMA

TMPRSS2-ERG GENE FUSIONS IN MINIMAL PROSTATIC CARCINOMA

814 THE JOURNAL OF UROLOGY® 2247 TMPRSS2-ERG GENE FUSIONS IN MINIMAL PROSTATIC CARCINOMA Roula Albadine, Mathew Latour, Elizabeth Platz, Alan Meeker...

47KB Sizes 0 Downloads 30 Views

814

THE JOURNAL OF UROLOGY®

2247 TMPRSS2-ERG GENE FUSIONS IN MINIMAL PROSTATIC CARCINOMA Roula Albadine, Mathew Latour, Elizabeth Platz, Alan Meeker, Angelo Demazo, George J Netto*, Baltimore, MD INTRODUCTION AND OBJECTIVE: Minimal or insignificant prostatic adenocarcinoma (MinPCa) is defined as tumors with insufficient virulence to threaten survival. Given recent suggestion of TMPRSS2ERG gene fusion association with aggressive PCa phenotype, we aimed to evaluate incidence of TMPRSS2-ERG fusion in MinPCa in comparison with grade matched non-minimal size PCa. METHODS: All 33 prostatectomies classified as containing MinPCa (2002-2003) were retrieved. Diagnosis of MinPCa (Gleason Score 6 PCa with total tumor volume < 0.5 CC, single section) was confirmed by a urologic pathologist. Tissue microarray (TMA) was constructed from the 33 cases where each tumor and paired benign tissue was represented by up to triplicate,1mm, spots. TMA sections of 59 additional archival PCa were used as controls (26 pT2 non-minimal in size, 31 pT3a and 2 pT3b). FISH analysis was performed using breakapart probes for 5 and 3 regions of ERG. Each spot was scored for presence of TMPRSS2-ERG fusion through deletion or translocation as well as for polyploidy ( 3 copies) at the ERG locus. At least 50 cells were scored per tumor. RESULTS: MinPCa: TMPRSS2-ERG fusion was identified in 46% (16/35) of MinPCa. In 87% (14/16) of positive tumors, fusion was due to deletion. The remaining 13% (2/16) of fusions were based on the demonstration of a split in the two juxtaposed probe signals. Ch21 polyploidy fusion and duplication of ERG deletion were not observed in any MinPCa case. Control group: TMPRSS2-ERG fusion was identified in 59% (35/59) of tumors. In 77% (27/35) of positive tumors, fusion was due to deletion. Ch21 polyploidy with fusion was present in 13/59 (22%) while polyploidy with duplicate ERG deletion was found in 6/59 (10%) of control tumors. On statistical analysis, there was no significant difference in TMPRSS2-ERG fusion incidence between the MinPCa and control groups (p=0.2). Statistically significant higher rates of ch 21 polyploidy fusion was present in control group (p=0.0002). A trend approaching statistical significance for higher incidence of ch21 polyploidy with duplicate deletion was also present in the control group (p=0.052). CONCLUSIONS: TMPRSS2-ERG fusion rate of 46% is present in MinPCa. The latter is not significantly different from rate of fusion in control group of non-minimal pT2 and pT3 PCa. A higher rate of Ch21 polyploidy is detected in the control group compared to MinPCa. Our finding of a comparable rate of TMPRSS2-ERG fusion in MinPCa and non-minmal PCa argues against its value as a marker of aggressive PCa phenotype. Source of Funding: Patrick C Walsh Foundation

2248 FEASIBILITY AND CLINICAL UTILITY OF A TMPRSS2:ERG GENE FUSION URINE TEST Jack Groskopf*, San Diego, CA; Javed Siddiqui, Laurie SeftonMiller, Ann Arbor, MI; Sheila M j Aubin, Amy Blase, San Diego, CA; Sooryanarayana Varambally, Ann Arbor, MI; Kyoko Sakamoto, San Diego, CA; Yves Fradet, Quebec City, QC; Jack Schalken, Nijmegen, Netherlands; Harry Rittenhouse, San Diego, CA; Arul Chinnaiyan, Ann Arbor, MI INTRODUCTION AND OBJECTIVE: Fusions between TMPRSS2 (T2) and the oncogenic transcription factor ERG are found in ~50% of prostate tumors. Adverse clinical outcome has been associated with the presence of T2:ERG gene fusions and/or specific mRNA isoforms. To determine the feasibility and clinical utility of a T2:ERG urine test, specimens are being collected prior to prostate biopsy or prostatectomy; urine T2:ERG mRNA results are then correlated with biopsy outcome, as well as gene fusion status and pathologic features from biopsy and prostatectomy tissue. Here we present interim results from this multicenter collaborative study. METHODS: Urine specimens were prospectively collected

Vol. 181, No. 4, Supplement, Wednesday, April 29, 2009

following digital rectal exam from men scheduled for prostate biopsy. Collection sites were San Diego VA Medical Center and Universite Laval (SD/UL, n = 258 subjects), or University of Michigan (UM, n = 81). For each subject, whole urine was processed by mixing with an equal volume of detergent solution, and urine sediments prepared by centrifugation. Urine sediment and biopsy tissue samples were tested with a multiplex qualitative assay that detects TMPRSS2exon1-ERGexon4 (1-4), 1-2 and 2-4 mRNAs. Whole urine specimens were tested with quantitative assays for each mRNA isoform. Urine T2:ERG results were correlated with biopsy outcome. Biopsy tissue T2:ERG mRNA results were correlated with gene fusion status determined using a FISH break-apart assay. RESULTS: Prostate cancer was detected in 113/258 and 37/81 men in the SD/UL and UM studies. Qualitative T2:ERG mRNA testing of SD/UL urine sediments yielded 86% specificity and 43% sensitivity relative to biopsy outcome; these results were reproduced in the UM cohort (specificity 91%, sensitivity 50%). Quantitative testing of whole urine specimens (n=127) yielded performance equivalent to qualitative results on matched sediments; at 93% specificity, sensitivity was 30 and 32%, respectively. In biopsy tissue (n = 19), T2:ERG mRNA was detected in 11/12 (92%) of specimens that were gene fusion positive by FISH. CONCLUSIONS: The T2:ERG mRNA urine test yielded high specificity relative to biopsy outcome; sensitivity was consistent with the ~50% prevalence of gene fusions in prostate tumors. In biopsy tissue, T2:ERG mRNA results showed high concordance with FISH. Based on previous publications and the hormonal regulation of T2:ERG expression, the urine assay may have additional utility for prognosis and treatment selection. Source of Funding: Gen-Probe Incorporated, NIH/NCI Early Detection Research Network

2249 CHROMOSOME 21 COPY NUMBER BUT NOT TMPRSS2ERG FUSION PREDICTS OUTCOME IN PROSTATIC ADENOCARCINOMA: A LARGE CASE-CONTROL RADICAL PROSTATECTOMY COHORT ANALYSIS. George J Netto*, Antoun Toubaji, Roula Albadine, Baltimore, MD INTRODUCTION AND OBJECTIVE: TMPRSS2-ERG gene fusion has been reported to be present in up to 70% of all PCa with hereto-conflicting findings in relation to its clinical significance. In the current study, we aimed to evaluate the role of TMPRSS2-ERG fusion in predicting disease progression using FISH analysis applied to a wellcharacterized set of PCa case-control cohort. METHODS: 9 TMAs containing paired tumor and normal tissues of 158 cases and 158 control radical prostatectomies matched for grade, stage, ethnicity and p stage. All RRPs were performed in our institution between1993-2000. Progression was defined as development of biochemical recurrence, clinical evidence of metastasis or death of PCa. FISH analysis was performed using break-apart probes for the 5 and 3 regions of ERG. Each spot was scored for presence of TMPRSS2-ERG gene fusion through deletion or split as well as for polyploidy ( 3 copies) at the ERG locus. At least 50 cells were scored per spot. RESULTS: Findings are summarized in table 1. Fusion Deletion Cases 51.27% (progressed)

Split

Duplicated Polyploidy Polyploidy ch21 Polyploidy Deletion + Deletion +Split without fusion

36.1%

26%

9.5%

10.1%

11.4%

29.8%

Controls

57.6%

46.2%

30.4%

8.9%

8.9%

9.5%

17.1%

p value

NS

NS

NS

NS

NS

NS

0.01

Presence of fusion due to either a deletion or split event was not associated with progression. Likewise, neither the presence of duplicated ERG deletion, duplicated split, presence of ch21 polyploidy with single allele ERG deletion nor the presence of ch21 polyploidy with single allele ERG split were associated with progression. ERG gene polyploidy without fusion was significantly associated with progression in our cohort