Evaluation of Novel ERG/SPINK1 IHC and 4-Color Quantum-Dot Based ERG/PTEN FISH in Radical Prostatectomy Specimens
Kyung Park, Nallasivam Palanisamy, Theresa MacDonald, Javed Siddiqui, Arul M Chinnaiyan, Martin G Sanda, Huihui Ye, Mark A Rubin, Juan Miguel Mosquera. Weill Medical College of Cornell University, New York, NY; University of Michigan, Ann Arbor, MI; Beth Israel Deaconess Medical Center and Harvard Med School, Boston, MA
Background: TMPRSS2-ERG gene fusions and SPINK1 overexpression have been reported in approximately 50% and 10% of prostate cancer (PCa). We investigated tissue extent of both biomarkers in radical prostatectomy (RP) specimens using novel dual ERG/SPINK1 IHC and 4-color quantum-dot based ERG/PTEN FISH. The study was conducted with tissue from patients enrolled in the Early Detection Research Network (EDRN), a well-characterized clinical cohort suitable for further validation studies.
Design: 236 H&E slides corresponding to different areas of 101 dominant and 122 secondary tumor nodule(s) from 101 RP specimens were reviewed. ERG/SPINK1 IHC and ERG/PTEN FISH were performed on Discovery and Benchmark XT platforms (Ventana). Semi-quantitative evaluation of nuclear ERG and cytoplasmic SPINK1 expression was performed: >20% staining extent and 2-3 intensity (scale 0 to 3) was considered positive. ERG (break-apart assay; red-5' and green-3') and PTEN (orange-PTEN; yellow-Centromere 10) signals were automatically deconvolved using an interferometer, and each gene status was determined using spectral imaging software.
Results: 24/101 cases had one dominant tumor nodule only, 31/101 had one dominant and one secondary nodule, and 46/101 had one dominant and two secondary nodules. By IHC, 59/101 (58%) RP cases demonstrated at least one ERG-positive tumor nodule, and 8/101 (8%) cases showed at least one SPINK1-positive tumor.
Nuclear ERG staining was always strong-diffuse, and cytoplasmic SPINK1 staining was strong-focal. ERG and SPINK1 expression was mutually exclusive in all cases. 34/101 (34%) RP specimens demonstrated neither ERG nor SPINK1 expression. By FISH, ERG rearrangement was observed in 98% of ERG positive tumors by IHC, approximately 30% of which demonstrated ERG translocation through deletion. Homozygous and hemizygous PTEN deletions were observed in both ERG-rearranged and non-rearranged tumors. However, PTEN deletions were not observed in SPINK1-positive cases.
Conclusions: To our knowledge, this is the first study designed to validate the feasibility of multiplex biomarker tissue-assays on all distinct tumor nodules in RP specimens from a clinical cohort. We demonstrate that both ERG/SPINK1 IHC and 4-color quantum-dot based FISH assays show sensitive and specific detection of dual protein expression and multiple gene rearrangements, respectively. Further studies to validate clinical correlation of these assays are ongoing.
Category: Genitourinary (including renal tumors)
Monday, March 19, 2012 1:00 PM
Poster Session II # 164, Monday Afternoon