Automated Bright-Field Dual Color, Dual Hapten HER2 In Situ Hybridization (DISH) Assay: An Alternative Method for Evaluation of HER2 Gene Amplification in Breast Cancer
LJ Tafe, J Teruya-Feldstein, A McElhinny, J Ranger-Moore, J Emch, C Schemp, V Barbashina. Memorial Sloan-Kettering Cancer Center, New York City, NY; Ventana Medical Systems, Inc., Tucson, AZ
Background: HER2 positivity characterizes approximately 20% of invasive breast cancers and is evidenced by protein overexpression and/or gene amplification. Trastuzumab, an anti-HER2 monoclonal antibody, inhibits HER2-related tumor proliferation thus increasing patient survival. FISH is currently the standard method for assessment of HER2 gene amplification. Recently developed chromogenic detection systems such as silver ISH offer complete automation, fast turnaround times and bright-field microscopy, presenting an attractive alternative to FISH. We evaluated the performance of a new fully automated dual color HER2 in situ hybridization assay (DISH, Ventana) in comparison with FISH.
Design: FFPE blocks from 68 invasive breast cancers were tested by FISH (PathVysion, Vysis/Abbott) and DISH (Ventana). DISH was performed on Ventana BenchMark® series instrument. The assay allows visualization of the HER2 and chromosome 17 (chr 17) probes on the same slide. The HER2 probe is detected via silver deposition and the chr 17 probe is detected with a fast red and naphthol phosphate reaction. The cases were independently scored by 2 pathologists. The signals were enumerated in at least 20 tumor nuclei at 40X, and scored using the 3-tiered system recommended by ASCO/CAP (HER2/chr 17 > 2.2: amplified; 1.8 - 2.2: equivocal; < 1.8: non-amplified), and the 2-tiered system detailed in the DISH assay manual (HER2/chr 17 ≥ 2: amplified; < 2: non-amplified).
Results: The 68 cases consisted of 61 primary tumors and 7 metastases, 85% were resection specimens. Sixty-four (94%) were successfully analyzed. The overall concordance between DISH and FISH was 98% with the 3-tiered scoring system and 97% with the 2-tiered system. Repeat rate for DISH was similar to FISH (15% vs. 13%). Lack of interpretable signals on repeat analysis was recorded as failure and was slightly higher for DISH compared to FISH (6% vs. 2%). Some difficulty in DISH scoring was noted in cases with polysomy 17 and low-level amplification due to overlapping HER2 and chr 17 signals. Overall, DISH interpretation was straightforward and reproducible (96% interobserver concordance).
Conclusions: We found good concordance between the DISH and FISH in unequivocally non-amplified and amplified tumors. The DISH assay can be considered a reliable technique for the assessment of HER2 status.
Tuesday, March 23, 2010 1:00 PM
Poster Session IV # 7, Tuesday Afternoon