HER2 Status in Breast Cancer as Determined by Three ASCO/CAP Recognized Reporting Techniques Reveals Testing Discordance, and Emphasizes the Utility of Co-Testing with FISH and IHC
Michael R Clay, Edward A Gilbert, Charles D Bangs, Athena Cherry, Kristin C Jensen. Stanford Hospital and Clinics, Stanford, CA; Veterans Hospital, Palo Alto, CA
Background: HER2 positivity is found in 12-15% of newly diagnosed breast carcinomas. Standard practice includes immunohistochemistry (IHC) or fluorescent in situ hybridization (FISH) for primary HER2 status determination, and stipulates that regardless of whether FISH is reported as a ratio, or a raw count, all three of these reporting techniques are acceptable. We wanted to determine how often these three reporting techniques yielded discrepant results in a large number of cases, and determine what if any information was gained by utilizing all three. We also evaluated HER2 heterogeneity by IHC.
Design: We selected the most recent 300 cases of breast carcinoma in our institution to have had primary HER2 determination by FISH. All of these cases were also tested by IHC, and FISH results were re-evaluated to include both the ratio and raw scores. Heterogeneity on IHC staining was determined. The characteristics of discrepant cases were evaluated based on a number of clinicopathologic features.
Results: We evaluated 91 core needle biopsies, 103 lumpectomies, 98 mastectomies, and 8 metastasis resections. Overall, discordant findings were observed in at least one testing modality (IHC vs. FISHratio vs Fishraw) in 37 cases (12.33%). Discordance was associated with the presence of heterogeneity, some cancer subtypes, and ER negativity. Discordance was not more likely with biopsies. It takes 9 patients to be screened by all 3 ASCO/CAP recommended modalities in order to find one patient with discrepant testing (6.2-11.6 CI). Two cases were what we have termed completely discordant, in that they had different results for each of the 3 modalities (figure 1).
Conclusions: We have shown a significant difference between the three testing modalities, with a relatively high percentage of discordant results but with few "complete discordances”. With a number needed to screen of 9, we feel the potential information gained by reporting all 3 results is significant enough to warrant this practice. Heterogeneity by IHC is associated with discrepant results, and may be utilized as an indicator for triple testing.
Monday, March 4, 2013 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 17, Monday Morning