Background: Assessment of HER2 by immunohistochemistry (IHC) or fluorescence in situ hybridization (FISH) is a standard practice for breast carcinomas. Testing is associated with a 20% disagreement between laboratories. The College of American Pathologists (CAP) guidelines for HER2 testing include validation of HER2 test methods by achieving 95% concordance with another validated method. Our laboratory requires IHC 3+ FISH nonamplified specimens to undergo retesting by polymerase chain reaction (PCR). A random sample of IHC 2+ cases are routinely tested by PCR. We found this practice useful for resolving discrepancies in HER2 testing.
Methods: At clinician request, seventy-nine 3+ and one hundred forty-eight 2+ cases were tested by FISH. In 22 cases, IHC was 3+ but FISH was nonamplified. These 22 cases underwent HER2 LightCycler monoplex polymerase chain reaction (MPCR) testing. Seventeen 2+ nonamplified cases were tested by MPCR.
Results: Twenty-one 3+, FISH nonamplified cases were found to be MPCR nonamplified. One IHC 3+, FISH nonamplified case was MPCR amplified. Seventeen 2+, FISH nonamplified cases were MPCR nonamplified. In all but one case, FISH and MPCR were concordant.
Discussion: American Society of Clinical Oncology/CAP guidelines propose validation of testing procedures by showing 95% concordance with a validated test for positive and negative assays. Specific actions are not recommended to resolve discordances between tests. Our laboratory uses 3 different modalities for HER2 testing. We have found that our 2 methods for testing gene amplification status show a higher degree of concordance between themselves than either did with IHC. Review of the 3+ IHC nonamplified cases showed them to have a dark, granular circumferential staining pattern.