Abstract
Newly diagnosed invasive breast cancers should be evaluated for Human Epidermal Growth Factor Receptor 2 (HER2) status by immunohistochemistry (IHC) and/or in situ hybridisation (ISH) to determine eligibility for trastuzumab or other HER2-targeted therapies. Previous reports of high discordance rates between IHC and ISH have raised concerns over the accuracy of HER2 testing, especially when IHC is conducted locally. This study aimed to determine the rate of false-negative IHC results at three pathology centres (one central, two local) in Greece by central retesting of 240 prospectively collected invasive breast cancers scored as IHC 0/1+ at initial testing. All samples were from female patients (median age 58.0 years). Initial IHC tests utilised either the CB11 (159/239; 66.5%) or 4B5 (80/239; 33.5%) antibodies and were scored as 0 in 105/240 cases (43.8%) and 1+ in 135/240 cases (56.3%). All samples were centrally retested by automated silver in situ hybridisation (SISH). Of 237 samples with SISH staining suitable for assessment, 223 (94.1%; 95% confidence interval 90.3–96.5%) were classed as SISH-negative (HER2:chromosome enumeration probe 17 (CEP17) <1.8). Eight tested SISH-positive (HER2:CEP17 >2.2), providing a false-negative rate of 3.4%. A further four samples (1.7%) exhibited equivocal amplification status (HER2:CEP17 1.8–2.2) and two (0.8%) showed polysomy of chromosome 17. The proportion of SISH-negative results did not significantly differ between the IHC 0 and 1+ subgroups (95.2% vs. 93.2%; p=0.505). In conclusion, the low observed rate of false-negative IHC results in this study supports the use of IHC for initial HER2 status assessment in local or central laboratories in Greece.