Immunohistochemistry and fluorescence in situ hybridization assessment of HER2 in clinical trials of adjuvant therapy for breast cancer (NCCTG N9831, BCIRG 006, and BCIRG 005)

A comprehensive, blinded, pathology evaluation of HER2 testing in HER2-positive/negative breast cancers was performed among three central laboratories. Immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) analyses were performed on 389 tumor blocks from three large adjuvant trials: N9831, BCIRG-006, and BCIRG-005. In 123 cases, multiple blocks were examined. HER2 status was defined according to FDA-approved guidelines and was independently re-assessed at each site. Discordant cases were adjudicated at an on-site, face-to-face meeting. Results across three independent pathologists were concordant by IHC in 351/381 (92 %) and FISH in 343/373 (92 %) blocks. Upon adjudication, consensus was reached on 16/30 and 18/30 of discordant IHC and FISH cases, respectively, resulting in overall concordance rates of 96 and 97 %. Among 155 HER2-negative blocks, HER2 status was confirmed in 153 (99 %). In the subset of 102 HER2-positive patients from N9831/BCIRG-006, primary blocks from discordant cases were selected, especially those with discordant test between local and central laboratories. HER2 status was confirmed in 73 (72 %) of these cases. Among 118 and 113 cases with IHC and FISH results and >1 block evaluable, block-to-block variability/heterogeneity in HER2 results was seen in 10 and 5 %, respectively. IHC-/FISH- was confirmed for 57/59 (97 %) primary blocks from N9831 (locally positive, but centrally negative); however, 5/22 (23 %) secondary blocks showed HER2 positivity. Among 53 N9831 patients with HER2-normal disease adjudicated as IHC-/FISH-(although locally positive), there was a non-statistically significant improvement in disease-free survival with concurrent trastuzumab compared to chemotherapy alone (adjusted hazard ratio 0.34; 95 % CI, 0.11-1.05; p = 0.06). There were similar agreements for IHC and FISH among pathologists (92 % each). Agreement was improved at adjudication (96 %). HER2 tumor heterogeneity appears to partially explain discordant results in cases initially tested as positive and subsequently called negative.