Central pathology laboratory review of HER2 and ER in early breast cancer: an ALTTO trial [BIG 2-06/NCCTG N063D (Alliance)] ring study.

Choice of therapy for breast cancer relies on human epidermal growth factor receptor-2 (HER2) and estrogen receptor α (ER) status. Before randomization in the phase III Adjuvant Lapatinib and/or Trastuzumab Treatment Optimisation (ALTTO) trial for HER2-positive disease, HER2 and ER were centrally reviewed by Mayo Clinic (Rochester, MN, and Scottsdale, AZ) for North America and by the European Institute of Oncology (IEO; Milan, Italy) for the rest of world (except China). Discordance rates (local vs. central review) differed between Mayo and IEO. Among locally HER2-positive cases, 5.8 % (Mayo) and 14.5 % (IEO) were centrally HER2 negative. Among locally ER-positive cases, 16.2 % (Mayo) and 4.2 % (IEO) were centrally ER-negative. Among locally ER-negative cases, 3.4 % (Mayo) and 21.4 % (IEO) were centrally ER-positive. We, therefore, performed a ring study to identify features contributing to these differing discordance rates. Mayo and IEO exchanged slides for 25 HER2 and 35 ER locally/centrally discordant cases. Both laboratories performed IHC and FISH for HER2 using the HercepTest(®) and PathVysion HER2 DNA probe kit/HER2/centromere 17 probe mixture. IHC for ER was tested centrally using the monoclonal ER 1D5 antibody (Mayo) or the DAKO cocktail of ER 1D5 and 2.123 antibodies (IEO). Mayo and IEO confirmed the central HER2-negative result in 100 % of 25 cases. Mayo and IEO confirmed the central ER result in 29 (85 %) of 34 evaluable cases. The five Mayo-negative/IEO-positive cases were ER-positive when retested at Mayo using the DAKO ER cocktail. In this ring study, ALTTO ineligibility did not change when HER2 testing was performed by either IEO or Mayo central laboratories. However, a dual antibody ER assay had fewer false-negative test results than an assay with a single antibody, and there was more discordance between the two ER reagents than has been previously reported. Using even slightly different assay methods yielded different results, even between experienced central laboratories.