Emerging data show that patients with operable, HER-2/neu overexpressed/amplified breast carcinomas have significantly better responses (more frequently obtaining pathologic complete response and greater percent disease-free survival) when treated with trastuzumab (Herceptin) simultaneously with neoadjuvant chemotherapy than with chemotherapy alone. With the increasing use of neoadjuvant therapies, clinicians require information on biomarkers including HER-2/neu status at the time of needle core biopsy. Concordance rates between fluorescence in situ hybridization (FISH)-determined HER-2/neu status on needle core biopsies and on subsequent excisional biopsies of the same tumor have not been well studied. Moreover, the practice of automatically performing ("reflexing") 2+ immunohistochemical (IHC) staining needle core biopsies for FISH analysis on the same sample needs to be validated. In this study, we set out to (1) determine the accuracy of HER-2/neu status as determined by FISH on needle core biopsy material compared with FISH on the subsequent excisional biopsy of the same tumor with special consideration of IHC 2+staining cases and (2) determine the constancy of HER-2/neu status in pre-neoadjuvant and post-neoadjuvant chemotherapy-treated tumor in the form of needle core biopsy and excisional biopsy samples, respectively.
100 patients whose needle core biopsies and subsequent excisional biopsy samples were pathologically evaluated at our institution were studied. For each patient, unstained sections from both specimens were prepared and used for IHC or FISH. IHC was carried out using the HercepTest kit (DAKO, Carpinteria, CA). Parallel unstained slides were used to carry out FISH (dual probe, Vysis). Statistical analyses were carried out on the resulting data generated after interpretation.
The concordance rate between FISH results determined on the needle core biopsy and subsequent excisional biopsy of the same tumor was 86% (kappa=0.56, P=2x10). If equivocal FISH cases (> or =1.8 to < or =2.2 amplification ratio) in a needle core biopsy or excisional biopsy specimen or both, were excluded, the concordance rate increased to 95% (kappa coefficient=0.86, P=2x10). Fourteen of 100 (14%) cases showed 2+ IHC staining in the needle core biopsy specimen with good concordance of FISH-determined HER-2/neu status between the needle core biopsy and excisional biopsy specimens (79% agreement and kappa=0.512, P=0.05). Nine, 3, and 2 cases of the 14 cases were amplified, equivocal, and negative on the excisional biopsy specimens, respectively. Of the 15 patients who received neoadjuvant chemotherapy, 93% and 87% had no change in HER-2/neu status as determined by IHC or FISH, respectively, in the excisional biopsy specimen when compared with that determined on the prior core biopsy sample.
Excellent overall concordance was achieved between FISH-determined HER-2/neu status on the needle core biopsy and that determined on the subsequent excisional biopsy of the same tumor. These results suggest that intratumoral heterogeneity of HER-2/neu assessed by FISH is not a significant confounding factor when analyzing smaller sized samples. Furthermore, 79% of 2+IHC staining needle core biopsy cases showed concordant FISH results in the needle core biopsy and subsequent excisional biopsy specimens. Our results show good concordance, however, larger cohorts need to be studied to verify this finding. HER-2/neu status remains unchanged in the majority of cases when comparing pre-neoadjuvant and post-neoadjuvant chemotherapy-treated specimens.