Background. Retrospective reviews and small prospective studies have shown that hormone receptor (HR) and HER2 status may change between primary breast cancer and recurrences in 15 to 40% and 0 to 34% respectively. This may lead to changing treatment in 15 to 20% of patients (pts). The primary aim of this study is to evaluate the discordance in HR and HER2 status between primary tumors, loco-regional recurrences (LRR) and distance metastasis (mets, DM). Secondary objective is to identify predictors of discordance.

Patients and Methods. Retrospective analysis of patients with invasive breast cancer with biopsy of suspicious metastatic lesion. HR status (estrogen and/or progesterone) was defined as positive (HR+ve) when immunohistochemistry (IHC) staining was ≥10% of cells. HER2 positive (HER2+ve) was defined by IHC 3+ or amplified by fluorescence in situ hibridization (FISH).

Results. Two hundred and fifty one pts were evaluated with two hundred biopsies available for analysis. Median follow-up was 77 months (range, 2.5 — 415.1). 59% of primary tumors were HR+ve/HER2-ve, 27% triple negative (TN), 9% triple positive (TP) and 5% HR-ve/HER2+ve. 75.5% of patients received adjuvant chemotherapy, 5.0% primary chemotherapy, 78% adjuvant radiation and 44% hormonal therapy. 12 pts developed secondary breast cancer before recurrence. 50% (131) of recurrences were LRR and 50% were DM (17% visceral, 27% non-visceral and 2% central nervous system mets). LRR or DM surgery was performed in 33% of patients. We observed a 21% of discordance for HR (18% negative to positive and 28% positive to negative) and 6% of discordance for HER2 (5% negative to positive and 6% positive to negative). The overall discordance in HR/HER2 subgroups was 22%. No statistical differences (p .27) were observed in HR/HER2 subgroups’ discordance between LRR (median time to LRR, 43.5 months) and DM (median time to DM, 58 months). Greatest discordance was seen in TN tumors (31%), follow by HR-ve/HER2+ve, TP and HR+ve/HER2-ve with 28.5%, 25% and 17% respectively. Neither of TP tumors nor of 3 patients treated with trastuzumab became negative for HER2. At multivariate analysis, recurrence free interval longer than 24 months was the only statistical predictor of discordance (p .02).

Conclusions. Considering the high percentage of receptors status discordance and current options of systemic treatments, including target therapies, tissue confirmation and HR/HER2 analysis should be recommended for patients with suspicious of loco-regional or distant mets.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr PD10-06.