Assessment of HER2 Status of Metastatic Breast Carcinoma on Cell Block Preparations of Fine Needle Aspirates is Unreliable

Objectives: HER2 status of breast carcinoma is a powerful prognostic and predictive biomarker-particularly in the metastatic setting. Limited data is available regarding assessment of HER2 on cell block preparations (CBP). The primary objective of this study was to assess correlation between HER2 results obtained via immunohistochemistry (IHC) and fluorescence in-situ hybridization (FISH) in cases of metastatic breast carcinoma (MBC) on CBP. Secondary objectives included study of inter-observer variability in interpretation of HER2 on IHC, and concordance between HER2 results on CBP and formalin-fixed paraffin embedded material (FFPEM). Materials and Methods: Cases of MBC diagnosed on fine needle aspirates (FNA) with HER2 testing performed via IHC and FISH on CBP over 5 years (2010-2015) were reviewed. CBP material was fixed in an ethanol-based fixative (CytoRich Red Fixative system, BD). HER2 IHC was performed using polyclonal antibodies against Cerb-2 (Dako 0485). HER2 FISH testing was performed using the LSI HER2/neu/CEP17 probes (Vysis/Abbott Molecular Inc., Des Plaines, IL). Results: Seventeen cases (all female, median age: 59) were analyzed. 41% of CBP were products of bone FNA (7/17). Other sites included lymph node (3), lung (2), pleural fluid (2), liver (1), skeletal muscle (1) and mesentery (1). Median interval between diagnosis of primary carcinoma and FNA of metastasis was 5 years (range: 10 months-32 years). FISH was inconclusive due to suboptimal specimen quality in 2 cases. Correlation between IHC and FISH results was as follows: IHC 0/1+ (0/2; 0% amplification), IHC 2+ (2/12; 16.7% amplification) and IHC 3+ (0/1; 0% amplification). Inter-observer agreement of IHC scoring between 2 pathologists who independently reviewed IHC slides was fair (66.7% agreement, κ =  0.31). Comparison of HER2 results on CBP with FFPEM (primary carcinoma or metastasis) showed a high discordance and slight agreement (discordance rate = 37.5%; κ =  0.02). Conclusions: In this study, (a) 16.7% of MBC cases that scored 2+ on IHC showed amplification on FISH, (b) there was poor inter-observer agreement in HER2 scoring of IHC on CBP, and (c) there was high discordance between HER2 results obtained on CBP and FFPEM. Our results indicate that HER2 testing of MBC on CBP may be unreliable.