Online-onlyTable A
Studies of HER2 in invasive breast cancer and paired metastases:HER2testing methods and interpretation criteria
1
Study(year published) / Sample / Immunohistochemistry (IHC) / In Situ Hybridization (FISH or CISH)Type of specimen, processing and fixation (where reported) / Assay & method / Scoring† & interpretation criteria / Were equivocal results tested with FISH? / Assay & method / Scoring & criteria forgene-amplified
Idirisinghe (2010) / Tissue sections of formalin-fixed and paraffin-embedded primary and paired metastasis or breast recurrence / SP3 (Neomarkers), rabbit monoclonal Ab, diluted 1:200; Ag retrieval / Non-standard (0-3+) scoring; positive 3+ (≥30% membrane staining), equivocal 2+ (≥10% staining) / NA (there were no equivocal results in dataset) / NA / NA
Aoyama (2010) / Tissue from primary and node metastases (not further described); all involved nodes examined / NA / NA (cohort selected to represent 4 groups of 0, 1, 2, 3 scores on IHC: all tested using FISH in the study) / NA / PathVysion HER2 DNA kit*(Vysis, USA) / HER2/CEP17 ratio ≥2; at least 60 tumour cells scored
Aitken (2010) / TMA sections of formalin-fixed and paraffin-embedded primary and paired node metastases / Dako K5207 (Herceptest diluted 1:50 after antigen retrieval), monoclonal Ab / Automated quantitative image analysis scoring of immuno-fluorescent-stained sections (negative set as the mean +2 SD scores for HercepTest ≤ 2+) / All had FISH (99% concordance with IHC) but not used to verify equivocal results / HER2 FISH pharmDx kit (Dako), dual DNA probe / According to ASCO/CAP1 guide: HER2/CEP17 ratio > 2.2;20 nuclei per core scored
Lower (2009) / Surgical or core biopsy tissue, formalin-fixed and paraffin-embedded specimens / Ventana immunostainer; CB11 monoclonal Ab / Standard (0-3+) scoring; positive 2+ or 3+ / No, only some cases of 2+ / NR / NR
Azam(2009) / Tissue sections of formalin-fixedprimary tumour from mastectomy specimens and node metastases dissected from axillary tissue / ND / Non-standard scoring: positive >10% cells had moderate or strong membranous staining; negative if weak or incomplete membrane staining, or cytoplasmic staining / Not done / NA / NA
Simmons (2009) / Core biopsy samples from metastases, fixed in formalin and paraffin-embedded; tissue from primary re-tested / ND / ND / NA / FISH, not further described / NR
Santiago (2009) / Tissue sectionsfrom primary and metastasis that had been fixed informalin (up to 24 hours) and embedded inparaffin / HercepTest (Dako) & Autostainer Plus (Dako); polyclonal Ab with epitope retrieval / Standard (0-3+) scoring, evaluated according to ASCO/CAP guide; positive 3+ or 2+ if FISH-amplified / Yes / HER2 FISH pharmDx kit (Dako), dual DNA probe / According toASCO/CAP1 guide: HER2/CEP17 ratio > 2.2
Santinelli (2008) / Tissue samples fixed (maximum of 48 hours) in 10% buffered formalin and embedded in paraffin / HercepTest (Dako, Canada); polyclonal Ab / Standard (0-3+) scoring; positive defined as suitability for trastuzumab therapy: 3+; 1+ or 2+ if gene-amplified / Yes, all had FISH / PathVysion HER2 DNA kit* (Vysis, USA) / HER2/CEP17 ratio > 2.2 (1.8-2.2 equivocal and re-tested); 100 invasive tumour cells scored
Guarneri (2008) / Tissue fixed in buffered formalin before processing and embedding / Novocastra clone CB11 (Novocastra, UK); avidin–biotin method, monoclonal Ab with epitope retrieval / Standard (0-3+) scoring; positive 3+ or 2+ if FISH-amplified / Yes / PathVysion HER2 DNA kit* (Vysis, USA) / HER2/CEP17 ratio ≥ 2; at least 60 non-overlapping nuclei scored
Cho (2008) / Representative tissue cores (2 for each tumour) from formalin-fixed paraffin-embedded tissue, arrayed in a recipient paraffin block / Novocastra CB11 (Novocastra, UK); avidin-biotin method, monoclonal Ab with Ag retrieval / Standard (0-3+) scoring; positive 2+ or 3+ / Yes, all had CISH / CISH Detection Kit (Zymed); digoxigenin- labeled DNA probe to HER2, diamino-benzidine (DAB) as chromogen / Amplified >10 copies per nucleus or large gene copy clusters in >50% of nuclei; low-level amplification 6-10 copies per nucleus in >50% of cells
Tapia
(2007) / TMA sections from primary cancers, cytological specimens from metastases (smears or cytospins). Tissue sections of primary tumour and cytology re-tested (and rescored) for cases with discordant initial results / NA / NA / NA / PathVysion HER2 DNA kit* (Abbott/Vysis, USA) / HER2/CEP17 ratio ≥2; in most of the cytological specimens ratio was calculated based on 60 scored cells
Gong (2005) / Paraffin-embedded tissue orfine needle aspirate(Diff-Quik-stained smears) / ND / NA / NA / PathVysion HER2 DNA kit* (Vysis) / HER2/CEP17 ratio ≥ 2; at least 60 cells scored
Zidan
(2005) / Formalin-fixed paraffin-embedded tissue from primary and metastasis / HercepTest (DAKO); CB11 monoclonal Ab with Ag retrieval / Standard (0-3+) scoring: positive 3+, or 2+ if FISH-amplified / Yes / NR / NR
Carlsson (2004) / Formalin-fixed paraffin-embedded tissue from primary and metastasis / Dako anti-HER2 Ab (Dako, Denmark); rabbit monoclonal Ab diluted 1:500 / Standard (0-3+) scoring (all samples had >50% cells stained); positive 2+ or 3+ if FISH-amplified / Yes all 2+ and 3+ were verified using ISH / PathVysion HER2 DNA Kit* (Vysis, USA)or CISH: digoxigenin- labeled DNA probe to HER2 / FISH amplified where HER2 signals per cell significantly higher than those for CEP17 (reported as ‘according to manufacturer’)
Regitnig (2004) / Representative tissue sections and TMA from formalin-fixed paraffin-embedded surgical specimens (TMA and sections had consistent results) / HercepTest (Dako, Denmark); polyclonal Ab / Standard (0-3+) scoring; exact scores were reported: 3+ was applied as positive in our analysis and correlated with FISH data / All evaluable samples tested with FISH / PathVysion HER2 DNA kit* (Vysis, USA) / HER2/CEP17 ratio ≥ 2 and cells had at least a median of 4 HER2 gene signals; ≥ 60 tumour cells scored
Edgerton (2003) / Formalin-fixed paraffin-embedded tissue from primary and metastasis / Compares several assays±: CB11 (Neo Markers, USA), Dako PAb (DAKO, USA) & HercepTest (DAKO) / Non-standard scoring: percentage of cells showing membrane staining for CB11; positive minimum 10% staining (had FISH) / Yes, IHC ≥10% expression tested with FISH / Inform HER2 Kit (Ventana); single DNA probe to the HER2gene / Amplified average ≥ 4.5
gene copies per cell (3.5-4.5 considered equivocal and further tested)
Sekido (2003) / Tumour sections from formalin-fixed paraffin-embedded tissue (fixed within 48 hours) / HercepTest (Dako, Denmark); polyclonal Ab / Standard (0-3+) scoring; positive 2+ or 3+ / No, only discordant results / PathVysion HER2 DNA kit* (Vysis, USA) / HER2/CEP17 ratio > 2; 60tumour cells scored
Gancberg (2002) / Paraffin-embedded tissue from primary and metastasis fixed in formalin or bouin / HercepTest (Dako, Denmark); polyclonal Ab / Standard (0-3+) scoring; positive 3+ / All evaluable samples had FISH / Vysis multi-colour probe (Vysis); dual DNA probe applied for HER2 testing / HER2/CEP17 ratio > 2; 60 nuclei scored
Tsutsui (2002) / Formalin-fixed paraffin-embedded tissue from primary and metastasis / cerbB2 monoclonal Ab (Ab-2, clone 9C6, diluted 1:100) no Ag retrieval (Oncogene Science, USA) / Non-standard scoring (0-2+): positive 1+ or 2+ if definite membrane staining. / No / NA / NA
Vincent-Salomon (2002) / Tissue sections from primary tumour and biopsy specimens from metastasis / Novocastra clone CB11 (Novocastra, UK); monoclonal Ab with Ag retrieval / Non-standard semi-quantitative scoring based on % of labelled tumor cells and membrane staining intensity; positive when ≥60% of cells were stained – HER2 over-expressed (strong or moderate staining) or not over-expressed / No / ND¶ / NA
Xu (2002) / Tissue sections fixed in formalin (4-12 hours) and paraffin-embedded / ND / NA / PathVysion HER2 DNA kit* (Vysis, USA) / HER2/CEP17 ratio > 1.5; at least 50 cells scored
Cardoso (2001) / Freshly cut sections from paraffin-embedded blocks of tissue from primary and metastasis fixed in formalin or bouin / Ventana Nexes autostainer (Ventana, USA); CB11 (Novocastra UK) monoclonal Ab / Non-standard semi-quantitative scoring system for focality staining, and intensity staining; HER2 positive ≥1% of cells positive / No / ND / NA
Simon (2001) / Several representative sets of TMA for each subject§ from tissue fixed in formalin and embedded in paraffin / HercepTest (Dako, Denmark); polyclonal Ab / Standard (0-3+) scoring; positive 2+ or 3+ [based on concordance between IHC & FISH, reported overall HER2 status where positive on IHC and/or gene-amplified]§ / All had FISH: 94% concordance for IHC & FISH / PathVysion HER2 DNA kit* (Vysis, USA) / HER2/CEP17 ratio of at least 3 or tight clusters of more than five HER2 signals in > 10% of tumour cells
Tanner (2001) / Formalin-fixed paraffin-embedded tissue from primary and metastasis / Novocastra clone CB11 (Novocastra, UK); avidin–biotin method, monoclonal Ab / Standard (0-3+) scoring; intense cell membrane staining in > 50% of cancer cells (3+) counted positive (only 1 case was 2+ and not amplified on CISH) / All had CISH†† / CISH Kit (Zymed, USA); digoxigenin- labeled genomic probe for HER2 / Amplified ≥ 6 copies per nucleus or large gene copy clusters in>50% of cells
Shimizu (2000) / Samples from surgical biopsy of primary and metastasis (most had paraffin-embedded blocks) / Commercial assays (Nichirei Co, Japan); polyclonal Ab in most cases (4 cases tested with enzyme immuno-assay) / Non-standard scoring: surface of membrane distinctly stained (strongly positive,over-expression) or surface of cell membrane and cytoplasm faintly stained (weakly positive, no over-expression) / No / ND / NA
Masood & Bui (2000) / Freshly cut sections from formalin-fixed paraffin-embedded tissue blocks from primary and metastasis / HercepTest Kit (DAKO); polyclonal Ab / Standard (0-3+) scoring; positive 2+ or 3+ / No / ND / NA
1
Key: HER2 = human epidermal growth factor receptor 2; IHC = immunohistochemistry; ISH = in situ hybridization;FISH = fluorescence in situ hybridization;
CISH = chromogenic in situ hybridization; NR = not reported; NA = not applicable; ND = not done; Ab= antibody; Ag = antigen; TMA = tumour micro-arrays;
CEP 17 = chromosome 17 centromere.
† Standard scoring for IHC was defined in terms of 0-3+ semi-quantitative scoring based on HercepTest criteria and published recommendations1;2 and requiring more than 10% of the tumour cells to be stained; non-standard scoring refers to other semi-quantitative methods not meeting this definition (further details on scoring methods in the primary studies included in this review are available from the authors).
* FISH method using the PathVysion HER2 DNA Kit: Hybridization of dual fluorescent DNA probes to HER2 gene (labelled with Spectrum Orange) and chromosome 17 centromere (CEP17) (labelled with Spectrum Green)
± CB11 data used for IHC in our analysis since it provided the most complete data-set for cohort
¶ Study referenced 98% concordance between protein expression (IHC) and gene status (FISH), however FISH data not reported
§Data included in our review from Simon et al were based on 125 subjects (lymph node-positive for metastatic cancer) in whom 16 arrayed tissue samples (4 samples each of the primary and of three different node metastases) yielded HER2 results. Because each of primary tumour and 3 corresponding node metastases had 4 samples tested, a small number of cases were reported as having heterogeneous HER2 status (a mix of HER2-positive and HER2-negative tissue samples for each tumor) – we have counted heterogeneous expression as positive in pooled analysis.
††Study by Tanner et al performed both IHC and CISH with complete concordance: paired HER2 data for primary and metastasis (in pooled analysis) are for CISH
References
(1) Wolff AC, Hammond ME, Schwartz JN et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. Journal of Clinical Oncology 2007;25:118-145.
(2) Bilous M, Dowsett M, Hanna W et al. Current perspectives on HER2 testing: a review of national testing guidelines. Modern Pathology 2003;16:173-182.
1