• human epidermal growth factor receptor 2 (HER2);
  • ERBB2;
  • concordance;
  • fluorescence in situ hybridization;
  • HER2 gene amplification;
  • HER2 expression;
  • HER3/ERBB3;
  • esophageal cancer;
  • esophageal adenocarcinoma;
  • esophagogastric cancer


  1. Top of page
  2. Abstract


Human epidermal growth factor receptor 2 (HER2) is a therapeutic target in patients with esophageal adenocarcinoma (EAC), with gene amplification used as a selection criterion for treatment, although to the authors' knowledge the concordance between amplification and HER2 protein expression remains undefined in EAC. Furthermore, the association between HER2 and its interacting partner, human epidermal growth factor receptor 3 (HER3), is unknown yet appears to be of potential therapeutic relevance.


Patients with untreated EACs (N = 673) were analyzed for HER2 amplification and polysomy 17 by fluorescence in situ hybridization in parallel with immunohistochemistry (IHC) (IHC scores of 0-1+, 2+, and 3+). Amplification was defined as HER2/CEP17 ≥ 2. HER3 expression by IHC was analyzed in randomly selected cases (n = 224). IHC and fluorescence in situ hybridization results were compared using least squares linear regression.


Overall, 17% of the EACs (116 of 673 EACs) were HER2-amplified with an amplification frequency that was highest among IHC3+ cases (89%) and declined among IHC2+ cases (13%) and IHC0 to IHC1+ cases (4%). Among HER2-amplified cases, the level of amplification increased linearly with HER2 membranous expression (HER2/CEP17 ratio: 7.9 in IHC3+ and 5.5 in IHC2+ vs 2.8 in IHC0 to IHC1+ [P < .0001]), with 14% of amplified tumors demonstrating absent/faint expression (IHC0 to IHC1+). Polysomy 17 was not found to be associated with HER2 expression. Cytoplasmic HER3 expression was detected in 87% of tumors (195 of 224 tumors) and was found to be significantly associated with better differentiation (P < .0001). Stepwise increases in HER3 expression were associated with higher HER2 expression levels (P = .0019).


Levels of HER2 protein expression and amplification were found to be linearly associated and highly concordant. Among amplified tumors with absent/faint expression, the level of amplification was low. Frequent expression of HER3 suggests its relevance as a therapeutic target, and its significant association with HER2 supports ongoing efforts to inhibit HER2/HER3 in patients with EAC. Cancer 2014;120:415–424. © 2013 American Cancer Society.


  1. Top of page
  2. Abstract

The incidence of adenocarcinomas of the esophagus, gastroesophageal junction (GEJ), and gastric cardia (collectively referred to as EAC) has risen 6-fold in the past 30 years, and continues to increase.[1] The 5-year survival remains < 25%, including those patients with regional disease who undergo surgery.[2] The human epidermal growth factor 2 (HER2/ErbB2) oncoprotein is a member of a family of transmembrane receptor tyrosine kinases (HER1/epidermal growth factor receptor, HER3/ErbB3, and HER4/ErbB4) that regulate cell growth, survival, differentiation, and migration.[3] Homodimerization or heterodimerization among HER receptors is essential for their function and signaling activity. The HER2-HER3 heterodimer is considered the most potent HER pair with respect to strength of interaction, ligand-induced tyrosine phosphorylation, and downstream signaling.[4] HER3 overexpression has been associated with HER2 overexpression[5] and with adverse survival in multiple tumor types including gastric cancer,[6-8] but to the best of our knowledge has not been evaluated in patients with EAC. Given its important role as a signaling hub, HER3 is a candidate target for drug discovery that is being actively investigated.[9]

HER2 is an established target in human breast and GEJ/gastric cancers. A recent phase 3 trial (Trastuzumab for GAstric cancer [ToGA]) of patients with advanced HER2-overexpressing GEJ or gastric adenocarcinoma demonstrated that adding trastuzumab, a HER2-targeted monoclonal antibody, to cytotoxic chemotherapy improved overall survival (OS).[10] In the ToGA trial, the benefit of trastuzumab was limited to tumors that demonstrated strong or intermediate HER2 immunoreactivity as confirmed by gene amplification.[10-12] Based on this study, trastuzumab is now approved worldwide for the treatment of patients with advanced esophagogastric cancer, and determination of HER2 status has become an integral part of the clinical-pathological workup. HER2 overexpression is usually a direct consequence of gene amplification, although overexpression can occur by other mechanisms.[4, 13] The importance of understanding the association between HER2 expression and amplification is underscored by the finding that no trastuzumab benefit was observed in gene-amplified cases demonstrating only absent/faint expression, which comprised approximately 25% of the ToGA study population. To our knowledge to date, the concordance between HER2 expression and amplification has been inadequately studied in patients with EAC.

HER2 and other key genes (P53, BRCA1, and TOP2A) reside on chromosome 17, and the effect of excess chromosome 17 (polysomy 17) on HER2 expression and response to HER2-targeted therapy remains unknown.[14] Although there are conflicting data, an increase in copies of chromosome 17 in non–HER2-amplified breast cancers has been associated with increased HER2 protein expression,[14] and preliminary data have suggested that polysomy 17 may enhance trastuzumab sensitivity, even in the absence of HER2 gene amplification.[15] It remains unknown whether polysomy 17 leads to HER2 overexpression in patients with EAC.

We determined the association between HER2 protein expression and the frequency and level of gene amplification and with polysomy 17 in patients with EAC, applying interpretive criteria specific for esophagogastric cancers that are used in clinical practice. We examined the full consecutive series of patients with EAC, which enabled the determination of HER2 test parameters such as false negativity and specificity that are critical to inform the selection of patients for trastuzumab therapy. We also examined HER3 expression and its association with HER2 amplification and expression. Given the potential effects of chemo(radio)therapy on tumor viability and HER2 expression, we studied patients before the routine use of neoadjuvant therapy.


  1. Top of page
  2. Abstract

Study Cohort

The parent cohort (n = 787) was derived from the Mayo Esophageal Cancer Outcomes Database comprising consecutive patients with newly diagnosed, invasive adenocarcinoma of the esophagus, GEJ, or gastric cardia who underwent surgical resection with negative margins at the Mayo Clinic between 1980 and 1997, as described.[16] Patients with subcardial gastric cancers and those with tumors with only nonadenocarcinoma histology were excluded, as were 9 patients who received neoadjuvant therapy. A total of 703 cases had sufficient tumor for evaluation. Regions of invasive carcinoma were identified using a hematoxylin and eosin-stained slide, and sequential whole-tissue sections were cut from formalin-fixed, paraffin-embedded surgical resection blocks, as previously described.[17, 18] The value of evaluating whole-tissue surgical sections is underscored by recent data indicating a significant false-negative rate for detecting HER2 aberrations in tissue microarrays and/or biopsy specimens.[19] The Institutional Review Board at the Mayo Clinic approved this research and waived specific informed consent.

HER2 Gene and Chromosome 17 Copy Number

HER2 amplification in tumor cells was assessed using a US Food and Drug Administration-approved test (HER2 and centromere 17 [CEP17] probes: PathVysion probe kit; Abbott Molecular, Des Plaines, Ill) as described.[17, 18] For each case, the number of HER2 or CEP17 copies was determined, as described.[17, 18] Briefly, with reference to a parallel hematoxylin and eosin-stained slide that identified regions of invasive carcinoma, 60 representative nuclei from the invasive tumor were scored for each tumor. A specimen with a mean HER2/CEP17 ratio of ≥ 2.0 was classified as being HER2 amplified, which is consistent with the definition used in the ToGA trial, in accordance with criteria developed for the classification of HER2 and CEP17 abnormalities as described.[17, 18, 20, 21] Chromosome 17 gain was determined using CEP17 signal patterns based on methodology and cutoff values that we previously validated using 2 large independent breast cancer sets.[20] Accordingly, polysomy 17 (gain) was defined as 3 CEP17 signals in > 30% of nuclei, as previously described[18, 20]; this cutoff clearly distinguished chromosome 17 polysomic cancers from cancers without chromosome 17 centromere anomalies.[20]

HER2 Protein Expression

HER2 expression was assessed using the US Food and Drug Administration-approved HercepTest (Dako, Carpinteria, Calif) as described.[10, 22] Each case was analyzed using criteria specific to upper digestive cancer,[10, 22] and the following data were recorded: 1) the intensity of complete, basolateral, or lateral membrane staining (none, faint, weak to moderate, or strong); and 2) the percentage of cancer cells with that intensity. Strong immunohistochemistry (IHC) intensity in ≥ 10% of cancer cells was considered positive (IHC3+), weak to moderate intensity in ≥ 10% of cancer cells was considered equivocal (IHC2plus;), and faint intensity in ≥ 10% or absent staining was considered negative (IHC1+ or IHC0, respectively). Interpretation of HER2 IHC and fluorescence in situ hybridization (FISH) was performed in parallel in a blinded manner.

HER3 Protein Expression

HER3 expression was analyzed by IHC in a randomly selected subset of the overall cohort (n = 224). Antibody staining was performed using the following procedure after antibody optimization using non-study tumor samples. After deparaffinization in xylene, slides were rehydrated and washed with distilled water. After heat-induced epitope retrieval, endogenous peroxidase activity was quenched using a 3% peroxidase block (Dako S2001; Dako) and a serum-free protein block (Dako X0909; Dako). Slides were incubated with an anti-HER3 antibody (1:100; rabbit polyclonal C-17, sc285; Santa Cruz Biotechnology, Santa Cruz, Calif.) followed by rinsing and the application of a secondary antibody (Dual Envision Link, Dako K4061; Dako). Slides were subsequently counterstained in modified Schmidt hematoxylin. For each case, the intensity of HER3 cytoplasmic staining (absent, weak, moderate, or strong) and the percentage of immunopositive tumor cells (as a continuous variable) with that intensity was determined and recorded. Slides were scored with blinding to HER2 expression/amplification and clinical data.

Statistical Analysis

Median values (with interquartile ranges) for HER2 and CEP17 variables were determined. The chi-square or Fisher exact test and Wilcoxon rank sum test were used to compare variables (ie, overall test). The Cochran-Armitage test for trend was used to determine whether significant monotonic associations existed between variables. Least squares linear regression was used to determine linearity between IHC groups and the HER2/CEP17 ratio. For Cox proportional hazards models, OS and disease-specific survival (DSS) were endpoints; events beyond 5 years were censored. All P values were 2-sided. A P value < .05 was considered to be statistically significant. Analyses were conducted using SAS statistical software (version 9.1; SAS Institute Inc, Cary, NC).


  1. Top of page
  2. Abstract

A total of 703 patients with EAC for whom there was sufficient tumor underwent HER2 testing by both IHC and FISH. Although IHC was successful in all cases, the failure rate for FISH was 4% (29 of 703 cases); 1 case had missing FISH data. Accordingly, both IHC and FISH were evaluable in 673 tumors. Table 1 shows key clinicopathologic variables.[23] The median age of the patients was 65 years; the majority of patients were male (88%) and had locally advanced disease (68% with T3-T4 disease and 73% with lymph node-positive disease). By anatomic site, 316 cases were located in the esophagus, 321 were located at the GEJ, and 36 were limited to the gastric cardia. No patient received neoadjuvant or HER2-targeted therapy.

Table 1. HER2 and HER3 Protein Expression and HER2 Gene Amplification in Relation to Clinicopathologic Characteristics in a Patient Cohort With Resected Esophageal Adenocarcinoma
  HER2 Protein Expression N = 673HER2 Gene Amplification N = 673HER3 Protein Expression n = 224a
 Total N = 6730 or 1+ (n = 417)2+ (n = 167)3+ (n = 89) Negative (n = 557)Positive (n = 116) Absent (n = 29)Anyb (n = 195) 
VariableNo. (%)No. (%)No. (%)No. (%)PNo. (%)No. (%)PNo. (%)No. (%)P
  1. Abbreviations: HER2, human epidermal growth factor receptor 2; HER3, human epidermal growth factor receptor 3; LN, lymph node.

  2. a

    Randomly selected cases.

  3. b

    Weak, moderate, or strong HER3 expression.

  4. c

    Pathologic tumor stage according to the American Joint Committee on Cancer, which incorporates histologic differentiation.[23]

Median age, y           
Male595 (88)369 (62)149 (25)77 (13).810494 (83)101 (17).62026 (13)171 (87).762
Female78 (12)48 (61)18 (23)12 (15) 63 (81)15 (19) 3 (11)24 (89) 
Depth of invasion           
T1-2217 (32)130 (60)51 (24)36 (16).217169 (78)48 (22).0247 (10)62 (90).537
T3-4452 (68)283 (63)116 (26)53 (12) 384 (85)68 (15) 20 (13)133 (87) 
Missing data4400 40 20 
Median no. of LN metastases   
Well to moderate397 (59)209 (53)116 (29)72 (18)<.0001305 (77)92 (23)<.00015 (4)121 (96)<.0001
Poor270 (41)205 (76)49 (18)16 (6) 249 (92)21 (8) 22 (23)73 (77) 
Missing6321 33 21 
I-II235 (35)140 (60)56 (24)39 (16).174185 (79)50 (21).0326 (8)72 (92).197
III428 (65)270 (63)109 (26)49 (11) 365 (85)63 (15) 19 (13)122 (87) 
Missing data10721 73 41 

Concordance Between HER2 Expression and Gene Amplification

Using the HER2 protein expression level, 62% of cases demonstrated negative expression (39% were IHC0 and 23% were IHC1+), 25% demonstrated equivocal expression (IHC2+), and 13% demonstrated positive expression (IHC3+).

As shown in Table 2, 17% of the cases (116 of 673 cases) demonstrated HER2 amplification. HER2 amplification was detected in 89% of IHC3+ cases, 13% of IHC2+ cases, and 4% of IHC0 to IHC1+ cases. As shown in Table 3, using FISH as the reference standard, the positive predictive value of IHC3+ was 89% and the negative predictive value of IHC0 to IHC1+ was 96%. The negative predictive value of an IHC0 or IHC1+ score was similarly high (97% [256 of 264 cases] or 95% [145 of 153 cases], respectively) (Table 3). Importantly, the positive predictive value of IHC2+ expression was 13% (21 of 167 cases). The specificity of IHC3+ expression was 98% (Table 3). The concordance rate between the IHC and FISH results was 74% (501 of 673 cases) when IHC scores of 2+ and 3+ were grouped together. However, when the IHC2+ group (167 tumors) was excluded, the concordance rate increased to 95% (480 of 506 cases).

Table 2. Frequency of HER2 Gene Amplification in Relation to HER2 Protein Expression
 No. of Patients (%)
  HER2 Gene Amplification (FISH)
HER2 Protein Expression (Immunohistochemistry)Total N = 673Not Amplified n = 557Amplified n = 116
  1. Abbreviation: FISH, fluorescence in situ hybridization; HER2, human epidermal growth factor receptor 2.

Positive (3+)89 (13)10 (11)79 (89)
Equivocal (2+)167 (25)146 (87)21 (13)
Negative (0 or 1+)417 (62)401 (96)16 (4)
Table 3. Sensitivity, Specificity, and Predictive Values of HER2 Protein Expression for HER2 Gene Amplification
  1. Abbreviation: HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry.

All cases (IHC2+ and IHC3+ grouped together)501/67374%
Excluding IHC2+ cases480/50695%
Sensitivity (including IHC2+ and IHC3+)100/11686%
Specificity (IHC3+ only)547/55798%
Predictive value  
Positive (IHC3+ only)79/8989%
Positive (IHC2+ only)21/16713%
Negative (IHC0 or IHC1+)401/41796%

HER2-Amplified Subgroup: Level of Amplification and Protein Expression

Among HER2-amplified tumors, the level of gene amplification increased linearly with protein expression level, and was found to be highest among IHC3+ cases, intermediate in IHC2+ cases, and lowest among IHC0 to IHC1+ cases (median gene/chromosome ratio of 7.9 vs 5.5 vs 2.8, respectively; P < .0001 for linearity) (Fig. 1). The level of gene amplification was similar between IHC0 versus IHC1+ tumors (median gene/chromosome ratio of 2.8 vs 2.9; P = .92), but differed significantly between IHC0 to IHC1+ versus IHC2+ cases (P = .0070), and between IHC2+ versus IHC3+ cases (P = .0060). Accordingly, the number of HER2 gene copies increased with HER2 expression levels (P < .0001), although the number of chromosome 17 copies was similar (P = .14) (Fig. 1).


Figure 1. Level of human epidermal growth factor receptor 2 (HER2) gene amplification in relation to protein expression is shown among HER2-amplified cases (n = 116). For each cytogenetic abnormality, the mean measurements per nucleus were determined for each case, and the median values (with interquartile ranges) across patients are shown for (A) the ratio between the HER2 gene copy number and chromosome 17 copy number (P < .0001 for linearity), (B) HER2 gene copy number, and (C) chromosome 17 copy number. NS indicates not significant.

Download figure to PowerPoint

HER2 Expression and Polysomy 17

Overall, 73% of EACs (491 of 673 EACs) demonstrated polysomy 17. The frequency of polysomy 17 was higher in non–HER2-amplified cases compared with HER2-amplified cases (75% [417 of 556 cases] vs 64% [74 of 115 cases]; P = .024), as reported.[18] However, polysomy 17 rates did not differ by HER2 protein expression overall (P = .33) or among non–HER2-amplified cases (data not shown).

HER2 and Clinicopathologic Characteristics

As shown in Table 1, HER2 amplification and expression were each found to be significantly associated with well to moderate (vs poor) tumor differentiation (P < .0001 for both). Fewer regional lymph node metastases were present in cases with HER2 amplification (vs those without) (P = .0034) and in cases with HER2 overexpression (vs no overexpression) (P = .063).

HER3 Expression

HER3 immunoreactivity was detected in 87% of randomly selected EACs (195 of 224 EACs), with 8%, 40%, and 40% of the cases, respectively, demonstrating strong, moderate, and weak expression, and expression was absent in 13% of cases (Fig. 2). Among EACs showing weak or moderate HER3 expression, 99% of tumors (176 of 178 tumors) demonstrated expression in ≥ 10% of tumor cells, with a relatively even distribution noted across quartiles of tumor cell percentage (ie, 10%-25%, 26%-50%, 51%-75%, and > 75%) generated from continuous data. Among EACs demonstrating strong HER3 expression, 82% of cases (14 of 17 EACs) demonstrated expression in ≥ 10% of tumor cells, with the majority showing expression in 26% to 50% of tumor cells. As shown in Table 1, HER3 expression was more common in tumors with well to moderate versus poor differentiation (96% vs 77%; P < .0001).


Figure 2. Representative images of human epidermal growth factor receptor 3 (HER3) protein expression by immunohistochemistry in esophageal adenocarcinomas are shown demonstrating (A) strong, (B) moderate, (C) weak, and (D) absent expression (magnification ×400).

Download figure to PowerPoint

Association Between HER2 and HER3

HER2 and HER3 expression was found to be positively associated (Table 4) (Fig. 3). Specifically, the frequency of HER2 expression was higher in tumors demonstrating any HER3 expression (ie, weak, moderate, or strong) compared with tumors lacking HER3 expression (P value for trend, .0019; overall P = .0089) (Table 4). Moreover, a positive stepwise gradient was observed between HER2 and HER3 expression levels (overall P = .0139) (Fig. 3). As shown in Figure 3, HER2 IHC3+ expression was observed in 24% of tumors demonstrating strong HER3 expression, but HER2 IHC3+ expression was not detected in any tumors that lacked HER3 expression. Likewise, HER2 was absent (HER2 IHC0 expression) in 24% of tumors with strong HER3 expression compared with 83% of tumors lacking HER3 expression (Fig. 3). The frequency of any HER3 expression was not found to be significantly higher in HER2-amplified versus non–HER2-amplified tumors (93% vs 86%, respectively; P = .235).

Table 4. Positive Association Between HER3 and HER2 Immunohistochemical Protein Expression
 HER3 Expression % (No.)
  Any (n = 195)
HER2 Expression N = 224Absent (n = 29)Weak, Moderate, or Strong
  1. Abbreviation: HER2, human epidermal growth factor receptor 2; HER3, human epidermal growth factor receptor 3; IHC, immunohistochemistry.

IHC0 (n = 123)83% (24)51% (99)
IHC1+ (n = 36)7% (2)17% (34)
IHC2+ (n = 39)10% (3)19% (36)
IHC3+ (n = 26)013% (26)
P for trendReference.0019
P overall.0089

Figure 3. Association between human epidermal growth factor receptor 3 (HER3) and human epidermal growth factor receptor 2 (HER2) protein expression is shown in randomly selected cases (n = 224). HER3 status is shown in relation to HER2 scores (immunohistochemistry [IHC]0 indicates absent; IHC1+, faint in ≥10% of cancer cells; IHC2+, weak to moderate in ≥10% of cancer cells; and IHC3+, strong in ≥10% of cancer cells). Histogram bars represent the percentage of cases.

Download figure to PowerPoint

Association of HER2 and HER3 Expression With Prognosis

The median follow-up for vital status for surviving patients was 12.6 years. HER2 positivity (defined as IHC3+ or IHC2+ with gene amplification) was found to be significantly associated with favorable OS (hazards ratio [HR], 0.76; 95% confidence interval [95% CI], 0.59-0.96 [P = .024]) and with a trend toward favorable DSS (HR, 0.79; P = .066), but the associations were not found to be statistically significant after adjustment for pathologic tumor stage (Table 5[23]) (Fig. 4), as previously reported.[17] Similarly, HER3 expression (any vs absent) was not prognostic for either OS (univariate HR, 0.66; 95% CI, 0.43-1.03 [P = .069]) or DSS (HR, 0.72; 95% CI, 0.45-1.15 [P = .16]) (Table 5) (Fig. 4).[23]

Table 5. Multivariable Cox Proportional Hazards Models Showing the Association of HER3 or HER2 With Patient Survivala
VariableDisease-Specific SurvivalOverall Survival
  1. 95% CI, 95% confidence interval; FISH; fluorescence in situ hybridization; HER2, human epidermal growth factor receptor 2; HER3, human epidermal growth factor receptor 3; HR, hazards ratio; IHC, immunohistochemistry.

  2. a

    HRs and 95% CIs were adjusted for American Joint Committee on Cancer pathologic tumor stage.[23]

Model 1      
HER3 any0.870.52-
HER3 absentReferenceReference
Model 2      
HER2-positive (IHC3+ or IHC2+/FISHpositive)0.890.69-
HER2-negative (IHC0-1+ or IHC2+/FISHnegative)ReferenceReference

Figure 4. Prognostic impact of human epidermal growth factor receptor 2 (HER2) and human epidermal growth factor receptor 3 (HER3) is shown. Kaplan-Meier survival curves are shown for (A and B) HER3 expression (any [n = 29] vs absent [n = 195]) and (C and D) HER2 positivity (immunohistochemistry [IHC]3+ or IHC2+ with HER2 gene amplification [n = 119] vs IHC2+ without amplification or IHC0 to IHC1+ [n = 589]) with HER2 analysis including cases with IHC data alone or with concurrent amplification data. Endpoints were (A and C) disease-specific survival and (B and D) overall survival. Hazards ratios (HRs) and 95% confidence intervals (95% CIs) from univariate Cox proportional hazards models are shown.

Download figure to PowerPoint


  1. Top of page
  2. Abstract

In the current study, we determined the frequency and level of HER2 gene amplification (defined as a HER2/CEP17 ratio ≥ 2) and chromosome 17 copy number in relation to HER2 protein expression in a consecutive series of surgically resected EACs. We found highly concordant results for HER2 expression and gene amplification at the extremes of IHC scores (IHC0-IHC1+ or IHC3+), but not in equivocal (IHC2+) cases, which had a 13% amplification rate. A negative IHC score (IHC0-IHC1+) was found to accurately predict a lack of HER2 amplification in 96% of tumors, as has been shown in patients with breast cancer.[24, 25] A HER2 score of IHC3+ expression was highly specific (98%) for gene amplification. Restricting analysis to HER2-amplified cases demonstrated a positive linear association between the level of amplification and protein expression. It is important to note that the level of amplification in IHC0 to IHC1+ tumors, despite being gene amplified, was low (median gene/chromosome ratio, 2.8) compared with tumors with IHC2+ (median gene/chromosome ratio, 5.5) or IHC3+ (median gene/chromosome ratio, 7.9) HER2 expression. The clinical relevance of this finding derives from the ToGA study, in which patients with HER2-amplified IHC0 to IHC1+ tumors did not benefit from treatment with trastuzumab, in contrast to those with HER2-amplified IHC2+ or IHC3+ tumors, who experienced a relative risk reduction for death of 22% or 34%, respectively.[10, 26] Our finding of low-level amplification in amplified IHC0 to IHC1+ cases provides a biologic rationale for the reported lack of benefit noted with trastuzumab. The data from the current study can assist in patient selection for HER2-targeted therapy by excluding patients with IHC0 to IHC1+ tumors and supporting FISH analysis in patients with IHC2+ tumors.

We found that HER2 amplification and overexpression were each associated with less aggressive tumor characteristics, including favorable tumor differentiation, and were not prognostic, as was previously reported by ourselves[17, 18] and in clinical trial cohorts of patients with gastroesophageal cancer.[27-31] We found that polysomy 17 was significantly more common in non–HER2-amplified compared with HER2-amplified tumors, but was not associated with HER2 protein expression. This finding suggests that extra copies of the HER2 gene due to polysomy 17 do not lead to HER2 overexpression.

HER3 was overexpressed in the majority of cases of EAC. To our knowledge, these data represent the first description of HER3 expression in human EACs. HER3 expression has been frequently detected in gastric cancers,[7, 32] in which it has not been found to be associated with numeric alterations of the HER3 gene or chromosome 12.[7] We found that HER3 expression was significantly more common in EACs with well to moderate versus poor tumor differentiation and, similar to HER2, was not prognostic. HER3-expressing tumors demonstrated a higher frequency of HER2 expression. HER2 forms heterodimers with HER3 and epidermal growth factor receptor, with each independently implicated as a key coreceptor that drives HER2-amplified breast cancer.[33, 34] Data have indicated that HER3 is as critical as HER2 to maintaining cell proliferation in breast cancer cells, and a HER2 antibody that inhibits HER3 signaling by blocking ligand-induced heterodimerization can induce tumor regression with effects that are distinct from those of trastuzumab.[33] These and other data[35] have suggested that blocking HER3 in addition to HER2 may augment therapeutic efficacy by inhibition of HER2/HER3 signaling. While this hypothesis awaits further study in patients with esophagogastric cancer, our finding of frequent HER3 expression and its coexpression with HER2 suggests that dual targeting may be a clinically relevant strategy in this malignancy and it is currently being pursued in ongoing clinical trials .

To our knowledge, the current study is the largest to date to examine HER2 by both IHC and FISH in whole-tissue sections of esophageal or gastric cancer. Compared with our prior report that examined HER2 expression and prognosis,[17] the current study analyzed gene amplification in the full study cohort to enable a more robust estimation of HER2 test concordance and examined the association between HER2 and HER3. It is important to examine EACs separately from subcardial gastric cancer because these diseases are different with regard to HER2 frequency and epidemiology.[10, 17, 36, 37] The HER2 amplification rate observed in the current study is comparable to those reported in patients with breast carcinoma. An important limitation of the current study is its retrospective nature and the finding that response data were not available to correlate with HER2 or HER3 expression levels, although trial data in which this question can be addressed are limited. In addition, the current study cohort was limited to patients with resectable disease, and it is possible that these findings may not be generalizable to those with metastatic disease.

HER2 protein expression and gene amplification were found to be highly concordant and, among amplified cases, the level of amplification increased linearly with protein expression. Equivocal (IHC2+) HER2 expression, which occurs in approximately 25% of patients, warrants analysis by FISH to identify those patients with gene amplification who may benefit from treatment with trastuzumab. We found frequent expression of HER3 among patients with EAC and a significant association with HER2 expression. These data suggest the clinical relevance of HER3 in the treatment of patients with EAC and support ongoing efforts to therapeutically target HER3 and its interaction with HER2.


  1. Top of page
  2. Abstract

Dr. Yoon-s research was supported by a Young Investigator Award from the American Society of Clinical Oncology, grant K12CA90628-10U from the National Institutes of Health (to Dr. Lynn Hartman; Dr. Yoon was a scholar on the award), a grant from Roche/Genentech, and a Charles F. and Marcia L. Forcey Career Development Award in Esophageal Cancer Research. Dr. Sinicrope was supported by K05CA-142885 (Senior Scientist Award). The Mayo Clinic received grants UL1 RR024150 and CA-114740 from the National Institutes of Health.


  1. Top of page
  2. Abstract

Dr. Yoon has received honoraria and research funding from Roche/Genentech. Dr. Sinicrope has received honoraria from Roche/Ventana.


  1. Top of page
  2. Abstract
  • 1
    Thrift AP, Whiteman DC. The incidence of esophageal adenocarcinoma continues to rise: analysis of period and birth cohort effects on recent trends. Ann Oncol. 2012;23:3155-3162.
  • 2
    Hur C, Miller M, Kong CY, et al. Trends in esophageal adenocarcinoma incidence and mortality. Cancer. 2013;119:1149-1158.
  • 3
    Spector NL, Blackwell KL. Understanding the mechanisms behind trastuzumab therapy for human epidermal growth factor receptor 2-positive breast cancer. J Clin Oncol. 2009;27:5838-5847.
  • 4
    Baselga J, Swain SM. Novel anticancer targets: revisiting ERBB2 and discovering ERBB3. Nat Rev Cancer. 2009;9:463-475.
  • 5
    Jacome AA, Wohnrath DR, Scapulatempo Neto C, et al. Prognostic value of epidermal growth factor receptors in gastric cancer: a survival analysis by Weibull model incorporating long-term survivors [published online ahead of print February 28, 2013]. Gastric Cancer.
  • 6
    Ocana A, Vera-Badillo F, Seruga B, Templeton A, Pandiella A, Amir E. HER3 overexpression and survival in solid tumors: a meta-analysis. J Natl Cancer Inst. 2013;105:266-273.
  • 7
    Begnami MD, Fukuda E, Fregnani JH, et al. Prognostic implications of altered human epidermal growth factor receptors (HERs) in gastric carcinomas: HER2 and HER3 are predictors of poor outcome. J Clin Oncol. 2011;29:3030-3036.
  • 8
    Zhang XL, Yang YS, Xu DP, et al. Comparative study on overexpression of HER2/neu and HER3 in gastric cancer. World J Surg. 2009;33:2112-2118.
  • 9
    Campbell MR, Amin D, Moasser MM. HER3 comes of age: new insights into its functions and role in signaling, tumor biology, and cancer therapy. Clin Cancer Res. 2010;16:1373-1383.
  • 10
    Bang YJ, Van Cutsem E, Feyereislova A, et al;ToGA Trial Investigators. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687-697.
  • 11
    Holden J, Garrett Z, Stevens A. NICE guidance on trastuzumab for the treatment of HER2-positive metastatic gastric cancer. Lancet Oncol. 2011;12:16-17.
  • 12
    Ruschoff J, Hanna W, Bilous M, et al. HER2 testing in gastric cancer: a practical approach. Mod Pathol. 2012;25:637-650.
  • 13
    Zuo T, Wang L, Morrison C, et al. FOXP3 is an X-linked breast cancer suppressor gene and an important repressor of the HER-2/ErbB2 oncogene. Cell. 2007;129:1275-1286.
  • 14
    Reinholz MM, Bruzek AK, Visscher DW, et al. Breast cancer and aneusomy 17: implications for carcinogenesis and therapeutic response. Lancet Oncol. 2009;10:267-277.
  • 15
    Kaufman PA, Broadwater G, Lezon-Geyda K, et al. CALGB 150002: correlation of HER2 and chromosome 17 copy number with trastuzumab efficacy in CALGB 9840, paclitaxel with or without trastuzumab in HER2+ and HER2- metastatic breast cancer [abstract]. J Clin Oncol. 2007;25(18s). Abstract 1009.
  • 16
    Yoon HH, Khan M, Shi Q, et al. The prognostic value of clinical and pathologic factors in esophageal adenocarcinoma: a Mayo cohort of 796 patients with extended follow-up after surgical resection. Mayo Clin Proc. 2010;85:1080-1089.
  • 17
    Yoon HH, Shi Q, Sukov WR, et al. Association of HER2/ErbB2 expression and gene amplification with pathologic features and prognosis in esophageal adenocarcinomas. Clin Cancer Res. 2012;18:546-554.
  • 18
    Yoon HH, Shi Q, Sukov WR, et al. Adverse prognostic impact of intratumor heterogeneous HER2 gene amplification in patients with esophageal adenocarcinoma. J Clin Oncol. 2012;30:3932-3938.
  • 19
    Warneke VS, Behrens HM, Boger C, et al. Her2/neu testing in gastric cancer: evaluating the risk of sampling errors. Ann Oncol. 2013;24:725-733.
  • 20
    Perez EA, Reinholz MM, Hillman DW, et al. HER2 and chromosome 17 effect on patient outcome in the N9831 adjuvant trastuzumab trial. J Clin Oncol. 2010;28:4307-4315.
  • 21
    Wolff AC, Hammond ME, Schwartz JN, et al;American Society of Clinical Oncology;College of American Pathologists. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol. 2007;25:118-145.
  • 22
    Hofmann M, Stoss O, Shi D, et al. Assessment of a HER2 scoring system for gastric cancer: results from a validation study. Histopathology. 2008;52:797-805.
  • 23
    Rice TW, Blackstone EH, Rusch VW. 7th edition of the AJCC Cancer Staging Manual: esophagus and esophagogastric junction. Ann Surg Oncol. 2010;17:1721-1724.
  • 24
    Press MF, Sauter G, Bernstein L, et al. Diagnostic evaluation of HER-2 as a molecular target: an assessment of accuracy and reproducibility of laboratory testing in large, prospective, randomized clinical trials. Clin Cancer Res. 2005;11:6598-6607.
  • 25
    Yaziji H, Goldstein LC, Barry TS, et al. HER-2 testing in breast cancer using parallel tissue-based methods. JAMA. 2004;291:1972-1977.
  • 26
    National Cancer Institute.FDA Approval for Trastuzumab: HER2-overexpressing Metastatic Gastric or Gastroesophageal (GE) Junction Adenocarcinoma. Accessed November 5, 2010.
  • 27
    Terashima M, Kitada K, Ochiai A, et al. Impact of expression of human epidermal growth factor receptors EGFR and ERBB2 on survival in stage II/III gastric cancer. Clin Cancer Res. 2012;18:5992-6000.
  • 28
    Waddell T, Chau I, Cunningham D, et al. Epirubicin, oxaliplatin, and capecitabine with or without panitumumab for patients with previously untreated advanced oesophagogastric cancer (REAL3): a randomised, open-label phase 3 trial. Lancet Oncol. 2013;14:481-489.
  • 29
    Van Cutsem E, de Haas S, Kang YK, et al. Bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a biomarker evaluation from the AVAGAST randomized phase III trial. J Clin Oncol. 2012;30:2119-2127.
  • 30
    Okines AF, Thompson LC, Cunningham D, et al. Effect of HER2 on prognosis and benefit from peri-operative chemotherapy in early oesophago-gastric adenocarcinoma in the MAGIC trial. Ann Oncol. 2013;24:1253-1261.
  • 31
    Lordick F, Kang YK, Chung HC, et al;Arbeitsgemeinschaft Internistische Onkologie and EXPAND Investigators. Capecitabine and cisplatin with or without cetuximab for patients with previously untreated advanced gastric cancer (EXPAND): a randomised, open-label phase 3 trial. Lancet Oncol. 2013;14:490-499.
  • 32
    Hayashi M, Inokuchi M, Takagi Y, et al. High expression of HER3 is associated with a decreased survival in gastric cancer. Clin Cancer Res. 2008;14:7843-7849.
  • 33
    Lee-Hoeflich ST, Crocker L, Yao E, et al. A central role for HER3 in HER2-amplified breast cancer: implications for targeted therapy. Cancer Res. 2008;68:5878-5887.
  • 34
    Holbro T, Beerli RR, Maurer F, Koziczak M, Barbas CF 3rd, Hynes NE. The ErbB2/ErbB3 heterodimer functions as an oncogenic unit: ErbB2 requires ErbB3 to drive breast tumor cell proliferation. Proc Natl Acad Sci U S A. 2003;100:8933-8938.
  • 35
    McDonagh CF, Huhalov A, Harms BD, et al. Antitumor activity of a novel bispecific antibody that targets the ErbB2/ErbB3 oncogenic unit and inhibits heregulin-induced activation of ErbB3. Mol Cancer Ther. 2012;11:582-593.
  • 36
    Heidl G, Langhans P, Mellin W, Bunte H, Grundmann E. Adenocarcinomas of esophagus and cardia in comparison with gastric carcinoma. J Cancer Res Clin Oncol. 1993;120:95-99.
  • 37
    Shah MA, Khanin R, Tang L, et al. Molecular classification of gastric cancer: a new paradigm. Clin Cancer Res. 2011;17:2693-2701.