Background
The majority of tumors that arise in the head and neck region are squamous cell carcinomas arising from the upper aerodigestive tract epithelium. Progressive local spread of head and neck squamous cell carcinoma (HNSCC) affects the highly critical functions of speech, swallowing and respiration. HNSCC has a 50% disease specific mortality in the USA[
1], claiming 11,000 lives a year, and also represents one of the top ten cancers worldwide[
2]. Despite significant advances in the medical and surgical treatment of these cancers, this statistic has remained stable for decades. Novel, more effective therapeutic strategies to improve overall survival are urgently needed.
Recently, the use of targeted agents against molecular markers belonging to the human epidermal growth factor receptor (HER) family has been integrated into the treatment of protocols for many malignancies. HER family consists of four homologue members (EGFR/Her1/erbB1, Her2/erbB2, Her3/erbB3, and Her4/erbB4). All share a common structure, with an extracellular ligand-binding domain, a transmembrane domain, and an intracytoplasmic tyrosine kinase domain[
3‐
5]. Ligand binding to these receptors induces the formation of receptor homodimers and heterodimers, and thereby activates numerous downstream pathways regulating diverse processes including differentiation, migration, proliferation, and survival.
Her2 has an extracellular domain, but appears to lack ligand-binding activity, while Her3 has a non-functional kinase domain and has no catalytic activity. Her2-Her3 function by formation of a heterodimeric complex which actives an oncogenic signaling pathway (e.g. PI3/AKT pathway)[
6]. Even in its over-expressed and oncogenic state Her2 does not escape its dependency on HER family partners, and Her3 plays an important and necessary function in Her2-mediated tumorigenesis[
7]. As the HER pathway contributes significantly to progression of cancers, its family members serve as a group of anti-cancer drug target with great clinical potential. Current therapeutic efforts against the HER family are focused on small molecule tyrosine kinase inhibitors (TKIs) and humanized or chimeric monoclonal antibodies (mAbs)[
8,
9]. Recent studies revealed that Her3 is the principle mediator of TKI resistance. TKIs effectively prevent auto-phosphorylation of EGFR and Her2 in tumor cells, however, the transphosphorylation of Her3 is only transiently suppressed and Her3 ultimately escapes inhibition by TKIs in Her2 over-expressing tumor cells[
10]. Consequently, the Her3 resistance causes PI3/Akt pathway resistance, tumor survival, and escape from proapoptotic consequences of the loss of oncogenic Her2 signaling.
Her2 over-expression in HNSCC has been reported, [
11] but there are few studies on Her3 expression in HNSCC[
12]. Clinical studies with agents targeting HER proteins have been performed in patients with HNSCC, with promising results[
13‐
15]. However, the prognostic significance and the potential as biomarkers of Her2 and Her3 in HNSCC remains undetermined. In the present study, protein expression levels of Her2 and Her3 were interrogated on a tissue microarray (TMA) of surgically removed samples of HNSCC by immunohistochemistry (IHC). The relationships between protein expression and clinicopathological parameters/survival of HNSCC patients were also analyzed.
Materials and methods
Patients and tumor samples
A total of four hundred twenty four formalin fixed and paraffin embedded tumor specimens with HNSCC were obtained from the archives of the Pathology Department of the University of Virginia Health System and were assembled into TMA blocks containing: 387 primary HNSCC tissues, 20 matching metastatic tissues and 17 recurrent HNSCC tissues. The clinical information of these patients was obtained from the University of Virginia Cancer Registry. Material was obtained with appropriate human protection approvals from the institutional review board of University of Virginia Health System and office of Human Subjects Research at the NIH. Information on post-operative radiation and/or chemotherapy, and performance status of patients was unavailable for analysis.
Tissue microarray construction
TMAs were constructed from archival formalin fixed, paraffin embedded tissue blocks. For each tumor, a representative tumor area was carefully selected from a hematoxylin and eosin stained section of a donor block which as previously described[
16]. Four 0.6 mm diameter cores were retrieved from selected regions of donor blocks from each case and transplanted to the recipient block using a manual tissue arrayer (Beecher Instruments, Silver Spring, MD). Multiple 5-μm thick sections were cut with a microtome and H&E staining of TMA slides were examined every 50
th sections for the presence of tumor cells.
Western blot analysis of Her3 antibody
For three cell lines, A549, MCF7 and BxPC3, a total 4 × 107 cells were rinsed twice with ice-cold PBS and added 0.5 ml of the Protein Extraction Solution RIPA (Pierce Biotechnology, Rockford, IL). After incubation for 30 min on ice, cells were scraped and centrifuged. Protein concentrations were measured by the BCA protein assay kit (Pierce Biotechnology). To determine the specificity of anti-Her3 antibody, 30 μg of protein were separated by 4-12% NuPAGE®Novex Bis-Tris polyacrylamide gel electrophoresis and transferred to nitrocellulose membrane (Invitrogen, Carlsbad, CA). The membranes were blocked with 5% nonfat dry milk in TBST (50 mM Tris, pH 7.5, 150 mM NaCl, 0.05% Tween-20) for 1 h, washed, and subsequently incubated overnight at 4°C in TBST with 5% BSA containing anti-Her3 antibodies (RTJ.2, mouse monoclonal; Santa Cruz Biotechnology, Santa Cruz, CA; dilution 1:1000). Her3 expressional signals were detected with horseradish peroxidase-labeled anti-mouse antibodies (Chemicon International) and enhanced with SuperSignal Chemiluminescence kit (Pierce Biotechnology).
Immunohistochemistry and scoring
To investigate the significance of Her2 and Her3 expression in HNSCC, 4-micron histologic sections of the TMAs were stained by IHC. Briefly, tissue sections were deparaffinized and hydrated in xylene and serial alcohol solutions, respectively. Endogenous peroxidase was blocked by incubation in 3% H
2O
2 for 10 min. Antigen retrieval was performed in a steam pressure cooker with prewarmed antigen retrieval buffer pH 6 (DakoCytomation, Carpinteria, CA) at 95°C, for 10 min and 40 min, for Her3 and Her2 staining respectively. To minimize non-specific staining, the section was incubated with protein block (DakoCytomation) for 15 min. After washing with TBST, the specimen was incubated with anti-Her3 antibodies (RTJ.2, mouse monoclonal; Santa Cruz Biotechnology, Santa Cruz, CA; dilution 1:500) overnight at 4°C, anti-Her2 antibodies (c-erbB2, A0485, rabbit polyclonal: Dako; dilution 1:750) at room temperature for 30 min. Antigen-antibody reactions were detected with DAKO LSAB
®+ peroxidase kit (Dako). The stain visualized using 3,3'-diaminobenzidine plus (Dako) and was lightly counterstained with hematoxylin, dehydrated in ethanol, and cleared in xylene. Appropriate negative controls were concurrently performed, and the TMAs included appropriate positive control tissues. The slides were covered and observed under a light microscope (Axioplot, Carl Zeiss, Jena, Germany). Her3 assessment included manual qualitative interpretation of both membranous and cytoplasmic staining. Her2 (membranous) staining and Her3 membranous staining were scored according to the commonly applied criteria of Her2 membranous staining (0,+1, +2, +3) and further dichotomized as either negative (score 0) or positive (+1, +2, or +3)[
17]. For assessment of Her3 cytoplasmic staining, two scores were assigned to each core. (a) the staining intensity [categorized as 0 (absent), 1 (weak), 2 (moderate), or 3 (strong)] and (b) the percentage of positively stained epithelial cells [scored as 0 (0% positive), 1 (1-25%), 2 (26-50%), 3 (51-75%), and 4 (76-100%)]. An overall protein expression score was calculated by multiplying the intensity and positivity scores (overall score range, 0-12). This overall score for each patient was further simplified by dichotomizing it to negative (overall score < = 3) or positive (score of > = 4). Consensus review by two pathologist (MT and SMH) was conducted.
Statistical analysis
The Chi-square test was applied to test the possible association between the expression of Her2/Her3 and the clinicopathologic parameters. The Mann-Whitney U-test was used for the analysis of the relationship between Her2/Her3 expression and the patient's age. Kaplan-Meier curves were plotted to assess the effect of Her2/Her3 expression on overall survival. Different survival curves were compared using the log-rank test. Multivariate proportional Cox models were applied to assess the prognostic significance of Her2, Her3, primary tumor sites, histological grading, gender and age. P< 0.05 was regarded as statistically significant. All statistical analyses were performed using the SPSS for Window (16.0) package (SPSS, Chicago, IL).
Discussion
The human epidermal growth factor receptor (EGFR) family of receptor tyrosine kinases, including EGFR, Her2, and Her3, is a potent target for antitumor strategies as it plays a critical role in HNSCC tumor cell growth, survival, invasion, metastasis and angiogenesis. Numerous pharmaceutical approaches have been undertaken to treat various human cancers using drugs that target EGFR family and more than 10 agents are in clinical trials[
18‐
20]. However, current EGFR-targeted therapeutics have had much narrower efficacy than initially predicted based on preclinical models. Due to the limited clinical benefit of current anti-EGFR family therapies, better understanding of EGFR family members is required to develop improved clinical benefit for cancer patients.
Expression of EGFR family members is highly regulated, and outside of the bone marrow, expression is generally low, with increased expression in tumors commonly characterized as over-expression, as it reflects an increase above the baseline expression encountered in the majority of tissues. We attempted to apply a scoring system that reflected the tumor-associated increase in expression of Her2 and Her3, using cut-offs of greater than or equal to +1 (range 0 to +3)for membranous staining of Her2 and Her3, and greater than or equal to 3 (range 0 to 12) for cytoplasmic expression of Her3.
Her2 gene amplification and over-expression has been reported in approximately 30% of breast cancers and in several other tumors, including ovarian, gastric, colorectal cancers[
21‐
24]. In HNSCC, Her2 over-expression has been described previously, although reports on its clinical relevance are less conclusive[
11,
25‐
34]. In the present study, we analyzed the expression of Her2 in primary HNSCC and corresponding metastatic tissues by IHC techniques. With a cut-off level between score 0 (negative) and score 1, 2, and 3 (positive) and found Her2 protein expression in 26.9% of primary tumor cases, which is relatively consistent with previous reports[
11,
30]. However, the frequency of Her2 over-expression decreased from 26.9% to 9.3% if we set cut-off level between score 0 and 1 (negative) and score 2 and 3 (positive), which is conventionally used in breast cancer. The scoring system for Her2 expression in breast cancer does not necessarily translate effectively to other tumor types, and alternative approaches to scoring may be more efficacious[
35].
We also found that Her2 expression in our samples was mostly detected in the membrane, and there was lack of cytoplasmic staining. Although Her2 cytoplasmic expression in HNSCC and other cancers has been reported in the previous studies, its interpretation is currently not clear[
27,
30,
36]. In case of breast cancer, membranous staining is the criterion for positivity[
17] Another interesting finding is that Her2 expression was associated with histological grade, with the most frequent Her2 expression observed in grade 2 tumors, when cut-off levels are set between score 0 (negative) and score 1, 2, and 3 (positive). However, the association disappears using in HNSCC the conventional cut-off levels for breast cancer, which is consistent with the previous study[
11]. These results support the approach that scoring parameters should be carefully considered depending on types of cancers.
The prognostic significance of Her2 expression in HNSCC remains to be elucidated. Some investigators have shown that there was no significant correlation between Her2 over-expression and clinicopathological factors[
26,
27,
34]. The findings of current study are consistent with those previous reports, although conflicting outcomes have been also reported. High frequency of Her2 expression was reported in the patients with HNSCC and it was significantly associated with positive lymph node status and advanced stage[
11,
28,
29]. As there was no correlation between Her2 expressions and most of the clinicophaological parameters, and there was no relation between Her2 over-expression and worse survival of the patients, this suggests that the expression status of Her2 alone might not be a good prognostic predictor in HNSCC.
Another EGFR family, Her3, is one of the most interesting targets for inhibition of EGFR signaling because Her2-Her3 heterodimer constitutes the most active signaling dimer in this family[
37]. Nevertheless, the efforts to develop new therapeutic agents that target Her3 in HNSCC or other cancers have lagged behind because of its impaired kinase activity. The dimerization of Her3 with other EGFR family members is required for activating signal pathways. Her2 is regarded as a preferred partner, and requires Her3 to promote cell proliferation. As Her3 and Her2 are mutually dependent proteins and function in complementary manner, but the combination of Her2 and Her3 expressions may be a potentially more useful biomarker in HNSCC than the status of Her2 or Her3 expression alone.
The effect of Her3 expression in HNSCC has been studied previously, [
12] but its significance as biomarker had remained undetermined. Several studies showed significant correlation between Her3 over-expression and decreased survival of patients with colorectal, gastric, lung, ovarian and breast cancer,[
17,
38‐
40] although conflicting results have been also reported in breast cancer[
41‐
44]. In the present study, Her3 expression was observed predominantly in the cytoplasm (77.5% of primary tumors) and less frequently (8.8%) in the cell membrane of tumor cells. Moreover Her3 over-expression significantly increased in metastatic tissues (30.0%) compared to primary tumors (8.8%). We also found that the Her3 membranous over-expression was significantly correlated with worse survival and was an independent predictive factor in multivariate analysis[
45]. Combining both Her2 and Her3 staining result, the patients with Her2 positive and Her3 negative tumors had significantly long survival (
P = 0.020). We are limited by the lack of information on the staging of primary HNSCC. Tumor staging is important because the stage at diagnosis is the most powerful predictor of survival. However, our findings are,, to our knowledge, the first report of the relationship between both Her2 and Her3 to survival in HNSCC.
The staining pattern of Her3 is not entirely clear, although membranous expression of EGFR and Her2 is regarded as an important parameter. Some investigators have reported predominant cytoplasmic Her3 staining in esophageal,[
46] ovarian,[
47] whereas cytoplasmic and membranous expression pattern have been reported in colorectal,[
48] gastric[
40] and breast cancer[
43]. In the case of HNSCC, Wei
et al. reported that Her3 staining was restricted in cytoplasm in laryngeal carcinoma[
12]. The discrepancies may be partially explained by the difference of antibodies used for staining or method of assessment for determination of Her3 status. So far, there are no standardized methods of Her3 staining and scoring, our findings suggest the importance of membranous expression of Her3.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
J-YC and SMH conceived of the study and devised the experimental design. SMH and CM designed and build the tissuemicroarrays. MT, J-YC, and YK performed experiments. RX, HC, J-YC and SMH performed data analysis for experiments and clinical records. MT, RX and J-YC drafted the final version of the manuscript and figure legends. SMH revised the figures, added critical content to the discussion and was responsible in revising all portions of the submitted portion of the manuscript. All authors read and approved the final manuscript.