One hundred and thirty-nine Chinese female patients with breast cancer who underwent surgery at the Department of Breast Oncology, Sun Yat-Sen University Cancer Center, from Jan 2010 to May 2011 were selected. The cases with primary breast cancer were randomly selected from the archives of our Department of Pathology based on the availability of blocks and sufficient tissues. Additionally, only cases with available EGFR FISH results, mutational status, and immunostaining were analyzed. Clinical information included age, disease stage, tumor type, mass size, and axillary lymph node metastasis status. The age interval of the patients was 25-75 years, with a mean of 50.8 years. The TNM Cancer Staging Manual 7th edition of the American Joint Committee on Cancer (AJCC)  was used to classify the cancer staging: stage 0, 2 cases; stage I, 28 cases; stage II, 78 cases; stage III, 30 cases; and stage IV, 1 case. Using immunohistochemistry as a surrogate definition of intrinsic subtypes for expression profiling, cases that were ER and/or PR positive, HER2 negative, and Ki-67 low (< 14%) were classified as luminal A cancers; cases that were ER and/or PR positive, HER2 negative, and Ki-67 high were classified as luminal B (HER2 positive) cancers; cases that were ER and/or PR positive, any Ki-67, and HER2 overexpressed or amplified were classified as luminal B (HER2 positive); cases that were HER2 overexpressed or amplified and ER and PR absent were classified as HER2 positive (non-luminal) cancers; cases that were ER and PR absent, HER2 negative, and CK5/6 and/or EGFR positive were classified as basal-like cancers; and cases that lacked expression of ER, PR, HER2, CK5/6, and EGFR were considered unclassified [34–36]. The histopathological classification of breast cancer in these cases was performed by an experienced pathologist of our pathology department. Ethics approval was obtained from the Ethical Review Committee of Sun Yat-Sen University Cancer Center, and informed consent was obtained from all patients.
Tissue microarrays were constructed using 0.6-mm tissue cores as previously described . One core from the central and the other from the peripheral part of the tumor were sampled for each tumor. For each case, 4-5 μm sections of formalin-fixed and paraffin-embedded tissues were stained with Hematoxylin and Eosin for the establishment of the histopathological tumor type and differentiation grade.
Fluorescence In-Situ Hybridization Analysis of EGFR Gene Copy Number
FISH analysis for EGFR gene copy number was performed according to the manufacturer's protocol using the GLP EGFR/CSP 7 probe (GP Medical Technologies, Beijing, China). Simply, the tissue microarray sections were incubated at 65°C overnight. The slides were deparaffinized in dimethyl benzene at room temperature for 10 minutes and dehydrated in 100% ethanol. After incubation in 30% sodium bisulfite at 50°C for 20-30 minutes, the sections were incubated in 2× saline sodium citrate buffer (2× SSC; pH 7.0) at 75°C for 5 minutes. The smears were digested with proteinase K (0.20 mg/ml in 2× SSC; pH 7.0) for 20-30 minutes at 37°C, followed by a rinse in 2× SSC at room temperature for 5 minutes and then dehydration in 70%, 85%, and 100% ethanol solutions in sequence. The solvents were changed frequently and regularly so that all traces of residual paraffin were removed in the above processes . After amplification of the GLP EGFR/CSP 7 probe set, the slide was covered with a coverslip and sealed with Indian rubber. The slides were heated at 80°C for 8-10 minutes and were then hybridized at 42°C overnight. After a post-hybridization wash and dehydration of the samples, 4', 6'-diamidino-2-phenylindole (DAPI) was applied for chromatin counterstaining. At least 100 nuclei were scored for both EGFR gene signals and chromosome 7 signals under a magnification of 1000×.
The ratio was defined as: the total number of red signals (EGFR copy number) divided by the number of green signals (chromosome 7 copy number) in 100 nuclei. Tumors were scored as EGFR amplified when the EGFR FISH-positive results were: a) ratio ≥2.0, b) ≥15 copies of the red signals per cell in ≥10% of total cells, or c) the presence of EGFR gene clusters; or scored as EGFR polysomy when: ratio < 2.0, but ≥40% of cells display ≥4 copies. Meanwhile, EGFR FISH-negative include disomy (≥90% of cells display ≤2 copies), low trisomy (≥40% of cells display ≤2 copies, 10-40% of cells display 3 copies, and < 10% of cells display ≥4 copies), high trisomy (≥40% of cells display ≤2 copies, ≥40% of cells display 3 copies, and < 10% of cells display ≥4 copies), and low polysomy (10-40% of cells display ≥4 copies) [38–40]. Healthy cells were used as controls.
Fluorescent PCR Method for Analysis of EGFR Gene Mutations
The 5-μm formalin-fixed and paraffin-embedded tissues were assayed for the presence of the most common EGFR mutations in exon 19 (E746-A750 and L747-P753insS short in-frame deletions) and exon 21 (L858R and L861Q point mutations) using a RT-PCR detection kit with the Taqman probe technique (GP Medical Technologies, Beijing, China). Genomic DNA was extracted from tissues using a TIANamp Genomic DNA kit (Tiangen Biotech, Beijing, China). Two of the exon 19 deletions and the exon 21 mutations of the EGFR gene were analyzed using fluorescently labeled RT-PCR products.
Amplification reactions were setup using reagents included in the Real Time PCR Detection Kit (GP Medical Technologies, Beijing, China), in accordance with the manufacturer's protocol. Essentially, the exon 19 PCR reaction consisted of 5.4 ul deionized water, 7.5 ul 2 × PCR pre-mix, 0.15 ul forward primers-1, 0.15 ul reverse primers-1, 0.15 ul probe-1 (delE746-A750), and 0.15 ul probe-2 (delL747-P753insS) in a total volume of 13.5 ul. The exon 21 PCR reaction consisted of 5.3 ul deionized water, 7.5 ul 2 × PCR pre-mix, 0.15 ul forward primers-2, 0.15 ul reverse primers-2, 0.2 ul probe-3 (L858R), and 0.2 ul probe-4 (L861Q) in a total volume of 13.5 ul. The PCR cycling program was as follows: 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 sec, 62°C for 1 min. PCR analysis was performed by using an ABI Prism 7500 Real-Time PCR equipment (Applied Biosystems, Foster City, CA, USA) as previously described [41, 42].
Immunohistochemistry of EGFR Protein Expression
Immunohistochemical staining for EGFR was performed on the 5-μm formalin-fixed, paraffin-embedded tissue slide. EGFR was stained using the pharmDx kit according to manufacturer's instructions (DAKO). EGFR was scored positive if any membranous tumor cell staining was observed, whether or not it was completely circumferential. Staining intensity was scored as follows: 0, no membrane staining; 1+, weak staining intensity; 2+, moderate staining intensity; 3+, strong staining intensity. The staining intensity was multiplied by the percentage of tumor cells be stained to obtain a total score, resulting in a possible range 0 to 300. Samples with an overall score of 200 and higher were considered positive for EGFR-overexpression.
SPSS version 13.0 software (SPSS, Chicago, IL, USA) was used to analyze the data. Associations between EGFR gene amplification and protein expression were evaluated using Pearson's chi-square test with cross tables. Differences of P < 0.05 were considered significant. Pearson's Chi-squared test or Fisher's exact test was also applied for evaluation of multiple comparisons between EGFR amplification and expression and age, disease stage, tumor type, axillary lymph node metastasis status, and immunohistochemical index (i.e., ER, PR, and HER2). A P value < 0.05 was considered statistically significant.