EGFR copy number alterations in primary tumors, metastatic lymph nodes, and recurrent and multiple primary tumors in oral cavity squamous cell carcinoma

This study was approved by the Institutional Review Board of Chang Gung Memorial Hospital. One hundred and seventy oral cancer patients treated at Chang Gung Memorial Hospital, Lin-Kuo, were recruited for participation in this study. All patients gave informed consent for participation and were interviewed uniformly before surgery by a well-trained interviewer. The questionnaire used in the interview sought detailed information on general demographic data, current and past cigarette smoking, alcohol consumption, areca-quid (AQ) chewing, and a history of family disease (Additional file 1). All patients received curative intent surgery as an initial treatment. In the surgeries, the primary tumors were excised with safety margins greater than or equal to 1 cm (for both peripheral and deep margins). The tumor margin tissue was cryosectioned to ensure that the margin was free of tumor. For each patient, clinical histological parameters were scored according to the recommendations for the reporting of specimens containing oral cavity and oropharynx neoplasms by the Association of Directors of Anatomic and Surgical Pathology (ADASP) [ 16].

For patients who received radical surgeries, neck dissection was performed according to the tumor stage of the patients. Two types of neck dissections were used in our patients: one was a dissection of level I-III lymph nodes (supraomohyoid neck dissection) for nodal negative patients; and the other was a dissection of level I-V lymph nodes (usually a modified radical neck dissection) for nodal positive patients. We selected pathologically proven metastatic lymph nodes from the neck dissection specimens.

Patients with advanced tumor status (T3 or T4), lymph node extracapsular spread, tumor depth ≥ 10 mm or poor differentiation, adjuvant radiotherapy or cisplatin-based concomitant chemoradiotherapy would be given after surgery.

After radical surgeries with or without adjuvant chemo-radiotherapy, the patients received regular follow-up visits. For tumors growing nearby the primary tumor, in the neck or distant sites, the lesions were recorded as recurrences. In the head and neck region, the mucosa carries similar risks for developing malignancies. Lesions that were located in different tumor subsites from the primary tumor or a 2 cm distance from the primary lesion in the mucosa were recorded as secondary primary lesions [ 17]. The secondary lesions could occur simultaneously with the primary lesion (synchronous) or be found during regular follow-up appointments in the clinic after surgeries (metachronous).

EGFR gene copies were investigated with FISH using the LSI EGFR SpectrumOrange/CEP 7 SpectrumGreen probe (Vysis; Abbott Laboratories, Downers Grove, IL) according to the manufacturer’s instructions and our previous report [ 7]. In brief, section slides were incubated at 56 °C overnight, deparaffinized, dehydrated, treated with 0.2 N HCl (pH 2.5) for 20 min, and treated with 1 M sodium thiocyanate (Sigma-Aldrich Corp., St. Louis, MO) in 1 M Tris (pH 8.0) at 82 °C for 20 min. Then the specimens were digested with 0.4% pepsin (Sigma-Aldrich Corp., St. Louis, MO) in 0.9% NaCl (pH 2.35) for 15 min. The samples were briefly rinsed in ddH ₂O and 2 × SSC between steps. After fixation in 4% formaldehyde for 5 min, each slide had the probe set applied to a selected area, and the hybridization area was covered with a plastic coverslip and sealed with a glue gun before the slides were heated at 75 °C for 10 min with OmniGene (Hybaid Ltd., Middlesex, United Kingdom) to promote co-denaturation of chromosomal and probe DNAs. Hybridization was carried out in a humidified oven at 37 °C for 18 h, followed by post-washing in 0.3% Nonidet P40 (BDH, England) in 2 × SSC at 45 °C for 4 min, in 2 × SSC at 45 °C for 5 min, and finally twice in 2 × SSC at room temperature for 5 min. After being counterstained with DAPI for 5 min, the slides were mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA) and scored under an fluorescent microscope using a Plan Neofluar 100× objective (Axiophot, Zeiss, Germany) with dual and triple pass filters (Chroma Technology Corp., Rockingham, VT). At least 100 non-overlapping nuclei per case were scored independently by two independent observers who followed strict scoring guidelines and used constant adjustment of the microscope’s focus because signals were located in different focal planes. In each nucleus, the number of EGFR copies and chromosome 7 probes were assessed independently.

FISH patterns were classified into 3 strata based on the number of copies of the EGFR gene per cell as described in previous studies [ 7, 18, 19]. The strata were normal disomy, ≤ two copies in more than 90% of analyzed cells (Fig.  1a); and low amplification/polysomy (LA/Poly), ≥ three copies in more than 40% of analyzed cells. Gene amplification was defined as the presence of tight EGFR gene clusters, a ratio of gene/chromosome per cell ≥2, or ≥15 copies of EGFR per cell in ≥ 10% of analyzed cells (Fig.  1b). Tumors with LA/Poly or gene amplification were considered to be FISH positive.

Statistical analysis was performed using the SPSS statistical package (SPSS, Chicago, IL). Correlations between the frequency of EGFR FISH status and age, TNM stage, cigarette smoking, alcohol consumption, and AQ chewing were examined with the χ ² test or Fisher’s exact test. The concordance of EGFR gene copy alterations between primary tumors, metastatic lesions, recurrences and SPTs was analyzed with the McNemar test. Disease-free survival (DFS) was defined as the time from diagnosis to recurrence or metastasis. Overall survival (OS) was defined as the time from diagnosis to death. Survival curves were constructed using the Kaplan-Meier method, and the curves were compared using the log-rank test. A two-sided value of p < 0.05 was considered to be statistically significant.

The clinicopathological features of the 170 OSCC male patients between Mar 1997 and Jul 2004 who took part in this study are listed in Table  1. The major primary sites were the bucca (39.4%, 67/170) and the tongue (34.1%, 58/170). Overall, 90.6% (154/170) of the patients were cigarette smokers, 68.2% (116/170) were alcohol drinkers and 91.2% (155/170) were AQ chewers. All 170 patients received surgery as their initial treatment, and 88 (51.8%) and 30 (17.6%) patients underwent additional radiation therapy and chemoradiotherapy, respectively. The median follow-up was 57.50 months.

In our patients, 69 had lymph node metastasis identified in neck dissection specimens and 57 positive lymph nodes from neck dissection specimens were available for FISH assays. Fourteen patients had gene copy number amplification (24.6%, 14/57) and nine (15.8%, 9/57) patients had EGFR gene polysomy. During follow-up, 87 patients had recurrence and 41 recurrence tissues were used for analyzing the EGFR gene copy number. Five (12.2%, 5/41) had EGFR gene amplification and five (12.2%, 5/41) had increased gene copy number (LA/polysomy). Twenty-six patients had secondary primary tumors. A total of 17 secondary primary lesions were suitable for FISH analysis. The results showed that two (11.8%, 2/17) had gene amplification and three (17.6%, 3/17) had an increase in gene copy number. The concordance rate of EGFR gene copy number in primary tumors and lymph node metastasis was 68.4% (McNemar test: p = 0.389). The concordance rate between primary tumors and recurrence tumors was 65.9% (McNemar test: p = 0.510), and the concordance rate between primary tumors and SPTs was 70.6% (Table  2).

In four patients with multiple primary cancers, the concordance rate of EGFR gene copy number was 100% (Table  3). In one patient (No. 2) with multiple recurrences, the EGFR copy number increased in the recurring tumor. The EGFR gene polysomy was maintained in the second recurrence.

As shown in Table  4, EGFR gene amplification was significantly more prevalent in tumors at an advanced stage than tumors at early stages. Younger patients had a higher risk of EGFR gene amplification. Tumors with high levels of tumor invasion, lymph node metastasis, bone invasion and perineural invasion had a significantly higher frequency of EGFR gene amplification than tumors without those characteristics. However, EGFR gene amplification was not associated with subsites, skin invasion, AQ chewing, cigarette smoking, and alcohol consumption. We analyzed other factors that may predict EGFR gene amplification in metastatic lymph node and found no clinicopathological factors related to amplification.

The Kaplan-Meier survival curves for patients with different EGFR gene copy numbers are shown in Fig.  2. Patients showing an EGFR FISH pattern were not significantly associated with either DFS or OS (Fig.  2a, p = 0.692 and Fig.  2b, p = 0.444). The EGFR gene amplification in metastatic lymph nodes was not associated with patient survival (DFS and OS, Fig.  3a, p = 0.872, and Fig.  3b, p = 0.618, respectively). Furthermore, the EGFR FISH pattern in recurrence tumors did not predict patient survival from recurrence to death (Fig.  4, p = 0.868).