The clinical impact of epidermal growth factor receptor (EGFR) (E746–A750del) mutation and human papillomavirus (HPV) in oral squamous cell carcinoma (OSCC) is unclear. EGFR (E746–A750del) expression was analyzed in OSCC specimens ( n = 161) by immunohistochemistry. The expression results were correlated with clinical characteristics and impact on survival. Using INNO-LiPA Extra, high-risk HPV types were genotyped and analyzed in 211 OSCC specimens. Positive EGFR (E746–A750del) expression ( n = 40/161, 25%) was not associated with any clinicopathological characteristics, prognostic factors, social habits (smoking, alcohol consumption), or tumour-specific survival. HPV16 DNA was detected in three out of 211 samples (HPV16-positive: n = 3/211, 1.4%). This study shows that mutation-specific EGFR (E746–A750del) expression and HPV do not appear to be relevant to the survival of patients with OSCC.
Topographically different subgroups of head and neck squamous cell carcinomas (HNSCC) harbour different chromosomal aberrations, which can play a role in the initiation of tumours and/or tumour progression. In this study, we focused on oral squamous cell carcinoma (OSCC), in which site-specific prognostic biomarkers may also occur. The exact molecular mechanism of carcinogenesis in OSCC has yet to be clarified. The identification of potential oncogenes in OSCC, such as epidermal growth factor receptor (EGFR) family members, and a better understanding of the molecular mechanisms, may provide more useful prognostic markers and probably a more effective treatment.
EGFR is a transmembrane tyrosine kinase receptor and is expressed throughout development in a variety of cell types. Several ligands, like epidermal growth factor (EGF) and transforming growth factor alpha (TGF-α), can bind to the cell-surface of EGFR, resulting in activation of its intrinsic kinase activity. Although clinical targeting of EGFR with cetuximab – a chimeric anti-EGFR monoclonal antibody (mAb) with multiple approved indications in HNSCC – has been established, whether or not EGFR is an appropriate therapeutic target specifically in OSCC remains controversial.
A previous study by Gröbe et al. revealed a statistically significant association between strong EGFR immunohistochemical expression and advanced lymph node involvement, but not patient survival. On the other hand, the work by Laimer et al. demonstrated high EGFR expression in patients with SCC of the oral cavity and oropharynx to be an independent prognostic marker for reduced survival. Therefore, the EGFR antigen represents an attractive target for targeted therapies with monoclonal antibodies or specific tyrosine-kinase inhibitors (TKI) in these patients; decreased EGFR expression might be favourable for a better survival in OSCC patients.
Smoking and alcohol consumption are well-described risk factors in OSCC. Some studies have demonstrated EGFR mutations in a subset of patients with non-small-cell lung cancer (NSCLC) that were associated with ‘never smoking’. Studies have shown a significant association between the presence of EGFR-activating mutations in never-smoking patients and lung tumours, and their sensitivity to EGFR TKIs like gefitinib and erlotinib.
Most of the EGFR mutations occur in exons 18–21 in the tyrosine kinase domain. Nearly 90% of all EGFR mutations are deletions localized in exon 19, such as E746–A750del, and the L858R point mutation in exon 21. Patients with these mutations are highly responsive to EGFR inhibitors like gefitinib and erlotinib. EGFR mutations, especially E746–A750del and the L858R point mutation, are closely associated with favourable clinical outcomes in nearly 80% of patients with NSCLC when treated with a TKI. However, the aetiology of these mutations remains unknown. Recently, EGFR mutations have been described in OSCC by direct polymerase chain reaction (PCR)-based sequencing of EGFR tyrosine kinase domains.
The immunohistochemical detection of biomarkers in tumour tissue sections is an essential and powerful technique to determine the malignancy of the tumour and to stratify cancer patients for treatment. Kawahara et al. demonstrated a simple immunohistochemical diagnosis using an antibody for the detection of the E746–A750del mutation that can provide important quantitative and tissue-specific expression data to complement DNA sequence results. They showed that the sensitivities of the immunohistochemical and DNA tests were comparable and that the two methods show good correlation in determining the EGFR mutation.
A study from Taiwan suggested that human papillomavirus (HPV) infection (HPV16/18) could be associated with NSCLC in non-smoking patients. It remains controversial whether HPV infection is related to carcinogenesis in OSCC. Laco et al. reported the association of HPV infection and non-smoker status in patients with OSCC. HPV-positive tumours have fewer mutations than tobacco-associated cancers. They also have a different spectrum of genetic changes because HPV will automatically target the p53 pathway through E6, and the retinoblastoma tumour suppressor will be influenced through E7. Rampias et al. showed that E6 and E7 oncogene repression restores p53 and retinoblastoma tumour suppressor pathways and induces apoptosis in HPV16-positive oropharyngeal cancer cell lines. Therefore, patients with HPV-positive tumours tend to have a better prognosis because they respond more frequently to radiotherapy or chemo-radiation and they have substantially better survival rates. In the context of EGFR expression and HPV infection, low EGFR expression in HPV-related HNSCC might be associated with a favourable patient outcome. However, the clinical impact of HPV and EGFR mutations in OSCC remains unclear.
The aim of the present study was to determine whether HPV infection is related to EGFR mutation or non-smoker status in OSCC. We also sought to analyze the potential prognostic and predictive influence of E746–A750del mutation expression in our study cohort with OSCC.
Materials and methods
Patients and tumour specimens
A retrospective review of the records of 211 patients with OSCC after primary radical R0 tumour resection over a period of 10 years and of healthy individuals (normal oral mucosa tissues, n = 10) was carried out. Clinicopathological and follow-up data were available for 191 of the 211 patients. Out of these 191 patients, formalin-fixed paraffin-embedded (FFPE) blocks were available for 161 patients for representative immunohistochemical staining. Patients with non-resectable disease, inadequate follow-up data, and those who had undergone preoperative antineoplastic therapies (chemo-radiation/chemotherapy) were excluded from the study. The diagnosis of SCC was confirmed by the department of pathology, and the specimens were retrieved retrospectively from the pathology archives. The material was archival FFPE tissue from routine histopathological work-up. The study was approved by the local ethics committee; informed consent was obtained from the patients prior to surgical resection. Follow-up data were obtained from the local tumour registry. The last follow-up was recorded as the last outpatient visit or the date of loco-regional recurrence or tumour-specific death. Tumour and patient characteristics are summarized in Table 1 .
|Characteristics||Number of patients||P -value|
|Total n = 161||EGFR E746–A750del negative (<10%) n = 121 (75%)||EGFR E746–A750del positive (>10%) n = 40 (25%)|
|<60 ± 11.8 SD||80 (49.7%)||55 (69%)||25 (31%)|
|≥60 ± 11.8 SD||81 (50.3%)||66 (81%)||15 (19%)|
|Male||125 (77.6%)||92 (74%)||33 (26%)|
|Female||36 (22.4%)||29 (81%)||7 (19%)|
|Site distribution of OSCC||0.2026|
|Lips||10 (6.2%)||10 (100%)||0 (0%)|
|Tongue||36 (22.4%)||28 (78%)||8 (22%)|
|Floor of the mouth||66 (41.0%)||48 (73%)||18 (27%)|
|Palate||15 (9.3%)||10 (67%)||5 (33%)|
|Buccal mucosa||9 (5.6%)||8 (89%)||1 (11%)|
|Alveolar ridge||25 (15.5%)||17 (68%)||8 (32%)|
|Histological grading||0.9888 a|
|G1||39 (24.2%)||30 (77%)||9 (23%)|
|G2||108 (67.1%)||80 (74%)||28 (26%)|
|G3||13 (8.1%)||10 (77%)||3 (23%)|
|G4||1 (0.6%)||1 (100%)||0 (0%)|
|Tumour size||0.5263 b|
|pT1||64 (39.8%)||48 (75%)||16 (25%)|
|pT2||40 (24.8%)||28 (70%)||12 (30%)|
|pT3||17 (10.6%)||14 (82%)||3 (18%)|
|pT4||40 (24.8%)||31 (78%)||9 (22%)|
|Cervical lymph node metastasis||0.7363|
|pN0||118 (73.3%)||90 (76%)||28 (24%)|
|pN1–3||43 (26.7%)||31 (72%)||12 (28%)|
|UICC stage||0.8927 c|
|I||48 (29.8%)||39 (81%)||9 (19%)|
|II||36 (22.4%)||24 (67%)||12 (33%)|
|III||31 (19.3%)||25 (81%)||6 (19%)|
|IV||46 (28.6%)||33 (72%)||13 (28%)|
|Never-smoker||44 (27.3%)||34 (77%)||10 (23%)|
|Smoker||117 (72.7%)||87 (74%)||30 (26%)|
|Never||55 (34.2%)||38 (69%)||17 (31%)|
|Ever||106 (65.8%)||83 (78%)||23 (22%)|
|No||117 (72.7%)||87 (74%)||30 (26%)|
|Yes||44 (27.3%)||33 (75%)||11 (25%)|