Mutant epidermal growth factor receptor contributes to head and neck cancer growth and resistance to EGFR targeting

Epidermal growth factor receptor (EGFR) is overexpressed in head and neck squamous-cell carcinoma (HNSCC) and its expression levels correlate with decreased patient survival. Nonetheless, therapies aiming at blocking EGFR has shown limited efficacy in a proportion of patients with HNSCC in clinical trials. Sok et al. in a recent paper (Clin Cancer Res, 2006, 12:5064-5073 ) attempted to ascertain whether it is due to mutation of EGFR. As the most common form of mutation of EGFR seen in several other types of cancer is a truncation mutation,

Epidermal growth factor receptor （EGFR） is overexpressed in head and neck squamous-cell carcinoma （HNSCC） and its expression levels correlate with decreased patient survival. Nonetheless, therapies aiming at blocking EGFR has shown limited efficacy in a proportion of patients with HNSCC in clinical trials. Sok et al. in a recent paper （Clin Cancer Res, 2006, 12：5064-5073） attempted to ascertain whether it is due to mutation of EGFR. As the most common form of mutation of EGFR seen in several other types of cancer is a truncation mutation, EGFR variant Ⅲ （EGFRvm）, a study was undertaken to determine the frequency of EGFRvm expression in patients with HNSCC and its biological consequences on tumor growth in response to EGFR targeting. Thirty three human HNSCC samples were evaluated by immunostaining with an EGFRv Ⅲ-specific antibody （L8A4）, and RT-PCR for EGFRvm expression. EGFRvm expression was detected in 14 of 33 （42%） of the HNSCC samples. It was only expressed by tumor cells but not by submucosa or normal adjacent mucosa. By RT-PCR, EGFRvm expression was found in nearly half of the HNSCC samples examined. Since human tumors expressing EGFRv Ⅲ do not retain expression of the mutant receptor when cultured in vitro, a representative HNSCC cell line （UM-22B） was stably transfected with an EGFRv Ⅲ expression construct. The EGFRv Ⅲ-expressing cells and vector-transfected controls were compared for growth rates. When these paired cells were grown in vitro, the population doubling time was shortened in EGFRvⅢ-expressing cells when compared with the vector-transfected control cell population. In in vivo experiments, the tumor volume of the growing EGFRv Ⅲ-expressing cells in athymic nude mice was 〉 7 fold greater in size in comparison with that of the vector-transfected control cells by day 24 after cell inoculation. When the xenografts derived from EGFRv Ⅲ-transfected HNSCC cells were immunohistochemically stained and by Western blot for VEGF, its level was found nearly doubled. When these paired cells were treated with cisplatin, the percentage of cells undergoing apoptosis in vector-transfected control cells was 43% whereas that in EGFRv Ⅲ-expressing cells was 19.6% （P 〈 0. 001 ）. When these paired cells were treated in vitro with monoclonal anti-EGFR antibodies （C225/cetuximab）, the percentage of killing was 33% in EGFRv Ⅲ-expressing cells and that in vector-transfected control cells was 54%.