Martine De Herdt

141 MET, ECD shedding, and loss of E-cadherin Introduction Approximately 30% of head and neck squamous cell carcinomas (HNSCCs) originate in the oral cavity (1). Depending on disease stage and histopathological features, treatment of oral squamous cell carcinoma (OSCC) consists of single modality surgery, radiotherapy, or a combination of both with or without adjuvant systemic therapy (chemotherapy and/or targeted therapies) (1-4). Further improvement using more aggressive (chemo)radiotherapy might be outweighed by increased toxicity (5). Therefore, interest exists to implement targeted therapies in the management of these cancers. An interesting, yet elusive, target for therapy is the receptor tyrosine kinase (RTK) MET (5-7). This transmembranous (TM) protein facilitates invasive growth by orchestrating a program similar to epithelial-to-mesenchymal transition (EMT) and is recognized as a negative prognostic factor for HNSCC (8, 9). During EMT, epithelial cells obtain a mesenchymal phenotype by downregulation of epithelial proteins (such as E-cadherin), induction of mesenchymal proteins, and invasion of the extracellular matrix (10, 11). Unfortunately, research into therapies directed against MET has not yet resulted in major survival benefits (12-14). This might be due to a lack of companion diagnostics (CDx), for which development is challenging for several reasons (9, 12, 13, 15). Some are of technical nature, ie, absence of specific antibodies and reliable evaluation of immunohistochemistry. Others are related to biology, ie, MET processing and specifically its degradation. MET can be subjective to presenilin-regulated intramembrane proteolysis. This process encompasses initial cleavage by membrane metalloproteases resulting in shedding of the ectodomain (ECD) from the membrane and subsequent cleavage of the remaining membrane-anchored C-terminal fragment by the γ-secretase complex (16). Theoretically, proteolytic processing of MET results in four different states of the receptor with respect to the cell membrane: no receptor (no MET), a membrane-anchored N-terminal fragment without the catalytic domain (decoy MET), the complete receptor (MET) and a TM C-terminal fragment with the catalytic domain (TM C-terminal MET). MET processing has necessitated novel approaches to categorize MET immunoreactivity. Using C- and N-terminal MET antibodies and a tissue microarray (TMA), it was established that C-terminal MET immunoreactivity either is homogeneous (uniform negative or positive staining) across oral and human papillomavirus (HPV-)negative 5

RkJQdWJsaXNoZXIy MTk4NDMw