Biomarker discovery for head and neck cancer, a proteomics approach

Abstract

The prognosis of head and neck squamous cell carcinomas (HNSCC) has only moderately improved during the last decades. Major causes for this are the high number of patients that present with advanced stage of disease and the high frequency of second primary tumors. This emphasizes the need for early diagnosis of both primary and secondary tumors. Head and neck tumors arise in a field of premalignant cells, i.e. a precursor field. Early detection of these fields, in particular those with a high risk of progression, might improve the overall prognosis of HNSCC patients. Proteomics is the field of research that aims to study all proteins present in a cell or tissue and thereby enables the identification of protein biomarkers, i.e. those proteins, present in tissue or body fluids, that are indicative of the presence, state or behavior of a tumor or precancerous lesion. The development of protein biomarkers is characterized by different phases, starting with an initial discovery of the possible protein candidates, to the final thorough clinical validation of these candidates in prospective multicenter trials. The studies described in this thesis report the discovery and initial clinical validation of protein biomarkers that can predict malignant progression of HNSCC precursor fields. We used genetically characterized normal, precursor and tumor tissue from eight patients to identify proteins with a different expression level. We demonstrated that expression of four proteins, keratin 4, keratin 13, cornulin and small proline-rich protein 3, was not only significantly decreased in tumors, but also in an independent series of severe dysplasias, mucosal tissue at a high risk of malignant progression. This suggested that we found potential promising candidate biomarkers, and we decided to investigate whether these proteins might predict malignant progression of precancerous tissue in different patient groups. Our data revelaed that these candidate biomarkers could not be used to discriminate progressing from non-progressing leukoplakias, i.e. white lesions in the mouth that carry a potential risk of malignant progression. However, keratin 4 and cornulin expression in surgical margins of patients treated for HNSCC strongly correlated with the occurrence of local recurrence. In the future, patients at high risk for local recurrence can be considered for increased surveillance or tertiary treatment to eradicate precursor fields. Taken together, the work in this thesis describes different approaches to discover and validate protein biomarkers for head and neck cancer and has provided new input for the early detection of head and neck cancer.