In search of diagnostic and prognostic markers for thyroid cancer : A proteomics approach

Abstract: Early diagnosis of cancer can influence the therapeutic strategy increasing the chances of cure or, at least, prolonged survival. It is equally important to be able to detect prognostic signs, i.e. tumor features that characterize its behavior. In the case of thyroid tumors, which usually debut as palpable nodules, fine needle aspiration biopsy is that very stage when one should ideally set a secure diagnosis and gather prognostic information (such as tumor aggressiveness, metastatic potential, response to treatment). This sampling technique has improved its performance to a considerable degree, especially when guided by ultrasonography. Nevertheless, it still fails to conclusively diagnose certain thyroid tumors as the whole tissue structure, rather than just a few hundreds of cells, is necessary for diagnosis. Therefore, there is an imperative need for molecular diagnostic and prognostic markers, which could be easily traced on cellular material. We investigated the potential of MIB-1, an index of cellular proliferation which is part of the cytological evaluation of other tumors, in facilitating the diagnosis or prognosis of thyroid tumors (Article I). Our conclusion was that MIB-1 is not particularly effective as a diagnostic marker for thyroid tumors, but it can serve as a marker of worse prognosis for patients with papillary thyroid carcinoma. In an attempt to start exploring the field of thyroid proteomics, we began by developing a protein pre-fractionation protocol suitable for using archival frozen tissue and compatible with proteomics methodology, such as two-dimensional electrophoresis (2DE) and mass spectrometry (MS) (Article II). This protocol enables us to obtain cytosolic and nuclear/nuclear membrane enriched protein fractions which comprised the starting material in our proteomics studies. S100A6, a Ca+2-binding protein evidently associated with p53 and ?-catenin, was found to be significantly over-expressed in papillary thyroid carcinomas as compared to normal tissue or follicular thyroid tumors (Article III). Besides that and by applying 2DE-coupled MS, we were able to come up with two, partly overlapping sets of proteins which seem to be robust enough in distinguishing follicular thyroid carcinoma from its benign counterpart as well as from papillary thyroid carcinoma (Article IV). This study gives new perspective in pursuing the discovery of novel thyroid cancer markers. We believe that by carrying out prospective studies particularly focused on the molecules presented in this thesis, thyroid diagnostics and prognostics can be greatly facilitated in the future for the sake of cancer patients, but also the mechanisms behind thyroid tumor development can be elucidated.