Defining the molecular pathology of epithelial thyroid tumours

2017-02-22T01:43:53Z (GMT) by Mond, Michael
Thyroid cancer is the most common endocrine malignancy and its incidence has been increasing over the last few decades. Most patients with differentiated thyroid cancer respond well to standard therapies including surgery and radioactive iodine. However, for patients with persistent or recurrent disease, treatment options are limited. Owing to the lack of effective therapies, advanced thyroid cancer is associated with significant mortality. Consequently, there is a need for new therapies targeted at the underlying biology. In addition, while current staging systems allow for an estimation of the risk of recurrence and disease-specific mortality, they cannot reliably predict outcomes in individual patients. There has therefore been interest in the potential of additional markers to refine current predictive models. In this study three aspects of the molecular pathogenesis of thyroid cancer were examined: the incidence and prognostic significance of BRAF and RAS mutations; the potential role of nuclear receptors; and the presence of somatic mutations in the thyroid transcription factors, FOXE1 and NKX2-1. Differentiated thyroid tumours were profiled for known mutations; clinicopathological parameters and follow-up data were correlated with BRAF mutation in papillary carcinoma (PTC) and RAS mutations in follicular carcinoma (FC). BRAF mutation was not associated with more aggressive features at initial surgery nor with the risk of clinical recurrence. However, RAS mutations were associated with widely invasive tumour histology and distant metastases in FC. Nuclear receptors (NR) play a key role in endocrine signaling and metabolism and are important therapeutic targets. I undertook the first systematic profiling of all 48 human NR in cohorts of differentiated thyroid cancer using low density gene expression arrays. The pattern of NR expression differed by tumour type and also by BRAF mutation status in PTC. Among the novel findings was the marked overexpression of Rev-erbα in BRAFV600E PTC. Rev-erbα is a key component of the circadian clock and an important metabolic regulator. A further gene expression study suggested dysregulation of other circadian genes in PTC including overexpression of Bmal1 and downregulation of clock output genes. This has potential implications for cell cycle regulation and other aspects of tumour biology. Thyroid transcription factors are critical to the normal development of the thyroid gland and to thyroid cell differentiation. Previous studies have found associations between loci near NKX2-1 and FOXE1 and the risk of developing thyroid cancer. We identified novel somatic mutations in FOXE1. Cell transfection and western blot studies revealed marked reductions in transcriptional activity on FOXE1 responsive promoters, suggesting a potential pathogenic role. This thesis adds to current knowledge regarding the prognostic significance of existing genetic markers. It also provides novel insights into the pathobiology of thyroid tumours and suggests directions for further mechanistic studies. It is hoped that this will ultimately lead to the identification of new therapies for advanced thyroid cancer.