Search

Conjunctival melanoma (CoM) is a rare, but potentially fatal, malignancy constituting about 2% of all ocular cancers. Despite successful local primary CoM treatment, an estimated 60% recur locally within 5 years, and are fatal in 26% at 10 years. Because of CoM rarity, management of patients varies between centres, and the protocols are based on limited case series and reports rather than randomised clinical trials. Further, the mechanisms behind disease pathogenesis and genetics are largely unknown. The scope of the work presented in this thesis was to understand the genetic aberrations in CoM in a large series within a mutlicentre collaboration. Chapter 1 provides an overview of CoM. The aim of chapter 2 was to determine gene copy number variations (CNVs) and their correlation with clinical or histological CoM characteristics, in an attempt to identify potential genetic biomarkers for pathogenesis and metastasis. Regional and focal gains of 6p and losses of 7q were the most common aberrations in CoM, whether primary or recurrent, but not related to any particular tumour features. Most important were the 10q deletions, which commonly recurred in metastasising CoM, and were also associated with lymphatic invasion and increased thickness. In chapter 3, BRAF and NRAS mutations, which are the most common in cutaneous melanoma, were investigated in addition to their associated CNVs. BRAF and NRAS mutational frequencies in CoM were 34% and 13%, respectively, similar to skin melanoma. Apart from BRAF-mutant CoM tending to occur in younger patients (median age 58 years) than NRAS-mutant or Wild-type tumours, there were no other significant associations between BRAF/NRAS mutational status and the tumour features and/or the clinical outcome. In addition, 10q and 19q deletions were commonly seen in the BRAF- and NRAS-mutant tumours, respectively. The focus of chapter 4 was to assess the prognostic value of the AJCC/Tumor-NodeMetastasis (TNM) system for a cohort of 220 CoM patients (follow up 0.36-35 years). This was achieved by comparing the 6th and 7th TNM staging systems. The 7th TNM staging was more accurate than the 6th, and better conformed to the literature-reported patients' outcomes. However, some deficiencies of the 7th staging system were also highlighted in the chapter, and suggestions were proposed for improvements in future editions. Notably the 8th TNM staging system for all cancers was only very recently released, and some relevant alterations have been made. In conclusion, the multicentre study on CoM presented in this MD thesis provides further insight into CoM genetics. It is a first step towards bridging the significant knowledge gap in CoM (when compared to its skin and intraocular counterparts). The work presented here provides basis for several future projects outlines potential future work that could be taken forward to improve this rare cancer's treatment & prognostication.

Uveal melanoma (UM) is the most common primary intraocular cancer of the eye in adults. Despite successful treatment of the primary tumour, approximately half of patients diagnosed with UM will develop metastases. To date, treatments for metastatic UM have had only limited effect, hence prognosis remains poor. Metastatic UM occurs predominantly in the liver. The mechanisms responsible for this remain largely unknown. Investigations into the molecular and cellular processes underlying UM development and metastasis are essential to develop therapeutic approaches directed against dissemination of this disease. In vitro and in vivo models play a crucial role in basic and translational oncology research. The scope of the work presented in this thesis was to develop innovative and clinically relevant in vitro and in vivo preclinical models that reproduce various aspects of metastatic processes in UM. In Chapter 2, primary UM cells were used to establish 3D tumour spheroids which retained many morphological and molecular characteristics of the original tumour. These 3D tumour spheroids more accurately mimic the 3D in vivo environment in which cancer cells reside and as a result can be used in co-culture assays, drug assays, and further transferred to an animal model to examine metastatic progression. Chapter 3 outlined one such in vivo model; the chick embryo. This model represents an accessible and economical in vivo model that has long been used in developmental biology. A panel of UM cell lines that represent the characteristic genetic and morphological profile in UM were used to form tumour nodules on the chorioallantoic membrane (CAM) and metastatic colonisation in internal organs following intravenous injection. Cells predominantly homed to the chick embryo eye spreading widely within the uveal tract, and to the liver, representative of the disease in patients allowing for a better understanding in the spread of this disease. Furthermore, cell lines forming tumour nodules on the CAM can be further used to assess candidate therapeutic drugs in an in vivo setting. Chapter 4 aimed to investigate such drug candidates by assessing the effects of targeted clinical inhibitors on UM cell lines. One particular inhibitor, the bromodomain PPI inhibitor JQ1, caused cell cycle arrest in two UM cell lines paving the way for many further insightful investigations. The models established in this thesis represent valuable, inexpensive and rapid in vitro and in vivo systems that can improve our understanding of the metastatic process in UM and provide a platform for the preliminary testing of potential new agents to aid molecular-based cancer therapy.