Use this URL to cite or link to this record in EThOS:
Title: Enhancing prognostication in uveal melanoma
Author: Thornton, S.
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
Uveal melanoma (UM) is the most common primary intraocular malignancy occurring in adults, with a propensity for liver metastases, which occur in ~50% of patients. Despite successful treatment of the primary tumour, there is currently no effective systemic therapy to treat metastatic disease. UM patient management is dependent upon accurate stratification into high or low risk of metastatic progression. Prognostication of UM patients currently involves the integration of clinical, histological, and genetic features of the tumour obtained by multiplex ligation-dependant probe amplification (MLPA) or microsatellite analysis (MSA), in order to obtain a refined predicted metastatic risk. The aims of this thesis were to examine prognostication performed in Liverpool and improve the way patients are stratified according to metastatic risk. In chapter 2 the genetic characteristics of the rarest form of UM, iris melanoma (IM) was examined. Molecular analysis was performed on archival IM samples collected over a twenty year period and demonstrated that a low-metastatic genotype was most commonly observed; this correlates with the improved survival of this UM subtype in comparison to ciliary body and choroidal melanomas. Chromosomal aberrations often seen in posterior UM are also seen in IM, in both treatment-naïve and irradiated tumours. MSA is a common method of genotyping for small UM with low DNA yields, i.e. fine-needle aspirate biopsy samples (FNAB). In chapter 3, I demonstrated that MSA can accurately determine chromosome 3 copy number in FNAB samples. I found no demonstrable differences in the quality or clarity of the genotypes generated in both treatment-naïve and irradiated UM. By examining metastasis-free survival in these samples, I established that there was no evidence that surgical intervention causes iatrogenic dissemination. Loss of 3q was identified as an indicator of poor prognosis and allelic imbalance as an indicator of good prognosis. In chapter 4, in the largest study to date, I demonstrated that loss of nuclear BAP1 protein expression (nBAP1) was a significant independent prognostic parameter associated with metastatic risk and reduced survival, with a stronger association with poor outcome than monosomy 3 (M3). Additionally I identified two subgroups within M3 UM whereby M3 nBAP1+ cases were associated with a prolonged survival in comparison to nBAP1- UM. To address the clinical need to integrate copy number analysis of chromosomes 1, 3, 6 and 8 and gene mutation status of GNAQ, GNA11, BAP1, SF3B1, EIF1AX, CYSTLR2, PLCB4, TTC28, KTN1, TP53BP1, and CSMD1 in chapter 5, I designed, compared and validated targeted next-generation sequencing (NGS) panels for UM. I demonstrated the superiority of hybrid capture as an enrichment method, and was able to successfully analyse copy number and single nucleotide variants in both fresh and formalin fixed tissue. Mutations in BAP1 were found to be the most important prognostic factor in terms of survival, and gains of 1q were associated with an increased incidence of metastatic death. I identified novel mutations in PLCB4, TTC28, CSMD1, KTN1 and TP53BP1 that require further evaluation. In chapter 6 I investigated protein expression levels of a recently identified molecule RasGRP3 in GNAQ/GNA11 mutant UM cell lines and primary tumours and compared this with BRAF mutant conjunctival melanoma specimens. I demonstrated that RasGRP3 was significantly and selectively overexpressed in response to GNAQ/11 mutations. This demonstrates that activation of the MAP-kinase pathway occurs through a different mechanism to that occurring in conjunctival melanoma. In conclusion, I have examined in detail ways of enhancing our current practice for prognostication in UM. There is a need for BAP1 mutations and nBAP1 negativity to be implemented into routine prognostication. This may be achievable by replacing current MLPA and MSA testing with the targeted UM NGS panel.
Supervisor: Coupland, S. E. ; Kalirai, H. ; Taktak, A. F. G. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral