Author Identifiers

Aaron Beasley
ORCID: 0000-0001-5727-7463

Date of Award


Degree Type


Degree Name

Doctor of Philosophy


School of Medical & Health Sciences

First Advisor

Associate Professor Elin Gray

Second Advisor

Dr Fred Chen

Third Advisor

Dr Samuel McLenachan

Fourth Advisor

Dr Jason Waithman

Fifth Advisor

Dr Weitao Lin


Uveal melanoma (UM) is the most common intraocular malignancy in adults, affecting about 200 Australians per annum. Unfortunately, up to 50% of patients will develop incurable metastatic disease despite effective treatment of the localised disease. Once metastases are detected, 92% of patients will die within two years. Specific genetic mutations, somatic copy number alterations (SCNAs), and gene expression profiles can accurately stratify patients who will develop metastatic disease. However, determination of prognosis requires an intraocular biopsy which is not always performed outside of centres of excellence. This thesis describes a series of investigations in UM research, including an epidemiological study, a methodological study to improve molecular characterisation of the tumour, and clinical studies to evaluate the clinical validity of circulating tumour cells (CTCs) and circulating tumour DNA (ctDNA) as biomarkers. This thesis consists of eight chapters: a literature review (Chapter One), results chapters (Chapter Two to Seven), and a general discussion of the main findings and future directions (Chapter Eight). Chapter One introduces the history and clinical landscape of UM, while further introducing the topic of liquid biopsy.

Chapter Two describes an investigation of the incidence and mortality of ocular melanoma (OM) from years 1982 to 2015 using data extracted from the Australian Cancer Database. The study demonstrated that the incidence of OM significantly increased from 1982 to 1993, followed by a significant decrease until 2014. The results underscore the fact that the survival of patients with OM has not changed from 1982 to 2015. Notably, we found that residence in Western Australia was a predictor of significantly worse survival.

In Chapter Three we aimed to improve the prognostic yield from low DNA content samples. This is necessary for cases where the biopsy yields are below what is needed for current clinical standards. Whole genome amplification (WGA) combined with low-pass whole genome sequencing (LP-WGS) was used to detect SCNAs associated with UM prognosis. This technique was highly concordant with one of the gold standard methods used for UM prognostication, multiplex ligation-depended probe amplification (MLPA). The results support a strategy of using WGA/LP-WGS a surrogate in cases where the amount of DNA recovered from the biopsy is too low to input into the MLPA assay or similar assays used for UM prognostication.

In Chapters Four to Six, the potential of CTCs to serve as a liquid biopsy in UM was investigated. In Chapter Four, we show that CTCs were readily detectable in patients with primary UM. CTCs harbour SCNAs similar to patient’s primary tumour and could serve as a surrogate for prognostication. Building upon these findings, Chapter Five describes a multimarker approach to improve the capture of CTCs in primary UM by evaluating the expression of melanoma and melanocytic markers on a primary UM tissue microarray (TMA) and UM cell lines. We found that membranous markers, ABCB5, MCAM, MCSP, and gp100 were suitable for cell isolation, and that the intracellular markers MART1, gp100, and S100β were suitable for cell identification. Using this panel of markers, the capture rate of CTCs was improved from 69% to 86% per 8 mL of whole blood. In Chapter Six, the genetic characteristics of CTCs isolated from patients with localised UM are described. Of the 9 patients analysed, only three had SCNAs associated with known markers of prognosis. The remainder harboured no SCNAs – further investigation is needed to determine whether this is a true biological phenomenon or that further refinement of the CTC isolation methodology is required to avoid DNA contamination.

Chapter Four also included a study of ctDNA in UM patients. These results indicated that ctDNA is generally undetectable in primary disease but is highly detectable in metastatic disease. Additionally, a case study is presented in which ctDNA detected metastases earlier than radiological scans. Following this theme, Chapter Seven further describes a prospective study evaluating ctDNA as a method for early detection of metastases. We found that of those who developed metastases, 50% had detectable ctDNA at the time of or prior to radiological detection and ctDNA increased with disease progression. Those who did not have detectable ctDNA had longer overall survival. Furthermore, we found in a different cohort of treatment naïve metastatic patients, ctDNA reduced after treatment with combination immunotherapy.

Lastly, Chapter Eight highlights how the results of these studies fit within the wider clinical and scientific literature. In particular, we highlight examples of how liquid biopsy can supplement the current standard of care. The limitations of each study and suggestions for improvements are comprehensively discussed. The chapter concludes by proposing future avenues for studies that can further demonstrate the clinical utility of liquid biopsies for UM.

Access Note

Chapters 1, 2, 3, 5 and 6 are not available in this version of the thesis.

Available for download on Saturday, May 18, 2024