Author Identifier

Muhammad Khattak

Date of Award


Document Type

Thesis - ECU Access Only


Edith Cowan University

Degree Name

Doctor of Philosophy


School of Medical and Health Sciences

First Supervisor

Elin Gray

Second Supervisor

Leslie Calapre


Advanced melanoma is an aggressive cancer with a poor prognosis once disseminated to other organs. However, survival outcomes have improved recently with the introduction of new immune modulating agents, more commonly known as immune checkpoint inhibitors (ICI), with some patients deriving durable long-term benefit. The latest data indicate that the 5-year overall survival for advanced melanoma patients is now approaching 50%. These agents are very expensive and can lead to substantial toxicity that can place a huge economic burden on the health system. Moreover, different combinations of these drugs are becoming available for those less likely to respond to single-agent regimens. However, biomarkers identifying this group of patients prior to treatment, are lacking.

Tissue biopsy is the current gold standard for cancer diagnosis and for determining prognosis. However, this can be invasive, uncomfortable to the patient and lead to complications. Liquid biopsies through blood sampling provide us a useful alternative or adjunct to tissue biopsy being non-invasive and more importantly enabling analysis of the tumour at the time that therapeutic decisions are to be made. Moreover, other biomarkers representative of the immunological status of the patients could provide additional information in relation to the response to ICI therapy.

Here we explore three distinct blood-based biomarkers in patients with advanced (unresectable stage III/stage IV) melanoma treated with ICI therapy, as anti-PD1 monotherapy (pembrolizumab/nivolumab) or combination anti-CTLA-4 and anti-PD-1 (ipilimumab/nivolumab). The overarching aim of these studies is to identify and develop blood based molecular markers that can assist clinicians in decision making in their routine clinics. These markers might help oncologists in identifying patients most suitable for treatment with ICI therapy and save toxicity to those that are less likely to benefit from such drugs. Additionally, they might identify patients best suited for treatment with combination ICI therapy that is much more efficacious but much more toxic.

Chapter 2 describes analysis of the circulating tumour cells (CTCs) isolated through liquid biopsies that can provide useful molecular information regarding the tumours. In contrast to tumour tissue-based PD-L1 expression, melanoma CTCs provide an alternative tumour derived material to evaluate PD-L1 expression. This pilot study indicated that in patients with detectable CTCs, expression of PD-L1 in CTCs was found in 9/23 melanoma patients prior to therapy initiation and was significantly associated with response to pembrolizumab. The 12-month PFS rates were 76% vs. 22% in the PDL-1+ vs. PDL-1- CTCs groups (P = 0.012), respectively. These results indicate that PD-L1 expression on CTCs may be predictive of response to pembrolizumab in patients with advanced melanoma.

In Chapter 3, the effect of HLA genotypes on survival in advanced melanoma patients undergoing treatment with pembrolizumab was investigated. In contrast with previous findings from a large pool study of different cancers and immunotherapy modalities, in our cohort, we did not find any impact of HLA-I or II homozygosity on progression free survival or overall survival in terms of response to pembrolizumab. Notably, melanoma patients with HLA-B27 supertype had worse prognosis compared to non-carriers and may predict poor response to single agent ICI therapy. This was also confirmed in the multivariate analysis and further validated in an independent analysis of 114 cases from the MSK-IMPACT cohort. Our observation requires further validation in a larger independent prospective study, as it may identify melanoma patients with certain HLA supertypes associated with poorer OS that could be offered treatment with combination ICI therapy to improve their long-term outcomes as they might not derive significant benefit from single agent anti-PD-1 therapy.

In Chapter 4, the predictive value of serum vascular endothelial growth factor (VEGF) in melanoma patients undergoing treatment with single agent anti-PD-1 therapy (pembrolizumab) or anti-CTLA-4 (ipilimumab) or combination immunotherapy with anti-PD1/anti-CTLA4 agents (ipilimumab and nivolumab) was evaluated. The predictive value of serum VEGF level has been previously evaluated in patients with advanced melanoma treated with interleukin 2 (IL-2). In our study, serum VEGF level was significantly higher in the non-responders (82.15 ng/mL) vs responders (60.40 ng/mL) in the ipilimumab monotherapy cohort in line with previous reports, but such difference was not seen between non-responders and responders in the pembrolizumab and ipilimumab/nivolumab combination groups. Combination anti-PD-1 and anti-VEGF agents have demonstrated good efficacy across a range of tumour types including melanoma, renal cancer, and endometrial cancer. Therefore, a blood-based biomarker evaluating the VEGF pathway should be investigated in the future due to its potential to identify the subset of patients who might be better suited to dual therapy rather single agent ICI therapy.

The results presented are from a discovery phase research and need to be validated in large clinical studies. Chapter 5 constitutes a critical overview of the findings and their potential clinical applications. Overall, a more personalised approach to treatment with ICI therapy might better stratify melanoma patients into certain prognostic groups and save the significant long-term and sometimes irreversible toxicity to others especially if they are less likely to respond to these agents.