Autoantibodies as prognostic biomarkers for cutaneous melanoma

Author Identifiers

Désirée Sexauer


Date of Award


Degree Type


Degree Name

Master of Medical and Health Science by Research


School of Medical and Health Sciences

First Advisor

Pauline Zaenker

Second Advisor

Elin Gray

Third Advisor

Johnny Lo


Cutaneous melanoma is the most aggressive type of skin cancer that is responsible for more than 80% of skin cancer-related deaths. The incidence of cutaneous melanoma, especially thin ( < 1.0 mm) melanomas, continues to increase nationally and globally. Although thick ( ≥ 4.0 mm) melanomas are correlated with a worse prognosis, thin melanomas account for the majority of melanoma deaths due to the high volume of the disease. Therefore, there is currently an urgent need to identify melanoma-associated prognostic biomarkers for the effective monitoring of patients that are predicted to have an increased risk of tumour progression. This will enable timely therapeutic intervention and ultimately decrease melanoma attributed morbidity and mortality.

The utilisation of autoantibodies as melanoma-associated biomarkers is a promising avenue towards personalised medicine. Autoantibodies are generated by the adaptive immune system in cancer patients towards autologous antigens and may provide biological information of the tumour. Additionally, autoantibodies have desirable biomarker properties such as persistent concentrations and long half-lives due to a limited proteolysis and clearance from the blood.

The overarching aim of this thesis was to identify serum autoantibody biomarkers from blood that are indicative of disease progression. Early-stage cutaneous melanoma patients provided a blood sample at the point of diagnosis of their primary melanoma. Follow-up data (median of 6.2 years) of melanoma progression and patient survival was then received from the Western Australian Clinical Cancer Registry.

This Masters by Research thesis is comprised of 6 chapters: a literature review introducing the research topic on cutaneous melanoma and prognosis (Chapter 1), a comprehensive published literature review discussing the current evidence for the utilisation of tumour-associated autoantibodies in the prognostication of cutaneous melanoma, as well as other cancers (Chapter 2), three research results chapters (Chapters 3-5); and a final chapter with a general discussion of main findings and future perspectives (Chapter 6).

A total of 104 sera were previously screened against a functional high-throughput microarray platform containing 1627 functional proteins to measure IgG autoantibody levels. Microarray autoantibody profiles were used in chapter 3 to 5. For these chapters, follow-up data (median 6.2 years, range 5.5-7.9 years) from the Clinical Cancer Registry was used to group the early-stage cutaneous melanoma cohort into a progression versus non-progression group. In chapter 3, the prognostic utility of tumour protein 53 (p53) autoantibody as a prognostic biomarker was assessed because extensive literature suggests its potential as a prognostic biomarker. Additionally, this chapter presents an extensive description of the methodology and optimisation procedure for bead-based immunoassays, which have been followed in the subsequent chapters. Carboxylated magnetic beads have been identified as the superior bead type compared to non-magnetic beads due to their decreased level of non-specific binding and were thus used for the following chapters. The microarray and bead-based immunoassay results indicated that p53 autoantibodies were found not to be of significant prognostic value as a single prognostic marker. Its prognostic value as part of a larger autoantibody panel remains to be validated.

For chapters 4 to 5, previously obtained microarray data was used to identify potential prognostic autoantibodies by using an analytical approach that takes into account both the frequency and strength of signal of autoantibodies in a high-risk versus a low-risk group. In chapter 4, superoxide dismutase 1 (SOD1) autoantibodies were identified as a potential marker for melanoma progression. The subsequent investigation of SOD1 antibodies via a bead-based immunoassay has shown that SOD1 autoantibodies might be of potential significant value for the prognostication of early-stage cutaneous melanoma patients. There was, however, a degree of discordance identified between the two testing platforms.

For chapter 5, the Nomogram tool, developed by the Melanoma Institute of Australia, was used to investigate whether autoantibodies detected at the point of diagnosis might be valuable in predicting whether a melanoma patient has an increased risk of developing SLN metastasis. The tumour-necrosis factor alpha-induced protein 8 (TNFAIP8) was the top prognostic marker based on the microarray data and was therefore further investigated. The subsequent bead-based immunoassay to validate the finding has shown that TNFAIP8 autoantibodies as a single biomarker has limited value in discriminating predicted high-risk from predicted low-risk patients. There was a weak correlation and a poor level of agreement identified between the two testing platforms.

Finally, chapter 6 provides a general discussion of the studies covered in this thesis. The limitations of each study and suggestions for improvement are comprehensively discussed. Importantly, future avenues for the utilisation of autoantibodies as prognostic biomarkers are highlighted.

Access Note

Access to this thesis is embargoed until 12 July 2024.

Access to this thesis is restricted. Please see the Access Note below for access details.