Epigenetic dysregulation of novel candidate genes in melanoma

  • Hexiao Wang

    Student thesis: Doctoral ThesisDoctor of Philosophy


    Cutaneous melanoma (CM) is an aggressive neoplastic disorder of melanocytes accounting for 4% of skin cancers, but over 75% of skin cancer-related deaths. The outlook for advanced melanoma is typically poor, particularly for patients with distant and/or visceral metastatic disease. Although new melanoma therapies including vemurafenib and ipilimumab confer significant clinical benefit, accumulating evidence suggests that they are less effective when tumour load is high and patient performance status poor, as is often the case in advanced disease. New strategies for the early detection of sub-clinical relapsed and/or metastatic disease are therefore urgently required. Development of such biomarkers, combining sensitivity and specificity, would allow the deployment of these treatments when tumour volume is low thereby maximising anti-melanoma efficacy. Cancer-associated gene promoter methylation is recognized as an important mechanism in tumour initiation and progression. Exploring the epigenetic profiles of melanoma will lead to the generation of new strategies for molecular subtyping, discovery of diagnostic and prognostic biomarkers and development of novel therapeutic targets.

    In the present study, I have identified a panel of promising epigenetic melanoma biomarkers by exploiting both a systematic approach and candidate gene approach. Biomarkers TFPI2 and P4HA3 showed a high frequency and high level of methylation in melanoma cell lines. Significantly increased methylation in melanoma suggests a role for TFPI2 and P4HA3 as epigenetic biomarkers that can potentially predict poor prognosis of melanoma. Furthermore, in my work a highly significant correlation has been established between detection of methylated genomic DNA of TFPI2 and P4HA3 in melanoma patients’ sera and metastatic disease. This implies potential clinical utility for TFPI2 and P4HA3 as serum biomarkers for the detection of metastatic melanoma. In contrast, the presence of NT5E and DUSP2 methylation is likely to be a good prognostic biomarker for both primary melanoma and metastatic disease. NT5E methylation was associated with early stage melanoma but became undetectable when melanoma metastasis progressed to visceral sites. Similarly, the significantly higher methylation level of DUSP2 in primary melanomas than in metastatic disease implies that DUSP2 methylation may also be a biomarker of good prognosis in melanoma.

    In summary, my study proposes that epigenetic dysregulation by CpG island methylation of tumour-related genes TFPI2, P4HA3, NT5E and DUSP2 is important in melanoma development and progression and that detection of methylated DNA from these genes may provide novel biomarkers for predicting tumour progression and patient outcome. Molecular functions of these biomarkers in melanoma should be investigated in future work for the purpose of better understanding the biology of this highly aggressive tumour. Such studies may also lead to new therapeutic strategies for advanced melanoma. For example, antibodies to NT5E show efficacy in animal models of metastatic breast cancer and merit assessment in melanoma. Moreover, detection of circulating methylated genomic DNA from TFPI2 and P4HA3 in patient sera may be useful for predicting prognosis and for the early diagnosis of metastatic disease. Given the emerging evidence that new melanoma therapeutics, particularly ipilimumab, are most effective when used in low volume metastatic disease, such epigenetic biomarkers are likely to be of value in informing the clinical deployment of such agents.
    Date of Award2013
    Original languageEnglish
    SupervisorCharlotte Proby (Supervisor), Tim Crook (Supervisor) & Gareth Inman (Supervisor)

    Cite this