The influence of pregnancy, parity, and sex on melanoma prognosis

Byrom, Lisa (2015). The influence of pregnancy, parity, and sex on melanoma prognosis MPhil Thesis, UQ Diamantina Institute, The University of Queensland. doi:10.14264/uql.2015.872

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s33638120_mphil_submission.pdf Thesis (open access) application/pdf 4.54MB 1

Author Byrom, Lisa
Thesis Title The influence of pregnancy, parity, and sex on melanoma prognosis
School, Centre or Institute UQ Diamantina Institute
Institution The University of Queensland
DOI 10.14264/uql.2015.872
Publication date 2015-08-31
Thesis type MPhil Thesis
Supervisor Kiarash Khosrotehrani
Adele Green
Total pages 112
Language eng
Subjects 1103 Clinical Sciences
1112 Oncology and Carcinogenesis
1117 Public Health and Health Services
Formatted abstract

The influence of pregnancy, parity, and sex on melanoma outcome is an important question as melanoma is one of the few cancers that affect women during their childbearing years and is one of the main cancers occurring during gestation. Since the 1950s, there have been concerns that pregnancy-associated hormones,immunological suppression, and increased lymphangiogenesis might lead to negative outcomes for pregnancy associated melanomas. However, to date there has not been conclusive evidence. Additionally, the timing of a pregnancy is of particular importance to melanoma survivors of reproductive age. Despite this proposed detrimental effect of pregnancy on melanoma outcome overall females have demonstrated superior survival rates over males. This advantage cannot be explained solely by behavioural factors like earlier diagnosis and management. The role of reproductive factors, such as parity and sex hormones, in relation to sex differences in melanoma survival has not been fully investigated.


My study objectives were firstly to evaluate the possible effect of a coinciding pregnancy on cause-specific death and recurrence of melanomas; secondly to evaluate the risk of cause-specific death and recurrence in women who have a pregnancy following their melanoma diagnosis; thirdly to evaluate the effect of parity versus nulliparity before melanoma diagnosis on cause-specific death; and lastly to evaluate the risk of cause-specific death in women with melanomas compared to men according to age group and stage of disease.


To address my first and second aims I conducted a systematic review and meta-analysis according to PRISMA guidelines. Five databases (Cochrane Database, MEDLINE, EMBASE, CINAHL, and PUBMED) were searched firstly for studies assessing the effect of a coinciding pregnancy on melanoma prognosis and secondly for studies assessing the effect of a subsequent pregnancy on the prognosis of a previous melanoma. Individual study effect estimates were pooled using the weighted average method. To address my third objective, melanoma survival rates were analysed after a 30-year follow-up in a small Queensland cohort of 232 melanoma patients diagnosed in 1979-1980 according to sex and parity at time of diagnosis. To address my fourth objective, registry-based studies were conducted to compare melanoma survival rates between males and females. This was done across two population cohorts of melanoma diagnoses between 1995 and 2008 in Australia (Queensland Cancer Registry) and the United States (the Surveillance, Epidemiology, and End Results Program) analysing outcomes at 10-year follow-up. The patients were grouped according to sex, melanoma stage, and patient age at diagnosis. Flexible parametric survival models compared survival between groups.


Meta-analysis of 4 eligible studies demonstrated that a female diagnosed with a melanoma during pregnancy has a 56% increased risk of death compared with those diagnosed outside of pregnancy, though this was not fully supported by the systematic review. The second systematic review and meta-analysis found no firm grounds for a melanoma survivor to delay pregnancy. The pooled estimate for risk of melanoma death was 0.81 (95% CI 0.60-1.09). However, given the wide confidence interval around the pooled estimate for risk of melanoma death, ill effect of subsequent pregnancy cannot be entirely ruled out. The small Queensland cohort study involved 232 patients grouped according to parity. This exploratory study lacked statistical power and matching was not adequate so full evaluation was not feasible. In the analysis of two large population cohorts in Australia and USA it was demonstrated that the female survival advantage was evident across nearly all melanoma stages, including stage 1 melanomas. This advantage was dependent on patient age at diagnosis. These findings support overall the importance of host tumour interactions in determining melanoma prognosis.


Across multiple studies, it is shown that sex and reproductive history can affect melanoma outcome. A pregnancy coinciding with a melanoma diagnosis in a woman has a detrimental influence on survival. Despite this, overall women have a survival advantage for melanoma compared to men. The underlying mechanisms are still being investigated.
Keyword Melanoma
Systematic review

Document type: Thesis
Collections: UQ Theses (RHD) - Official
UQ Theses (RHD) - Open Access
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Mon, 17 Aug 2015, 22:16:06 EST by Lisa Byrom on behalf of Scholarly Communication and Digitisation Service