Hormone replacement therapy and the risk of melanoma in post-menopausal women

STUDY QUESTION: Is hormone replacement therapy (HRT) associated with an increased risk of melanoma skin cancer or prognostic outcomes amongst post-menopausal women? SUMMARY ANSWER: Whilst we found evidence of an association with melanoma risk, the lack of dose-response and associations observed with recent use, localised disease and intravaginal oestrogens suggests this is a non-causal association. WHAT IS KNOWN ALREADY: Evidence on HRT and melanoma risk remains inconclusive, with studies providing conflicting results. Furthermore, evidence on melanoma survival is sparse, with only one previous study reporting protective associations with HRT use, likely attributable to immortal time bias. STUDY DESIGN, SIZE, DURATION: We conducted a nation-wide population-based case-control study and a retrospective cohort study utilising the Danish healthcare registries. Case-control analyses included 8279 women aged 45-85 with a first-ever diagnosis of malignant melanoma between 2000 and 2015, matched by age and calendar time to 165 580 population controls. A cohort of 6575 patients with a diagnosis of primary malignant melanoma between 2000 and 2013 and followed through 2015 was examined to determine if HRT use had an impact on melanoma survival outcomes. PARTICIPANTS/MATERIALS, SETTING, METHODS: Based on prescriptions dispensed since 1995, ever-use of HRT was defined as having filled at least one prescription for HRT prior to the index date. In total, 2629 cases (31.8%) and 47 026 controls (28.4%) used HRT. Conditional logistic regression was used to calculate odds ratios (ORs) for melanoma risk according to HRT use, compared with non-use, adjusting for potential confounders. For cohort analyses, Cox proportional hazards models was used to estimate adjusted hazard ratios (HRs) with 95% confidence intervals (CIs) for second melanoma incidence and all-cause mortality associated with HRT. MAIN RESULTS AND THE ROLE OF CHANCE: High use of HRT was associated with an OR of 1.21 (95% CI 1.13-1.29) for melanoma risk, with no evidence of a dose-response pattern. Results were most pronounced amongst recent high users (OR, 1.28; 95% CI 1.17-1.41), for localised disease (OR, 1.25; 95% CI 1.15-1.36) and for intravaginal oestrogen therapy (OR, 1.38; 95% CI 1.13-1.68). Compared with non-use, there was no evidence of an association for secondary melanoma for post-diagnostic new-use (fully adjusted HR, 1.56; 95% CI 0.64-3.80) or continuous HRT use (fully adjusted HR, 1.26; 95% CI 0.89-1.78). Similar associations were observed for all-cause mortality. LIMITATIONS, REASONS FOR CAUTION: Despite the large sample size and the use of robust population-based registries with almost complete coverage, we lacked information on some important confounders including sun exposure. WIDER IMPLICATIONS OF THE FINDINGS: Whilst we cannot rule out an association between HRT use and melanoma risk, the associations observed are also compatible with increased healthcare utilisation and thus increased melanoma detection amongst HRT users. No association between HRT use and melanoma survival outcomes was observed. This should provide some reassurance to patients and clinicians, particularly concerning the use of HRT in patients with a history of melanoma. STUDY FUNDING/COMPETING INTEREST(S): B.M.H. is funded by a Cancer Research UK Population Research Postdoctoral Fellowship. The funding source had no influence on the design or conduct of this study. A.P. reports participation in research projects funded by Alcon, Almirall, Astellas, Astra-Zeneca, Boehringer-Ingelheim, Servier, Novo Nordisk and LEO Pharma, all with funds paid to the institution where he was employed (no personal fees) and with no relation to the work reported in this article. The other authors have no competing interests to declare. TRIAL REGISTRATION NUMBER: N/A.