Work

Association of IRF 4 single‐nucleotide polymorphism rs12203592 with melanoma‐specific survival

Year: 2020

Type: article

Abstract: Dear Editor, We have previously shown that the functional melanoma risk variant rs12203592 in the IRF4 gene (interferon regulatory factor‐4) was associated with increased Breslow thickness, an important prognostic indicator for melanoma.1 In line with this, a recent study found evidence for an association between the minor T allele of this single‐nucleotide polymorphism (SNP) and melanoma‐specific survival in patients from two European hospitals.2 We therefore sought to extend our previous work and replicate these hospital‐based findings by investigating the association of rs12203592 with melanoma‐specific survival, overall and by anatomical site. We used 3303 melanoma cases from individuals with European ancestry from two large and representative population‐based studies: the Genes, Environment and Melanoma (GEM) study and the Western Australian Melanoma Health Study (WAMHS), both of which have been described in detail previously.3,4 The GEM cases (n = 2298) were diagnosed with an incident, first primary melanoma between 2000 and 2001 in Australia, Canada, Italy or the U.S.A. (mean follow‐up 7·4 years). DNA was extracted from buccal brushes and genotyping was conducted using the MassArray iPLEX chemistry and platform (Agena Bioscience, San Diego, CA, U.S.A.). Histopathology data were confirmed by pathologist's review of haematoxylin–eosin‐stained tumour sections. The WAMHS cases (n = 1005) were diagnosed in Western Australia between 2006 and 2009, also with an incident, first primary melanoma (mean follow‐up 8·8 years). DNA was obtained from peripheral blood samples and genotyping was performed using the Illumina OmniExpressExome v1 chip (Illumina, San Diego, CA, U.S.A.). Histopathology data were obtained from pathology reports. Cox proportional hazards regression analyses estimated the hazard ratios (HRs) and 95% confidence intervals (CIs) for the association of IRF4 rs12203592 (per minor T allele) with melanoma‐specific survival. Analyses were conducted in each separate study sample and then in the combined GEM–WAMHS sample, with baseline adjustments made for age at diagnosis, sex and study centre. A second model also adjusted for Breslow thickness, which was log transformed to normalize the left‐skewed distribution. Analyses were further stratified by anatomical site of tumour occurrence, categorized as melanoma of the head and neck, trunk, or extremities. In the combined GEM–WAMHS sample, a total of 208 melanoma‐specific deaths had occurred. The male to female ratio was 1·05 : 1 and the mean ± SD age was 56·0 ± 15·0 years. The mean Breslow thickness was 1·22 ± 1·87 mm and the majority of individuals had melanoma of either the upper and lower limbs (42·3%) or the trunk (42·1%). In each separate study, a significantly increased risk of melanoma‐specific death was observed for each copy of the T allele in the baseline‐adjusted model (Table 1). In the combined sample, the HR estimate also indicated that rs12203592*T was significantly associated with increased melanoma‐specific death (HR 1·35, 95% CI 1·09–1·67; P = 0·006). Association between IRF4 single‐nucleotide polymorphism (SNP) rs12203592 and melanoma‐specific survival, in each individual study and in a pooled sample of participants with single primary melanoma from the Genes, Environment and Melanoma (GEM) study and the Western Australian Melanoma Health Study (WAMHS) HR, hazard ratio; CI, confidence interval. an alive number of patients still alive; n deaths, number of melanoma‐specific deaths. bAdjusted for baseline variables: age at diagnosis (years, continuous), sex and study centre. cAdjusted for baseline variables plus logged Breslow thickness. dCox proportional hazards models estimated the per‐allele HRs, 95% CIs and trend P‐value. P‐values < 0.05 are significant. Association between IRF4 single‐nucleotide polymorphism (SNP) rs12203592 and melanoma‐specific survival, in each individual study and in a pooled sample of participants with single primary melanoma from the Genes, Environment and Melanoma (GEM) study and the Western Australian Melanoma Health Study (WAMHS) HR, hazard ratio; CI, confidence interval. an alive number of patients still alive; n deaths, number of melanoma‐specific deaths. bAdjusted for baseline variables: age at diagnosis (years, continuous), sex and study centre. cAdjusted for baseline variables plus logged Breslow thickness. dCox proportional hazards models estimated the per‐allele HRs, 95% CIs and trend P‐value. P‐values < 0.05 are significant. After adjustment for Breslow thickness, no associations remained significant and the HRs were attenuated (combined sample: HR 1·12, 95% CI 0·87–1·45; P = 0·38). Further, no significant interactions for SNP‐by‐anatomical site were observed in any of the models. When stratified by anatomical site in the pooled cohort, the direction of association was the same for each site. Subdistribution HRs from the Fine–Gray5 model for competing risk also showed that rs12203592*T was significantly associated with melanoma death in the GEM–WAMHS sample, prior to adjustment for Breslow thickness (subdistribution HR 1·38, 95% CI 1·10–1·72; P = 0·02). Following adjustment, the association was just above the threshold for statistical significance (subdistribution HR 1·26, 95% CI 0·99–1·59; P = 0·057). A mediation analysis found that 70·2% of the observed association between rs12203592*T and melanoma‐specific survival was mediated through Breslow thickness, with significant natural indirect effects identified (HR 1·13, 95% CI 1·06–1·20; P < 0·001). Our results suggest that IRF4 rs12203592 is associated with poorer melanoma‐specific survival, largely mediated through Breslow thickness, reinforcing our previous findings related to rs12203592*T and tumour thickness.1 This variant's adverse effect on survival may occur through its functional effect on IRF4 expression. The SNP is known to modulate IRF4 enhancer‐mediated transcriptional regulation in melanocytes, and it may play a similar role in immune cells.6 Increased IRF4 expression in immune cells and the subsequent activation of the TERT promoter has been shown to increase telomerase activity, which in turn has been associated with increased Breslow thickness and tumour progression.7,8 In lymphocytes, rs12203592*T results in higher IRF4 expression, and it has been suggested that this may enable suppressive regulatory T cells to inhibit the immune response, enabling an acceleration in tumour growth and progression.2 In conclusion, IRF4 rs12203592 appears to play a role in melanoma prognosis, which is seemingly mediated by the SNP's association with Breslow thickness at diagnosis. The effect on survival does not appear to be site specific. Further studies are now required to confirm the pathway and mechanism by which this SNP affects survival, and to determine the potential clinical and prognostic utility of these findings. the authors gratefully acknowledge all participants for their time and involvement, as well as the Western Australian DNA Bank at The University of Western Australia, the Western Australian Cancer Registry, and the WAMHS and GEM study teams for their support. The members of the GEM study group are listed in Appendix S1 (see Supporting Information). E‐mail: [email protected] Funding sources: this work was supported by the National Health and Medical Research Council (Early Career Fellowship ID: 1121242 to S.V.W. and Career Development Fellowship ID: 1147843 to A.E.C.), the Cancer Institute New South Wales (Career Development Fellowship ID: 15/CDF/1‐14 to A.E.C.) and the National Cancer Institute (P01CA206980 to N.E.T. and M.B., R01CA112243 to N.E.T., U01CA83180 and R01CA112524 to M.B., R01CA098438 to C.B.B., R03CA125829 and R03CA173806 to I.O., and P30CA008748 to Memorial Sloan Kettering). Conflicts of interest: none to declare. Appendix S1 Members of the GEM study group.

Source: British Journal of Dermatology

DOI: 10.1111/bjd.18881