Development and validation of a melanoma risk score based on pooled data from 16 case-control studies

Inter-Rater Agreement in Assessing Risk of Bias in Melanoma Prediction Studies Using the Prediction Model Risk of Bias Assessment Tool (PROBAST): Results from a Controlled Experiment on the Effect of Specific Rater Training

Assessing the risk of bias (ROB) of studies is an important part of the conduct of systematic reviews and meta-analyses in clinical medicine. Among the many existing ROB tools, the Prediction Model Risk of Bias Assessment Tool (PROBAST) is a rather new instrument specifically designed to assess the ROB of prediction studies. In our study we analyzed the inter-rater reliability (IRR) of PROBAST and the effect of specialized training on the IRR. Six raters independently assessed the risk of bias (ROB) of all melanoma risk prediction studies published until 2021 (n = 42) using the PROBAST instrument. The raters evaluated the ROB of the first 20 studies without any guidance other than the published PROBAST literature. The remaining 22 studies were assessed after receiving customized training and guidance. Gwet's AC₁ was used as the primary measure to quantify the pairwise and multi-rater IRR. Depending on the PROBAST domain, results before training showed a slight to moderate IRR (multi-rater AC₁ ranging from 0.071 to 0.535). After training, the multi-rater AC₁ ranged from 0.294 to 0.780 with a significant improvement for the overall ROB rating and two of the four domains. The largest net gain was achieved in the overall ROB rating (difference in multi-rater AC₁: 0.405, 95%-CI 0.149-0.630). In conclusion, without targeted guidance, the IRR of PROBAST is low, questioning its use as an appropriate ROB instrument for prediction studies. Intensive training and guidance manuals with context-specific decision rules are needed to correctly apply and interpret the PROBAST instrument and to ensure consistency of ROB ratings.

Using the Prediction Model Risk of Bias Assessment Tool (PROBAST) to Evaluate Melanoma Prediction Studies

Simple Summary

The rising incidence of cutaneous melanoma over recent decades, combined with a general interest in cancer risk prediction, has led to a high number of published melanoma risk prediction models. The aim of our work was to assess the validity of these models in order to discuss the current state of knowledge about how to predict incident cutaneous melanoma. To assess the risk of bias, we used a standardized procedure based on PROBAST (Prediction model Risk Of Bias ASsessment Tool). Only one of the 42 studies identified was rated as having a low risk of bias. However, it was encouraging to observe a recent reduction of problematic statistical methods used in the analyses. Nevertheless, the evidence base of high-quality studies that can be used to draw conclusions on the prediction of incident cutaneous melanoma is currently much weaker than the high number of studies on this topic would suggest.

Abstract

Rising incidences of cutaneous melanoma have fueled the development of statistical models that predict individual melanoma risk. Our aim was to assess the validity of published prediction models for incident cutaneous melanoma using a standardized procedure based on PROBAST (Prediction model Risk Of Bias ASsessment Tool). We included studies that were identified by a recent systematic review and updated the literature search to ensure that our PROBAST rating included all relevant studies. Six reviewers assessed the risk of bias (ROB) for each study using the published “PROBAST Assessment Form” that consists of four domains and an overall ROB rating. We further examined a temporal effect regarding changes in overall and domain-specific ROB rating distributions. Altogether, 42 studies were assessed, of which the vast majority (n = 34; 81%) was rated as having high ROB. Only one study was judged as having low ROB. The main reasons for high ROB ratings were the use of hospital controls in case-control studies and the omission of any validation of prediction models. However, our temporal analysis results showed a significant reduction in the number of studies with high ROB for the domain “analysis”. Nevertheless, the evidence base of high-quality studies that can be used to draw conclusions on the prediction of incident cutaneous melanoma is currently much weaker than the high number of studies on this topic would suggest.

Genome-Wide Association Study Suggests the Variant rs7551288*A within the DHCR24 Gene Is Associated with Poor Overall Survival in Melanoma Patients

Simple Summary

The aim of this work was to investigate prognostic genetic factors in melanoma patients. Phenotypic and disease data as well as biomaterial were collected after informed consent from patients followed up in a Skin Cancer Center of a University clinic. Genome-wide analysis (GWAS) was performed with survival data of 556 melanoma patients and genetic data including more than 300,000 common polymorphisms. The SNP rs7551288 reached suggestive genome-wide significance (p = 2 × 10⁻⁶). This intronic variant of the DHCR24 gene is involved in the cholesterol synthesis pathway. Further analyses and a literature review suggest an important role of this locus for the clinical course of disease in melanoma patients.

Abstract

Melanoma incidence rates are high among individuals with fair skin and multiple naevi. Established prognostic factors are tumour specific, and less is known about prognostic host factors. A total of 556 stage I to stage IV melanoma patients from Germany with phenotypic and disease-specific data were analysed; 64 of these patients died of melanoma after a median follow-up time of 8 years. Germline DNA was assessed by the HumanCoreExome BeadChip and data of 356,384 common polymorphisms distributed over all 23 chromosomes were used for a genome-wide analysis. A suggestive genome-wide significant association of the intronic allele rs7551288*A with diminished melanoma-specific survival was detected (p = 2 × 10⁻⁶). The frequency of rs7551288*A was 0.43 and was not associated with melanoma risk, hair and eye colour, tanning and total naevus count. Cox regression multivariate analyses revealed a 5.31-fold increased risk of melanoma-specific death for patients with the rs7551288 A/A genotype, independent of tumour thickness, ulceration and stage of disease at diagnoses. The variant rs7551288 belongs to the DHCR24 gene, which encodes Seladin-1, an enzyme involved in the biosynthesis of cholesterol. Further investigations are needed to confirm this genetic variant as a novel prognostic biomarker and to explore whether specific treatment strategies for melanoma patients might be derived from it.

Reporting Quality of Studies Developing and Validating Melanoma Prediction Models: An Assessment Based on the TRIPOD Statement

Transparent and accurate reporting is essential to evaluate the validity and applicability of risk prediction models. Our aim was to evaluate the reporting quality of studies developing and validating risk prediction models for melanoma according to the TRIPOD (Transparent Reporting of a multivariate prediction model for Individual Prognosis Or Diagnosis) checklist. We included studies that were identified by a recent systematic review and updated the literature search to ensure that our TRIPOD rating included all relevant studies. Six reviewers assessed compliance with all 37 TRIPOD components for each study using the published “TRIPOD Adherence Assessment Form”. We further examined a potential temporal effect of the reporting quality. Altogether 42 studies were assessed including 35 studies reporting the development of a prediction model and seven studies reporting both development and validation. The median adherence to TRIPOD was 57% (range 29% to 78%). Study components that were least likely to be fully reported were related to model specification, title and abstract. Although the reporting quality has slightly increased over the past 35 years, there is still much room for improvement. Adherence to reporting guidelines such as TRIPOD in the publication of study results must be adopted as a matter of course to achieve a sufficient level of reporting quality necessary to foster the use of the prediction models in applications.

Risk Prediction Models for Melanoma: A Systematic Review on the Heterogeneity in Model Development and Validation

The rising incidence of cutaneous melanoma over the past few decades has prompted substantial efforts to develop risk prediction models identifying people at high risk of developing melanoma to facilitate targeted screening programs. We review these models, regarding study characteristics, differences in risk factor selection and assessment, evaluation, and validation methods. Our systematic literature search revealed 40 studies comprising 46 different risk prediction models eligible for the review. Altogether, 35 different risk factors were part of the models with nevi being the most common one (n = 35, 78%); little consistency in other risk factors was observed. Results of an internal validation were reported for less than half of the studies (n = 18, 45%), and only 6 performed external validation. In terms of model performance, 29 studies assessed the discriminative ability of their models; other performance measures, e.g., regarding calibration or clinical usefulness, were rarely reported. Due to the substantial heterogeneity in risk factor selection and assessment as well as methodologic aspects of model development, direct comparisons between models are hardly possible. Uniform methodologic standards for the development and validation of risk prediction models for melanoma and reporting standards for the accompanying publications are necessary and need to be obligatory for that reason.

The interplay of sun damage and genetic risk in Australian multiple and single primary melanoma cases and controls

BACKGROUND

Skin phenotype, host genotype and ultraviolet (UV) damage play a role in the development of melanoma.

OBJECTIVES

To ascertain whether the level of UV damage at the site of melanomas was associated with genetic polymorphisms.

METHODS

Deep phenotyping was performed on 1244 individuals; 281 with multiple primary melanomas (MPMs), 304 with single primary melanoma (SPM) and 659 convenience controls. Genotype data was generated using the Illumina CoreExome microarray platform, assaying over 500 000 single-nucleotide polymorphisms. A subset of variants were combined to assess a polygenic risk score (PRS) for melanoma.

RESULTS

Most MPM cases were diagnosed in patients aged > 40 years, in sites with visible chronic UV damage. Women and those diagnosed at age ≤ 40 years were less likely to have perilesional UV damage. Patients with MPM had higher frequencies of MITF E318K, MC1R R-alleles and the ASIP risk haplotype. Individuals who had melanoma in a visibly UV-damaged site were more likely to carry MC1R rs75570604 [odds ratio (OR) 2·5], 9q31.2 rs10816595 (OR 1·4) and MTAP rs869329 (OR 1·4). These same alleles were more common in patients with MPM who were diagnosed at age ≤ 40 years. The mean PRS was significantly higher in MPM than in SPM and controls. Naevus count was comparable in early-onset MPM cases and those diagnosed at age > 40 years.

CONCLUSIONS

Our cohort demonstrated higher frequencies of previously reported alleles associated with melanoma. MPM melanomas more commonly occur in UV-damaged areas, and these individuals are more likely to carry MC1R red hair colour alleles. Awareness of the interplay of genetic vulnerability with UV damage can stratify risk and guide recommendations for melanoma screening. What's already known about this topic? Skin phenotype, host genotype and ultraviolet (UV) damage all play a role in melanoma development. One of the main risk factors is a personal history of melanoma; second and subsequent primary melanomas account for over 20% of all melanomas registered in Queensland. Multiple loci are associated with melanoma risk, including many low-penetrance loci, which may have a cumulatively significant risk. Population-wide screening programmes for melanoma are not yet economically viable. What does this study add? Patients diagnosed with melanoma at age ≤ 40 years were more likely than older patients to have melanomas in non-UV-damaged sites. Patients with multiple melanomas had higher frequencies of MITF E318K, MC1R R-alleles, and the ASIP extended risk haplotype than patients with single melanoma. CDKN2A, MC1R and MTAP variants were more frequent in patients who developed melanomas at a younger age, but also in those whose melanomas were all on visibly UV-damaged sites. What is the translational message? Incorporating these genetic findings into the known risk factors of skin phenotype and visible UV damage may allow for a more customized and economically feasible approach to early detection of melanoma, particularly in younger patients. Plain language summary available online.

Genetic Associations with Indoor Tanning Addiction among non-Hispanic White Young Adult Women

BACKGROUND

Some young people may become addicted to indoor tanning in a manner similar to other forms of addiction, but research on genetic associations with indoor tanning addiction remains limited.

PURPOSE

To examine if liabilities in genetic addiction reward pathways and psychiatric comorbidity influence the risk of indoor tanning addiction.

METHODS

This was a cross-sectional study with a community sample of 292 non-Hispanic white young adult women aged 18-30 years who reported indoor tanning in the past year. Self-report measures included indoor tanning frequency, appearance orientation, depressive symptoms, and two screeners of tanning addiction. DNA samples were analyzed for 34 single nucleotide polymorphisms (SNPs) in candidate genes in addiction reward pathways.

RESULTS

No SNPs were significantly associated with tanning addiction in univariate analyses after multiplicity adjustment. In multivariable analyses adjusting for indoor tanning frequency, appearance orientation, and depressive symptoms, variant genotypes (CC or CT) in two DRD2 dopamine receptor gene SNPs were associated with increased odds of indoor tanning addiction (rs4436578, odds ratio [OR]: 2.30, 95% confidence interval [CI]: 1.11-4.77; rs4648318, OR: 1.95, 95% CI: 1.02-3.72). Variant SNP genotypes interacted with depressive symptoms to increase the risk of indoor tanning addiction: OR: 10.79, 95% CI: 3.25, 35.80, OR: 13.60, 95% CI: 4.13, 44.78, respectively.

CONCLUSIONS

This study provides preliminary evidence that DRD2 dopamine receptor gene SNPs are associated with indoor tanning addiction and young women with variant genotypes and elevated depressive symptoms may be at higher risk. These preliminary results support a reward-based model for indoor tanning addiction and warrant further investigation.

Melanoma signature in Brazil: epidemiology, incidence, mortality, and trend lessons from a continental mixed population country in the past 15 years

The current research aimed to understand melanoma epidemiology in Brazil and to evaluate temporal trends in incidence and mortality. The data came from Brazilian Hospital Cancer Registries, Population Based Cancer Registries, and the National Mortality Information System from 2000 to 2014. Descriptive statistics were used for epidemiological and clinical characteristics. To describe trends in change in incidence and mortality rates, the Average Annual Percentage Change (AAPC) was calculated. Between 2000 and 2013, in men, the median incidence rate rose from 2.52 to 4.84, with an AAPC of +21.5% [95% confidence interval (CI): 15.4-28] and in women from 1.93 to 3.22 per 100 000, with an AAPC of +13.9% (95% CI: 8.1-20). Regarding mortality, between 2000 and 2014, the rates went from 0.85 to 0.9 per 100 000 for men (AAPC=+0.8, 95% CI: 0.4-1.1) and from 0.56 for 0.53 per 100 000 for women (AAPC=-0.1, 95% CI: -0.2 to 0). From the database, a total of 28 624 patients with melanoma were included. Most of the patients were females (51.9%), White (75%) and with stage I or II (53.2%). Sex, ethnicity, education level, geographical area of the cancer center, topography, histology, time between diagnosis and treatment, and early death were significantly associated with distant metastases. Brazil is a large country with a very young population and a low rate of melanoma incidence and prevalence that should increase over the years. Understanding the trends attributed to melanoma is important for behavioral counseling interventions that focus on promoting skin cancer prevention.

Melanoma risk stratification of individuals with a high-risk naevus phenotype - A pilot study

BACKGROUND/OBJECTIVES

High a naevus counts and atypical naevi are risk factors for cutaneous melanoma. However, many individuals with a high-risk naevus phenotype do not develop melanoma. In this study, we describe the clinical and dermoscopic attributes of naevi associated with melanoma in a high-risk naevus phenotype population.

METHODS

This single-centre, hospital-based case-control study included 54 prospectively enrolled adult patients ≥18 years old with a high-risk naevus phenotype (18 cases with a history of melanoma and 36 age- and gender-matched controls without a history of melanoma). We analysed clinical and dermoscopic images of the 20 largest naevi for each participant.

RESULTS

Cases had a higher mean age than controls (48.2 vs. 39.1 years, P = 0.007) but there was no difference in the male-to-female ratio between groups. Nearly, all participants (97%) were Fitzpatrick skin type II or III. Naevi in cases were more likely to be truncal, (72.6% vs. 53.6%, P = 0.01), particularly anterior truncal, (29.2% vs. 14.4%, P < 0.001) and larger than 8 mm (17.4% vs. 7.8%%, P = 0.01) compared to controls. CASH score of naevi did not differ between groups. Naevi in cases were more likely to have a multicomponent dermoscopic pattern than in controls (18.4% vs. 12.6%, P = 0.02).

CONCLUSION

Larger naevi, truncal naevi, and naevi, with a multicomponent dermoscopic pattern may be risk factors for melanoma among individuals with a high-risk naevus phenotype. Further studies are needed to validate these findings.

Combining common genetic variants and non-genetic risk factors to predict risk of cutaneous melanoma

Melanoma heritability is among the highest for cancer and single nucleotide polymorphisms (SNPs) contribute to it. To date, only SNPs that reached statistical significance in genome-wide association studies or few candidate SNPs have been included in melanoma risk prediction models. We compared four approaches for building polygenic risk scores (PRS) using 12 874 melanoma cases and 23 203 controls from Melanoma Meta-Analysis Consortium as a training set, and newly genotyped 3102 cases and 2301 controls from the MelaNostrum consortium for validation. We estimated adjusted odds ratios (ORs) for melanoma risk using traditional melanoma risk factors and the PRS with the largest area under the receiver operator characteristics curve (AUC). We estimated absolute risks combining the PRS and other risk factors, with age- and sex-specific melanoma incidence and competing mortality rates from Italy as an example. The best PRS, including 204 SNPs (AUC = 64.4%; 95% confidence interval (CI) = 63-65.8%), developed using winner's curse estimate corrections, had a per-quintile OR = 1.35 (95% CI = 1.30-1.41), corresponding to a 3.33-fold increase comparing the 5th to the 1st PRS quintile. The AUC improvement by adding the PRS was up to 7%, depending on adjusted factors and country. The 20-year absolute risk estimates based on the PRS, nevus count and pigmentation characteristics for a 60-year-old Italian man ranged from 0.5 to 11.8% (relative risk  = 26.34), indicating good separation.

CM-Score: a validated scoring system to predict CDKN2A germline mutations in melanoma families from Northern Europe

BACKGROUND

Several factors have been reported that influence the probability of a germline CDKN2A mutation in a melanoma family. Our goal was to create a scoring system to estimate this probability, based on a set of clinical features present in the patient and his or her family.

METHODS

Five clinical features and their association with CDKN2A mutations were investigated in a training cohort of 1227 Dutch melanoma families (13.7% with CDKN2A mutation) using multivariate logistic regression. Predefined features included number of family members with melanoma and with multiple primary melanomas, median age at diagnosis and presence of pancreatic cancer or upper airway cancer in a family member. Based on these five features, a scoring system (CDKN2A Mutation(CM)-Score) was developed and subsequently validated in a combined Swedish and Dutch familial melanoma cohort (n=421 families; 9.0% with CDKN2A mutation).

RESULTS

All five features were significantly associated (p<0.05) with a CDKN2A mutation. At a CM-Score of 16 out of 49 possible points, the threshold of 10% mutation probability is approximated (9.9%; 95% CI 9.8 to 10.1). This probability further increased to >90% for families with ≥36 points. A CM-Score under 16 points was associated with a low mutation probability (≤4%). CM-Score performed well in both the training cohort (area under the curve (AUC) 0.89; 95% CI 0.86 to 0.92) and the external validation cohort (AUC 0.94; 95% CI 0.90 to 0.98).

CONCLUSION

We developed a practical scoring system to predict CDKN2A mutation status among melanoma-prone families. We suggest that CDKN2A analysis should be recommended to families with a CM-Score of ≥16 points.

Indoor Tanning Dependence in Young Adult Women

Background: There is mounting evidence that young people can develop a dependence on indoor tanning, but research on factors associated with indoor tanning dependence remains limited.Methods: This cross-sectional study investigated factors associated with indoor tanning dependence in a community sample of 389 non-Hispanic white young adult women ages 18 to 30 who had indoor tanned ≥1 time in the past year. Participants completed measures of indoor tanning dependence, including the modified CAGE and modified Diagnostic and Statistical Manual for Mental Disorders-IV psychiatric screening assessments, indoor tanning behavior and beliefs, and behavioral and psychiatric comorbidity.Results: Overall, 22.6% of the sample screened positive for indoor tanning dependence. In multivariable analyses, indoor tanning dependence was associated with younger age of indoor tanning initiation [adjusted odds ratio (aOR) = 0.79; P = 0.017], indoor tanning ≥20 times in the past year (aOR = 3.03; P = 0.015), stronger beliefs about the benefits of tanning (aOR = 2.15; P = 0.004), greater perceived susceptibility to indoor tanning risks (aOR = 2.72; P < 0.001), stronger beliefs about physical appearance (aOR = 1.73; P = 0.037), and depressive symptoms (aOR = 3.79; P < 0.001).Conclusions: Indoor tanning dependence among young, non-Hispanic white women is associated with behaviors that increase the risk of skin cancer, beliefs favoring the perceived benefits of tanning, and comorbid risks such as stronger beliefs about physical appearance and depressed mood.Impact: Comprehensive skin cancer prevention efforts should address indoor tanning dependence among young women and its leading risk factors. Cancer Epidemiol Biomarkers Prev; 26(11); 1636-43. ©2017 AACR.

Prediction of Melanoma Risk in a Southern European Population Based on a Weighted Genetic Risk Score

Many single nucleotide polymorphisms (SNPs) have been described as putative risk factors for melanoma. The aim of our study was to validate the most prominent genetic risk loci in an independent Greek melanoma case-control dataset and to assess their cumulative effect solely or combined with established phenotypic risk factors on individualized risk prediction. We genotyped 59 SNPs in 800 patients and 800 controls and tested their association with melanoma using logistic regression analyses. We constructed a weighted genetic risk score (GRSGWS) based on SNPs that showed genome-wide significant (GWS) association with melanoma in previous studies and assessed their impact on risk prediction. Fifteen independent SNPs from 12 loci were significantly associated with melanoma (P < 0.05). Risk score analysis yielded an odds ratio of 1.36 per standard deviation increase of the GRSGWS (P = 1.1 × 10(-7)). Individuals in the highest 20% of the GRSGWS had a 1.88-fold increase in melanoma risk compared with those in the middle quintile. By adding the GRSGWS to a phenotypic risk model, the C-statistic increased from 0.764 to 0.775 (P = 0.007). In summary, the GRSGWS is associated with melanoma risk and achieves a modest improvement in risk prediction when added to a phenotypic risk model.

Telomere length and the risk of cutaneous melanoma and non-melanoma skin cancer: a review of the literature and meta-analysis

There is much evidence supporting the role of telomeres in cancer pathogenesis, however the studies that investigated the association between telomere length and skin cancer risk provided inconsistent results. To help clarify this issue, we performed a systematic review and meta-analysis of published papers on the association between peripheral leukocytes telomere length (PLTL) and the risk of cutaneous melanoma and non-melanoma skin cancer (NMSC). We calculated summary relative risks (SRR) and 95% confidence intervals (95% CI) using random effect models with maximum likelihood estimates, and explored causes of between-studies heterogeneity of risk estimates. We included 1629 cutaneous melanoma and 1439 NMSC from eight independent studies published until March 2015. The SRR of cutaneous melanoma for those in the lowest (vs. highest) category of PLTL distribution was 0.25 (95% CI 0.09-0.67). The results were less clear for NMSC, with two studies reporting no association and one study showing an increase in risk for those in the lowest (vs. highest) category of PLTL distribution. For both cutaneous melanoma and NMSC, the between-studies heterogeneity was large, mainly due to inclusion of hospital-based case-control studies. Our meta-analysis shows evidence of an association between short PLTL and reduced risk for cutaneous melanoma.