Advanced

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

Davies, John R; Chang, Yu-mei; Bishop, D Timothy; Armstrong, Bruce K; Bataille, Veronique; Bergman, Wilma; Berwick, Marianne; Bracci, Paige M; Elwood, J Mark and Ernstoff, Marc S, et al. (2015) In Cancer Epidemiology Biomarkers & Prevention 24(5). p.24-817
Abstract

BACKGROUND: We report the development of a cutaneous melanoma risk algorithm based upon seven factors; hair color, skin type, family history, freckling, nevus count, number of large nevi, and history of sunburn, intended to form the basis of a self-assessment Web tool for the general public.

METHODS: Predicted odds of melanoma were estimated by analyzing a pooled dataset from 16 case-control studies using logistic random coefficients models. Risk categories were defined based on the distribution of the predicted odds in the controls from these studies. Imputation was used to estimate missing data in the pooled datasets. The 30th, 60th, and 90th centiles were used to distribute individuals into four risk groups for their age, sex,... (More)

BACKGROUND: We report the development of a cutaneous melanoma risk algorithm based upon seven factors; hair color, skin type, family history, freckling, nevus count, number of large nevi, and history of sunburn, intended to form the basis of a self-assessment Web tool for the general public.

METHODS: Predicted odds of melanoma were estimated by analyzing a pooled dataset from 16 case-control studies using logistic random coefficients models. Risk categories were defined based on the distribution of the predicted odds in the controls from these studies. Imputation was used to estimate missing data in the pooled datasets. The 30th, 60th, and 90th centiles were used to distribute individuals into four risk groups for their age, sex, and geographic location. Cross-validation was used to test the robustness of the thresholds for each group by leaving out each study one by one. Performance of the model was assessed in an independent UK case-control study dataset.

RESULTS: Cross-validation confirmed the robustness of the threshold estimates. Cases and controls were well discriminated in the independent dataset [area under the curve, 0.75; 95% confidence interval (CI), 0.73-0.78]. Twenty-nine percent of cases were in the highest risk group compared with 7% of controls, and 43% of controls were in the lowest risk group compared with 13% of cases.

CONCLUSION: We have identified a composite score representing an estimate of relative risk and successfully validated this score in an independent dataset.

IMPACT: This score may be a useful tool to inform members of the public about their melanoma risk.

(Less)
Please use this url to cite or link to this publication:
author
, et al. (More)
(Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Algorithms, Case-Control Studies, Humans, Melanoma, Research Design, Risk Factors, Skin Neoplasms, Journal Article, Research Support, N.I.H., Extramural, Research Support, Non-U.S. Gov't, Validation Studies
in
Cancer Epidemiology Biomarkers & Prevention
volume
24
issue
5
pages
8 pages
publisher
American Association for Cancer Research
external identifiers
  • pmid:25713022
  • wos:000353702800007
  • scopus:84938582672
ISSN
1538-7755
DOI
10.1158/1055-9965.EPI-14-1062
language
English
LU publication?
yes
id
ce2be338-b0b6-45bf-ac99-82d2779c2c31 (old id 5142727)
alternative location
http://www.ncbi.nlm.nih.gov/pubmed/25713022?dopt=Abstract
date added to LUP
2015-03-10 22:02:08
date last changed
2017-06-19 13:18:15
@article{ce2be338-b0b6-45bf-ac99-82d2779c2c31,
  abstract     = {<p>BACKGROUND: We report the development of a cutaneous melanoma risk algorithm based upon seven factors; hair color, skin type, family history, freckling, nevus count, number of large nevi, and history of sunburn, intended to form the basis of a self-assessment Web tool for the general public.</p><p>METHODS: Predicted odds of melanoma were estimated by analyzing a pooled dataset from 16 case-control studies using logistic random coefficients models. Risk categories were defined based on the distribution of the predicted odds in the controls from these studies. Imputation was used to estimate missing data in the pooled datasets. The 30th, 60th, and 90th centiles were used to distribute individuals into four risk groups for their age, sex, and geographic location. Cross-validation was used to test the robustness of the thresholds for each group by leaving out each study one by one. Performance of the model was assessed in an independent UK case-control study dataset.</p><p>RESULTS: Cross-validation confirmed the robustness of the threshold estimates. Cases and controls were well discriminated in the independent dataset [area under the curve, 0.75; 95% confidence interval (CI), 0.73-0.78]. Twenty-nine percent of cases were in the highest risk group compared with 7% of controls, and 43% of controls were in the lowest risk group compared with 13% of cases.</p><p>CONCLUSION: We have identified a composite score representing an estimate of relative risk and successfully validated this score in an independent dataset.</p><p>IMPACT: This score may be a useful tool to inform members of the public about their melanoma risk.</p>},
  author       = {Davies, John R and Chang, Yu-mei and Bishop, D Timothy and Armstrong, Bruce K and Bataille, Veronique and Bergman, Wilma and Berwick, Marianne and Bracci, Paige M and Elwood, J Mark and Ernstoff, Marc S and Green, Adele and Gruis, Nelleke A and Holly, Elizabeth A and Ingvar, Christian and Kanetsky, Peter A and Karagas, Margaret R and Lee, Tim K and Le Marchand, Loïc and Mackie, Rona M and Olsson, Håkan and Østerlind, Anne and Rebbeck, Timothy R and Reich, Kristian and Sasieni, Peter and Siskind, Victor and Swerdlow, Anthony J and Titus, Linda and Zens, Michael S and Ziegler, Andreas and Gallagher, Richard P and Barrett, Jennifer H and Newton-Bishop, Julia},
  issn         = {1538-7755},
  keyword      = {Algorithms,Case-Control Studies,Humans,Melanoma,Research Design,Risk Factors,Skin Neoplasms,Journal Article,Research Support, N.I.H., Extramural,Research Support, Non-U.S. Gov't,Validation Studies},
  language     = {eng},
  number       = {5},
  pages        = {24--817},
  publisher    = {American Association for Cancer Research},
  series       = {Cancer Epidemiology Biomarkers & Prevention},
  title        = {Development and validation of a melanoma risk score based on pooled data from 16 case-control studies},
  url          = {http://dx.doi.org/10.1158/1055-9965.EPI-14-1062},
  volume       = {24},
  year         = {2015},
}