Development and external validation of a melanoma risk prediction model based on self-assessed risk factors

Kylie Vuong, Bruce K. Armstrong, Elisabete Weiderpass, Eiliv Lund, Hans Olov Adami, Marit B. Veierod, Jennifer H. Barrett, John R. Davies, D. Timothy Bishop, David C. Whiteman, Catherine M. Olsen, John L. Hopper, Graham J. Mann, Anne E. Cust, Kevin McGeechan, Joanne F. Aitken, Graham G. Giles, Richard F. Kefford, Helen Schmid, Mark A. Jenkins & 1 others Australian Melanoma Family Study Investigators

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Importance: Identifying individuals at high risk of melanoma can optimize primary and secondary prevention strategies. Objective: To develop and externally validate a risk prediction model for incident first-primary cutaneous melanoma using self-assessed risk factors. Design, Setting, and Participants: We used unconditional logistic regression to develop a multivariable risk prediction model. Relative risk estimates from the model were combined with Australian melanoma incidence and competing mortality rates to obtain absolute risk estimates. A risk prediction model was developed using the Australian Melanoma Family Study (629 cases and 535 controls) and externally validated using 4 independent population-based studies: the Western Australia Melanoma Study (511 case-control pairs), Leeds Melanoma Case-Control Study (960 cases and 513 controls), Epigene-QSkin Study (44 544, of which 766 with melanoma), and Swedish Women's Lifestyle and Health Cohort Study (49 259 women, of which 273 had melanoma). Main Outcomes and Measures: We validated model performance internally and externally by assessing discrimination using the area under the receiver operating curve (AUC). Additionally, using the Swedish Women's Lifestyle and Health Cohort Study, we assessed model calibration and clinical usefulness. Results: The risk prediction model included hair color, nevus density, first-degree family history of melanoma, previous nonmelanoma skin cancer, and lifetime sunbed use. On internal validation, the AUC was 0.70 (95%CI, 0.67-0.73). On external validation, the AUC was 0.66 (95%CI, 0.63-0.69) in the Western Australia Melanoma Study, 0.67 (95%CI, 0.65-0.70) in the Leeds Melanoma Case-Control Study, 0.64 (95%CI, 0.62-0.66) in the Epigene-QSkin Study, and 0.63 (95%CI, 0.60-0.67) in the Swedish Women's Lifestyle and Health Cohort Study. Model calibration showed close agreement between predicted and observed numbers of incident melanomas across all deciles of predicted risk. In the external validation setting, there was higher net benefit when using the risk prediction model to classify individuals as high risk compared with classifying all individuals as high risk. Conclusions and Relevance: The melanoma risk prediction model performs well and may be useful in prevention interventions reliant on a risk assessment using self-assessed risk factors.

LanguageEnglish
Pages889-896
Number of pages8
JournalJAMA Dermatology
Volume152
Issue number8
DOIs
Publication statusPublished - 1 Aug 2016

Fingerprint

Melanoma
Women's Health
Area Under Curve
Case-Control Studies
Life Style
Western Australia
Cohort Studies
Calibration
Hair Color
Nevus
Skin Neoplasms
Primary Prevention
Secondary Prevention
Logistic Models
Outcome Assessment (Health Care)
Skin
Mortality
Incidence

Cite this

Vuong, K., Armstrong, B. K., Weiderpass, E., Lund, E., Adami, H. O., Veierod, M. B., ... Australian Melanoma Family Study Investigators (2016). Development and external validation of a melanoma risk prediction model based on self-assessed risk factors. JAMA Dermatology, 152(8), 889-896. https://doi.org/10.1001/jamadermatol.2016.0939
Vuong, Kylie ; Armstrong, Bruce K. ; Weiderpass, Elisabete ; Lund, Eiliv ; Adami, Hans Olov ; Veierod, Marit B. ; Barrett, Jennifer H. ; Davies, John R. ; Bishop, D. Timothy ; Whiteman, David C. ; Olsen, Catherine M. ; Hopper, John L. ; Mann, Graham J. ; Cust, Anne E. ; McGeechan, Kevin ; Aitken, Joanne F. ; Giles, Graham G. ; Kefford, Richard F. ; Schmid, Helen ; Jenkins, Mark A. ; Australian Melanoma Family Study Investigators. / Development and external validation of a melanoma risk prediction model based on self-assessed risk factors. In: JAMA Dermatology. 2016 ; Vol. 152, No. 8. pp. 889-896.
@article{a8e74e7607274cf5875904ad19292e66,
title = "Development and external validation of a melanoma risk prediction model based on self-assessed risk factors",
abstract = "Importance: Identifying individuals at high risk of melanoma can optimize primary and secondary prevention strategies. Objective: To develop and externally validate a risk prediction model for incident first-primary cutaneous melanoma using self-assessed risk factors. Design, Setting, and Participants: We used unconditional logistic regression to develop a multivariable risk prediction model. Relative risk estimates from the model were combined with Australian melanoma incidence and competing mortality rates to obtain absolute risk estimates. A risk prediction model was developed using the Australian Melanoma Family Study (629 cases and 535 controls) and externally validated using 4 independent population-based studies: the Western Australia Melanoma Study (511 case-control pairs), Leeds Melanoma Case-Control Study (960 cases and 513 controls), Epigene-QSkin Study (44 544, of which 766 with melanoma), and Swedish Women's Lifestyle and Health Cohort Study (49 259 women, of which 273 had melanoma). Main Outcomes and Measures: We validated model performance internally and externally by assessing discrimination using the area under the receiver operating curve (AUC). Additionally, using the Swedish Women's Lifestyle and Health Cohort Study, we assessed model calibration and clinical usefulness. Results: The risk prediction model included hair color, nevus density, first-degree family history of melanoma, previous nonmelanoma skin cancer, and lifetime sunbed use. On internal validation, the AUC was 0.70 (95{\%}CI, 0.67-0.73). On external validation, the AUC was 0.66 (95{\%}CI, 0.63-0.69) in the Western Australia Melanoma Study, 0.67 (95{\%}CI, 0.65-0.70) in the Leeds Melanoma Case-Control Study, 0.64 (95{\%}CI, 0.62-0.66) in the Epigene-QSkin Study, and 0.63 (95{\%}CI, 0.60-0.67) in the Swedish Women's Lifestyle and Health Cohort Study. Model calibration showed close agreement between predicted and observed numbers of incident melanomas across all deciles of predicted risk. In the external validation setting, there was higher net benefit when using the risk prediction model to classify individuals as high risk compared with classifying all individuals as high risk. Conclusions and Relevance: The melanoma risk prediction model performs well and may be useful in prevention interventions reliant on a risk assessment using self-assessed risk factors.",
author = "Kylie Vuong and Armstrong, {Bruce K.} and Elisabete Weiderpass and Eiliv Lund and Adami, {Hans Olov} and Veierod, {Marit B.} and Barrett, {Jennifer H.} and Davies, {John R.} and Bishop, {D. Timothy} and Whiteman, {David C.} and Olsen, {Catherine M.} and Hopper, {John L.} and Mann, {Graham J.} and Cust, {Anne E.} and Kevin McGeechan and Aitken, {Joanne F.} and Giles, {Graham G.} and Kefford, {Richard F.} and Helen Schmid and Jenkins, {Mark A.} and {Australian Melanoma Family Study Investigators}",
year = "2016",
month = "8",
day = "1",
doi = "10.1001/jamadermatol.2016.0939",
language = "English",
volume = "152",
pages = "889--896",
journal = "JAMA Dermatology",
issn = "2168-6068",
publisher = "American Medical Association",
number = "8",

}

Vuong, K, Armstrong, BK, Weiderpass, E, Lund, E, Adami, HO, Veierod, MB, Barrett, JH, Davies, JR, Bishop, DT, Whiteman, DC, Olsen, CM, Hopper, JL, Mann, GJ, Cust, AE, McGeechan, K, Aitken, JF, Giles, GG, Kefford, RF, Schmid, H, Jenkins, MA & Australian Melanoma Family Study Investigators 2016, 'Development and external validation of a melanoma risk prediction model based on self-assessed risk factors', JAMA Dermatology, vol. 152, no. 8, pp. 889-896. https://doi.org/10.1001/jamadermatol.2016.0939

Development and external validation of a melanoma risk prediction model based on self-assessed risk factors. / Vuong, Kylie; Armstrong, Bruce K.; Weiderpass, Elisabete; Lund, Eiliv; Adami, Hans Olov; Veierod, Marit B.; Barrett, Jennifer H.; Davies, John R.; Bishop, D. Timothy; Whiteman, David C.; Olsen, Catherine M.; Hopper, John L.; Mann, Graham J.; Cust, Anne E.; McGeechan, Kevin; Aitken, Joanne F.; Giles, Graham G.; Kefford, Richard F.; Schmid, Helen; Jenkins, Mark A.; Australian Melanoma Family Study Investigators.

In: JAMA Dermatology, Vol. 152, No. 8, 01.08.2016, p. 889-896.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Development and external validation of a melanoma risk prediction model based on self-assessed risk factors

AU - Vuong, Kylie

AU - Armstrong, Bruce K.

AU - Weiderpass, Elisabete

AU - Lund, Eiliv

AU - Adami, Hans Olov

AU - Veierod, Marit B.

AU - Barrett, Jennifer H.

AU - Davies, John R.

AU - Bishop, D. Timothy

AU - Whiteman, David C.

AU - Olsen, Catherine M.

AU - Hopper, John L.

AU - Mann, Graham J.

AU - Cust, Anne E.

AU - McGeechan, Kevin

AU - Aitken, Joanne F.

AU - Giles, Graham G.

AU - Kefford, Richard F.

AU - Schmid, Helen

AU - Jenkins, Mark A.

AU - Australian Melanoma Family Study Investigators

PY - 2016/8/1

Y1 - 2016/8/1

N2 - Importance: Identifying individuals at high risk of melanoma can optimize primary and secondary prevention strategies. Objective: To develop and externally validate a risk prediction model for incident first-primary cutaneous melanoma using self-assessed risk factors. Design, Setting, and Participants: We used unconditional logistic regression to develop a multivariable risk prediction model. Relative risk estimates from the model were combined with Australian melanoma incidence and competing mortality rates to obtain absolute risk estimates. A risk prediction model was developed using the Australian Melanoma Family Study (629 cases and 535 controls) and externally validated using 4 independent population-based studies: the Western Australia Melanoma Study (511 case-control pairs), Leeds Melanoma Case-Control Study (960 cases and 513 controls), Epigene-QSkin Study (44 544, of which 766 with melanoma), and Swedish Women's Lifestyle and Health Cohort Study (49 259 women, of which 273 had melanoma). Main Outcomes and Measures: We validated model performance internally and externally by assessing discrimination using the area under the receiver operating curve (AUC). Additionally, using the Swedish Women's Lifestyle and Health Cohort Study, we assessed model calibration and clinical usefulness. Results: The risk prediction model included hair color, nevus density, first-degree family history of melanoma, previous nonmelanoma skin cancer, and lifetime sunbed use. On internal validation, the AUC was 0.70 (95%CI, 0.67-0.73). On external validation, the AUC was 0.66 (95%CI, 0.63-0.69) in the Western Australia Melanoma Study, 0.67 (95%CI, 0.65-0.70) in the Leeds Melanoma Case-Control Study, 0.64 (95%CI, 0.62-0.66) in the Epigene-QSkin Study, and 0.63 (95%CI, 0.60-0.67) in the Swedish Women's Lifestyle and Health Cohort Study. Model calibration showed close agreement between predicted and observed numbers of incident melanomas across all deciles of predicted risk. In the external validation setting, there was higher net benefit when using the risk prediction model to classify individuals as high risk compared with classifying all individuals as high risk. Conclusions and Relevance: The melanoma risk prediction model performs well and may be useful in prevention interventions reliant on a risk assessment using self-assessed risk factors.

AB - Importance: Identifying individuals at high risk of melanoma can optimize primary and secondary prevention strategies. Objective: To develop and externally validate a risk prediction model for incident first-primary cutaneous melanoma using self-assessed risk factors. Design, Setting, and Participants: We used unconditional logistic regression to develop a multivariable risk prediction model. Relative risk estimates from the model were combined with Australian melanoma incidence and competing mortality rates to obtain absolute risk estimates. A risk prediction model was developed using the Australian Melanoma Family Study (629 cases and 535 controls) and externally validated using 4 independent population-based studies: the Western Australia Melanoma Study (511 case-control pairs), Leeds Melanoma Case-Control Study (960 cases and 513 controls), Epigene-QSkin Study (44 544, of which 766 with melanoma), and Swedish Women's Lifestyle and Health Cohort Study (49 259 women, of which 273 had melanoma). Main Outcomes and Measures: We validated model performance internally and externally by assessing discrimination using the area under the receiver operating curve (AUC). Additionally, using the Swedish Women's Lifestyle and Health Cohort Study, we assessed model calibration and clinical usefulness. Results: The risk prediction model included hair color, nevus density, first-degree family history of melanoma, previous nonmelanoma skin cancer, and lifetime sunbed use. On internal validation, the AUC was 0.70 (95%CI, 0.67-0.73). On external validation, the AUC was 0.66 (95%CI, 0.63-0.69) in the Western Australia Melanoma Study, 0.67 (95%CI, 0.65-0.70) in the Leeds Melanoma Case-Control Study, 0.64 (95%CI, 0.62-0.66) in the Epigene-QSkin Study, and 0.63 (95%CI, 0.60-0.67) in the Swedish Women's Lifestyle and Health Cohort Study. Model calibration showed close agreement between predicted and observed numbers of incident melanomas across all deciles of predicted risk. In the external validation setting, there was higher net benefit when using the risk prediction model to classify individuals as high risk compared with classifying all individuals as high risk. Conclusions and Relevance: The melanoma risk prediction model performs well and may be useful in prevention interventions reliant on a risk assessment using self-assessed risk factors.

UR - http://www.scopus.com/inward/record.url?scp=84982163221&partnerID=8YFLogxK

UR - http://purl.org/au-research/grants/nhmrc/1063593

UR - http://purl.org/au-research/grants/nhmrc/107359

UR - http://purl.org/au-research/grants/nhmrc/211172

UR - http://purl.org/au-research/grants/nhmrc/402761

U2 - 10.1001/jamadermatol.2016.0939

DO - 10.1001/jamadermatol.2016.0939

M3 - Article

VL - 152

SP - 889

EP - 896

JO - JAMA Dermatology

T2 - JAMA Dermatology

JF - JAMA Dermatology

SN - 2168-6068

IS - 8

ER -