Melanoma risk prediction models
Abstract
Background/Aim. The lack of effective therapy for advanced stages of melanoma emphasizes the importance of preventive measures and screenings of population at risk. Identifying individuals at high risk should allow targeted screenings and follow-up involving those who would benefit most. The aim of this study was to identify most significant factors for melanoma prediction in our population and to create prognostic models for identification and differentiation of individuals at risk. Methods. This case-control study included 697 participants (341 patients and 356 controls) that underwent extensive interview and skin examination in order to check risk factors for melanoma. Pairwise univariate statistical comparison was used for the coarse selection of the most significant risk factors. These factors were fed into logistic regression (LR) and alternating decision trees (ADT) prognostic models that were assessed for their usefulness in identification of patients at risk to develop melanoma. Validation of the LR model was done by Hosmer and Lemeshow test, whereas the ADT was validated by 10-fold cross-validation. The achieved sensitivity, specificity, accuracy and AUC for both models were calculated. The melanoma risk score (MRS) based on the outcome of the LR model was presented. Results. The LR model showed that the following risk factors were associated with melanoma: sunbeds (OR = 4.018; 95% CI 1.724–9.366 for those that sometimes used sunbeds), solar damage of the skin (OR = 8.274; 95% CI 2.661–25.730 for those with severe solar damage), hair color (OR = 3.222; 95% CI 1.984–5.231 for light brown/blond hair), the number of common naevi (over 100 naevi had OR = 3.57; 95% CI 1.427-8.931), the number of dysplastic naevi (from 1 to 10 dysplastic naevi OR was 2.672; 95% CI 1.572–4.540; for more than 10 naevi OR was 6.487; 95%; CI 1.993–21.119), Fitzpatricks phototype and the presence of congenital naevi. Red hair, phototype I and large congenital naevi were only present in melanoma patients and thus were strongly associated with melanoma. The percentage of correctly classified subjects in the LR model was 74.9%, sensitivity 71%, specificity 78.7% and AUC 0.805. For the ADT percentage of correctly classified instances was 71.9%, sensitivity 71.9%, specificity 79.4% and AUC 0.808. Conclusion. Application of different models for risk assessment and prediction of melanoma should provide efficient and standardized tool in the hands of clinicians. The presented models offer effective discrimination of individuals at high risk, transparent decision making and real-time implementation suitable for clinical practice. A continuous melanoma database growth would provide for further adjustments and enhancements in model accuracy as well as offering a possibility for successful application of more advanced data mining algorithms.
References
Rigel DS. Trends in dermatology: melanoma incidence. Arch Dermatol 2010; 146(3): 318.
Hollestein LM, Akker SA, Nijsten T, Karim-Kos HE, Coebergh JW, Vries E. Trends of cutaneous melanoma in The Netherlands: increasing incidence rates among all Breslow thickness catego-ries and rising mortality rates since 1989. Ann Oncol 2012; 23(2): 524−30.
Giblin AV, Thomas JM. Incidence, mortality and survival in cu-taneous melanoma. J Plas Reconstr Aesthet Surg 2007; 60(1): 32−40.
Rigel DS, Robinson JK, Ross M, Friedman RJ, Cockerell CJ, Lim HW, et al. Cancer of the skin. 2nd ed. Philadelphia, PA: Elsevier Saunders; 2011.
Jemal A, Siegel R, Ward E, Murray T, Xu J, Smigal C, et al. Cancer statistics, 2006. CA Cancer J Clin 2006; 56(2): 106−30.
Joshua AM. Melanoma prevention: are we doing enough? A Canadian perspective. Curr Oncol 2012; 19(6): e462−7.
Chen ST, Geller AC, Tsao H. Update on the Epidemiology of Melanoma. Curr Dermatol Rep 2013; 2(1): 24−34.
Cho E, Rosner BA, Feskanich D, Colditz GA. Risk factors and individual probabilities of melanoma for whites. J Clin Oncol 2005; 23(12): 2669−75.
Xu LY, Koo J. Predictive value of phenotypic variables for skin cancer: risk assessment beyond skin typing. Int J Dermatol 2006; 45(11): 1275−83.
Walls AC, Han J, Li T, Qureshi AA. Host Risk Factors, Ultra-violet Index of Residence, and Incident Malignant Melanoma In Situ Among US Women and Men. Am J Epidemiol 2013; (In Press)
Chen J, Chi M, Chen C, Zhang XD. Obesity and melanoma: possible molecular links. J Cell Biochem 2013; 114(9): 1955−61.
Li X, Liang L, Zhang M, Song F, Nan H, Wang LE, et al. Obesi-ty-related genetic variants, human pigmentation, and risk of melanoma. Hum Genet 2013; 132(7): 793−801.
Kareus SA, Figueroa KP, Cannon-Albright LA, Pulst SM. Shared predispositions of parkinsonism and cancer: a population-based pedigree-linked study. Arch Neurol 2012; 69(12): 1572−7.
Reddy KK. Vitamin D level and basal cell carcinoma, squamous cell carcinoma, and melanoma risk. J Invest Dermatol 2013; 133(3): 589−92.
Kubica AW, Brewer JD. Melanoma in immunosuppressed pa-tients. Mayo Clin Proc 2012; 87(10): 991−1003.
Faisal RA, Lear JT. Melanoma in organ transplant recipients: incidence, outcomes and management considerations. J Skin Cancer 2012; 2012: 404615.
Engels EA, Pfeiffer RM, Fraumeni JF, Kasiske BL, Israni AK, Snyder JJ, et al. Spectrum of cancer risk among US solid organ transplant recipients. JAMA 2011; 306(17): 1891−901.
Hammer GP, Blettner M, Zeeb H. Epidemiological studies of cancer in aircrew. Radiat Prot Dosimetr 2009;136(4): 232−9.
Koomen ER, Joosse A, Herings RM, Casparie MK, Guchelaar HJ, Nijsten T. Estrogens, oral contraceptives and hormonal re-placement therapy increase the incidence of cutaneous mela-noma: a population-based case-control study. Ann Oncol 2009; 20(2): 358−64.
Gandini S, Iodice S, Koomen E, Di PA, Sera F, Caini S. Hormonal and reproductive factors in relation to melanoma in women: current review and meta-analysis. Eur J Cancer 2011; 47(17): 2607−17.
Thrift AP, Whiteman DC. Can we really predict risk of cancer. Cancer Epidemiol 2013; 37(4): 349−52.
Mar V, Wolfe R, Kelly JW. Predicting melanoma risk for the Australian population. Australas J Dermatol 2011; 52(2): 109−16.
Fortes C, Mastroeni S, Bakos L, Antonelli G, Alessandroni L, Pilla MA, et al. Identifying individuals at high risk of melanoma: a simple tool. Eur J Cancer Prev 2010; 19(5): 393−400.
Williams LH, Shors AR, Barlow WE, Solomon C, White E. Identi-fying Persons at Highest Risk of Melanoma Using Self-Assessed Risk Factors. J Clin Exp Dermatol Res 2011; 2(6): pii: 1000129.
Fargnoli MC, Piccolo D, Altobelli E, Formicone F, Chimenti S, Peris K. Constitutional and environmental risk factors for cutaneous melanoma in an Italian population. A case-control study. Me-lanoma Res 2004; 14(2): 151−7.
Ballester I, Oliver V, Bañuls J, Moragón M, Valcuende F, Botella-Estrada R, et al. . Multicenter case-control study of risk factors for cutaneous melanoma in Valencia, Spain. Actas Dermosifi-liogr 2012; 103(9): 790−7. (English, Spanish)
Bakos L, Mastroeni S, Mastroeni S, Bonamigo RR, Melchi F, Pasquini P, et al. A melanoma risk score in a Brazilian population. An Bras Dermatol 2013; 88(2): 226−32.
Fears TR, Guerry D, Pfeiffer RM, Sagebiel RW, Elder DE, Halpern A, et al. Identifying Individuals at High Risk of Melanoma: A Practical Predictor of Absolute Risk. J Clin Oncol 2006; 24(22): 3590−6.
Grimes DA, Schulz KF. Making sense of odds and odds ratios. Obstet Gynecol 2008; 111(2 Pt 1): 423−6.
Bauer A, Diepgen TL, Schmitt J. Is occupational solar ultraviolet irradiation a relevant risk factor for basal cell carcinoma? A systematic review and meta-analysis of the epidemiological lite-rature. Br J Dermatol 2011; 165(3): 612−25.
Fartasch M, Diepgen TL, Schmitt J, Drexler H. The relationship between occupational sun exposure and non-melanoma skin cancer: clinical basics, epidemiology, occupational disease evaluation, and prevention. Dtsch Arztebl Int 2012; 109(43): 715−20.
Gallagher RP, Lee TK, Bajdik CD, Borugian M. Ultraviolet radia-tion. Chronic Dis Can 2010; 29 (Suppl 1): 51−68.
Surdu S, Fitzgerald EF, Bloom MS, Boscoe FP, Carpenter DO, Haase RF, et al. Occupational exposure to ultraviolet radiation and risk of non-melanoma skin cancer in a multinational European study. PLoS One 2013; 8(4): 623−59.
Chang Y, Barrett JH, Bishop TD, Armstrong BK, Bataille V, Bergman W, et al. Sun exposure and melanoma risk at different lati-tudes: A pooled analysis of 5700 cases and 7216 controls. Int J Epidemiol 2009; 38(3): 814−30.
Boniol M, Autier P, Boyle P, Gandini S. Cutaneous melanoma at-tributable to sunbed use: systematic review and meta-analysis. BMJ 2012; 345: e4757.
Lazovich D, Vogel RI, Berwick M, Weinstock MA, Anderson KE, Warshaw EM. Indoor tanning and risk of melanoma: a case-control study in a highly exposed population. Cancer Epide-miol Biomarkers Prev 2010;1 9(6): 1557−68.
International Agency for Research on Cancer, Working Group: On ar-tificial ultraviolet (UV) light and skin cancer. The association of use of sunbeds with cutaneous malignant melanoma and other skin cancers: A systematic review. Int J Cancer 2007; 120(5): 1116−22.
Csoma Z, Erdei Z, Bartusek D, Dosa-Racz E, Dobozy A, Kemeny L, et al. The prevalence of melanocytic naevi among schoolchild-ren in South Hungary. J Eur Acad Dermatol Venereol 2008; 22(12): 1412−22.
Fehér K, Cercato MC, Prantner I, Dombi Z, Burkali B, Paller J, et al. Skin cancer risk factors among primary school children: inves-tigations in Western Hungary. Prev Med 2010; 51(3−4): 320−4.
Pesch B, Ranft U, Jakubis P, Nieuwenhuijsen MJ, Hergemöller A, Un-fried K, et al. Environmental arsenic exposure from a coal-burning power plant as a potential risk factor for nonmelano-ma skin carcinoma: results from a case-control study in the district of Prievidza, Slovakia. Am J Epidemiol 2002; 155(9): 798−809.
Gandini S, Sera F, Cattaruzza MS, Pasquini P, Abeni D, Boyle P, et al. Meta-analysis of risk factors for cutaneous melanoma: I. Common and atypical naeci. Eur J Cancer 2005; 41(1): 28−44.
Ahmed K, Emran AA, Jesmin T, Mukti RF, Rahman MZ, Ahmed F. Early detection of lung cancer risk using data mining. Asian Pac J Cancer Prev 2013; 14(1): 595−8.
Singleton KW, Hsu W, Bui AA. Comparing predictive models of glioblastoma multiforme built using multi-institutional and lo-cal data sources. AMIA Annu Symp Proc 2012; 2012: 1385−92.
Kurosaki M, Hiramatsu N, Sakamoto M, Suzuki Y, Iwasaki M, Tamori A, et al. Data mining model using simple and readily available factors could identify patients at high risk for hepato-cellular carcinoma in chronic hepatitis C. J Hepatol 2012; 56(3): 602−8.
Amini L, Azarpazhouh R, Farzadfar MT, Mousavi SA, Jazaieri F, Khorvash F, et al. Prediction and control of stroke by data min-ing. Int J Prev Med 2013; 4(Suppl 2): 245−9.
Briones N, Dinu V. Data mining of high density genomic va-riant data for prediction of Alzheimer's disease risk. BMC Med Genet 2012; 13: 7.