Importance
In Denmark, the incidence of malignant melanoma (MM) has doubled during the past 25 years, with an incidence of 29.5 and 31.7 per 100 000 person-years in 2012 for men and women, respectively. Understanding the nature of this increase in incidence is important to optimize prevention, early diagnosis, and treatment of in situ and invasive melanoma in Denmark.
Objective
To describe changes over time in the incidence and clinical and pathologic characteristics of in situ and invasive melanoma in Denmark from 1985 through 2012.
Design, Setting, and Participants
We used the official national Danish Melanoma Group database to describe all eligible, prospectively registered cases of in situ and invasive melanoma in Denmark from January 1, 1985, through December 31, 2012. Data analyses were performed from April 1, 2012, through January 31, 2013.
Main Outcomes and Measures
Estimated annual percentage changes (EAPCs) for men and women in European age-standardized incidence, age at diagnosis, and tumor region for in situ melanoma and MM. For MM only, melanoma type, Breslow thickness, ulceration, and mortality.
Results
We included 3299 cases of in situ melanoma and 20 760 cases of MM. The incidence (95% CI) of MM increased by 4.5% (3.6%-5.3%) for men and 4.3% (3.5%-5.2%) for women, which was especially pronounced in patients older than 60 years (EAPCs, 5.8% [4.7%-6.8%] and 4.8% [3.8%-5.9%], respectively), in thin (Breslow thickness, <0.75 mm) melanoma (EAPCs, 6.6% [5.0%-8.2%] and 6.1% [6.0%-7.1%], respectively), and in superficially spreading MM (EAPCs, 5.2% [4.3%-6.2%] and 4.7% [3.9%-5.7%], respectively). We found no significant EAPC in the incidence of melanomas with Breslow thickness greater than 2.00 mm in women, and relative ulceration rates (95% CI) declined in both sexes (EAPCs, −3.3% [−4.0% to −2.6%] in men and −3.4% [−4.0% to −2.8%] in women). More proximal tumor location occurred over time (P < .001). Incidence of in situ melanoma (95% CI) greatly increased (EAPCs, 14.0% [12.2%-15.8%] in men and 11.6% [10.2%-13.2%] in women) with changes over time in age and region (defined by codes in the International Statistical Classification of Diseases, Tenth Revision) similar to those for MM. Mortality related to MM increased in men (EAPC, 0.6% [0.1% to 1.2%]), whereas mortality in women (EAPC, −0.4% [−1.0% to 0.3%]) remained stable.
Conclusions and Relevance
This study confirms a worldwide increase in melanoma incidence. Results may indicate the importance of secondary melanoma prevention in Denmark. Future efforts could intensify primary prevention aimed at young adults, adolescents, and children and maintain and target secondary prevention at the population older than 60 years.
Denmark has the fifth highest incidence of malignant melanoma (MM) in the world.1 Despite increased awareness and primary interventions, the incidence of MM is growing worldwide.2-7 Although the overall cancer incidence in Denmark is plateauing, the incidence of MM has more than doubled during the past 25 years (from 12.1 and 9.6 per 100 000 person-years in 1985 to 29.5 and 31.7 per 100 000 person-years for men and women, respectively). Five-year survival rates were 84% for men and 91% for women in 2012.8 Malignant melanoma is the most common cancer in young women and the second most common cancer in young men (aged 15-39 years), with an incidence of 11.9 and 22.6 per 100 000 person-years in 2012 for men and women, respectively.8 The cause of the rising incidence of MM is debated, and some investigators6,9-11 have suggested that the increase is merely a result of better diagnostic procedures and enhanced focus, leading to increased diagnoses. Knowledge of the current patient profile and stage of primary disease in the Danish population are essential to provide documentation for the needed capacity and organization of melanoma treatment in the Danish health care system. Furthermore, understanding how the characteristics of MM at the time of the primary diagnosis and treatment have changed during the past 3 decades can provide clues to prevention and understanding of the nature of this disease. Our aim in this study was to describe changes over time in the incidence of in situ and invasive melanoma and in the distribution of histologic prognostic factors and clinical characteristics of melanoma in the Danish population from 1985 through 2012.
We performed a descriptive study of all prospectively registered cases of in situ and invasive melanoma in the Danish Melanoma Group (DMG) database from January 1, 1985, through December 31, 2012. The DMG database contains prospectively registered histologic and clinical data on approximately 80% to 90% of the Danish population with in situ and invasive melanoma since 1985 (Figure 1).12 The current data completion rate is approximately 90%.12 All submitted data were anonymized. The Danish Data Protection Agency approved this study.
Inclusion–Exclusion Criteria
We included all cases of in situ and invasive melanoma in the DMG database during the study period. Cases were excluded owing to a missing date of first contact, missing data on Breslow thickness (BT), and mismatch between reported BT and Clark level. Cases were excluded from specific subgroup analyses if they lacked data on the variable in question. Furthermore, in the subgroup analyses of tumor localization, all cases after December 31, 2010, were excluded owing to a sudden decrease in reported rates on this variable during 2011 and 2012. The decrease is likely to be technical because database reporting was converted to online forms in 2011 and because the decrease was not observed in any of the other variables. The inclusion and exclusion process is shown in eFigure 1 in the Supplement.
Our primary end point consisted of changes in European age-standardized incidence of MM and in situ melanoma and changes in European age-standardized mortality rates for MM over time. Mortality rates and corresponding estimated annual percentage changes (EAPCs) for MM were obtained from the NORDCAN (Cancer Incidence, Mortality, Prevalence and Survival in the Nordic Countries) database.8
Our secondary end points included changes in clinical and histopathologic variables at the time of diagnosis during the study period. Clinical end points included age and localization according to codes from the International Statistical Classification of Diseases, Tenth Revision (ICD-10); histopathologic end points for MM included BT, ulceration, and histopathologic melanoma subtype.
Cases were categorized by age at first contact into the following 6 age groups: 14 to 20, 21 to 30, 31 to 40, 41 to 50, 51 to 60, and older than 60 years. Localization was grouped by ICD-10 codes into the following 10 regions: head and neck, upper arm, lower arm, hands, feet, anterior trunk, posterior trunk, thigh, lower leg, and perineum. In situ melanoma was defined as a BT of 0.00 mm and a Clark level of 1; and invasive melanoma was defined as a BT of greater than 0.00 mm and a Clark level other than 1. For MM, the BT was divided into the following 5 thickness groups: greater than 0.00 to 0.74, 0.75 to 1.00, 1.01 to 2.00, 2.01 to 4.00, and thicker than 4.00 mm. Ulceration was reported as a relative rate. Histopathologic types of melanoma registered in the DMG were superficially spreading MM, lentigo MM, nodular MM, acral-lentiginous MM, and unknown primary MM.
Data analyses were performed from April 1, 2012, through January 31, 2013. Annual European age-standardized incidence rates and EAPCs were calculated for all outcome variables. The EAPC was calculated using the following formula:
Ln (European age-standardized incidence) = a (year) + b,where the incline of the line a obtained when linear regression is performed on the natural logarithm of the incidence expresses the EAPC through the following formula:
EAPC = [(ea) − 1] × 100,where e is the base of the natural logarithm. Furthermore, secondary end points were analyzed for changes in the distribution of characteristics during 6 periods (1985-1987, 1988-1992, 1993-1997, 1998-2002, 2003-2007, and 2008-2012) using the χ2 test with the Kendall tau-β for direction of distribution. Finally, we performed an age-period-cohort analysis for the overall incidence of MM to investigate the effects of age, period, and cohort on the risk for developing MM (eFigure 2 in the Supplement).
Results were reported as mean European age-standardized incidence for each period with corresponding EAPCs and P values for changes in distribution during the full study period. Data were analyzed using commercially available software (SPSS Statistics, version 20.0; IBM).
We included 20 760 cases of MM and 3299 cases of in situ melanoma for a total of 24 059 cases. Of these cases, subsets were available for the analyses of age (20 755 MM cases and 3299 in situ cases), tumor localization (17 420 MM cases and 2370 in situ cases), and, in MM cases, tumor thickness (20 760 cases), relative ulceration (20 084 cases), and histologic subtype (20 054 cases). The inclusion and exclusion process is summarized in eFigure 1 in the Supplement.
The incidence of MM increased significantly for both sexes. Female incidences were higher than male incidences, but the male EAPC was higher throughout the study period (Table 1 and Figure 1).
Male mortality increased throughout the study period, but we found no significant change over time in female mortality (Table 1 and Figure 2). Because the rates were obtained from a different registry (NORCAN database), the number of cases is not comparable to the number of cases included from the DMG.
The incidence increased in all age groups for both sexes, except for the youngest male participants (14-20 years). The increase was especially marked in the population older than 60 years and in young women (21-30 years) (Table 1 and eFigure 3 in the Supplement). We found a greater age at diagnosis (P < .001; Table 1) and a marked period effect on the incidence of MM, indicating that the later a person is born, the greater their risk for developing MM becomes (eFigure 2 in the Supplement). The incidence for all BT groups in men increased, but we found no significant change in melanomas with a BT greater than 2.00 mm in women. The incidence of thin MM (<0.75 mm) increased relatively more than that for other BT thickness groups, and we detected thickness distribution for both sexes that favored thinner tumors (P < .001) (Table 1 and eFigure 4 in the Supplement). The incidence of all MM subtypes, except for nodular MM, increased for both sexes, with a decrease in cases of unknown primary MM in (EAPC [95% CI], −2.7% [−8.9% to 4.0%]) and a shift in distribution toward spreading MM (P < .001 for both sexes) (Table 1 and eFigure 5 in the Supplement).
The incidence increased in all ICD-10 localized codes for both sexes, except for the hands, feet, and perineum, with especially marked increases in MM of the anterior and posterior trunk; in women, the increase occurred in the thighs and upper arms (Table 1 and eFigure 6 in the Supplement). We found a shift toward more proximal tumor localization in both sexes (P < .001) (Table 1). The relative ulceration rate decreased significantly during the study period for both sexes (Table 1 and eFigure 7 in the Supplement).
The incidence of in situ melanoma increased for both sexes and was slightly greater in women throughout the study period (Table 2 and Figure 1). The incidence increased in all age groups for both sexes, except for the youngest group (14-20 years), in whom no significant change occurred. Increases were especially marked in the population older than 60 years and in women aged 21 to 30 years (Table 2 and Figure 3). We found an overall greater age at diagnosis in men, but not in women (P = .007 vs P = .08; Table 2).
The incidence increased in all ICD-10 localized codes for both sexes, except for the feet and lower legs in men and the hands, feet, and perineum in women. We found a slight decrease in in situ melanomas of the hands and perineum in men. As in MM, we observed more proximal localization over time (P < .001 and P < .002 for women and men, respectively; Table 2 and Figure 3).
This study is, to our knowledge, the first national study to examine changes over time in the incidence and characteristics of the Danish population with in situ melanoma and the most extensive study of changes over time in the incidence, mortality, and characteristics of the Danish population with MM to date. Results of changes over time for MM correspond well with those reported in other studies13 because they confirm an increase in the incidence of MM during the period from 1985 through 2012, with a relatively greater increase from the early 2000s, but also contribute new, important clues about the cause of this rapid increase.
Some investigators propose that the increase in MM incidence currently seen around the world is caused by a greater awareness of MM, leading to increased diagnosis, overdiagnosis, and diagnostic drift.9,11 If these consequences of greater awareness were true in Denmark, we would see an increase in thin and in situ melanomas and a decrease in thick melanomas and mortality, as shown after a large screening program.14 Most studies3,6,7,10,15 report an increase in melanomas of all BT, but with a relatively greater increase in in situ melanoma and thin MM compared with thick MM. In most countries, mortality rates are rising, except in the United States and Spain, where mortality has remained stable and even decreased for patients younger than 65 years.5-7,10,16,17 From 1985 through 2011, MM-related mortality in Denmark increased for men, whereas MM-related mortality for women remained stable, and 5-year survival rates increased.8,13 In our study, the incidence of MM seemed to increase relatively faster from the early 2000s, which is also seen in data from the Danish Cancer Registry.8,13 This increase, which is especially pronounced for in situ melanoma and thin melanoma (BT, <0.75 mm), may reflect the introduction of the routine use of dermoscopy into the diagnostic process because the use of dermoscopy facilitates the discovery of early neoplasms.18 The increase in incidence was greater in patients older than 60 years relative to that of the other age groups. This increase could reflect a part of our population who received early UV exposure at a time when we had no primary interventions, such as public UV-protection campaigns, reaching an age at which they begin to develop neoplasms. This observation and the persisting increase in male mortality and incidence of thick and intermediate MM (BT, >1.00 mm) suggests that the increase in incidence is not caused solely by increased diagnoses but is, at least in part, a true phenomenon. However, the increase in 5-year survival and the proportion of melanomas with more favorable prognostic characteristics, such as in situ melanoma and thin MM, suggests that secondary prevention efforts, such as early detection and increased public awareness, also play a part.8
Ultraviolet radiation obtained by sun exposure and especially intense intermittent UV radiation exposure and sunburns is a key factor in melanoma development.3,19-21 In our study, the greatest increase in incidence was seen in areas that are often subject to intermittent sun exposure and sunburn (trunk, thighs, and upper arms), but no increase was seen in the hands, feet, or perineum, areas with constant or very low UV exposure. Changes in fashion that favor less covering clothing and sun-tanned skin and changed sun-exposure and vacation habits have influenced melanoma incidence and characteristics. Our study showed a marked period effect of MM (ie, the later one is born, the greater the risk for developing MM becomes). This effect may suggest that lifestyle changes are increasing our population’s general risk for developing MM. According to the Danish Cancer Society,22 public health campaigns are effective because the use of sunscreen has increased and sunbathing has decreased since 2006. However, despite primary interventions, 41% of the Danish population reported at least 1 episode of sunburn during the summer of 2010. These episodes were especially marked in the group aged 15 to 19 years, among whom 73% were sunburned.22 This finding is troubling because UV exposure early in life is a relatively greater risk factor for MM development than accumulated and late UV exposure.23 The incidence of MM likely will continue to rise in Denmark, and we may not see the full effect of primary interventions for at least another decade.
Our study design has some limitations. The DMG report rate ranges from approximately 80% to 90%, which is comparable to the rates of the Swedish and Dutch databases.10,12,24 Data completion is approximately 90% at present; however, data completion has varied in the past.12 Owing to missing data from the early study period, we were not able to include melanoma staging in our study, and the DMG database has only recently begun to register data on the mitotic rate, which would have been interesting to examine. Melanoma cases in the DMG database have been prospectively and consecutively registered and represent cases diagnosed and treated mainly in major hospital melanoma centers. The sampling error is probably very small but could result in skewed rates if the underreporting was systematic. However, because our overall incidence rates correspond fairly well with data from the Danish Cancer Registry, this result is unlikely (Figure 1).8
Future efforts should focus on minimizing early UV exposure by intensifying primary interventions, such as public health campaigns and regulations aimed at children, adolescents, and high-risk groups of patients. For instance, one could propose mandatory use of sun protection in day-care centers and schools or to plan schedules to avoid outdoor activities in these institutions from noon to 3 pm. Because the increase in incidence in the older population is likely caused by previous sun exposure, primary intervention is not likely to have much effect for this group.
The study found evidence in favor of an actual increase in melanoma incidence. A relative increase in in situ melanoma, thin MM, and 5-year survival rates with a decrease in ulcerated tumors indicates that secondary interventions, such as early detection and treatment, are most likely effective. Meanwhile, increasing incidence of thick MM and mortality in men and an increase in the incidence in sun-exposed areas suggest that primary interventions need to be intensified. This conclusion is further supported by the lack of a decrease in thick melanomas and mortality in women. Future efforts should intensify primary prevention, which could be specifically aimed at young adults, adolescents, and children. Efforts in secondary prevention should be maintained and targeted at the population older than 60 years.
Accepted for Publication: April 16, 2015.
Corresponding Author: Neel Maria Helvind, MD, Department of Plastic Surgery, Breast Surgery, and Burns, University Hospital Rigshospitalet, Istedgade 120, 2.tv, 1650 Copenhagen V, Denmark (neelhelvind@gmail.com).
Published Online: June 10, 2015. doi:10.1001/jamadermatol.2015.1481.
Author Contributions: Dr Helvind had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Helvind, Hölmich, Glud, Drzewiecki.
Acquisition, analysis, or interpretation of data: Helvind, Hölmich, Smith, Andersen, Dalton, Drzewiecki.
Drafting of the manuscript: Helvind, Drzewiecki.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Helvind, Smith, Andersen.
Obtained funding: Glud.
Administrative, technical, or material support: Helvind, Glud, Drzewiecki.
Study supervision: Hölmich, Glud, Dalton, Drzewiecki.
Conflict of Interest Disclosures: None reported.
Additional Contributions: We thank the departments of Plastic Surgery, Pathology, and Oncology in Denmark for consecutively and systematically contributing cases to the DMG database.
1.Ferlay
J, Soerjomataram
I, Ervik
M,
et al.
GLOBOCAN 2012 v1.0: Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11.
http://globocan.iarc.fr. Accessed July 28, 2014.
2.Pruthi
DK, Guilfoyle
R, Nugent
Z, Wiseman
MC, Demers
AA. Incidence and anatomic presentation of cutaneous malignant melanoma in central Canada during a 50-year period: 1956 to 2005.
J Am Acad Dermatol. 2009;61(1):44-50.
PubMedGoogle ScholarCrossref 3.Whiteman
DC, Bray
CA, Siskind
V, Green
AC, Hole
DJ, Mackie
RM. Changes in the incidence of cutaneous melanoma in the west of Scotland and Queensland, Australia: hope for health promotion?
Eur J Cancer Prev. 2008;17(3):243-250.
PubMedGoogle ScholarCrossref 4.Downing
A, Newton-Bishop
JA, Forman
D. Recent trends in cutaneous malignant melanoma in the Yorkshire region of England: incidence, mortality and survival in relation to stage of disease, 1993-2003.
Br J Cancer. 2006;95(1):91-95.
PubMedGoogle ScholarCrossref 5.Marcos-Gragera
R, Vilar-Coromina
N, Galceran
J,
et al. Rising trends in incidence of cutaneous malignant melanoma and their future projections in Catalonia, Spain: increasing impact or future epidemic?
J Eur Acad Dermatol Venereol. 2010;24(9):1083-1088.
PubMedGoogle Scholar 6.Chellini
E, Crocetti
E, Carli
P, Martini
A, Giovannetti
L. The melanoma epidemic debate: some evidence for a real phenomenon from Tuscany, Italy.
Melanoma Res. 2007;17(2):129-130.
PubMedGoogle ScholarCrossref 7.Linos
E, Swetter
SM, Cockburn
MG, Colditz
GA, Clarke
CA. Increasing burden of melanoma in the United States.
J Invest Dermatol. 2009;129(7):1666-1674.
PubMedGoogle ScholarCrossref 8.Engholm
G, Ferlay
J, Christensen
N,
et al.
NORDCAN: Cancer Incidence, Mortality, Prevalence and Survival in the Nordic Countries, Version 6.1 (25.04.2014).http://www.ancr.nu. Accessed July 20, 2014.
10.Hollestein
LM, van den Akker
SA, Nijsten
T, Karim-Kos
HE, Coebergh
JW, de Vries
E. Trends of cutaneous melanoma in the Netherlands: increasing incidence rates among all Breslow thickness categories and rising mortality rates since 1989.
Ann Oncol. 2012;23(2):524-530.
PubMedGoogle ScholarCrossref 13.Bay
C, Kejs
AM, Storm
HH, Engholm
G. Incidence and survival in patients with cutaneous melanoma by morphology, anatomical site and TNM stage: a Danish population-based register study 1989-2011.
Cancer Epidemiol. 2015;39(1):1-7.
PubMedGoogle ScholarCrossref 14.Schneider
JS, Moore
DH
II, Mendelsohn
ML. Screening program reduced melanoma mortality at the Lawrence Livermore National Laboratory, 1984 to 1996.
J Am Acad Dermatol. 2008;58(5):741-749.
PubMedGoogle ScholarCrossref 15.Nørgaard
C, Glud
M, Gniadecki
R. Are all melanomas dangerous?
Acta Derm Venereol. 2011;91(5):499-503.
PubMedGoogle Scholar 16.Karim-Kos
HE, de Vries
E, Soerjomataram
I, Lemmens
V, Siesling
S, Coebergh
JW. Recent trends of cancer in Europe: a combined approach of incidence, survival and mortality for 17 cancer sites since the 1990s.
Eur J Cancer. 2008;44(10):1345-1389.
PubMedGoogle ScholarCrossref 18.Nachbar
F, Stolz
W, Merkle
T,
et al. The ABCD rule of dermatoscopy: high prospective value in the diagnosis of doubtful melanocytic skin lesions.
J Am Acad Dermatol. 1994;30(4):551-559.
PubMedGoogle ScholarCrossref 19.Gilchrest
BA, Eller
MS, Geller
AC, Yaar
M. The pathogenesis of melanoma induced by ultraviolet radiation.
N Engl J Med. 1999;340(17):1341-1348.
PubMedGoogle ScholarCrossref 23.Mack
TM, Floderus
B. Malignant melanoma risk by nativity, place of residence at diagnosis, and age at migration.
Cancer Causes Control. 1991;2(6):401-411.
PubMedGoogle ScholarCrossref 24.Lyth
J, Hansson
J, Ingvar
C,
et al; Swedish Melanoma Study Group. Prognostic subclassifications of T1 cutaneous melanomas based on ulceration, tumour thickness and Clark’s level of invasion: results of a population-based study from the Swedish Melanoma Register.
Br J Dermatol. 2013;168(4):779-786.
PubMedGoogle ScholarCrossref