Cutaneous malignant melanoma (CMM) is the most rapidly increasing form of cancer in white populations worldwide 1, and the risk is highly related to exposure to ultraviolet (UV) radiation 2. In 1992, the International Agency for Research on Cancer (IARC) classified solar radiation as carcinogenic to humans, causing CMM and nonmelanoma skin cancer 2. Later, IARC reported a huge and consistent increase in the risk of melanoma associated with artificial UV radiation 3, and in 2009, sun bed use was classified as carcinogenic 4.
In Oceania and Northern Europe, where incidence rates are high, a stabilization or even a decrease in the incidence trends of CMM has been observed during the 1990s in age groups younger than 70 years of age 5–7. In southern European countries, however, the increase in incidence seems to continue 5,8. The incidence rate in Norway is amongst the highest in Europe as well as worldwide 9, and the 1292 reported cases of CMM in 2008 represented 5% of all new invasive cancer diagnoses this year 10.
In 1973, Magnus presented the increasing incidence rates of CMM in Norway for the period 1955–1970, according to age, sex, topography and geographical regions 11, which was followed up in 1981 12. In 1991, Magnus published a comparative study of incidence of CMM in the three Nordic countries, covering cancer data from 1955 until 1988 13. Since then, a similar time-trend analysis on CMM in Norway has not been reported. The present study aims to describe the incidence trends of CMM in Norway during the 50-year period of 1954–2008, according to sex, age, stage of disease, primary site of the tumour and geographical regions, using data from the population-based Cancer Registry of Norway. The results are discussed in light of sun exposure habits and sun awareness over time.
Materials and methods
The Cancer Registry of Norway has registered all cancer cases nationwide since its establishment in 1953. Patients are identified through a unique 11-digit personal identification number, assigned to all newborns and individuals residing in Norway 14. Mandatory reporting from several independent sources ensures completeness and high data quality 14. The present study comprises all new cases of histologically verified invasive CMM diagnosed in the Norwegian population between 1 January 1954 and 31 December 2008. A total of 31 783 cases were registered.
Information on sex, age, diagnostic date, stage of disease, primary site of the tumour (topography) and place of residence was retrieved from the Cancer Registry. The period of follow-up was divided into 5-year diagnostic time periods, and age at diagnosis was divided into four age groups (0–29, 30–49, 50–69, 70+). On the basis of information on metastases at the time of diagnosis, the melanoma patients were categorized into one of three stage groups: localized stage (invasive cancer without any metastases), advanced stage (any infiltration into surrounding structures, regional or distant metastases) and unknown stage (status of metastasis is unknown). A change in the stage-coding practice was introduced at the Cancer Registry in 1991, which implied a stricter information basis for the stage classification. Since then, CMM cases with any uncertainty as to whether metastases were present have been coded as unknown with respect to stage. A stage migration has occurred, from localized stage to unknown stage, indicating that most of these cases have a real localized disease. A small proportion of unknown stage cases may, however, have a more advanced disease (regional proliferation).
The following categories were used on the basis of the primary anatomical site of the tumour: head (ICD7 1900), trunk (ICD7 1901), leg (ICD7 1904), arm (ICD7 1902) and unspecified site (ICD7 1905–1909). Because of the coding practice, further division of the data in the head category was not possible. Norway was divided into five geographical regions on the basis of the degree of latitude and index of UV radiation 15. The Norwegian population is principally white, with an immigrant population comprising only 11.4% of the 4.9 million inhabitants 16, of whom only a minority are non-Caucasians. Although Oslo is located in the eastern region, it is the capital and the largest city in Norway, with a higher percentage of immigrants than in the surrounding area. In Oslo, immigrants from Asia, Africa and South America comprise 19% of the total population, and in some parts of the city, immigrants comprise 40% of the population 16. Accordingly, separate analyses were carried out for Oslo.
The cancer rates were calculated according to sex, age, stage of disease, primary anatomical site of the tumour and by geographical region. The age-adjusted incidence rates are based on direct standardization according to the European Standard Population.
A joinpoint regression model was used to identify the changes in the age-adjusted incidence rates. A joinpoint constitutes the point in time where two regression lines with different slopes meet. The program fits the simplest model specifying zero joinpoints (a straight line) and then uses a series of permutation tests to determine whether additional joinpoints are statistically significant and should be included in the model. Computation of the joinpoints – calendar years for which significant changes in the overall log-linear trend were detected – was based on the best-fitting regression models allowing 0, 1, 2 and 3 joinpoints. The annual percent changes (APC) for the entire time period 1954–2008 are presented together with the APC for the linear parts of the period identified by the model (Table 1).
An age–period–cohort model was used to explore the nonlinear effects of calendar time and birth cohort on CMM incidence trends, for all melanomas, and separately for melanomas at trunk and leg subsites. The incidence rates were modelled as a function of age, diagnostic periods and birth cohorts using a log-linear regression model on the basis of the Poisson distribution. The model can be expressed as follows:
where Rapc is the incidence rate in age group a in birth cohort c, Aa is the age component for age group a, D is the common drift parameter, Pp is the nonlinear period component of period P(P=a+c) and Cc is the nonlinear cohort component of cohort c. The data were organized into 12 5-year age groups (30–34, …, 80–84 and 85+), 11 5-year diagnostic periods (1954–1958, …, 1999–2003 and 2004–2008) and 24 synthetic (overlapping) 10-year birth cohorts (<1869, 1865–1874, …, 1965–1974, 1969–1978). Cohort effects are relative to the 1919–1928 cohorts, assuming no linear period effect. Period effects are relative to the period 1979–1983, assuming no linear cohort effect. Birth cohort was calculated as the age at diagnosis subtracted from the calendar period. Because of the low number of cases when analysing by anatomical subsite, the three oldest age groups (>74 years) and the three first calendar periods (<1968) were excluded from the analysis for trunk and leg subsites.
During the first calendar period, 1954–1958, the age-adjusted incidence rate was less than three per 100 000, whereas the rate in 2004–2008 was 21 per 100 000. Figure 1 shows the increase of the age-adjusted incidence rates of CMM in Norway during the 50-year period 1954–2008 and Table 1 presents the APC in the rates of CMM. The most striking increase in incidence occurred from 1954 and throughout the 1980s for both sexes, and then the rates stabilized. A new increase in incidence followed from the late 1990s, but was most pronounced in men.
Time trends by age
Figure 2 shows the incidence curves according to the four age groups: <30, 30–49, 50–69 and 70+ years. For the youngest age group, a significant increase in incidence was observed throughout the 1980s, followed by decreasing rates in both sexes (Table 1). For the age group 30–49, the rate increased significantly during the entire period. The most striking increase, however, occurred during 1954–1969 and 1954–1978 in men and women, respectively, and after 1990, the rate in men decreased (Table 1). A significant increase in incidence was also observed for the age group 50–69 years, in particular, throughout the 1980s. The most pronounced increase in rate was observed in men, older than 70 years, showing an annual percentage increase of 5.2 (Table 1). The corresponding number for women was 3.9%. The incidence rates in men exceed those in women at the age of 50 (data not shown).
Figure 3a–c presents the cohort and period effects from the age–period–cohort modelling for all anatomical subsites combined, trunk and leg subsites, respectively. For all melanomas combined, the relative cohort-specific risk increased considerably up to around 1945, after which it levelled off in men. For women, the cohort-specific rates continued to increase also in the younger cohorts. In terms of subsites, the pattern for legs resembles the pattern for all sites combined, but for the trunk, the cohort-specific risks are higher for both men and women for the youngest cohorts. For younger cohorts, the risk curve for the trunk is thus increasing, and is steeper in women than in men.
The temporal time trends for the CMM rates were levelling off in the early 1990s, followed by an increase after 2000, and this pattern was found to remain after adjusting for the cohort effects. The recent increase in rates is most distinct for CMM at leg subsites for men.
In the Poisson regression model, both the nonlinear cohort and the period components were significant for both sexes, for all melanomas combined and in the separate analyses for trunk and leg subsites (Table 2).
There is a south–north gradient of CMM occurrence in Norway. The incidence rate in the southern part of the country is twice the rate of the northern part, and this pattern has persisted during the entire period of follow-up (Fig. 4). Roughly, the age-adjusted curves have similar patterns in the southern, eastern, middle and western parts of the country during the entire period of follow-up.
The age-adjusted incidence rates by anatomical subsites are presented in Fig. 5. Throughout the 1990s, the leg was the most common melanoma location in women, with an increasing trend throughout the 1980s. From 1990, however, the rate has decreased (Table 1). In men, an increasing trend was observed for the entire time period. The most striking increase in rate was observed for trunk melanomas, in both sexes, although the rates have levelled off after 1990 (Table 1). For the entire period of follow-up, the strongest increase of both head, trunk and arm melanomas was observed in men (Table 1). Although the southern region has higher rates, the trend patterns according to anatomical subsites were similar in all the geographical regions (data not shown).
Stage of disease
The incidence of localized melanomas increased considerably until the early 1990s for both sexes; thereafter, a stage migration from the localized stage to an unknown stage occurred because of changes in the coding practice at the Cancer Registry of Norway in 1991 (Fig. 6, upper part). In the bottom part of Fig. 6, the unknown stage melanomas are included in the localized stage category, as the major part of melanomas coded as unknown with respect to stage are in a localized stage. The annual increase in the percentage change during the entire period was shown to be 4.1 and 3.5 for men and women, respectively, but the strongest increase in rate was observed during the period 1954–1974 (Table 1). An increase in advanced-stage melanomas was also observed (Fig. 6), with an APC of 1.6 and 1.5% for men and women, respectively (Table 1).
During this 50-year period of follow-up, the incidence of CMM in Norway has increased more than all other cancer types, except for female lung cancer 10. It is worth mentioning that both the definition and the classification of CMM, and the accuracy and specificity of the coding have been consistent over time (except for the changes in the practice of stage coding). Since the 1960s, Norway has had the highest rates of CMM in the Nordic countries, except for Icelandic women, who had higher rates during 1998–2003, and Danish women, who had rates that exceeded those of Norwegian women in 2003 17.
During the last decades, Norway has experienced huge economical progress, with more than a tripled Gross National Product since 1970 16. This financial growth influences sun exposure habits, and holidays abroad to sunny areas have become common, irrespective of socioeconomic level and age group 18,19. Hence, risk-related behaviour that was associated previously with high-income groups has been adopted by the general population.
Since 1990, the CMM rates in men were found to exceed the rates in women at the age of 50 years (not shown), which has also been observed in the other Nordic countries 20. The question has been raised as to whether any sex-dependent factor can explain the different curves for men and women before and after the age of 50 years. Recent studies, investigating the potential role of female sex hormones, have not found any significant association, either with the risk or the prognosis of CMM 21,22. As outdoor workers have a reduced risk of CMM 23, a lower proportion of outdoor working men could have contributed to the increase in the rate. In Norway, the strongest decrease in the number of outdoor workers occurred during the 1960s, which mainly involves cohorts born after 1930–1940. The present cohort analyses show, however, a flattening trend for these cohorts (Fig. 3), which indicate that changes in the occupational structure can hardly explain the high CMM rates in men. Another factor that might be involved is sunbathing behaviour, which has changed during the last decades, in particular, in men older than 50 years of age. The grandparent generation is now frequently seen sunbathing both in Norway and abroad. Recently, American men were reported to have a higher rate of sunburns and less use of sunscreen compared with women 24, which also seems to be the situation in Norway. In a survey investigating sun exposure habits in Norway, 50% of the respondents reported sunburn episodes during the last year, and the frequency in men was higher than that in women 25. The survey also indicated less use of sunscreens in men, particularly in older age groups. Thus, an increase in intermittent sun exposure, because of changed sun exposure habits, could be the major reason for the high risk of CMM in men after the age of 50.
As also discussed by Magnus in the 1970s 11, we are faced with the question of whether the increase in incidence is true or an artefact because of advances in the detection of tumours. The steep increase in the rate of localized melanomas could have resulted from an increased public knowledge of signs and symptoms, in addition to improved and more intense diagnostic facilities. The Norwegian Cancer Society, a nationwide and non-profit voluntary organization, has implemented comprehensive prevention campaigns. However, these were conducted after 1990 (personal communication) and after the period with the most pronounced increase in incidence. In 2002, a Swedish study concluded that prevention campaigns have a minor impact on the melanoma trends 26. An increased awareness might, however, result in a screening effect, with an increase in incidence, followed by a flattening rate. As women are expected to be most concerned about health, and thus possibly more attentive to the risk of melanoma, we would expect the highest increase in the rate in women. The steepest increase, however, was observed in men. Also, the rate of advanced-stage melanomas has increased, and is most pronounced in men, which corresponds to a similar increase in the mortality rates 17.
Moreover, the different time trends for CMM according to anatomical sites support the fact that the increase in CMM rates in Norway is real. The subsite distribution observed during the recent decades is quite similar to what was observed 30 years ago 11, although the rates are significantly higher. There is no reason to believe that the diagnostic intensity and increased focus on moles should create this pattern. Recently, a Norwegian study, focusing on the association between latitude gradient and anatomical location of melanomas, presented similar results 27, and it can be assumed that an increase in intermittent sun exposure explains the increase in incidence observed.
Sex differences according to the anatomical location of the tumours have frequently been described, and are suggested to result from clothing differences 28. During these 50 years of follow-up, the clothing fashion has changed and areas of uncovered skin have increased for both sexes. In a Norwegian survey conducted in 2004, both sexes reported that the intention of sunbathing was to get tanned, as a symbol of health and success 25, and more similar sun exposure habits for men and women may explain the subsite trends. Moreover, UV irradiance from indoor tanning also influences the risk of melanoma 3,29–33. During the 1980s, indoor tanning beds became popular in Norway and, particularly, girls in the age group 15–24 years are reported to be common users 34. Thus, an increasing use of sun beds may have contributed towards the increased rate of female trunk tumours observed. This also corresponds with the upwards trend in the cohort-specific risks for women born after 1955 (Fig. 3b). However, when calculating the APC for trunk melanomas by age groups, the strongest increase after 1990 was observed in women older than 60 years (data not shown). As these women are too old to be among those most exposed to indoor tanning, the increase may not result from sun bed use.
The cohort-specific pattern can be used to speculate about the future rates of melanoma 35. In a previous study, we observed that sun exposure at both younger and older ages influenced the lifetime risk of melanoma 36. Therefore, it is reasonable to assume that at least a part of the cohort-specific patterns will also remain in the future. Men born after 1945 experience a levelling in the rates compared with men born earlier. These men will turn 65 years of age in 2010, most likely leading to a levelling of the rates in those aged older than 65 years in the coming years.
Throughout the follow-up, the CMM rate in the southern part was twice as high as in the northern part of the country. Long-term UV data have been reconstructed for Norway, which were shown to fit well with the geographical variation in the risk of melanoma 37. In addition, geographic differences in the mean summer-temperature may strengthen the geographic rate difference as it influences clothing habits and, hence, the degree and area of skin exposure. In 2001, we investigated whether regions of residence during childhood and adulthood, among domestic migrants, impact the risk of CMM 36. The results showed that the Norwegian inhabitants utilize the available UV radiation and adapt sun exposure habits in the area in which they reside. UV exposure at any age seems to be important for the lifetime risk of CMM.
In Oslo, a lower increase in the rates was observed compared with the eastern region (located at the same latitude with the same climate). The relatively high proportion of non-Western immigrants residing in Oslo 16 might explain this difference, and when calculating the incidence rate excluding city parts with the highest density of immigrants, the rate became similar to that in the eastern region (data not shown).
In Europe, the latitude gradient is in an opposite direction compared with Norway. Northern and western European countries have higher incidence rates than countries in the south and east 9. This disparity might be because of differences in skin pigmentation and genetic factors, with higher tolerance to UV light in the south of Europe 38, as well as climatic differences between the south and the north, which impact the intermittent exposure pattern.
Throughout the 1980s, a steep increase in the rate of melanoma was observed in both sexes, which continued to increase from the late 1990s. The steepest increase has occurred in age groups older than 50 years of age, and was most pronounced in men. In Norway, there are distinct seasonal variations in UV radiation. The long and cold winters makes the Norwegians quite willing to get heavily sun exposed during the short summer period. During this long-term follow-up, the clothing habits have changed, to uncover more skin, in particular, in age groups greater than 50 years. Because of the economic progress in Norway, there are increased possibilities for vacations abroad in sunny areas. The net result seems to strengthen the pattern of intermittent UV exposure and may explain the increasing incidence rates in Norway. The higher increase in incidence in older men corresponds with their less protective behaviour compared with women.
We sincerely thank the Norwegian Cancer Society for thoroughly reviewing their public activities and for their information campaigns for awareness and prevention of malignant melanoma.
Conflicts of interest
There are no conflicts of interest.
1. Garbe C, Leiter U. Melanoma epidemiology and trends
. Clin Dermatol. 2009;27:3–9
2. IARC monographs on the evaluation of carcinogenic risks to humans: solar and ultraviolet radiation. 1992;Vol. 55 France Lyon
3. IARC working group reports: exposure to artificial UV radiation and skin cancer. 2006;Vol. 1 France Lyon
4. El Ghissassi F, Baan R, Straif K, Grosse Y, Secretan B, Bouvard V, et al. WHO international agency for research on cancer monograph working group: a review of human carcinogens – part D: radiation. Lancet Oncol. 2009;10:751–752
5. de Vries E, Bray FI, Coebergh JW, Parkin DM. Changing epidemiology of malignant cutaneous melanoma in Europe 1953–1997: rising trends
in incidence and mortality but recent stabilizations in western Europe and decreases in Scandinavia. Int J Cancer. 2003;20:119–126
6. Liang JJC, Robinson E, Martin RCW. Cutaneous melanoma in New Zealand: 2000–2004. ANZ J Surg. 2010;80:312–316
7. Coory M, Baade P, Aitken J, Smithers M, McLeod GR, Ring I. Trends
for in situ and invasive melanoma in Queensland, Australia, 1982–2002. Cancer Causes Control. 2006;17:21–27
8. Crocetti E, Carli P, Miccinesi G. Melanoma incidence in central Italy will go on increasing also in the near future: a registry-based, age-period-cohort analysis. Eur J Cancer Prev. 2007;16:50–54
9. De Vries E, Coebergh JW. Cutaneous malignant melanoma
in Europe. Eur J Cancer. 2004;40:2355–2366
10. Cancer in Norway 2008 – cancer incidence, mortality, survival and prevalence in Norway. 2009 Oslo Cancer Registry of Norway
11. Magnus K. Incidence of malignant-melanoma of skin in Norway, 1955–1970 – variations in time and space and solar-radiation. Cancer. 1973;32:1275–1286
12. Magnus K. Habits of sun exposure and risk of malignant-melanoma – an analysis of incidence rates in Norway 1955–1977 by cohort, sex
, age, and primary tumor site. Cancer. 1981;48:2329–2335
13. Magnus K. The Nordic profile of skin cancer incidence. A comparative epidemiological study of the three main types of skin cancer. Int J Cancer. 1991;47:12–19
14. Larsen IK, Småstuen M, Johannesen TB, Langmark F, Parkin DM, Bray F, et al. Data quality at the cancer registry of Norway: an overview of comparability, completeness, validity and timeliness. Eur J Cancer. 2009;45:1218–1231
15. Johnsen B, Mikkelborg O, Hannevik M, Nilsen LT, Saxebøl G, Blaasaas KG The Norwegian UV monitoring program, period 1995/96 to 2001. 2002 Østerås, Norway Norwegian Radiation Protection Authority
16. Annual statistics 2010 [in Norwegian]. 2010 Norway Oslo-Kongsvinger
17. Tryggvadottir L, Gislum M, Hakulinen T, Klint Å, Engholm G, Storm H, et al. Trends
in the survival of patients diagnosed with malignant melanoma
of the skin in the Nordic countries 1964–2003 followed up to the end of 2006. Acta Oncol. 2010;49:665–672
18. Bentham G, Aase A. Incidence of malignant melanoma
of the skin in Norway, 1955–1989: associations with solar ultraviolet radiation, income and holidays abroad. Int J Epidemiol. 1996;25:1132–1138
19. Holiday survey 2005 [in Norwegian]. 2007 Norway Statistics Norway, Oslo-Kongsvinger
20. Engholm G, Ferlay J, Christensen N, Johannesen TB, Klint Å, Køtlum JE, et al. NORDCAN: Cancer Incidence, Mortality, Prevalence and Survival in the Nordic Countries, Version 5.1 (March 2012).Association of the Nordic Cancer Registries. Danish Cancer Society
21. Lens M, Bataille V. Melanoma in relation to reproductive and hormonal factors in women: current review on controversial issues. Cancer Causes Control. 2008;19:437–442
22. Gupta A, Driscoll MS. Do hormones influence melanoma? Facts and controversies. Clin Dermatol. 2010;28:287–292
23. Pukkala E, Martinsen JI, Lynge E, Gunnarsdottir HK, Sparén P, Tryggvadottir L, et al. Occupation and cancer – follow-up of 15 million people in five Nordic countries. Acta Oncol. 2009;48:646–790
24. Reuter NP, Bower M, Scoggins CR, Martin RCG, McMaster KM, Chagpar AB. The lower incidence of melanoma in women may be related to increased preventative behavious. Am J Surg. 2010;200:765–769
25. Survey of sun exposure habits [in Norwegian]. 2004 Oslo, Norway The Norwegian Cancer Society & Norwegian Radiation Protection Authority
26. Månsson-Brahme E, Johansson H, Larsson O, Rutquist LE, Ringborg U. Trends
in incidence of cutaneous malignant melanoma
in a Swedish population 1076–1994. Acta Oncol. 2002;41:138–146
27. Cicarma E, Juzeniene A, Porojnicu AC, Bruland ØS, Moan J. Latitude gradient for melanoma by anatomic site and gender in Norway 1966–2007. J Photochem Photobiol B. 2010;101:174–178
28. Giblin AV, Thomas JM. Incidence, mortality and survival in cutaneous melanoma. J Plast Reconstr Aesthet Surg. 2007;60:32–40
29. . The association of use of sunbeds with cutaneous malignant melanoma
and other skin cancers: a systematic review. Int J Cancer. 2007;120:1116–1122
30. Faurschou A, Wulf HC. Ecological analyses of the relation between sunbeds and skin cancer. Photodermatol Photoimmunol Photomed. 2007;23:120–125
31. Cust AE, Armstrong BK, Goumas C, Jenkins MA, Schmid H, Hopper JL, et al. Sunbed use during adolescence and early adulthood is associated with increased risk of early-onset melanoma. Int J Cancer. 2010;128:2425–2435
32. 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 Epidemiol Biomarkers Prev. 2010;19:1557–1568
33. Veierød MB, Adami H-O, Lund E, Armstrong BK, Weiderpass E. Sun and solarium exposure and melanoma risk: effects of age, pigmentary characteristics, and nevi. Cancer Epidemiol Biomarkers Prev. 2010;19:111–120
34. Wichstrøm L Sun exposure habits among teenagers
[in Norwegian]. 1995 Oslo The Norwegian Cancer Society
35. Bray F, Møller B. Predicting the future burden of cancer. Nat Rev Cancer. 2006;6:63–74
36. Robsahm TE, Tretli S. Cutaneous malignant melanoma
in Norway: variation by region of residence before and after the age 17. Cancer Causes Control. 2001;12:569–576
37. Medhaug I, Olseth JA, Reuder J. UV radiation and skin cancer in Norway. J Photochem Photobiol B. 2009;96:232–241
38. Sturm RA. Human pigmentation genes and their response to solar UV radiation. Mutat Res. 1998;422:69–76