Trends in Skin Melanoma Burden: Findings From the Global Burden of Disease Study

Cutaneous melanoma is a cancer of the skin caused by unchecked proliferation of the melanin-producing cells (melanocytes) of the dermis. Melanoma is the third most common skin cancer and the leading cause of skin cancer mortality.¹ Amongst the population with the highest risk for melanoma, males of older age with fair skin, it is believed that the mortality rate has increased over time.²

Much research has been conducted on the burden of melanoma, which is a particularly relevant topic given the increasing incidence.³ The Global Disease Burden study of 2015 demonstrated 351,880 cases worldwide (incidence of ~5 of 100,000), with the highest incidences of melanoma occurring in Australasia and North America.⁴ Beyond the incidence of melanoma, many reports have attempted to identify years of life lost (YLL), disability adjusted life years (DALYs), and economic loss associated with this cancer.⁵ Research focused specifically in North America has demonstrated 1,800,000 YLLs accrued over an 18-year period accounting for $66.9 billion in economic productivity loss.⁶ Despite this proliferation of information on the burden of melanoma there remains a paucity of synthesized information regarding trends on a global scale and information related to population socioeconomic status and melanoma burden.

The past several decades have been marked by large changes in the treatment and prevention of melanoma care worldwide. Beginning in the 1980s, there have been many public health initiatives to improve the prevention and detection of melanoma, most notably the “Slip! Slop! Slap!” campaign of Australia.⁷ The 1990s and 2000s were marked by the advent and adoption of sentinel lymph node biopsy and clinical trials that advanced surgical care, with the Multicenter Selective Lymphadenectomy Trials (MSLT-1 and MSLT-2) and Dermatologic Cooperative Oncology Group trial (DeCOG) being some of the most important.^8,9 Finally, over the past several years there have been innovations in the use of immunotherapies for the treatment of Stage 3 and 4 melanoma.^10,11 Assessing the data for trends in burden of disease will provide insight into the global effects of the aforementioned interventions. Further, assessment of recent data will set a baseline as new therapies are developed and become widely available.

This study aims to utilize the Global Disease Burden database to evaluate the trend of melanoma incidence, DALYs, mortality, and mortality-to-incidence ratio (M/I) between 1990 and 2017. Further, it compares melanoma burden and mortality based on global socioeconomic status.  This study hyphothesizes an increase in melanoma incidence and a decrease in DALYs and mortality. Information on trends of the global burden of melanoma may provide insight to areas requiring focused intervention.   

Data extraction

This study evaluated the data of the Global Burden of Disease Study 2017 (GDB 2017) of the Institute for Health Metrics and Evaluation (IHME) of the University of Washington, Seattle.¹² GDB 2017 is the result of a collaborative effort of more than 3600 researchers to collect and evaluate information on the incidence, mortality, and disability associated with more than 350 diseases and injuries in 195 countries since 1990. GDB 2017 prospectively captures and monitors relevant data. Specific methodology has been described in previous publications.¹³ The study follows the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER).¹⁴

This study queried information on incidence (per 100,000 people), mortality, M/I, and DALYs of melanoma from 1990-2017, in the US overall, and in countries with low, middle, or high sociodemographic index (SDI), for males and females of all ages. The authors also abstracted associated 95% uncertainty interval (UI) for each value.

Statistical analysis

Descriptive statistics were employed, and the rate ratios were calculated to assess differences in rates of incidence, mortality, M/I, and DALYs from 1990-2017. Similar statistics were employed to analyze differences in rates for male versus female patients and patients among different age groups. Comparison of the trends in the US were compared with relevant values for countries with low, middle, and high SDI. Statistical comparisons were performed for values in 1990 and 2017 by employing t-test. Rate ratio function in R statistical language was employed.¹⁵ All other analyses were performed in STATA (v.13.0, SataCorp).

Incidence

During the study period, melanoma incidence increased on average by 1.6 times (95% UI 0.75 to 3.41); however, this difference was not statistically significant (Table 1). A subgroup analysis based on the age showed that the incidence increased with age. Patients over the age of 60 had a statistically significant increase of incidence by 1.72 (age group: 60–64) to 164.6 (age group: 95 plus) times (Table 1). A subgroup analysis based on the sex revealed no difference in terms of incidence in 2017 and no statistically significant changes were observed between 1990 and 2010 incidence rates (Table 1). The comparison of US to other countries showed that US had statistically significantly higher incidence rate in 2017 than other countries. However, no statistically significant differences were observed for any country group between the rates reported in 1990 and 2010 (Figure 1, Table 1).

Mortality

The analysis on mortality showed a relatively stable mortality rate in the US. Increasing rates of mortality were observed for patients over 65 years of age out of the groups examined; however, this difference was only statistically significant for patients over 95 years of age (Table 2). A subgroup analysis based on the sex, did not reveal statistically significant changes across the study period in terms of mortality or in between sex groups in 2017 (Table 2). The US was found to have statistically significantly higher rates of mortality in 2017. However, no statistically significant differences were observed for any country group between the rates reported in 1990 and 2010 (Table 2).

M/I

The calculated average M/I was relatively stable across the study period and across age groups, sex, and countries (Figure 2, Table 3).

DALYs

Average DALYs were relatively unchanged during the studied period. However, patients over the age of 75 years demonstrated statistically significantl increased DALYs from 1990 to 2010 by 1.34 to 1.85 times (Table 4). No statistically significant changes were observed for males or females over time; however, males were found to have statistically significantly higher DALYs in 2017. Once again, the US was found to have statistically significantl higher rates of DALYs in comparison with other groups. However, no statistically significant differences were observed for any country group between the rates reported in 1990 and 2010 (Figure 3, Table 4).

Discussion

This study utilized the Global Disease Burden database to characterize trends in population burden of melanoma with a focus on the subgroups of sex, age, and socioeconomic status as measured by SDI. The purpose was to assess for the trends in the melanoma burden on a population level in the US with comparisons to populations of various SDI worldwide. Regular surveillance and assessment of melanoma trends is vital for the assessment of large-scale interventions. Global melanoma care can advance if doctors and public health officials have information necessary to change practice management and allocate resources and healthcare workers.¹⁶ Surveillance of disease burden can help guide this decision-making.

This study found a significant increase in incidence of melanoma overall between 1990 and 2017 in patients over 60 years of age. These findings could be attributed to an increasing rate in the general population but is certainly influenced by increased public awareness of melanoma and improved access to screening. Previous research has demonstrated that the incidence of both thin and thick melanomas has increased in the United States.¹⁷ On a global scale, the increase in incidence is not as readily apparent. In fact, analysis of variances of incidence difference between 1990 and 2017 does not demonstrate any difference among SDI groups. However, it is clear the low SDI countries have not experienced the same increase in incidence of melanoma that has occurred in those countries of higher SDI. Possibilities for this finding include that higher SDI countries have the resources to provide increased screening and diagnosis of melanoma or that higher SDI countries tend to have a greater proportion of Caucasians who are at highest risk for melanoma (ie, United States, Australia, Western European nations). Further investigation into the low SDI subgroup is warranted to determine whether there is a truly slower increase in incidence or whether screening and access to care are affecting the rate.

This study did not demonstrate any improvement in the mortality rates between 1990 and 2010. Conversely, increased mortality was found for patients over 95 years of age. Also, the US was found to have higher rates of mortality than other country groups. Interestingly, an inverse trend was observed between SDI level and mortality. This suggests a disparity that requires further evaluation and intervention. The reason for this disparity could have many etiologies. It is likely that improved documentation in higher SDI countries may lead to “falsely” elevated mortality.¹⁸ In addition, increased life expectancy in higher SDI countries and decreased mortality related to other common causes of death in lower SDI countries (eg, injuries, labor-related conditions, cardiovascular events) may also be related to this “falsely” elevated mortality related to melanoma in higher SDI countries. It is reasonable to assume that with the advent of effective immunotherapy and targeted chemotherapy for melanoma in the early 2010s, a decline in mortality will likely be evident in the next few years. No significant changes were found for the M/I.

DALYs were included in the analysis to attempt to more completely capture the trend in human toll of melanoma. There was a statistically significant increase of DALYs in the US for patients over 75 years of age, concerning for increasing burden of melanoma among those age groups. The study also found higher DALYs for male patients, indicating higher burden of disease for male patients. Sex differences in melanoma are not striking, as it has been previously supported that males may have disadvantage in comparison to females,² which may be related to suboptimal skin care and examination¹⁹ or estrogen protective effects for women.²⁰ Strikingly, once again it was found that the US and countries with high SDI suffer higher DALYs. This can be easily explained if someone reviews the definition of DALYs. One can understand that DALYs are related to years of life lost. With the longer life expectancy in countries with SDI, mortality by melanoma will directly lead to higher DALYs.

This study was limited by the features innate to the GDB database and the retrospective study design. Specifically, the database does not provide information regarding location, depth, or stage of diagnosis of melanoma thath could provide further description into the burden of the disease. Direct comparison of various groups was limited by a paucity of demographic data that would allow for potential propensity matching or more direct comparison between individual countries.

This study highlights differences in melanoma incidence and mortality from 1990 to 2017. A statistically significant increase of incidence and DALYs for patients over the age of 60 and 75 years, respectively, was observed. Physicians involved in melanoma care should be aware of these changes in order to anticipate care needs.

Affiliations: ¹Division of Plastic and Reconstructive Surgery, Rush University Medical Center, Chicago, IL; ²Department of Surgery, Rush University Medical Center, Chicago, IL; ³Department of Mechanical and Industrial Engineering, University of Illinois at Chicago (UIC), Chicago, IL

Correspondance: Deana S. Shenaq, MD; deana_shenaq@rush.edu

Disclosures: AH Dorafshar received indirect research grant support from De Puy Synthes and royalties from KLS Martin and Elsevier. The authors have no other relevant financial or non-financial interests to disclose.