Years of life lost due to metastatic melanoma in 12 countries

Abstract

Background:

Regionally or distantly metastatic melanoma (stages IIIB/C and IV) place a high burden on society. To quantify this burden, this study estimated years of life lost (YLL) per patient for adults with metastatic melanoma in 12 countries in 2014.

Methods:

General population growth and life expectancy were estimated from the Organization for Economic Co-operation and Development data and life tables for each country. Incidence of melanoma and mortality rates for the disease were based on GLOBOCAN and US registry data. The prevalence of metastatic melanoma was calculated using mortality rates and survival data from patients with melanoma. YLL per patient was estimated by the difference between the disease-free life expectancy and the life expectancy with metastatic melanoma.

Results:

YLL per patient were as follows: Australia, men = 19.9 years, women = 22.7 years; Brazil, 16.3, 19.8; Canada, 19.4, 22.3; France, 18.8, 23.1; Germany, 18.3, 20.8; Italy, 19.3, 22.7; Mexico, 17.2, 19.0; the Netherlands, 18.5, 21.5; Spain, 19.2, 23.1; Sweden 19.4, 22.0; the UK, 18.7, 21.2; and the US, 17.9, 20.6.

Conclusions:

The burden of metastatic melanoma as measured by YLL is substantial in all 12 countries; although there is variation across countries and between men and women.

Keywords::

• Metastatic melanoma
• YLL
• Mortality
• Life expectancy
• Disease burden

Introduction

Regionally or distantly metastatic melanoma (stages IIIB/C and IV) is a life threatening malignant tumor1–4. Median overall survival (OS) varies by stage, ranging from ∼24 months for patients with stages IIIB/C to 5–10 months for patients with stage IV M1c5. Compared with other cancers, a relatively high proportion of people diagnosed with melanoma are younger adults and this age distribution is generally consistent across countries. Thirty-four per cent of all melanomas diagnosed in 2012 worldwide were in people younger than 55 years of age6^,7.

Little has been published to date on the burden of metastatic melanoma and in particular on the number of years of life lost (YLL) due to metastatic melanoma4^,8–10. Mortality rates have been reported;, however, crude mortality rates are not a sufficient indicator of the burden to society, particularly for diseases that affect younger people and result in premature death, and years of lost work productivity. YLL provides a more useful measure of disease burden by combining life expectancy and mortality data to estimate the average number of years that are lost because of premature death.

The aim of this study was to estimate YLL in patients with metastatic melanoma in 2014 in Australia, Brazil, Canada, France, Germany, Italy, Spain, Mexico, the Netherlands, Sweden, the UK, and the US. These 12 countries were selected to represent different regions and climates, and to include areas of significant disease prevalence.

Materials and methods

YLL is calculated by the difference in life expectancy with and without a given disease. At the population level, this is multiplied also by prevalence and population growth. All assumptions in this study were validated with clinical experts at several stages of the study.

Data sources

Data on general population growth, melanoma incidence, and mortality specific to each country were obtained using GLOBOCAN 200811. Disease-free life expectancy was obtained from statistical agencies in each country12–23. Survival by stage and the age distribution at melanoma diagnosis were based on a US database because no data could be identified specific to each country with the equivalent required level of detail. The only source of data on the stage distribution of melanoma at a detailed level was the American Joint Committee on Cancer (AJCC) reported by Balch et al.24; we were able to determine stage III and stage IV proportions using country-specific publications, but we assumed that the sub-stage distribution reported by Balch et al. was the same for all countries in order to proceed with the calculations. The age distribution at diagnosis of melanoma was obtained from the 2006–2010 estimates from the Surveillance, Epidemiology, and End Results Program (SEER)25. Therefore, the estimates here are more a reflection of general life expectancy with advanced melanoma rather than characteristics of life expectancy with advanced melanoma in each country.

The GLOBOCAN data were selected to ensure consistency of measurement across all countries; the SEER estimates and the work by Balch et al. represent the most up-to-date data on age distribution and survival by stage, respectively. Definition of adults was in accordance with the definitions used in the Organization for Economic Co-operation and Development (OECD) and GLOBOCAN 2008 databases, which was defined as individuals 15 years of age and over.

Prevalence of metastatic melanoma

Prevalence of melanoma was calculated by adding the estimated incident cases of melanoma in the current year and the number of patients surviving from melanoma from previous years24. Then, the distribution of stages for the incident cases was applied to the estimate of the prevalence of melanoma to estimate the number of patients with stage IIIB/C or IV melanoma.

To determine the estimated prevalence in 2014, we applied country-specific, age-adjusted mean annual population growth between 2000–2010 and assumed that incidence remained a constant proportion of the total population11.

Distribution of age at diagnosis

SEER data from 2006–201025 were used to estimate the distribution of patients with metastatic melanoma across the 5 year age groups that were used by GLOBOCAN. We assumed that the distribution of patients across age groups was similar for all stages of metastatic disease (IIIB–IV).

Disease-free life expectancy calculation

Estimates of disease-free life expectancy were obtained from life tables for each country12–23. Some life tables reported life-expectancy for exact ages, whereas others did so for various age ranges; we, therefore, aggregated the data into age groups that matched those for the prevalence of metastatic melanoma, using a published algorithm for interpolating life expectancy based on intervals between the mismatched thresholds26. We assumed that the maximum life expectancy was 100 years of age (i.e., life tables displaying 100 and above years of age were truncated at 100 years of age).

Life expectancy with metastatic melanoma

Life expectancy with metastatic melanoma was calculated by estimating an annualized mortality rate for disease-free life expectancy and then applying that mortality rate to calculate life expectancy with the disease.

Specifically, life expectancy with metastatic melanoma was calculated by the following three steps:

1. A weighted average (WA) annualized mortality rate for people with advanced melanoma was calculated for each of the following periods after diagnosis: 0–1 years, 1–2 years, 2–5 years, and 5–10 years. Estimates of 1, 2, 5, and 10 year survival24 for stages IIIB–IV were used for this.

2. Disease-free life expectancy (DFLE) was extracted from life tables for each country.

3. To estimate life expectancy with metastatic melanoma (LEMM) in each age group, their future life expectancy was reduced in line with the WA annualized mortality rate with advanced melanoma using the following formula:

i.e., LEMM was reduced, over time, by increasing mortality rates to reflect disease.

YLL per patient calculation

YLL per patient was estimated by calculating the difference between the disease-free life expectancy and the life expectancy with metastatic melanoma. This was done for both men and women and for each age group; the values were age-weighted using SEER data, taking into account the distribution of patients diagnosed with melanoma in each age group. YLL per patient was calculated for men and women, respectively. YLL per patient for all patients was calculated by using the ratios of men and women in each country21, using the following formula:

Results

The projected prevalence of metastatic melanoma in each country is shown in Table 1. The projected prevalence rate was highest in Australia (23.18 per 100,000 people) and lowest in Brazil (1.04 per 100,000).

Table 1. Projected prevalence of metastatic melanoma in 2014.

Country	Prevalence	Prevalence rate per 100,000 people
Australia	4516	23.18
US	26,207	10.11
UK	4606	8.89
Netherlands	1191	8.52
Sweden	672	8.29
Germany	5924	8.16
Canada	2440	8.05
Italy	3290	6.11
France	2172	4.06
Spain	1508	3.64
Mexico	1387	1.65
Brazil	1616	1.04

Table 2 reports the distribution of age at melanoma diagnosis across the 5 year age groups used by GLOBOCAN (which were also used to estimate prevalence), which was estimated using SEER data from 2006–201025.

Table 2. Distribution of age at diagnosis, SEER, 2006–201013.

Age group	Distribution of age at diagnosis
15–39	13.3%
40–44	5.5%
45–49	9.9%
50–54	9.9%
55–59	12.1%
60–64	12.1%
65–69	10.9%
70–74	10.9%
75+	15.4%

The estimated average survival rates, annualized mortality rates and the weighted and unweighted average annualized mortality rates are reported in Tables 3–5.

Table 3. Average survival rates for metastatic melanoma.

Duration of survival, years	1	2	5	10
	0.7307	0.5316	0.3224	0.2426

Table 4. Annualized mortality rates.

Duration of period, years	0–1	1–2	2–5	5–10
	0.7307	0.7275	0.8464	0.9447

Table 5. Average annualized mortality rates.

Unweighted average	Weighted average
0.8123	0.8721

YLL per patient

Table 6 shows the estimates of YLL per patient with metastatic melanoma for men, women, and all patients by country. Across all countries, those with metastatic melanoma are estimated to die between 18.1 years (Brazil and Mexico) and 21.3 years (Australia) prematurely.

Table 6. YLL per patient for metastatic melanoma across 12 countries, 2014.

Country	YLL per patient for metastatic melanoma (years)
	Men	Women	All patients
Australia	19.9	22.7	21.3
Brazil	16.3	19.8	18.1
Canada	19.4	22.3	20.9
France	18.8	23.1	21.1
Germany	18.3	20.8	19.6
Italy	19.3	22.7	21.0
Mexico	17.2	19.0	18.1
The Netherlands	18.5	21.5	20.0
Spain	19.2	23.1	21.2
Sweden	19.4	22.0	20.7
UK	18.7	21.2	20.0
US	17.9	20.6	19.3

YLL per patient for all patients was calculated using the ratios of men-to-women in each country21.

YLL per patient also differed between the sexes: women lose more years of life from metastatic melanoma than men, which is explained by the longer natural life expectancy for women.

Note that, although there is inherently variation or uncertainty in these estimates, confidence intervals (CIs) were not part of the available data, and so 95% CIs could not be included in these results (Tables 6 and 7, Figure 1).

Figure 1. YLL per patient due to metastatic melanoma, 2014.

Table 7. Total YLL for metastatic melanoma across 12 countries, 2014.

Country	Total YLL for metastatic melanoma (years)
	Men	Women	All patients
Australia	89,926	102,617	96,368
Brazil	21,543	26,520	24,121
Canada	47,409	54,518	50,999
France	40,927	50,192	45,630
Germany	108,164	123,119	115,459
Italy	63,389	74,572	69,037
Mexico	23,782	26,373	25,071
The Netherlands	18,258	21,502	19,888
Spain	28,966	34,808	31,916
Sweden	13,041	14,798	13,937
UK	86,162	97,820	92,020
US	468,140	540,944	505,096

Total YLL

Additionally, total YLL was calculated for each country. Estimates of total YLL in men with metastatic melanoma ranged from 13,041 in Sweden to 108,164 in Germany, and in women from 14,798 to 123,119 in the same countries. The total YLL values for the US were outliers: 468,140 for men and 540,944 for women (see Table 7), which were mainly driven by the size of patient population.

Regional variations

The estimates of YLL varied across countries (Figure 1). However, there was relatively small variations in terms of YLL at per patient level (Table 6). The variations in total YLL per country were likely due to difference in prevalence rates of metastatic melanoma and population size. Such as, a higher rate in Australia was observed and the US has a substantially larger population.

Discussion

The results show that the burden of metastatic melanoma is high; YLL per patient ranged from 18.1 (Brazil and Mexico) to 21.3 years (Australia). YLL per patient for men ranged from 16.3 (Brazil) to 19.9 years (Australia) and for women from 19.0 (Mexico) to 22.7 years (Australia). Compared to the estimates of YLL per patient due to various other cancers that have been reported in the literature26^,27, melanoma has one of the highest YLL per patients. YLL per patient is generally less than 15 years for breast cancer, lung cancer, colorectal cancer, and pancreatic cancer27^,28.

Salama et al.9 have recently reported a value for YLL per patient due to melanoma of 23.8 years in the 1990s in the US, compared with our US estimate for 2014 of 19.3 years, using a similar methodology. The different estimates are likely due to the positive impact of public health programs in earlier detection of melanoma, which have helped limit the progression of melanoma to advanced stages, both before detection and after treatment is initiated.

The methods for estimating YLL have some potential limitations that should be taken into account when considering the results. The distribution of age at diagnosis was based on SEER data for 2006–201025 relating to melanoma in general, and it was assumed that this age distribution was the same across all disease stages. Assuming that age and disease stage are independent is likely to cause some bias, but the direction of the bias cannot be predicted. Older patients and patients with more advanced disease are likely to be correlated; however, although more advanced disease causes greater mortality, older patients have shorter disease-free life expectancy. The study used the best data sources that were available when the study was being conducted, so the time periods covered by the data sources varied. Age distribution and survival by stage data was only available in the US and, thus, was applied to all other countries. However, no evidence exists to show substantial differences among these measures across countries7. All assumptions in this study were validated with clinical experts at several stages of the study.

The methodology and results have some strengths. Using life tables and a consistent method across 12 countries allows comparison between countries. The study provides estimates of YLL per patient and total YLL in a consistent way for many countries and, thus, contributes to a more global understanding of the burden of metastatic melanoma.

Conclusion

Measuring the burden of illness is important to policy-makers and administrators to inform decisions about resource allocation and healthcare priorities. The YLL estimates in 12 countries for 2014 show that the burden of metastatic melanoma is substantial. The significant burden of metastatic melanoma shows the need for safer and more effective treatment options for metastatic melanoma, as well as ongoing efforts to save years of life through improved public education and screening to detect melanoma in its earlier, potentially curable phases.

Transparency

Declaration of funding

This study was funded by Amgen Inc.

Declaration of financial/other relationships

AT and CQ consulted for Amgen. ZZ and BB were employed by Amgen. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

Writing assistance was provided by PRMA Consulting.