Increasing melanoma incidence and survival trend shifts with improved melanoma-specific survival between 1990 and 2020 in Sweden

Abstract

Background

Melanoma-specific survival (MSS) is heterogenous between stages and is highly dependent on the T stage for primary localized disease. New systemic therapies for metastatic cutaneous melanoma (CM) have been introduced since 2012 in Sweden.

Objectives

To analyse the incidence and MSS time trends between 1990 and 2020 in Sweden.

Methods

Nationwide, population-based and prospectively collected clinico-pathological data on invasive CM from the Swedish Melanoma Registry (SweMR) were analysed for survival trends between 1990 and 2020 using Kaplan–Meier curves and Cox proportional hazard ratios (HRs).

Results

In total, 77 036 primary invasive CMs were diagnosed in 70 511 patients in Sweden between 1990 and 2020. The 5-year MSS [95% confidence interval (CI)] was 88.9% (88.3–89.4) for 1990–2000, 89.2% (88.7–89.6) for 2001–2010 and 93.0% (92.7–93.9) for 2011–2020. The odds ratios for being diagnosed with nodular melanoma (vs. superficial spreading melanoma) was significantly reduced by 20% (2001–2010) and by 46% (2011–2020) vs. the reference period 1990–2000. Overall, the MSS improved over both diagnostic periods (2001–2010 and 2011–2020) vs. the reference period 1990–2000 among men and women, respectively [HR_men: 2001–2010: 0.89 (95% CI 0.82–0.96) and 2011–2020: 0.62 (95% CI 0.56–0.67); HR_women: 2001–2010: 0.82 (95% CI 0.74–0.91) and 2011–2020: 0.62 (95% CI 0.56–0.70)]. The risk of death from CM was significantly lower in all age groups for both men and women in the most recent diagnostic period (2011–2020 vs.1990–2000).

Conclusions

The results emphasize the improved MSS among men and women in Sweden. The MSS improvements, specifically for the period 2011–2020, may be correlated to the introduction of new systemic therapies and are here shown for the first time in detail for Sweden.

Linked Article: Claeson Br J Dermatol 2023; 189:650–651.

Plain language summary available online

Author Video: https://youtu.be/17gz7_AwQVw?feature=shared

What is already known about this topic?

• Melanoma survival is heterogeneous, both within and between stage categories.

• Survival trends vary between countries and time periods.

What does this study add?

• Improved melanoma-specific survival for both men and women may be correlated to efficient systemic therapies, which is shown for the first time in Sweden and in one of the first European studies.

Cutaneous melanoma (CM) is one of the most aggressive types of skin cancer with a rapid increase in the incidence in Western countries and with approximately 57 000 deaths worldwide yearly.¹ The global CM burden, including both CM incidence and mortality, is expected to increase until 2040, underlining the importance of effective prevention strategies.² In Sweden, the incidence has been moving towards thinner melanomas (≤1 mm) with a better prognosis while the incidence of intermediate CM (2.1–4.0 mm) has been stable. However, additionally, there is still an increase in thick (> 4 mm) melanomas.³

Tumour thickness (according to Breslow) and ulceration of the tumour are established prognostic factors in primary CM, in combination with sentinel node status.^4,5 The prognosis varies both within and between CM stages.⁶ The majority of the patients diagnosed with invasive CM have thin CM (T1, ≤ 1.0 mm) and for those patients, the melanoma-specific survival (MSS) is generally favourable with a reported 10-year MSS of over 97%.⁷ However, the MSS is heterogeneous depending on T subcategory (stage I–II), and the 5-year MSS varies between 99% for T1a disease to 82% for T4b CM. The corresponding MSS within stage III CM varies between 32% and 93%.⁶ Still, the disease-specific survival is considered to be dependent on early detection and complete removal of the tumour.⁸

In this study, we have analysed incidence and MSS trends for different T stages between 1990 and 2020 in Sweden using nationwide population-based data from the Swedish Melanoma Registry (SweMR).

Patients and methods

Since 1990, nationwide, population-based and prospectively collected clinico-pathological data on invasive CMs have been registered in the SweMR. The SweMR has a high coverage (99%) from 1996, comparable to the Swedish Cancer Register.^9,10 The patients were observed from the date of diagnosis until censoring (diagnosis of a second CM, death from causes other than CM), emigration or end of follow-up on 31 December 2020, whichever occurred first.

Information on clinico-pathological variables was assessed from the SweMR and included sex, age (< 40, 40–59, 60–69, 70–79, ≥ 80 years), histopathological subtype of CM [superficial spreading melanoma (SSM), nodular melanoma (NM), lentigo malignant melanoma (LMM), acral lentiginous melanoma (ALM), other, missing information], tumour site (head-neck, extremeties, trunk, palm/sole/subungual, missing information), (Breslow tumour thickness (continuous), T stage (TX, T1, T1a–b, T2, T2a–b, T3, T3a–b, T4, T4a–b, missing information), and ulceration status (absent, present, missing information). The T stage was defined according to the American Joint Committee on Cancer 8th edition staging for invasive CM:⁶ TX: primary CM thickness cannot be assessed; T1: ≤ 1.0 mm, ulceration unknown or unspecified and not classified as T1a or T1b; T1a: < 0.8 mm without ulceration; T1b: < 0.8 mm with ulceration and 0.8–1.0 mm with/without ulceration; T2: 1.1–2.0 mm, ulceration unknown or unspecified; T2a: 1.1–2.0 mm without ulceration; T2b: 1.1–2.0 mm with ulceration; T3: 2.1–4.0 mm, ulceration unknown or unspecified; T3a: 2.1–4.0 mm without ulceration; T3b: 2.1–4.0 mm with ulceration; T4: > 4.0 mm, ulceration unknown or unspecified; T4a: > 4.0 mm without ulceration; T4b: > 4.0 mm with ulceration.

Statistical analyses

To evaluate clinico-pathological characteristics and survival over time for CM, the data were analysed by full calendar years and categorized into three diagnostic time periods: 1990–2000 (period 1), 2001–2010 (period 2) and 2011–2020 (period 3).

To evaluate the change of tumour characteristics over time, a multinomial logistic regression was used with the time period 1990–2000 as a reference. The results were expressed as odds ratios (ORs) with 95% confidence intervals (CIs).

Kaplan–Meier (KM) curves were used to illustrate the MSS for different T subcategories stratified by time periods. KM curves were also used to illustrate the MSS for each time period. A multivariable Cox proportional hazard regression (with adjustments for age, histopathological subtype, tumour site, tumour ulceration and tumour thickness) were applied to show the prognostic impact for each time period compared with 1990–2000 and were stratified by different clinical and histopathological characteristics. We adjusted for known confounders such as age and sex, and adjustment for tumour data was made as we wanted to investigate if the prognosis was different over time independent of the tumour stage. The stratification was made to investigate if the potential change in prognosis was general or limited to certain groups. The results are presented as hazard ratios (HRs) with 95% CIs. The level of significance was 0.05 and all P-values were two-tailed. All analyses were performed using R Statistical Software (v4.0.3; R Core Team 2020; https://www.r-project.org/). We used three sensitivity analyses to test the robustness of KM estimates of MSS and HR in the Cox regression. In the first analysis, we increased the median follow-up period to 5 years in period 3 by excluding the last 2 years and with this have a longer follow-up to allow for deaths from melanoma to occur. In the second analysis, we did not censor for the diagnosis of a second CM. In the third analysis, we obtained subhazard ratios (sHRs) derived from a Fine–Gray subdistribution hazard model where we treated death by other cause than CM as a competing risk.

Results

Table 1 describes clinical and histopathological characteristics according to diagnostic periods. In total, 77 036 primary invasive CM were diagnosed in Sweden between 1990 and 2020 divided into three diagnostic time periods, 1990–2000 (period 1), 2001–2010 (period 2) and 2011–2020 (period 3). The absolute number of diagnosed CM increased over each diagnostic periods. However, the proportion of CMs for men and women showed a small increase in the proportion of CMs for men from 49.3% to 51.8% and a decrease from 50.7% to 48.2% for women. The proportion of younger patients (age groups < 40 years and 40–59 years) diagnosed with CM decreased over time, specifically among the youngest, whereas the proportion of CM in older age groups increased (most pronounced in ≥ 80 years) over time. The absolute number of patients diagnosed with CM increased in all age groups over time. All subtypes of CM increased in absolute numbers over time with the most common type, SSM, increasing from 8181 newly diagnosed cases in period 1, to 13 054 in period 2 and 26 202 in period 3. CMs per site also increased in numbers concerning all localizations. The number of CMs per tumour thickness group increased over time. In tumours with Breslow thickness ≤ 1 mm, period 1 had 7604 new cases, period 2 had 11 667 and period 3 had 23 906 new cases. The median age at diagnosis was 61 years during period 1, 63 years during period 2 and 67 years during period 3. The median follow-up time was 16.2 years for period 1, 11.2 years for period 2 and 3.6 years for period 3. Correspondingly, the 5-year MSS was 88.9% (95% CI 88.3–89.4), 89.2% (95% CI 88.7–89.6) and 93.0% (95% CI 92.7–93.9). In the first sensitivity analysis, where we limited period 3 to 2011–2018, the MSS was unchanged, at 92.9% (95% CI 92.5–93.2) and the adjusted HR was also the same for period 3. In the second sensitivity analysis, where we did not censor for the diagnosis of a new CM, the MSS was not changed – period 1: 88.6% (95% CI 88.1–89.2), period 2: 88.8% (95% CI 88.4–89.3) and period 3: 92.4% (95% CI 92.1–92.7) – and the adjusted HR was not significantly different from the main analysis. In the third sensitivity analysis, where, instead of censoring for death by other causes, we treated it as a competing risk using the Fine–Gray subdistribution hazard model, the sHR for period 3 compared with period 1 was not significantly different compared with the original HR values for neither men (0.58, 95% CI 0.53–0.63) nor women (0.57, 95% CI 0.51–0.64).

Table 1

Clinical and histopathological characteristics according to period of diagnosis for primary cutaneous melanoma (CM) in Sweden between 1990 and 2020

Clinical and histopathological characteristics	Period of diagnosis, n (%)
	1990–2000	2001–2010	2011–2020
Total number of cases of primary CM	14 325	22 489	40 222
Sex
Men	7056 (49.3%)	11 300 (50.2%)	20 819 (51.8%)
Women	7269 (50.7%)	11 189 (49.8%)	19 403 (48.2%)
Age at diagnosis (years)
< 40	2054 (14.3%)	2497 (11.1%)	3153 (7.8%)
40–59	4709 (32.9%)	6699 (29.8%)	10 952 (27.2%)
60–69	2734 (19.1%)	5096 (22.7%)	8828 (21.9%)
70–79	3047 (21.3%)	4490 (20.0%)	10 326 (25.7%)
≥ 80	1781 (12.4%)	3707 (16.5%)	6963 (17.3%)
Histopathological subtype
LMM	925 (6.5%)	1542 (7.0%)	2731 (6.8%)
SSM	8181 (57.1%)	13 054 (58.0%)	26 202 (65.1%)
NM	2733 (19.1%)	4190 (18.6%)	5157 (12.8%)
ALM	226 (1.6%)	271 (1.2%)	367 (0.91%)
Other	2105 (14.7%)	2956 (13.1%)	4162 (10.3%)
Missing	155 (1.1%)	476 (2.1%)	1603 (4.0%)
Tumour site
Head/neck	1897 (13.2%)	3008 (13.4%)	5050 (12.6%)
Extremities	5463 (38.1%)	9274 (41.2%)	17 519 (43.6%)
Trunk	6657 (46.5%)	9735 (43.3%)	17 178 (42.7%)
Palm/sole/subungual	289 (2.02%)	312 (1.4%)	429 (1.1%)
Missing	19 (0.13%)	160 (0.71%)	46 (0.11%)
Breslow thickness (mm)
≤ 1.0	7604 (53.1%)	11 667 (51.9%)	23 906 (59.4%)
1.1–2.0	2827 (19.7%)	4436 (19.7%)	7275 (18.1%)
2.1–4.0	2023 (14.1%)	3218 (14.3%)	4529 (11.3%)
> 4.0	1336 (9.3%)	2556 (11.4%)	4006 (10.0%)
Missing	535 (3.7%)	612 (2.7%)	506 (1.3%)
T stage
T1	1556 (10.9%)	1189 (5.3%)	716 (1.8%)
T1a	3706 (25.9%)	6944 (30.9%)	17 102 (42.5%)
T1b	2342 (16.3%)	3534 (15.7%)	6088 (15.1%)
T2	789 (5.5%)	553 (2.5%)	318 (0.8%)
T2a	1522 (10.6%)	3087 (13.7%)	5690 (14.1%)
T2b	516 (3.6%)	796 (3.5%)	1267 (3.2%)
T3	472 (3.3%)	348 (1.5%)	213 (0.5%)
T3a	750 (5.2%)	1464 (6.5%)	2389 (5.9%)
T3b	801 (5.6%)	1406 (6.3%)	1927 (4.8%)
T4	273 (1.9%)	288 (1.3%)	169 (0.4%)
T4a	315 (2.2%)	709 (3.2%)	1175 (2.9%)
T4b	748 (5.2%)	1559 (6.9%)	2662 (6.6%)
Missing	535 (3.7%)	612 (2.7%)	506 (1.3%)
Ulceration
No	7753 (54.1%)	15 024 (66.8%)	31 807 (79.1%)
Yes	2521 (17.6%)	4330 (19.3%)	6621 (16.5%)
Missing	4051 (28.3%)	3135 (13.9%)	1794 (4.5%)

ALM, acral melanoma; LMM, lentigo maligna melanoma; NM, nodular melanoma; SSM, superficial spreading melanoma

The ORs for CM per age category (vs. <40 years) increased markedly in the second and third periods, compared with the reference time period 1 (Table 2). For example, for age group 60–69 years, the odds for CM increased and were 1.63 times higher in period 2 and 2.35 higher in period 3, respectively, vs. period 1. However, the odds for being diagnosed with NM were significantly reduced by 0.80 times (period 2) and by 0.54 times (period 3 vs. period 1) in relation to period 1 with SSM being the reference (Table 2).

Table 2

Multinominal logistic regression of clinical and histopathological characteristics of primary cutaneous melanoma (CM) diagnosed between 1990 and 2020 in Sweden

Clinical and histopathological characteristics	2001–2010,a OR (95% CI)	2011–2020,a OR (95% CI)
Intercept	1.34 (1.23–1.45)	2.28 (2.11–2.46)
Sex
Men	1 (ref)	1 (ref)
Women	0.97 (0.92–1.02)	0.91 (0.87–0.95)
Age at diagnosis (years)
< 40	1 (ref)	1 (ref)
40–59	1.22 (1.13–1.32)	1.60 (1.48–1.72)
60–69	1.63 (1.49–1.78)	2.35 (2.17–2.55)
70–79	1.26 (1.15–1.37)	2.52 (2.32–2.73)
≥ 80	1.79 (1.62–1.97)	3.18 (2.90–3.49)
Histopathological subtype
SSM	1 (ref)	1 (ref)
LMM	0.95 (0.85–1.06)	0.72 (0.65–0.80)
ALM	0.78 (0.61–0.99)	0.61 (0.49–0.76)
NM	0.80 (0.74–0.87)	0.54 (0.50–0.58)
Other	1.14 (1.05–1.24)	1.01 (0.94–1.10)
Tumour site
Trunk	1 (ref)	1 (ref)
Head/neck	1.10 (1.01–1.19)	1.07 (0.99–1.16)
Extremities	1.23 (1.16–1.30)	1.36 (1.29–1.43)
Palm/sole/subungual	0.70 (0.57–0.87)	0.62 (0.50–0.75)
T stageb
T1a	1 (ref)	1 (ref)
T1b	0.97 (0.90–1.04)	0.75 (0.70–0.80)
T2a	1.09 (1.01–1.17)	0.84 (0.78–0.90)
T2b	0.84 (0.74–0.94)	0.56 (0.50–0.62)
T3a	1.07 (0.96–1.19)	0.76 (0.69–0.83)
T3b	0.99 (0.89–1.10)	0.60 (0.54–0.66)
T4a	1.23 (1.06–1.43)	0.89 (0.77–1.02)
T4b	1.16 (1.03–1.30)	0.90 (0.81–1.00)

^aThe period 1990–2000 was the reference group in all analyses; the table is based on CM cases. ^bTumours with missing information regarding ulceration status were excluded. 

ALM, acral melanoma; CI, confidence interval; LMM, lentigo maligna melanoma; NM, nodular melanoma; OR, odds ratio; SSM, superficial spreading melanoma

The MSS per diagnostic period for T stage is shown in the KM curves (Figure 1). The MSS for men and women, respectively, is shown in Figures S1 and S2 (see Supporting Information). As shown in these KM curves, the absolute differences were significantly better for women. Overall, the MSS improved over both diagnostic periods (periods 2 and 3) vs. the reference period 1 among men and women, respectively [HR_men: period 2: 0.89 (95% CI 0.82–0.96) and period 3: 0.62 (95% CI 0.56–0.67); HR_women: period 2: 0.82 (95% CI 0.74–0.91) and period 3: 0.62 (95% CI 0.56–0.70)] (Table 3 and Table S1; see Supporting Information). The risk of death from CM was significantly lower in all age groups for both men and women in the most recent diagnostic period (period 3 vs. period 1; Table 3). For diagnostic periods 2 and 3, respectively, the MSS significantly improved for CMs located on the trunk for both sexes. During the last period the MSS improved significantly for all T stages for men. A similar pattern was shown for the last period also for women, except for thicker melanomas, i.e. T2b, T3b and T4a CM.

Figure 1

Kaplan–Meier estimates showing melanoma-specific survival (MSS) in Sweden according to T stage for three diagnostic periods: (a) 1990–2000; (b) 2001–2010; (c) 2011–2020; and (d) MSS according to diagnostic periods.

Table 3

Melanoma-specific survival in primary cutaneous melanoma cases (n = 66 306) for men and women in Sweden according to diagnostic periods between 1990 and 2020 and stratified by clinical and histopathological characteristics

Clinical and histopathological characteristics	Mena–c		Womena–c
	HR for period 2001–2010 vs. 1990–2000	HR for period 2011–2020 vs. 1990–2000	HR for period 2001–2010 vs. 1990–2000	HR for period 2011–2020 vs. 1990–2000
All patients	0.89 (0.82–0.96)	0.62 (0.56–0.67)	0.82 (0.74–0.91)	0.62 (0.56–0.70)
Age at diagnosis (years)
< 40	1.01 (0.73–1.38)	0.45 (0.29–0.71)	1.02 (0.72–1.44)	0.70 (0.45–1.09)
40–59	0.85 (0.73–0.99)	0.56 (0.46–0.68)	0.71 (0.58–0.88)	0.55 (0.43–0.70)
60–69	1.03 (0.87–1.22)	0.63 (0.52–0.77)	0.77 (0.62–0.98)	0.52 (0.40–0.68)
70–79	0.80 (0.69–0.94)	0.60 (0.51–0.70)	0.84 (0.68–1.02)	0.57 (0.45–0.71)
≥ 80	0.90 (0.73–1.12)	0.72 (0.58–0.89)	0.88 (0.70–1.10)	0.78 (0.61–0.98)
Tumour site
Head/neck	1.01 (0.82–1.25)	0.82 (0.65–1.03)	0.99 (0.75–1.32)	0.61 (0.44–0.83)
Extremities	0.83 (0.71–0.98)	0.62 (0.53–0.74)	0.86 (0.74–1.00)	0.68 (0.58–0.80)
Trunk	0.87 (0.78–0.97)	0.55 (0.49–0.63)	0.72 (0.60–0.86)	0.56 (0.46–0.68)
Palm/sole/subungual	1.13 (0.73–1.75)	0.64 (0.38–1.06)	0.75 (0.48–1.18)	0.67 (0.41–1.09)
Tumour thickness (mm)
≤ 1.0	0.91 (0.72–1.15)	0.57 (0.43–0.76)	0.70 (0.55–0.89)	0.49 (0.36–0.67)
1.1–2.0	0.80 (0.67–0.95)	0.57 (0.47–0.71)	0.74 (0.59–0.92)	0.55 (0.42–0.73)
2.1–4.0	0.86 (0.74–0.99)	0.64 (0.54–0.75)	0.94 (0.78–1.14)	0.74 (0.60–0.91)
> 4.0	0.99 (0.87–1.14)	0.66 (0.57–0.77)	0.90 (0.75–1.08)	0.68 (0.56–0.82)
T staged
T1a	0.87 (0.61–1.25)	0.59 (0.39–0.90)	0.76 (0.51–1.12)	0.53 (0.32–0.86)
T1b	0.91 (0.67–1.24)	0.53 (0.36–0.78)	0.62 (0.45–0.84)	0.45 (0.30–0.68)
T2a	0.78 (0.63–0.96)	0.54 (0.42–0.70)	0.80 (0.61–1.04)	0.51 (0.36–0.72)
T2b	0.83 (0.60–1.15)	0.62 (0.43–0.90)	0.65 (0.44–0.96)	0.67 (0.44–1.04)
T3a	0.95 (0.76–1.20)	0.70 (0.54–0.90)	0.98 (0.74–1.29)	0.61 (0.44–0.84)
T3b	0.80 (0.67–0.96)	0.61 (0.50–0.74)	0.91 (0.70–1.19)	0.85 (0.64–1.14)
T4a	0.90 (0.69–1.18)	0.55 (0.40–0.74)	1.31 (0.86–2.00)	0.89 (0.57–1.37)
T4b	1.02 (0.87–1.20)	0.69 (0.59–0.82)	0.80 (0.65–0.99)	0.62 (0.50–0.76)

^aReference group: diagnostic period 1990–2000. ^bAdjusted for age, histopathological subtype, tumour site, tumour ulceration, tumour thickness; cases with missing data on any of these variables are excluded from the analysis. ^cStratified by age at diagnosis, tumour site, tumour thickness and T stage. ^dTumours with missing information regarding ulceration status were excluded.

HR, hazard ratio

Discussion

The present study was based on data from the population-based, nationwide SweMR, including patients with invasive CM diagnosed between 1990 and 2020. This report therefore uniquely reflects CM trends for different T stages over the last few decades on a population level in a country with a top-six world incidence of CM.¹¹ Importantly, we found a significant improvement in MSS for both men and women and in all age groups for the most recent time period. However, trends of improvements varied between diagnostic periods for different tumour sites and T stages.

Immune checkpoint inhibitors (ICIs) and targeted therapies (TTs) have ushered in a new era of effective systemic treatment for patients with metastatic CM, showing significantly improved clinical outcomes in both stage III and stage IV disease.^12–14 In Sweden, ICIs were approved and recommended in clinical routine from 2012 and TTs from 2013. Our results emphasize that long-term survival trends in Sweden have a temporal association with the approval of novel treatments for metastatic disease, which is in line with results from a recent study of CM mortality rate trends in the USA by Kahlon et al.¹⁵ In their report, they have shown that a significant reduction in mortality rate trends between 2013 and 2017 was seen for the first time in recent decades. These findings might be explained by the introduction of CM drugs (i.e. ICI and TT) in 2011 in the USA. Moreover, Tichanek et al. have previously shown increased CM survival from after the year 2000 up to the last period 2016–2020 in the Nordic countries, with the highest survival results for Denmark.¹⁶ The finding was most likely the effect of the introduction of ICI and TT treatments. However, clinico-histopathologcial details were not analysed. Accordingly, we report a general increase of the MSS over the last three decades in Sweden. This finding may also be correlated to a higher proportion of older individuals who are diagnosed with CM but survive longer, due to better basic health status, better primary management and better systemic treatments. A clear variation in CM mortality trends between countries, sexes and different age groups has been found in several other settings as described by Yang et al.¹⁷ Moreover, sex-specific survival differences are established in all stages of CM.^18,19 In Sweden, Lyth et al.²⁰ have reported improved MSS for men for the time period 2007–2011 vs. 1997–2001 and that women were diagnosed with thinner invasive tumours over these time periods. Moreover, although the improvements in the relative survival differences were comparable between sexes in the current study, the absolute differences were significantly better for women.

In several European countries, the 5-year relative CM survival rates have increased, which is related to the detection of early-stage CM with a more favourable prognosis.^21,22 Importantly, patients with T1 CM have always had a superior prognosis.^3,23 Therefore, it is challenging to further improve the MSS for this stage category of CM, as the survival rate has reached almost 100% in Europe. However, our results demonstrated a significantly lower proportion of NMs (as well as a lower proportion of both ALMs and LMMs), compared with the 1990s. During this analysis, SSM was used as a reference. This is the most common subtype and has increased both in proportion compared with other types, but also in absolute numbers of new cases diagnosed [from 8181 (57.1%) in period 1 to 26 202 (65.1%) in period 3, Table 1]. The proportion of increased SSMs impact the results of a significantly lower proportion of NMs, compared with the 1990s, which also could explain the present MSS trends in Sweden.²⁴ NMs, associated with a poor survival, are more often diagnosed among men and elderly individuals.²⁵ In addition, the CONCORD Working Group have to date performed the largest analysis of MSS between 2000 and 2014, specifically providing a global perspective on CM survival.²⁶ According to their analyses the distribution of CM morphology varies by continent and country and a lower survival for NM and ALM was found to persist after adjustments. Brunssen et al. showed that disparities in the relative survival between CM histological subtypes in Germany were significantly mediated by tumour thickness with the effect that the CM relative survival increased by the changes in T-stage distribution.²⁷ In Sweden, to prevent patient care inequalities and to minimize local differences in CM evaluation and management, CM treatment guidelines have been implemented on a national level. Moreover, a more widespread use of dermatoscopy since the early 2000s has been implemented in Sweden, further improving early detection and diagnoses of thinner CMs.^28–30 Thus, novel CM treatments and improved secondary prevention activities might explain the observed improved CM survival in our report.^23,31

The main strengths of the present study are the prospectively collected high-quality population-based data from the high-coverage national SweMR, with detailed clinico-­pathological information on all invasive CMs diagnosed in Sweden with complete follow-up. This allows the results to be generalizable to other populations dominated by fair-skinned individuals. As highlighted by Di Carlo et al.,²⁶ high-quality data from cancer registries are the gold standard for outcome comparisons, which can be provided by the SweMR.

Common for all registry-based studies, a possible weakness is the incomplete registration of some variables, which to some extent might reduce the study cohort. However, due to the high number of complete registrations, we do not think the missed data would have changed our findings.

In conclusion, population-based cancer survival analyses are crucial to evaluate the overall effectiveness of CM prevention activities as well as new treatment strategies. We found similar improvements in MSS between men and women over the time periods, which is promising, although the absolute survival effect was correlated to a female advantage. As one of the first European studies and the first Nordic study including detailed clinico-pathological data, we report an effect on the MSS that is most likely associated with the introduction of effective systemic therapies for patients diagnosed with high-risk CM that developed metastatic disease. The results highlight the positive survival trends in a country with increasing CM incidence as well as the important role of population-based high-quality data registration, including follow-up of systemic therapy outcomes in CM.

Acknowledgements

We thank the patients contributing to the Swedish national registries, the Swedish Melanoma Study Group and the national cancer centres collecting data to the SweMR.

Funding sources

To H.E.: The Swedish Research Council, The Swedish Cancer Society, The Radiumhemmet Research Fund, Region Stockholm, Cronqvist Foundation, Cancer- and Allergy Foundation, Folkfonden. To K.N.: John and Inga Hain Foundation, WelanderFinsen Foundation, grants from the Swedish state under the agreement between the Swedish government and the county councils, the ALF-agreement, The Gyllenstiernska Krapperup foundation. No sponsor had any role in study design, data collection, data analysis, manuscript preparation or publication decisions.

Data availability

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Ethics statement

The study was approved by the Ethical Review Board (Dnr 99160/1999; approved on 8 June 1999).

Supporting Information

Additional Supporting Information may be found in the online version of this article at the publisher’s website.

Author notes

Conflicts of interest K.N. has received speaker honoraria from Galderma Sweden, LEO pharma, Novartis Sweden and UCB Pharma, and has served on one advisory board for MSD. K.I. has received a speaker honorarium from Pierre Fabre. All other authors declare no conflicts of interest.