Skin cancers are divided into 2 categories according to epidemiological and prognostic differences: cutaneous melanoma, and nonmelanoma skin cancer (NMSC). The 2 main types of NMSC are basal cell carcinoma (BCC) and squamous cell carcinoma (SCC).

The incidence of cutaneous melanoma continues to rise in Europe, with considerable geographical differences in mortality.1,2 Owing to the higher incidence of NMSC compared to other skin tumors and the difficulty of quantifying the condition,3 NMSC tumors are not consistently reported in cancer registries.

Both cutaneous melanoma and NMSC represent a significant economic burden to the Spanish National Health Service. Thus, it is essential to know the true burden of disease attributable to skin cancer in Spain.

The main objective of this systematic review is to describe the incidence and mortality of skin cancer in Spain, including cutaneous melanoma, BCC, SCC, and Merkel cell carcinoma.

The search strategy identified 1694 articles. Of these, 89 were selected on the basis of a reading of the abstract. After a review of the full texts, 17 were selected for analysis. To these we added a further 15 articles identified by the search of the “gray literature” outside of PubMed and EMBASE. In total, 32 studies were included in the review (Fig. 1).

The selected studies, which are listed in Table 2, analyze data from different provinces, regions, or the whole country relating to the period from 1989 to 2015.

The score for methodological quality ranged from 2 to 9 points. Twenty-four studies (75%) achieved a high quality score, but none achieved the maximum score of 10. Most of the studies (84.8%) included histological confirmation. (Table 3)

Basal Cell Carcinoma

The overall incidence for BCC in the meta-analysis was 113.05 (95% CI, 89.03-137.08) cases per 100000 person-years for the studies based on the registration methodology normally used by registries (in which only 1 tumor with histological confirmation is counted per person) (Fig. 2). Only 1 study calculated the incidence using both clinical and histological criteria and counting tumors rather than patients, reporting an incidence of 253.23 (95% CI, 273.01-269.45) cases per 100000 person-years.6 The incidence rates for these studies standardized to the world population are shown in Fig. 1, Appendix 1.

Figure 2.

Incidence rate for basal cell carcinoma. All ages.

(0.27MB).

Squamous Cell Carcinoma

The overall incidence rate for SCC was 38.16 (95% CI, 29.82-46.49) cases per 100000 person-years in the meta-analysis (Figs. 2 and 3). The incidence rate standardized to the world population is shown in Figure 2, Appendix 1.

Figure 3.

Incidence of squamous cell carcinoma. All ages.

(0.22MB).

The mortality rate for NMSC was reported in 2 studies: 1 based on data from Navarre,5 with a mortality rate of 0.90 (95% CI, 0.54-1.25) cases per 100000 person-years; and the other on data for Spain as a whole, with a mortality rate of 1.10 (95% CI, 1.00-1.20) cases per 100000 person-years (Fig. 3, Appendix 1). Only the Navarre study reported the mortality rate standardized to the world population (Fig. 4, Appendix 1).

Melanoma

Melanoma has been the subject of more studies than any other skin cancer (n = 24). The meta-analysis yielded a crude incidence rate of 8.82 (95% CI, 7.59-10.04) cases per 100000 person-years (Fig. 4). Of the studies in which the diagnosis of melanoma was confirmed histologically, the study reporting the highest incidence rate was in Marbella (Malaga) between 2002 and 2011, with a rate of 17.53 (95% CI, 15.64-19.41) cases per 100000 person-years.9 The lowest incidence was reported in Zaragoza19 in 1986, with 3.63 (95% CI, 2.33-4.92) cases per 100000 person-years. The only study based on both clinical and histological confirmation reported an incidence rate in Guipúzcoa of 10.15 (95% CI, 9.08-11.23) cases per 100000 person-years for the period from 1998 to 2002.10 Based on both histological and clinical data, Cabanes et al.23 calculated a crude incidence rate for Spain in 2006 of 1.90 (95% CI, 1.77-2.03), but in a study with an high risk of bias. The data standardized to the world population are shown in Figure 5, Appendix 1.

Figure 4.

Incidence of melanoma. All ages.

(0.56MB).

The overall crude mortality rate for melanoma was 2.17 (95% CI: 1.78-2.55) cases per 100000 person-years for studies based on histological diagnoses in the provinces of Navarre,5 Girona,8 and Tarragona12 (Fig. 6, Appendix 1). Cayuela et al.,7 who analyzed data from death certificates, identified a crude mortality rate for Spain in 2001 of 1.82 (95% CI, 1.69-1.95) cases per 100000 person-years. The mortality rates for melanoma standardized to the world population are shown in Figure 7, Appendix 1.

The present systematic review included 32 studies on the incidence and mortality of skin cancer in Spain. This is the largest review on this subject to date.

The main objective of this review was to determine the burden of disease attributable to skin cancer in order to ensure adequate planning of available resources. The work was carried out as part of the White Book on Skin Cancer being prepared by the Foundation of the Spanish Academy of Dermatology and Venereology (FAEDV). In this discussion we will focus on the crude incidence rate, since the primary objective of age-standardized incidence rates is to allow comparison between groups with differing age structures, but such rates are less useful for planning purposes.

The results for BCC differ depending on the method used to calculate incidence. The overall crude incidence rate in Spain was 113.05 cases per 100000 person-years when only histologically confirmed tumors were included and only the first tumor for each patient was counted (the usual methodology used by cancer registries). However, when the primary objective is to evaluate the demand for the care of BCC it is preferable to also include tumors not confirmed histologically and all the tumors occurring in any patient. The only study that used this methodology was carried out in the Barcelonès Nord area and reported a crude incidence rate of 253.23 cases per 100000 person-years,6 more than twice the rate estimated by earlier studies. It is likely that this figure more reliably represents the real need for care.

The incidence rate of the meta-analysis standardized to the world population was 55.87 cases per 100000 person-years, similar to rates reported for Scotland, United Kingdom, and Germany, and much lower than the rate reported for Finland.3

The crude incidence rate for SCC was 38.16 (95% CI, 29.82-46.49) cases per 100000 person-years. As this result was also obtained using the methodology typical of registries, it is very likely that it underestimates the need for care.

The incidence rate standardized to the world population based on the results of the meta-analysis of available studies was 13.38 per 100000 person-years. These incidence rates are relatively low when compared to those of countries such as Germany and Slovakia, and are lower than countries such as Sweden, Switzerland and Wales.3

The mortality rate for NMSC is low, and predominantly associated with SCC, particularly because of the existence of a high-risk subgroup.25

In the case of melanoma, the crude incidence rate upon meta-analysis was 8.82 cases per 100000 person-years. Differences can be observed between the results obtained several decades ago, 3 to 4 cases per 100000 person-years,14,17–19 and those reported by studies carried out in the 1990s and later, with rates of 7 cases per 100000 person-years and higher.8,12,15,22,24,26 These differences may reflect an increase in the incidence of melanoma27 and the greater risk of bias in older studies. Of particular interest is the study by Fernández-Canedo et al.,9 who reported a high incidence rate of 17.56 cases per 100000 person-years in a study of Marbella in the province of Malaga. This result is attributed to the high proportion of people from central and northern Europe in the population of the study area.

The standardized incidence rate in Spain is low compared to other European countries, reflecting the recognized difference in Europe between north and south and between east and west, with a rate that varies from 6 cases per 100000 person-years in eastern Europe to 10 in southern Europe and up to 19 in northern Europe.28 The incidence rate in Spain is much lower than the more than 20 cases per 100000 person-years reported in the USA,29 and more than 60 reported for New Zealand and Australia.30

On the basis of the only 3 studies available, the crude mortality rate upon meta-analysis was 2.17 cases per 100000 person-years.5,8,12 While relatively low at present, this rate may continue to increase in coming years, particularly in the older population.7,12 Such an increase would be in contrast to the trend in more northern European countries, where the mortality rates have tended to plateau and even decrease. Among the reasons for this difference are the fact that the Spanish population is older and that melanomas occurring at a later age tend to be thicker and have a more aggressive phenotype.31

Finally, in the only study on Merkel cell carcinoma, Vilar-Coromina et al.21 reported an incidence rate for this tumor of 0.28 cases per 100000 person-years calculated on the basis of data from the Cancer Registry in the province of Girona. This rate is similar to that obtained in similar studies in Denmark32 and Finland,33 but lower than that reported in Australia, where the incidence rate adjusted to the US population is 0.8 cases in western Australia and 1.6 cases per 100000 person-years in north-eastern Australia (Queensland), rates 10 times those reported in European studies.34 This epidemiological difference is probably explained by the difference in population phenotype and levels of exposure to ultraviolet radiation.35

The present systematic review provides a good account of the incidence and mortality of skin cancer in Spain in recent years. The study has, however, a number of limitations. The main limitation of the study is that most of the data on which it is based were obtained using the methodology generally used by cancer registries, which is appropriate for the study of etiological factors but tends to underestimate the need for care, especially in the case of BCC, in which patients often present multiple tumors and are treated without histological confirmation. In light of our results, the approximate underestimation could be 50%. Other limitations of the review are that the studies analyzed do not cover the whole population of Spain or all reported tumors (no data are included on neural, hematolymphoid, adnexal, or soft tissue tumors), and that the data refer to different time periods. Finally, not all the studies based on data from cancer registries explicitly stated that in situ tumors were excluded, but such an exclusion was assumed for the purposes of this analysis.

In conclusion, the data indicate that BCC and SCC are very common tumors in Spain. At least in the case of BCC, the need for care is higher than the level that could be determined from registry data. The incidence rate for cutaneous melanoma in Spain is low compared to other European countries. Finally, the incidence rate for Merkel cell carcinoma is low and similar to that reported for other countries in Europe.

This work was funded by the AEDV Foundation and forms part of the White Paper on Skin Cancer in Spain.

The authors declare that they have no conflicts of interest.