Trends during a half century in relative squamous cell carcinoma distribution by body site in the Swedish population: Support for accumulated sun exposure as the main risk factor

Authors


Bernt Lindelöf, M.D., Ph.D., Department of Dermatology, Karolinska University Hospital, SE-17176 Stockholm, Sweden. Email: bernt.lindelof@karolinska.se

ABSTRACT

There is a strong relationship between squamous cell carcinoma (SCC) and exposure to ultraviolet radiation in terms of accumulated exposure. In this study, data from the Swedish Cancer Registry are surveyed to discern a reflection of behavioral and societal changes in relative distribution of SCC by body site. Data for the time period 1960–2004, including a total of 66 221 cases (56 669 people) were analyzed by body site for age and gender cohorts. The age-standardized (European population) incidence per 100 000 of SCC in the year 2004 was 30.4 in males and 15.4 in females. In the year 1960, the corresponding incidences were 7.7 and 3.8; that is, SCC has become four times more frequent in Sweden for both sexes during this period. The standardized incidence of SCC increased on all body sites except eyelids (men and women) and ears (women). Head tumors dominated among patients aged 70 years or more and diagnosed 1960–1964. Among patients less than 70 years old at diagnosis in 2000–2004, tumors of the trunk and limbs dominated. A relative increase of tumors of the scalp and neck was observed in all age groups (men), and of tumors of the trunk and upper limbs in all age groups and both sexes except among patients aged more than 90 years of age. In contrast, a relative decrease of tumors on the face (including the ears) was seen in all age groups. The relative increase of SCC of the trunk and upper limbs is a plausible reflection of intentional tanning.

INTRODUCTION

There is strong epidemiological evidence of a dose-response relationship between squamous cell carcinoma (SCC) and exposure to ultraviolet radiation (UVR) in terms of accumulated exposure,1 and exposure during childhood and adolescence,2 both from the sun and from sunbeds.3 This relationship applies in particular to sun sensitivity (skin type), and is also congruent with the latitudinal dependency of SCC.4 In the wavelength of UVR, the most critical action spectrum in the genesis of SCC is between 290–340 nm,5,6 thus ultraviolet (UV)-B and well into the wavelengths of UV-A (as emitted by sunbeds). The relationship between tanning behavior and relative body site distribution of SCC has been observed in several studies.7–11 As all kinds of exposure (intentional and unintentional, intermittent and continuous) contribute to accumulated UV exposure on a specific body site, the body site distribution of SCC may give a clue to patterns of both intentional and unintentional exposure to UVR, at least in older age groups, due to induction times which presumably may stretch over several decades. The low incidence of SCC in northern countries a half century ago indicates intentional and unintentional sun protective behavior, less leisure time, and absence of sunbeds. The change in body site distribution of SCC may reflect major behavior changes in Sweden during the latter part of the 20th century, the domination of outdoor work (farming and fishing industries) which has shifted to indoor work; and, on the other hand, intentional outdoor tanning patterns which have evolved in Sweden during the same time period (sun-seeking behavior with horizontal postures, fewer clothes, outdoor sports such as golf, soccer, alpine and aquatic sports, travel to seaside resorts, and the introduction of cosmetic sunbeds in 1978). We surveyed data from the Swedish Cancer Registry (SCR) of SCC by body site for age cohorts 1960–2004, in an attempt to discern a reflection of major behavioral and societal changes in Sweden relevant to UV exposure.

METHODS

Geographical data, study base and study population

Sweden in Northern Europe is situated between the 54th and the 69th parallels, with yearly ambient solar UVR of 270–420 kJ/m2 skin surface (horizontal surface) depending on the latitude.12 The recorded meteorological spring periods (diurnal mean temperature, 0–10°C) are short, usually lasting April–May.13

Of the 9 million inhabitants (2004), 85% live south of the 60th parallel, and almost half of those in one of the three metropolitan areas (Stockholm, Gothenburg, Malmö).14 During the 1990s, between one-third and half (depending on population segment) of the adolescents and young adults living in densely populated and/or metropolitan areas have been exposed to sunbeds.15–17 Regarding mobility, more than 7 million airline departures to foreign destinations were registered in 2004, of which 860 000 (12%) were departures to seaside resorts south of the 40th parallel.18

Cancer registry

In 1958, the reporting of cancers upon diagnosis to the SCR became compulsory by law. The register is based on a 10-digit national registration number (NRN) unique for every Swedish resident. The NRN is made up of birth date (year, month, day), plus two digits indicating birthplace, one indicating gender, and a control number. The NRN enables linkage of exposure and outcome data and can be used to test specific etiologic hypotheses relevant for epidemiological surveillance and public health promotion.

Malignancies are reported by clinicians, pathologists and cytologists, with most cancers thus being reported twice.19 The coverage of the SCR was reported to be 96% in the late 1970s, and close to 100% in 2000.19,20 At the time of the study, basal cell carcinoma (BCC) was not registered. Thus, the registered non-melanoma skin cancers are mainly SCC (92.1% SCC, 6.7% SCC/BCC type mixed, and 1.2% other primary malignant tumors of the skin).8

Statistical methods

The SCC cases (66 221 cases, 56 669 people), diagnosed 1960–2004, and coded according to the seventh revision of the International Classification of Diseases (ICD-7 code 191.0, 191.1–4 head, 191.5 trunk, 191.6 arms and shoulders, 191.7 legs and hips, 191.8–191.9 multiple parts or unspecified parts) were classified according to gender, age, calendar period (5-year intervals), and body site of tumor. The sites of the face and eyelids were joined in one category (191.1, 191.3) due to few cases on eyelids. The overall trend is presented by the age-standardized incidence per 100 000 (European population).

The relative site-specific distribution of diagnosed SCC was calculated for each gender and age group, and thus trends assessed by incidence per site, relative site distribution per age group and calendar period. The patients were divided into the following age groups according to age at diagnosis: 0–49 years; 50–69 years; 70–89 years; and 90 years or more. For each of the age-at-diagnosis groups, the absolute and relative distribution of body site over time was analyzed. When using age groups as narrow as the ones chosen, age-standardization was not considered necessary.

RESULTS

The age-standardized incidence per 100 000 of SCC in the year 2004 was 30.4 in males and 15.4 in females. In the year 1960, the corresponding incidences were 7.7 and 3.8; that is, SCC has become four times more frequent in Sweden for both sexes during this period. SCC increased on all body sites except on eyelids (men and women, data not shown), and ears (women) (Fig. 1a,b). The head tumors dominated among patients age 70 years or more and diagnosed 1960–1964, and tumors of the trunk and limbs dominated in patients diagnosed 2000–2004, and less than 70 years old at diagnosis.

Figure 1.

Incidence (cases/100 000, standard population) of squamous cell carcinoma by body site, Sweden 1960–2004.

Regarding the relative site distribution, an increase of tumors of the trunk and upper limbs was observed in all age groups and both sexes except among patients aged over 90 years, whereas there was a relative decrease of head tumors, except for those of the scalp in men. No increase was observed for tumors on lower limbs. Among patients less than 50 years of age diagnosed 2000–2004, 57% (men) and 65% (women) of the SCC occurred on the trunk and the limbs, compared to 34% and 44% respectively among patients less than 50 years of age diagnosed 1960–1964. In contrast, a relative decrease of tumors of the ears was observed in all age groups, particularly among men less than 60 years of age. However, across the lifespan, SCC of the upper limbs and trunk decreased relative to the head sites. This pattern was similar throughout the observation period (Fig. 2a–d).

Figure 2.

Figure 2.

Relative distribution of squamous cell carcinomas by age group and body site, Sweden 1960–2004.

Figure 2.

Figure 2.

Relative distribution of squamous cell carcinomas by age group and body site, Sweden 1960–2004.

The incidence of ear tumors has been the next to highest among men but extremely low in women during the whole registration period (Fig. 1a,b). Among patients aged less than 50 years at diagnosis, the relative body site distribution was similar between sexes, whereas tumors of the ears and scalp took a higher share among males in the other age groups, as did other parts of the face in females (Fig. 2a–d).

DISCUSSION

The main finding in this study was the relative increase of tumors of the trunk and upper limbs, and the relative decrease of SCC on head sites. As SCC occur on sun-exposed body sites, this study may give a clue to societal and behavioral factors related to the relative body site distribution of SCC during the last four decades. Lifestyles related to urbanization quickly became established during the period 1960–2000 when the percentage of the Swedish population living in densely populated areas rose from 60% to 85%.14

In men, the standardized incidence increase of ear tumors was not as rapid as that of the rest of the face (Fig. 1a). Possibly, protection of the head with poor protection of the ears (peaked caps, hats), was more common among outdoor workers and other persons with high sun exposure at the beginning of the 20th century, compared to people of today whose heads (i.e. scalp, neck, face, ears) are often altogether exposed to UVR due to outdoor work, sunbed use or leisure time (Figs 3,4). Removal of a cap or hat would cause a small increase of UVR exposure to the ears but a large increase in UVR exposure to the scalp, neck, forehead. However, tumors of the ears made a fair share of all male tumors, probably due to a generally shorter haircut compared to women. The similarity in body site distribution between men and women diagnosed under the age of 50 years may be a result of similar outdoor tanning habits and sunbed use, even though sunbed use occurs less frequently among men.15 On the other hand, changes in male hair fashion such as long hair may yield UVR protection. Also, because the protection of today's outdoor workers by roofed-over tractors and road-making machinery is widespread, most overexposure to UVR could probably be ascribed to leisure time activities outdoors.

Figure 3.

Swedish farmers in the early 20th century.

Figure 4.

Swedish golfers in the late 20th century.

The standardized incidence increase of tumors of the trunk and the upper limbs is roughly the same for men and women, which may reflect increased UVR exposure of these sites as a result of changed tanning habits (fewer clothes, intentional outdoor tanning, sunbed use), implying sitting and lying postures.21–23 SCC have not increased to the same extent on the lower limbs. A possible cause is that the legs have been covered by trousers (men, young women) or long skirts (older women). However, there is an exception; in women aged 50–69 years, SCC on the lower limbs increased proportionally in relation to the head sites. A possible explanation is the fashion of short skirts during the 1950–1970s, where after this period trousers became general fashion for women as well. Also, in men, extremely hairy skin on the legs may serve as an extra protection.

The relative body site distribution of SCC during the 1960s may reflect the sun protection of individuals who were children in the 1890–1920s, and usually dressed in the sun, whereby accumulation of lifetime sun exposure was curtailed from the start. It may also reflect the domination of outdoor work (farming and fishing industries) which has shifted to indoor work during the last half century. During the same period, roofed-over tractors and fishing boats, which were gradually introduced in farming and fishing industries, may have contributed to the fact that outdoor workers in rural industries are no longer at an increased risk of SCC.24,25 This may have contributed to a relative decrease of SCC of the ears. On the other hand, new leisure time habits may dilute the patterns of relative body site distribution; namely, the increasing number of departures to seaside resorts south of the 40th parallel and the use of sunbeds which quickly became widespread in Sweden after their introduction in 1978.15–17 It is reasonable to assume that intentional tanning related to leisure time travel and sunbed use involve more right-angled exposures to larger skin surfaces21–23 than unintentional tanning related to fishing and farming, and that the exposed skin surfaces consequently have increased during the observation period. Further, the use of sunscreens which at an early stage were promoted as imperative means of protection have been questioned due to scanty application (and probably even more scanty on large body surfaces) and a probable intention of the user to prolonged sun exposure,26 possibly in combination with a tendency to select peak UVR hours for sun exposure. Also, stable sunscreens sufficiently protecting against UV-A and UV-B are still under development.27,28 Another aspect in this context of possible increase in aquatic and alpine sports is the combination of chilled skin with high exposure to UVR, and the implication this has for DNA-repair, as there is evidence of poorer DNA-repair in chilled skin (1000 dimer repairs/sec vs 10 000 dimer repairs/sec).29

Episodes of ozone depletion (deviation of long range mean total ozone by mean value per month) have been recorded in Sweden throughout the 1990s, with peaks occurring in the spring months of 1993, 1996 and 2000.12 Even though it is unknown whether any cases of SCC in Sweden could be ascribed to ozone depletion, it may in the future have implications for both unintentional sun exposure to the head, and intentional sun exposure all over the body.

A limitation of this study is that the impact of certain chemical and microbiological factors relevant for the genesis of SCC (i.e. other than UVR) has not been considered as they are beyond the scope of this study. Arsenic and chromium in drinking water are known carcinogens of non-melanoma skin cancer,30,31 as well as the exposure to tar and X-rays.32 Exposures to these agents could separately, or in combination with each other or with UVR, in theory dilute or stress the footprint of behavior in the relative body site distribution of SCC. Further, an association between sun exposure and being seropositive for human papillomavirus on sun-exposed sites has been suggested.33,34 Immunosuppression in organ transplant recipients is another known risk factor for SCC.35

Patients aged 70 years or more have so far mainly been diagnosed with head-site tumors. However, a relative increase of tumors on other body sites among women aged 70–89 since 1990 may be a sign that risky tanning habits (in view of long induction times) have begun to show in older age groups.

Our conclusion is that the shift in relative body-site distribution observed during the past four decades of the relative increase of SCC is congruent with societal changes of migration from rural to urban areas, the introduction of sunbeds, and the upsurge in leisure time involving aquatic sports, and airline travel to seaside resorts. Considering the fact that younger age groups introduce new trends in fashion and life-style (clothes, hair-style, long-distance travelling) which may diffuse to older age groups (Fig. 4), together with a new trend among the elderly to move to warmer countries (e.g. Spain, Thailand) during the fall–winter period in Sweden, a rapid increase of SCC may be expected in Sweden unless powerful skin cancer control programs are implemented.

Ancillary