IJC International Journal of Cancer

Host characteristics, sun exposure, indoor tanning and risk of squamous cell carcinoma of the skin Marit B. Veierïd1, Elisabeth Couto1,2,3,4, Eiliv Lund5, Hans-Olov Adami3,6 and Elisabete Weiderpass2,3,5,7 1

Department of Biostatistics, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway Department of Etiological Research, Cancer Registry of Norway, Oslo, Norway 3 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden 4 Health Economics and Drugs Unit, Norwegian Knowledge Centre for the Health Services, Oslo, Norway 5 Department of Community Medicine, University of Tromsï, Tromsï, Norway 6 Department of Epidemiology, Harvard School of Public Health, Boston, MA 7 Department of Genetic Epidemiology, Folkh€ alsan Research Center, Helsinki, Finland 2

Cutaneous squamous cell carcinoma (SCC) incidence is increasing worldwide, but population incidence data are scarce in many countries due to registration practices.1 The national cancer registries in Norway and Sweden systematically collect non-melanoma skin cancer incidence (excluding basal cell carcinoma) of which about 95% are SCC.2 Nonmelanoma skin cancer was one of the most rapidly increasing malignant tumors in Sweden in the last decade, with an average annual increase of 4.7% in men and 6.7% in women.3 Norwegian female incidence rates were more than five times

Key words: squamous cell carcinoma, host risk factors, sun exposure, indoor tanning, prospective cohort Grant sponsors: Norwegian Cancer Society; Swedish Research Council; National Institutes of Health (USA) DOI: 10.1002/ijc.28657 History: Received 17 Sep 2013; Accepted 26 Nov 2013; Online 6 Dec 2013 Correspondence to: Marit B. Veierïd, Department of Biostatistics, Institute of Basic Medical Sciences, University of Oslo, P.O. Box 1122 Blindern, N-0317 Oslo, Norway, Tel.: 147-22-85-14-32/14741-63–99-13 (mob); Fax: 147-22-85-13-13, E-mail: [email protected]

C 2013 UICC Int. J. Cancer: 135, 413–422 (2014) V

higher in 2006–2010 compared to 1956–1960 and the male rates were nearly five times higher.4 Hair and eye color, the propensity to sunburn, and the ability to tan are generally accepted host risk factors for SCC,5,6 but pigmentation is less studied as a risk factor for SCC than for cutaneous malignant melanoma (MM). Likewise, nevi count, an important risk factor for MM, has been studied in relation to SCC risk in a few studies but no consistent association has been found.7–9 SCC risk is positively associated with total (lifetime, cumulative, chronic) sun exposure5,6,8,10,11 as well as with cumulative work-related sun exposure.5,12 Sun exposure during childhood and adolescence may affect SCC risk.6 While a causal positive association with use of indoor tanning devices is established for MM, the studies of indoor tanning and SCC are fewer and their results are not consistent.13–15 A recent meta-analysis included six studies and found a significant positive association between SCC risk and ever use of indoor tanning.15 However, half of these studies16–18 and a recent case-control study9 did not find a significant association between SCC risk and indoor tanning. Moreover, the effect of age at indoor tanning exposure on SCC risk is unclear.

Epidemiology

Use of indoor tanning devices increases risk of cutaneous malignant melanoma, but the association with risk of squamous cell carcinoma (SCC) of the skin is unclear. Cohort studies of SCC risk are rare and we aimed to assess the association between SCC risk and host characteristics, sun exposure, and indoor tanning in a population-based cohort of Norwegian and Swedish women conjunctly with SCC incidence data from national cancer registries. Host characteristics and exposure to sun and indoor tanning devices before 50-years old were recorded by questionnaire at inclusion (30–50 years) in 1991/92. Multivariable relative risks (RRs) and 95% confidence intervals (CIs) were estimated by Poisson regression. During follow-up of 106,548 women through December 2009, SCC was diagnosed in 141 women. Skin sensitivity to acute sun exposure was the most important pigmentation characteristic (RR 5 2.73, 95% CI 1.47–5.05, for red with pain/red with pain and blisters versus brown). We found no consistent associations with sunburns and bathing vacations in the first five age decades, but a significant positive trend for bathing vacations summarized over ages 10–49 years (Ptrend 5 0.02). We also found significantly increased risks of SCC following indoor tanning at age 40–49 years (RR 5 2.17, 95% CI 1.29–3.67, for  1 time/month versus never) and indoor tanning summarized over ages 10–49 years (Ptrend 5 0.001). RR for ever versus never use of indoor tanning over ages 10–49 years was 1.93 (95% CI 1.27–2.95). Propensity to burn was an important host characteristic, and bathing vacations and indoor tanning summarized over ages 10–49 years increased SCC risk.

414

Host factors, UV exposure and SCC

What’s new? Indoor tanning devices are known to increase the risk of malignant melanoma, but their impact on the risk of squamous cell carcinoma (SCC) of the skin is unclear. In this large, population-based, prospective cohort study, the authors found that skin sensitivity to acute sun exposure was the most important host risk factor for SCC, and that the use of indoor tanning devices was indeed associated with an increased risk of SCC. They conclude that the use of indoor tanning devices should therefore be discouraged.

Most previous studies on the effects of ultraviolet exposure on SCC risk used a case-control design, sometimes nested in a cohort, and more data from prospective studies are needed.14,19 We have previously studied host characteristics, sun exposure and indoor tanning in a large populationbased cohort study in Scandinavia, and found consistent effects on MM risk.20,21 We present here results of the association between SCC risk and host characteristics, sun exposure, and indoor tanning using the same cohort study merged with SCC incidence data from the national cancer registries.

Material and Methods

Epidemiology

Study population

The Norwegian-Swedish Women’s Lifestyle and Health Cohort study was established in 1991–1992.20,21 A nationwide random sample of 100,000 women (born 1943–1957) was drawn from the Norwegian National Population Register. Similarly, 96,000 women (born 1943–1962) residing in Uppsala Health Care Region (comprising about one-sixth of the Swedish population in the relevant age group) were randomly selected from the Swedish National Population Register. Women received an invitation letter that requested written informed consent, and a comprehensive questionnaire to be completed and returned in a prepaid envelope. Only those who consented were included in the study cohort. The national Data Inspection Boards and responsible Medical Ethical Committees approved the study. Host characteristics and ultraviolet exposure information

Variables relevant to this article were identical in the two countries. Participants reported natural hair color (dark brown/black, brown, blond/yellow or red), eye color (brown, gray/green or blue), and the number of asymmetric nevi larger than 5 mm on their legs from toes to groin (0, 1, 2–3, 4–6, 7–12, 13–24 or  25 nevi). Pictures of three examples of asymmetric nevi were enclosed in a color brochure. Because of small numbers, we categorized nevus counts as 0, 1 and  2. Questions were also asked about participants’ skin reaction to heavy (acute) sun exposure in the beginning of the summer (turns brown without first becoming red, turns red, red with pain, or red with pain and blisters), and their skin reaction to repeated and long-lasting (chronic) sun exposure (turns deep brown, brown, light brown, or never turns brown). In the analyses, we combined the upper two

(most sun sensitive) categories of skin reaction to acute and chronic sun exposure. Histories of ultraviolet (UV) exposure were reported for four age periods: 10–19, 20–29, 30–39 and 40–49 years old. For each period, women reported the number of times per year they had been burned by the sun so severely that it resulted in pain or blisters and subsequent peeling (never, 1, 2–3, 4–5 or 6 times/year), the average number of weeks per year spent on bathing vacations in southern latitudes or within Norway or Sweden (never, 1, 2–3, 4–6, or 7 weeks/ year), and the average use of an indoor tanning device (never, rarely, 1, 2, 3–4 times/month or >1 time/week). In addition, the Norwegian questionnaire included the above questions on sunburn, bathing vacations, and indoor tanning for the age period 0–9 years. We collapsed higher categories in the analyses of sunburns, bathing vacations, and use of a tanning device during different decades of life. In addition, for sunburns, bathing vacations, and indoor tanning, we combined the exposure across each of the three decades of life recorded for all women (10–19, 20–29 and 30–39 years). These combined variables were designed to study effects of age at exposure and of exposure accumulating over age decades.21 Additionally, because cumulative exposure is considered important for SCC risk, we summed the observed frequencies at ages 10–19, 20–29, 30–39 and 40–49 years for sunburns, bathing vacations, and indoor tanning. On the basis of these sums, we constructed a score considering the frequencies of sunburns, bathing vacations, and indoor tanning. Furthermore, scores were also calculated summarizing bathing vacations and indoor tanning together. A low score reflects a low frequency and a higher score higher frequencies of UV exposure and we refer to them as summarized variables hereafter. The summary variables were calculated for ages 10–39 (recorded for all women), and for ages 10–49 (includes women aged 40–49 who were recruited after age 40). The first age-decade of life was not included in the combined or summarized variables, since it was only recorded for Norwegian women.

Follow-up and endpoints

Follow-up started on the date of receipt of the returned questionnaire (May 15, 1991, to June 15, 1992). The end of follow-up was December 31, 2009. We calculated personyears from the start of follow-up to the date of first diagnosis C 2013 UICC Int. J. Cancer: 135, 413–422 (2014) V

415

Veierïd et al.

Statistical methods

We estimated the associations between SCC risk and host characteristics, sunburns, bathing vacations or indoor tanning by Poisson regression analysis, and present results as relative risks (RRs) with 95% confidence intervals (CIs). All analyses were adjusted for attained age categorized in 5-year intervals (for women aged 40 at inclusion we used 40–49, 50–54 and 55). The multivariable models included geographic region of residence (the southern, middle and northern regions of Norway, and the Uppsala Health Care Region in Central Sweden).20 In the analyses of host characteristics, we also included hair color and skin reaction to acute and chronic sun exposure (if not the characteristic under study). The correlation coefficients between the pigmentation characteristics were relatively low.22 We adjusted for hair color and skin reaction to acute and chronic sun exposure in the analyses of sunburns, bathing vacations and indoor tanning. Because further adjustments for eye color had no effect on the effect estimates, this variable was not included in the final model. Lastly, each age-specific model for indoor tanning included corresponding age-specific sunburns and bathing vacations. Trends in RRs were analyzed by a likelihood ratio test assigning equally spaced values (e.g., 1, 2, 3 or 4) to the categories and treating the variable as continuous in the regresC 2013 UICC Int. J. Cancer: 135, 413–422 (2014) V

sion analysis. Statistical interaction effects between dichotomized pigmentation characteristics and dichotomized cumulative UV exposure (ages 10–39) were also evaluated by a likelihood ratio test. We found no significant interaction effects and these results are not presented since the analyses were limited by low power. Calendar year of indoor tanning is important because exposure before 1982/83 is likely to be from tanning devices with UVB-rich mercury arc lamps whilst exposure after 1982/83 would be predominantly from modern sunbeds with UVA-rich fluorescent lamps. In analyses of MM, we used indoor tanning at ages 20–29 to investigate this hypothesis as indoor tanning was rare at younger age in our cohort and because only a small proportion of the cohort was 30–39 years old (or older) before 1993.23 However, due to fewer cases this analysis was uninformative for SCC and is not presented here. All tests were two sided, p < 0.05 was considered statistically significant, and the analyses were conducted in STATA (http://www.stata.com/).

Results We achieved complete follow-up of the 106,548 women in the cohort during 1,905,614 person years with a mean follow-up of 17.9 years (range, 0.01–18.6 years). A total of 141 incident first cases of SCC were reported to the cancer registries in Norway and Sweden during this follow-up. It was the first cancer diagnosis for 110 women, the second for 23, the third for 5, the fourth for 2, and the fifth for 1. The mean age at inclusion was 40.4 years (range, 30–50 years), and the mean age at SCC diagnosis was 56.1 years (range, 41.7–66.1 years). Head was the most common site of SCC (n 5 53), followed by the neck and trunk (n 5 41), lower limbs (n 5 22) and upper limbs (n 5 18). Host characteristics

Skin sensitivity to acute and chronic sun exposure and hair color were significantly associated with SCC risk in the age adjusted analyses (pheterogeneity  0.03; Table 1), while only skin reaction to acute sun exposure was significant in the mutually adjusted analysis (pheterogeneity 5 0.003). Eye color and the total number of large asymmetric nevi on the legs were not associated with SCC risk (pheterogeneity 5 0.71 and ptrend 5 0.65, respectively, multivariable models). Sunburns

Annual number of sunburns in each age category, sunburns combined for ages 10–39 years, and sunburns summarized over ages 10–39 and 10–49 years were not significantly associated with SCC risk in the multivariable analyses (0.09  ptrend  0.90; Table 2). Results were similar with no sunburns as the reference for sunburns combined (ages 10– 39 years), but are not presented due to sparse numbers of cases. When we dichotomized the variables summarizing sunburns over ages 10–39 years and ages 10–49 years (i.e.,

Epidemiology

of SCC, to the date of emigration or death, or to the end of follow-up, whichever occurred first. We identified cancer cases by linkage of the study database to the Cancer Registry of Norway and the Swedish Cancer Registry and obtained information on death and emigration by linkage to Statistics Norway and Statistics Sweden. Record linkages were ensured by the unique personal identification number assigned to all residents of the two countries. Cases of SCC (International Classification of Diseases, 7th revision, coding 191), diagnosed between recruitment (1991–1992) and December 31, 2009 were considered. Only invasive SCC cases were included. We excluded the following histological types recorded in the cancer registers: malignant tumor not otherwise specified, sarcomas not otherwise specified, metatypic carcinomas, adenocarcinomas, endocrine carcinomas, spindle cell carcinomas, malignant histyocytomas, dermatofibrosarcomas, leiomyosarcoma myoepithelial and emangiopericytoma. We previously reported that 57,584 (57.6%) of the Norwegian women, and 49,259 (51.3%) of the Swedish women returned completed questionnaires.20 In total, 106,820 women were available for analysis (23 Norwegian women have withdrawn from the study during follow-up). We excluded 17 women who were diagnosed with SCC prior to the start of follow-up, 1 woman with missing information on vital status, 59 women who had emigrated or died before or at the start of follow-up, and 195 women who did not answer the questions regarding sun exposure or host characteristics (sun sensitivity of skin, hair color, eye color and number of asymmetric nevi). Thus, 106,548 women were included in the analyses.

416

Host factors, UV exposure and SCC

Table 1. Relative risks1 (RRs) and 95% confidence intervals (CIs) of squamous cell carcinoma according to pigmentation factors Characteristic

Frequencies no. (%)

No. of cases

Age-adjusted RR (95% CI)

Multivariable2 RR (95% CI)

Skin color after heavy (acute) sun exposure at the beginning of the summer (n 5 102,169) Brown

25,487 (24.9)

16

1.00

1.00

Red

49,269 (48.2)

73

2.41 (1.40–4.14)

2.18 (1.23–3.86)

Red with pain/red with pain and blisters

27,413 (26.8)

50

3.18 (1.81–5.59)

2.73 (1.47–5.05)

Host characteristics, sun exposure, indoor tanning and risk of squamous cell carcinoma of the skin.

Use of indoor tanning devices increases risk of cutaneous malignant melanoma, but the association with risk of squamous cell carcinoma (SCC) of the sk...
179KB Sizes 0 Downloads 0 Views