Abstract
Background Sun exposure in combination with skin pigmentation is the main determinant for vitamin D status. Human skin color seems to be adapted and optimized for regional sun ultraviolet (UV) intensity. However, we do not know if fair, UV-sensitive skin is a survival advantage in regions with low UV radiation.
Methods A population-based nested case–control study of 29,518 Caucasian women, ages 25 to 64 years from Southern Sweden who responded to a questionnaire regarding risk-factors for malignant melanoma in 1990 and followed for 25 years. For each fair woman, defined as having red hair or freckles (n = 11,993), a control was randomly selected from all non-fair women from within the cohort of similar age, smoking habits, education, marital status, income, and comorbidity, i.e., 11,993 pairs. The main outcome was the difference in all-cause mortality between fair and non-fair women in a low UV milieu, defined as living in Sweden and having low-to-moderate sun exposure habits. Secondary outcomes were mortality by sun exposure, and among those non-overweight.
Results In a low UV milieu, fair women were at a significantly lower all-cause mortality risk as compared to non-fair women (log rank test p = 0.04) with an 8% lower all-cause mortality rate (hazard ratio [HR] = 0.92, 95% CI 0.84‒1.0), including a 59% greater risk of dying from skin cancer among fair women (HR 1.59, 95% CI 1.26‒2.0). Thus, it seem that the beneficial health effect from low skin coloration outweigh the risk of skin cancer at high latitudes.
Conclusion In a region with low UV milieu, evolution seems to improve all-cause survival by selecting a fair skin phenotype, i.e., comprising fair women with a survival advantage.
Discussion
Women with a fair UV-sensitive phenotype living in a low UV milieu had a significantly increased life expectancy as compared to non-fair women. Fair women were at an eight percent lower all-cause mortality rate, as compared to those with non-fair skin. There is a strong inverse dose-dependent risk between increasing sun-exposure habits and all-cause mortality.Strengths and limitations
Our large sample, comprising 20% of all women in the south Swedish region between 25 and 64 ages, as drawn by random selection from the population registry at the study inception 1990 is a strength. It was thus a representative sample of the South Swedish population at the time of recruitment before the large immigration of the 2000’s. It comprises almost exclusively European Caucasian women. Thus, the comparison between fair and non-fair was mainly a comparison between Fitzpatrick types 1 skin vs. type 2‒3 skin. Since the questionnaire was administrated at the inception of the study, there was no recall bias. Since we earlier have been criticized that our adjustments in Cox regression might not be adequate, we decided to perform a one-to-one matched design. Historically, during evolution there was no possibility to use solarium or to travel for sunbathing. Therefore, we were predetermined to make the main outcome comparison in a low UV milieu, i.e., among those with low-to-moderate sun exposure habits. As secondary outcome we assessed mortality by sun-exposure with adjustment for exercise or stratified for low BMI, only including the time period after year 2000. A major limitation is that the significance level of the lower risk of all-cause mortality among fair women was close to the 5% significance level in all analyses regarding skin type, but it was according to the predetermined hypothesis. Another strength is that the analyses from year 2000 including exercise habits, and BMI showed similar results, but with wider CIs. The results might not be generalized into regions with more intense UV radiation. The aim of the study was not to assess cause specific mortality. However, it is impossible to publish on beneficial effects by sun exposure without including data on skin cancer mortality. Thus, our study is in agreement with the large amount of papers showing an increased incidence of skin cancer with fair skin and we also showed increased mortality in skin cancer. Since fair skin is selected at high latitudes, an improved all-cause survival is also expected from an evolutionary perspective [2]. Frost and coworkers reported in an open internet-based study that red-haired women were particularly prone to ovarian-, uterus-, cervical, and colorectal cancer, our results could not reproduce these findings and we did not find an increased incidence of these groups among fair women in our study [15]. There has been somewhat conflicting evidence regarding sun exposure and all-cause mortality. The Swedish Women´s Lifestyle and Health Study reported that increased sun exposure (measured as sunbathing holidays, i.e., which was one of our four questions) was related to reduced HRs for all-cause mortality [16]. On the other hand, a large US epidemiological study based on regional, not personal, UV radiation reported a positive relation between increasing UV radiation and all-cause mortality [17]. A possible explanation for the opposing results might be the differences in latitude and, therefore, UV intensity (Sweden latitude 55o to 59o and continental US latitude 24o to 42o. While the mean level of the biomarker vitamin D for sun exposure was 48.6 (± 20.5) nmol/L in Sweden it was 77.0 (± 25.0) nmol/L in the US, indicating a greater problem with sun deficiency at high latitudes [9, 18]. Based on data from the Swedish Meteorological and Hydrological Institute (SMHI), in 2014 there was one day with strong UV exposure, i.e., UV-index ≥ 6.Skin cancer mortality
When we investigated whether the increased mortality associated with skin cancer influenced the strong inverse relationship between all-cause mortality and increasing sun exposure habits and found that this was not the case. Women with fair skin were at a 59% increased risk of death in skin cancer. This was counterbalanced by the health benefits, as measured by all-cause mortality, of fair skin and sun exposure. There is an increased risk of skin cancer with both fair skin and increasing sun exposure, but the prognosis of skin cancer seem to improve with increasing sun exposure [19, 20]. Thus, there seem to be a tradeoff between health benefits and skin cancer and in regions with scarcity of solar UV radiation fair skin have been selected [2]. In our modern society there is not unusual with a mismatch between skin coloration and geography/climate/ habits that might cause increased morbidity and mortality [2].Sun exposure and overweight
Overweight and obese women do not seem to obtain the same benefit from having fair skin or from sun exposure as non-overweight women. We have seen similar findings in prior studies, where the lower risk of type 2 diabetes mellitus and endometrial cancer after UV exposure was mainly seen in non-overweight women [21, 22]. Wortsman and coworkers have clearly demonstrated that obesity has a detrimental effect on vitamin D levels for a given amount of UV exposure [23]. Thus, lower sun exposure habits among overweight is not the cause. It appears that vitamin D is either produced in a smaller quantity or consumed/inactivated among overweight women. Further, a study using Mendelian randomization analysis showed that increasing BMI leads to lower vitamin D levels [24]. The differential impact of BMI by sun exposure on all-cause mortality is an area that would benefit from additional research. Since BMI seem to be in the causal pathway of sun exposure and all-cause mortality, we chose not to adjust for BMI and present only stratified analysis.It has been hypothesized that the inbreeding with Neanderthals some 47,000 to 65,000 years ago in northern Canaan might have helped Homo sapiens adjust to life beyond Africa [25–27]. Studies of the ancient Neanderthal genome have shown that Westerners carry approximately 1% to 3% of Neanderthal DNA [25, 26]. People of European origin are highly likely (≈ 60% to 70%) to have the Neanderthal DNA that affects keratin filaments, i.e., zinc finger protein basonuclin-2 (BNC2). The latter alleles are thought to be involved in the adaptive variation of skin coloration, influencing skin pigmentation towards fairer skin [6, 28]. With our finding of increased life expectancy with fair skin, we speculate that the preserved high carriership of the Neanderthal BNC2 allel might be an advantage at high latitudes.
We interpret our findings to support that a fair, UV-sensitive phenotype in Sweden seems to be related to prolonged life expectancy in a low UV milieu, but at the cost of an increased risk of death due to skin cancer. Over thousands of years a fair UV-sensitive phenotype has possibly been selected for optimal health at high latitudes
No comments:
Post a Comment