There has been considerable interest in recent years as to whether vitamin D inhibits breast cancer development [1-3]. Vitamin D is produced when the skin to solar is exposed ultraviolet B (UVB) radiation, which is the primary source of vitamin D for most people [2,4]. Previtamin D from both diet and sunlight is hydroxylated in the liver into 25-hydroxyvitamin D (25(OH)D). 25(OH)D is then further hydroxylated into 1,25-dihydroxyvitamin D (1,25(OH)2D). 1,25(OH)2D is the biologically active metabolite that binds to nuclear vitamin D receptors in the intestine and bone, as well as breast tissues [5]. Experimental studies have shown the anticarcinogenic potential of vitamin D in different cell types, including normal and malignant breast cells, by affecting the induction of cell differentiation and apoptosis, as well as the inhibition of cell growth [1,3].

Epidemiological studies have conveyed the protective effect of sun exposure on breast cancer risk. Breast cancer mortality rates are higher in the north-eastern than in the southern United States [6], and are inversely correlated with solar radiation [6-8]. Studies have shown the association of reduced breast cancer risk with high early-life residential solar radiation or a usual residence in a region of high solar radiation and frequent adult recreational or occupational sun exposure [7,8]. In addition, high sun exposure was found to be associated with a reduced risk of advanced breast cancer [9].

A substantial body of evidence indicated that among the risk factors for breast cancer, mammographic parenchymal patterns have the strongest association with breast cancer risk [10,11]. Tabár's mammographic patterns IV and V have been shown to be associated with the known breast cancer risk factors, as well as with increased risk of disease [12,13]. It is reported that dense mammographic tissue is associated with breast epithelial hyperplasia and with stromal fibrosis, which may be in response to growth factors induced by circulating levels of sex hormones (mitogenesis) [11,13].

Breast cancer is the most common cancer in Hong Kong women (approximately 2,100 new cases per year) and had the highest increase in incidence in 2007 among all cancers in Hong Kong [14]. There have no studies yet exploring whether sunlight exposure may reduce the risk of breast cancer in Asians. Recent evidence revealed that vitamin D deficiency (≤25 nmol/L) was indicated in 18% of Hong Kong women aged 18 to 40 years during the spring, and over 90% of women of reproductive age had insufficient levels of vitamin D (≤50 nmol/L) [15]. Therefore, this study aims to examine the relationship of sunlight exposure with breast cancer risk measured by breast density assessed from Tabár's mammographic pattern, in a population-based study of Hong Kong Chinese premenopausal women.

The 277 and 369 low (Tabár's pattern I, II, & III) and high (IV & V) breast cancer risk subjects were included in the analysis (Table 1). The subjects with a low risk pattern were slightly older than those with high risk (43.99 years vs. 43.48 years). Statistically significant differences between low and high risk groups were observed with respect to waist-to-hip ratio, body mass index, height, physical activities index, education, and sun protection use from 35 years to present. There were no statistically significant differences between the low and high risk groups in terms of age at menarche, hormonal use, pregnancy, breastfeeding, number of live births, and family history of breast cancer. The subjects were more likely to spend more time in the sun during their younger ages. Figure 1 showed a declining trend of the total hours spent in the sun with increasing age stages, however, women with high risk breast pattern (Tabár pattern IV & V) had spent fewer hours in the sun than those with low risk (I, II, & III) at the various age stages. Sunlight exposure was only significantly associated with age and education, but not with other factors such as body mass index, physical activity, sun protection use and others (data not shown).

In the unconditional regression model, a statistically significant inverse association between time spent in the sun and breast cancer risk (Tabár IV & V vs. I, II, & III) was observed. Women aged 40 to 44 years who were in the highest tertile of lifetime total hours spent in the sun had an OR of 0.41 (95% CI, 0.18-0.92) of having high risk breast pattern compared with those in the lowest tertile (>2.19 hr/day vs. <1.32 hr/day), with a significant linear trend across tertiles (p for trend=0.03) after adjusting for other breast cancer risk factors, including age at mammography screening, age at menarche, ever pregnancy, ever breastfeeding, education, body mass index, waist-to-hip ratio, height, family history of breast cancer, hormone replacement usage, physical activity and sun protection use (Table 2). Compared to those in the first tertile, subjects aged 40 to 44 years belonging to the highest tertile of hours spent in the sun during the ages of 6 to 12 years had a lower adjusted breast cancer risk (OR, 0.37; 95% CI, 0.15-0.91; p for trend=0.03). The associations between sunlight exposure across age groups 13 to 19, 20 to 34, and over 35 years and breast cancer risk were not observed in women aged 40 to 44 years. There were no apparent differences in sunlight exposure duration between the high and low breast pattern risk groups for women aged 37 to 40 and 44 to 50 years.

No apparent differences were found between the two breast pattern risk groups for sunlight exposure during different seasons, the latitude where the subjects lived, vitamin D intake and skin reaction to sun (data not shown). Only 1 subject (0.15%) used the sunbed once and very few participants below the age of 35 years ever had trips to summer climate locations during winter seasons; hence, the risks of these factors could not be compared between the two comparison groups.

In this population-based cross-sectional study of premenopausal women, we observed an inverse association between lifetime sunlight exposure and breast cancer risk measured by breast density assessed from Tabár's mammographic parenchymal pattern. The reduced risk was particularly apparent for subjects aged 40 to 44 years with high sunlight exposure during the ages 6 to 12 years. Sun exposure is a major source of vitamin D. Evidence from both in vitro and in vivo studies revealed that high levels of vitamin D can be obtained from less than a minimal erythemal dose (slight reddening of the skin) of sunlight [23,24]. Previtamin D is hydroxylated in the liver into 25(OH)D. 25(OH)D is then further hydroxylated into 1,25(OH)2D. 1,25(OH)2D is the biologically active metabolite that binds to nuclear vitamin D receptors in the intestine and bone, as well as breast tissues [5]. Experimental studies have shown the anticarcinogenic potential of vitamin D in diverse kinds of cell types, including normal and malignant breast cells, by affecting the induction of cell differentiation and apoptosis as well as inhibition of cell growth [1,3].

Previous studies have suggested an inverse relationship between vitamin D and breast cancer [7,19]. In the first National Health and Nutrition Examination Survey Epidemiologic Follow-up Study, measures of sunlight exposure and dietary vitamin D intake were consistently associated with reduced risk of breast cancer, although the power was limited and many of the estimates did not reach statistical significance [7]. One large case-control study [19] reported an association between reduced breast cancer risks and increasing sun exposure from ages 10 to 19 (OR, 0.65; 95% CI, 0.58-0.85); yet, there was weaker evidence for the associations from ages 20 to 29 and no evidence for ages 45 to 54 years. The present study found that exposure during childhood, a critical period before breast development, was protective of breast cancer risk.

To our knowledge, this is the first study to examine the association between sun exposure and mammographic parenchymal patterns. To date, no studies had yet explored the mechanism as to why sun exposure has an effect on breast density. It is reported that extensive mammographic density and delayed breast involution are both associated with an increased risk of breast cancer and cumulative exposure of breast tissue to hormones and growth factors that stimulate cell division, as well as the accumulation of genetic damage in breast cells, are the major determinants of breast cancer incidence [22]. The protective influence of sun exposure on breast cancer risk measured by mammographic parenchymal patterns might be related to the anticarcinogenic potential of vitamin D on breast cells by affecting the induction of cell differentiation and apoptosis as well as the inhibition of cell growth [1,3].

This study used self-reports of personal sunlight exposure as a surrogate for vitamin D status. Although this study had no quantitative information on the intensity and duration of sunlight exposure or on other factors that influence vitamin D synthesis (e.g., skin pigmentation, medical conditions, and medications) [24], self-reported sunlight exposure has been reported in other studies to be a good surrogate for serum levels of 25(OH)D [25,26], which is the major circulating metabolite of vitamin D. The data in this study group has indicated that the lifetime sunlight exposure questionnaire had sufficient content validity to measure the sun exposure of Hong Kong Chinese women [21]. These data support the validity of our exposure measure in ranking the individuals according to sunlight exposure.

The accuracy of recall is a concern in this study. The reproducibility of sun exposure-related questions was examined in a number of studies over different time periods ranging from a few weeks to several years [27,28]. In general, these studies have found that sun exposure and outdoor activities, whether in childhood and adolescence or recent, could be reasonably and consistently recalled. Some questionnaires tended to focus on the total time in the sun and sun damage to the skin for studies of melanoma [28] or other outcomes [18,20]. In our study, the questionnaire was designed specifically to evaluate lifetime sun exposure-the major source of vitamin D; the validated results showed adequate content validity and reproducibility [21]. Although individuals are unlikely to be able to recall exactly how much time they spent outside, especially during childhood (6-12 years), the questionnaire is still able to roughly distinguish those who spent very little time versus those who spent considerable time outdoors. Due to the recall bias, our results on the protective effect of sun exposure on breast cancer risk might be biased toward either direction. Similarly, specific sun protection methods for those aged less than 35 years were also not collected due to the difficulty to recall except for the qualitative information, according to the results from the pretests.

Another concern for our study is the possible misclassification of mammographic patterns. However, almost all mammograms were taken using the same equipment (Senographe DMR). Furthermore, the assessment of mammographic patterns was performed by one trained reader using a computer-assisted method. Thus, if present, the misclassification of mammographic patterns should be relatively small and most likely random. Another limitation of our study is that we did not obtain information of vitamin D supplementation alone other than multivitamins because most of the women did not take pure vitamin D supplements. Furthermore, our study used breast density measured by Tabár's mammographic pattern as a marker for breast cancer risk, which is not a direct association between sun exposure and the risk of breast cancer development. Nonetheless, our findings provide some insight into the potential for breast cancer prevention.

This study was cross-sectional in nature and did not allow us to determine the temporal associations. However, it is unlikely that mammographic patterns influence sunlight exposure in childhood, adolescent and young adulthood. The results might occur by chance because we have multiple comparisons; thus, the significance level of type I error would be smaller than 0.05 using Bonferroni correction. However, the correction will increase the probability of reducing "power" and result in the entire test to become irrelevant.

Other than influences on the vitamin D status, a number of other health benefits from sufficient levels of vitamin D have been identified. These include anticancer properties [8]; however, sun exposure has also been suspected of causing skin cancer [8].With respect to the risks and benefits of sun exposure, a balance is required between avoiding an increase in the risk of skin cancer by excessive sun exposure and achieving sufficient exposure in order to maintain adequate vitamin D levels [29]. Relatively short sun exposure time will maximize vitamin D generation, although the amount of vitamin D produced in a given time depends on the degree of skin coverage and skin color.

In conclusion, the present findings indicate that increased sunlight exposure throughout one's lifetime and across ages 6 to 12 years seem to be associated with a protective effect of breast cancer risk measured by breast density among Hong Kong premenopausal women. The research into the relationship between breast cancer and sunlight exposure create new opportunities in breast cancer prevention. Moreover, the timing of exposure seems to be crucial and additional work will be required in order to meet the challenge of assessing the prevention mechanism early in life. If confirmed, the results of this study would be particularly promising for the primary prevention of breast cancer due to the fact that sunlight exposure is a modifiable lifestyle factor.