Skip Navigation

Sunlight and Reduced Risk of Cancer: Is The Real Story Vitamin D?

Solar radiation is a well-established skin carcinogen, responsible for more cancers worldwide than any other single agent (1). Yet, evidence is beginning to emerge that sunlight exposure, particularly as it relates to the vitamin D synthesized in the skin under the influence of solar radiation, might have a beneficial influence for certain cancers.

In this issue of the Journal, two independent epidemiologic studies (2,3) suggest that sunlight may reduce the risk of non-Hodgkin lymphoma (NHL) and may be associated with increased survival rates in patients with early-stage melanoma. In a large population-based case–control study of more than 3700 patients with incident lymphoma and nearly 3200 control subjects in Sweden and Denmark, Smedby et al. (2) reported a 20% to 40% reduction in the risk of this cancer. The reduction in risk was dose-related with increasing indices of prior sun exposure. In a follow-up study of more than 500 melanoma patients who participated in a U.S. case–control study in the 1990s, Berwick et al. (3) noted that subsequent mortality from melanoma was approximately one-half as high among those with signs of solar elastosis assessed by means of a standardized physical examination as among those without solar elastosis. The association of this biomarker of sun exposure and melanoma mortality persisted after adjustment for potential confounding factors, including thickness of the lesion and other established melanoma prognostic factors. The association was not confounded by early detection or various screening behaviors, including skin awareness, and skin self- or physician examination, or by social class.

The notion that sun exposure might have a salutary influence for some types of cancers has been around for several decades. Among Caucasians in the United States, cancer mortality for several prominent cancers, including cancer of the breast, prostate, and colon, shows a striking latitudinal gradient, with increased mortality rates among individuals residing in the northern states compared with individuals residing the southern states (4). These patterns persist even after confounding variables like socioeconomic status, urban and rural residence, Hispanic heritage, and other risk factors are taken into account (5,6). In 1990, it was proposed that vitamin D, which is synthesized in the skin upon exposure to ultraviolet B light (UVB), might be the agent that accounts for these geographical patterns (7). 1,25-Dihydroxyvitamin D3 has multiple cellular effects, including inhibiting the G1/S cell cycle checkpoint, increasing expression of the cyclin-dependent kinase inhibitors p21 and p27, and potentially inducing apoptosis via a number of pathways (810). The ability to convert the provitamin to the active 1,25-dihydroxyvitamin D3 is much reduced at northern latitudes, and populations living far from the equator are at increased risk of vitamin D deficiency during the winter months (11).

The findings from Berwick et al. (3) are consistent with laboratory evidence and are in line with some previous epidemiologic observations. Melanoma cell lines express the vitamin D receptor and are responsive to the antiproliferative and prodifferentiation effect of 1,25-dihydroxyvitamin D3 (12). Even more provocative is the observation of low 1,25-dihydroxyvitamin D3 serum levels in patients with malignant melanoma (13) and an association in some studies of specific vitamin D receptor polymorphisms with the development and outcome of melanoma (14). The incidence of melanoma of the skin on intermittently exposed sites is reduced among outdoor workers compared with indoor workers (15). This pattern of incidence among workers has usually been attributed to the fact that skin tanning from habitual exposure may reduce the incidence of sunburning, a known melanoma risk factor. However, an alternative hypothesis is that outdoor workers are less likely to be deficient in vitamin D because of their more regular exposure to sunlight. This latter hypothesis might also help to explain the difficulties in demonstrating any clear protective influence from the use of sunblocking agents (16). High SPF sunscreens can completely block vitamin D synthesis in the skin (17) and thus persons that habitually use sunscreens may not reap the benefits of sunlight exposure with respect to the endogenous synthesis of vitamin D.

The study by Berwick et al. (3) involved mostly patients with early stage melanoma, and it is plausible that the known antiproliferative and antiangiogenic properties of vitamin D may be inversely associated with melanoma progression. However, it should be noted that solar elastosis and other sun exposure markers in the study reflected prediagnostic exposures to sunlight. Sun avoidance is the most predictable response to a melanoma diagnosis (18), so it is not clear whether the results of Berwick et al. (3) reflect benefits related to continued sunlight exposure, or (as seems more likely) a different pathogenesis in melanomas that arise in persons at risk to develop actinic skin damage. Melanomas are genetically heterogenous tumors and evidence is fast accumulating that these tumors can develop along different genetic pathways, in which different patterns of sun exposure may be involved (19). For example, recent studies show that B-raf mutations are common in melanomas that develop from intermittently sun-exposed sites, but not in tumors that develop from chronically sun-exposed (20) or consistently sun-protected (mucosal, uveal tract) sites (2023). Whether B-raf involvement affects the biologic behavior of melanomas is yet to be determined. The results of Berwick et al. (3) should also be interpreted with caution because we know that persons with one primary melanoma are at a statistically significantly increased risk to develop another melanoma, and continued intense and/or intermittent sun exposure will almost certainly increase the risk.

The results for the Scandinavian study (2) for lymphoma are of particular interest because these malignancies were suspected to have a common etiology with cancer of the skin. Incidence rates of non-Hodgkin lymphoma have increased worldwide in recent decades in parallel with increased melanoma incidence (24). Certain other descriptive features of the disease (higher rates in Caucasians and in males; increased rates before and after the diagnosis of a UV-related skin cancer) also seem to mirror the epidemiology of both melanoma and nonmelanoma skin cancer, which are both strongly linked to sunlight exposure. The only established risk factor for lymphoma, immune suppression, also fits with the idea that sunlight is a causal agent, because UVB is known to suppress the immune response (25). The findings from the Scandinavian study (2), however, which are consistent with a study published recently from Australia (26), cast serious doubt on the likelihood that heavy sun exposure might be a cause of lymphoma. The Scandanavian study (2) encompassed the entire population of Denmark and Sweden, and included a systematic evaluation of multiple measures of solar exposure in a population where sun bathing is still a common practice (approximately half the general population control subjects reported sun bathing at least twice per week at age 20 years, with 40% still reporting this frequency of exposure in recent years). Various proxies for sun exposure, including sunbathing habits and history of sunburn (though not constitutional factors like pigmentation and skin reaction to the sun) were consistently associated with a reduced incidence of lymphoma in this study. Likewise, in the population-wide Australian study (23), risk of non-Hodgkin lymphoma decreased with different measures of increasing self-reported sun exposure. The fact that a history of skin cancer was a statistically significant risk factor in the Scandinavian study (2), controlling for sunlight exposures, implies that some other risk factor (impairment of a specific DNA repair pathway, for example) may be common to both cancers.

Assuming the findings of Smedby et al. (2) and of Berwick et al. (3) are confirmed, what is the most rational public health message regarding safe levels of sun exposure? Ultraviolet radiation in the solar spectrum is an established human carcinogen, one of only approximately 60 agents so designated by the World Health Organization. More than 1 million cases of skin cancer diagnosed annually in the United States are attributable to this exposure (1). However, of these skin cancers, melanoma represents a small (an estimated 54 000 new cases are diagnosed per year in the United States) but important fraction because skin melanoma may be a fatal disease and because skin melanoma contributes disproportionately to cancer-related death rates among young adults (27). Thus, the risks and benefits of sun exposure have to be weighed against many factors, including medical history, personal attributes such as pigmentation and the proneness to sunburn, and family and personal history of cancer. Although it is clearly premature to make dietary recommendations based on the findings of these two studies (2,3), it is known that a substantial proportion of people residing in northern latitudes are deficient in vitamin D during the winter months (28). The few dietary sources of vitamin D include milk purchased in the United States, some brands of orange juice (which are now supplemented with vitamin D), cod liver oil and oily fish, such as salmon, mackerel, and sardines, and vitamin D supplements are available.

In summary, the two studies (2,3) in this issue of the Journal provide intriguing new evidence that solar radiation may have a beneficial influence in both the incidence and outcome of cancer. In view of the major potential public health consequence of these results, further studies of sunlight and the vitamin D connection to cancer are certainly warranted.

  1. William J. Blot
  1. Affiliations of authors: Vanderbilt University Medical Center, Vanderbilt-Ingram Cancer Center, Nashville, TN (KME, JAS, WJB); International Epidemiology Institute, Rockville, MD (WJB)
  1. Correspondence to: William J. Blot, PhD, International Epidemiology Institute, 1455 Research Blvd., Ste. 550, Rockville, MD 20850 (e-mail: blotw{at}


    1. (1)
    2. (2)
    3. (3)
    4. (4)
    5. (5)
    6. (6)
    7. (7)
    8. (8)
    9. (9)
    10. (10)
    11. (11)
    12. (12)
    13. (13)
    14. (14)
    15. (15)
    16. (16)
    17. (17)
    18. (18)
    19. (19)
    20. (20)
    21. (21)
    22. (22)
    23. (23)
    24. (24)
    25. (25)
    26. (26)
    27. (27)
    28. (28)
    | Table of Contents

    Navigate This Article