Factors Modifying Exposure and Susceptibility
One of the most overarching sets of such genetic differences is that conferring high or low degrees of pigmentation. Numerous studies have indicated that, in general, those of Negroid ancestry show a very much lower incidence of skin cancer (100 fold for NMSC, 10 fold for CM) than those of Caucasoid ancestry (Scotto et al., 1981). A more recent series of reports from Hawaii extended this observation to other races and found that Caucasians on the island of Kauai have about a 10 times higher incidence of NMSC compared to those of Japanese ancestry, who in turn have a five fold higher incidence than those of Philippine or Hawaiian ancestry (Reizner et al., 1993; Chuang et al., 1995). Interestingly, a similar degree of protection is not conferred by pigmentation for either for cataract (Hiller et al., 1983) or the immunological effects of solar exposure (Screibner et al., 1987).
Among the light-skinned races, certain qualitative differences in pigment also appear to be important to sun sensitivity, as well as to skin cancer risk. In particular, those with higher ratios of pheomelanin (the yellow/orange melanin found in red hair) to eumelanin (the brown/black melanin of brown/black hair) who rarely tan and almost always burn, appear to be at greatest risk (Gallagher et al., 1995a, 1995b, Holman et al., 1984b). Pheomelanin is known to generate reactive oxygen species (ROS) when irradiated with UV-B, whereas eumelanin appears to act protectively against ROS; it has been postulated that this difference may be the reason for the increased susceptibility to skin cancer of those with fair skin and red hair (USEPA 1987). Recent information suggests however, that multiple factors are important to susceptibility, and that different genes may be important to poor tanning ability and UV sensitivity. In the case of poor tanning ability, it has been shown that people with a poor tanning response show variations in a gene important to eumelanin synthesis in melanocytes (Valverde et al., 1995). Furthermore, some of these genetic variations may be associated with increased risk of melanoma (Valverde et al., 1996). In the case of sensitivity to the erythemal effects of UV-B, it has recently been shown that individuals showing the greatest inflammatory response to UV-B exposures lack a key detoxification enzyme for ROS (Kerb et al., 1997).
Another set of genes important to understanding susceptibility to the effects of solar exposures are those associated with repair of UV-B-induced alterations in DNA. Patients with xeroderma pigmentosum (XP), a rare genetic disease characterized by poor repair of UV-induced DNA damage, have a 2000-fold increased risk of developing skin cancer before the age of 20 (Kraemer et al., 1984). It has been suggested that many skin cancer patients suffer from similar albeit much less severe defects in DNA repair (Alcalay et al.1990; Wei et al., 1993); however, these findings have not been universal (Hall et al., 1994) and require additional investigation.
One interesting finding with regard to repair deficiency syndromes and skin cancer is the discrepancy observed between two such syndromes in the development of skin cancer. Trichothiodystrophy (TDD) is a DNA repair deficiency syndrome with many similarities to XP; indeed one form of XP, XP-D, has mutations in the same gene as that affected in TDD. However, only XP-D individuals are at increased risk of skin cancer. An explanation for this anomaly may be that cells from XP-D patients, but not from TDD patients, are more susceptible to one of the steps in UV-B induced immunosuppression (Ahrens et al., 1997). Thus, in addition to a faulty DNA repair process, XP-D patients also have a compromised immune response. The conclusion from this study is that faulty DNA repair alone may not be sufficient to cause the observed increase in skin cancer, it may need to be accompanied by a compromised ability to respond to UV-induced tumorigenesis (Ahrens et al., 1997).
Two additional sets of genes which appear to be able to influence the development of UV-B induced neoplastic responses are 1) those which affect the immune response and 2) those which act as growth regulators, either stimulating uncontrolled growth, e.g., proto-oncogenes, or restraining such growth, e.g., suppressor genes. In the case of those that influence the immune response, as demonstrated by the finding with XP-D patients, compromising the immune response to UV-induced tumors. In the case of growth regulators, a host of genes have been identified which when mutated (by UV or another insult), can result in the development of a tumor. Chief among these is the gene for p53 that has been discussed above in some detail. Mutations in p53 appear to be key to the development of both BCC and SCC, but not CM (Brash et al., 1996). Another gene, that which codes for cyclin dependent kinase inhibitor 2 (CDKN2) or p16, has been found to be very important in CM. Recent information suggests that CDKN2 is a melanoma tumor suppressor gene located on chromosome 9 which is particularly important to familial melanoma but may also have a role in sporadic melanoma (Naylor an Everett, 1996; Fountain 1998).
In the case of skin cancer, the epidemiologic evidence also appears to be somewhat conflicting. A summary of the information on NMSC presented in one recent review indicates that while one study found a protective effect of dietary factors, other studies found no significant benefit (Strom and Yamamura, 1997). In the case of melanoma, a number of different studies have examined either the consumption or serum levels of vitamin E, a -tocopherol, or b -carotene consumption and related them to risk. The results have been highly variable; e.g., in the case of consumption, vitamin E in foods, but not in supplements, was protective (Berwick, 1998). In the case of serum levels, some results suggest that low serum concentrations of a -tocopherol, or b -carotene were associated with higher risk (Armstrong and Kricker, 1995) whereas others studies found that plasma levels of a -tocopherol were not related to risk. Taken collectively, these results suggest that dietary interventions may be of little help in preventing or managing the risks of cataract and skin cancer from UV-B.
Given the importance of the immune response to the development of skin cancer, it was hypothesized early on that immunosuppression would have an impact on tumor development. Studies in renal transplant patients (whose immune response are suppressed in order to prevent rejection of a kidney transplant) confirmed this hypothesis by revealing a dramatic increase in warts and SCC on sun-exposed skin of these patients (Harteveld et al.,1990). The warts are known to be associated with human papilloma viruses (HPV), but the carcinomas and precursor lesions in these patients were also found to bear a great variety of HPV (Tieben et al., 1994), not generally found in SCC (Kawashima et al.,1990). It has, however, recently been found that HPV are commonly detectable in hair plucked from eye brows (Boxman et al., 1997). The question now is whether the HPV is merely a hitchhiker in proliferating carcinoma cells or whether it really plays a causal role in the development of these tumors. The skin carcinomas in people with renal transplants were also found to contain UV-related p53 gene mutations (McGregor et. al., 2997). These findings clearly show that a good immune system prevents the development of potential carcinomas on sun-exposed skin. As discussed above, the UV radiation from the sun can also exert immunosuppressive action and thus enhance the development of skin carcinomas.
Besides SCC, non-Hodgkin's lymphoma occurs much more frequently in people on immunosuppressive medication (Deeg et al., 1993). The risks of non-Hodgkin's lymphoma and skin cancer appear to be associated; people who were treated for skin carcinomas have an increased risk of non-Hodgkin's lymphoma (Frisch and Melbye, 1995; Frisch et al., 1994) and vice versa, i.e., people treated for non-Hodgkin's lymphoma have an increased risk of skin cancer (Adami et al., 1995). The persistent increase in non-Hodgkin's lymphoma over the last 4 to 5 decades parallel increases in skin cancer incidence and it is hypothesized that both these trends are due to increased exposure to sunlight (Cartwright et al., 1994). It has been speculated that the common factor in the etiologies of these cancers is the UV-induced immunosuppression (Goldberg, 1997) or more specifically, that cytokines (Vos et al., 1994) released upon UV exposure may stimulate the outgrowth of precursor cells of B lymphocytes to develop into a non-Hodgkin's lymphoma (Hirayama and Ogawa,1994). Non-Hodgkin's lymphomas have been reported to occur more frequently in the sunniest parts of Great Britain (Bentham et al., 1996), but in contrast to skin carcinomas, there is no increase in non-Hodgkin lymphomas toward the south in North America (Hartge et al.,1996). Interestingly, mice developed lymphomas as a consequence of exposure to UV radiation and a chemical carcinogen (7,12-dimethylbenz (a) anthracene), whereas treatment with each of these factors separately did not induce lymphomas (Husain et al.,1991). In sum, there appears to be an association between the risk of skin carcinomas and non-Hodgkin's lymphomas, but whether UV radiation is a risk factor for non-Hodgkin's lymphomas is not clear. More data are required to negate or confirm a direct relationship between UV radiation and non-Hodgkin's lymphoma.
The immune system can be compromised in many different ways, e.g. through medication or infections. An additional UV-induced suppression might then have more devastating impact than normal. A very timelytopical example of an affliction that cripples the immune system is AIDS (cause by HIV), and these patients could run an increased risk from UV exposures. However, AIDS patients with psoriasis have been treated with UV radiation and no aggravation of the AIDS was found (Adams et al., 1996; Meola et al., 1993). This result could be offset by the fact that psoriasis itself activates the immune system on which UV radiation then exerts a first dampening effect without further diminishing the resistance to HIV.
Certain medical treatments may add to the cancer risk from (solar) UV exposures, e.g., immunosuppressionradiation therapies for cancer (Berwick, 1998), PUVA, (a combination of UVA and oral dosing with 8-methoxypsoralen) treatments for psoriasis (Stern and Laird, 1994).
The PUVA treatment of psoriasis has
become very widely used but has been found to be associated with a substantial
increase in skin cancer risk in a long-term follow-up. Recently, it has
been reported that in the long run the risk of melanoma is also significantly
increased (Stern et al., 1997). The squamous cell carcinomas found on the
PUVA-treated individuals occurred frequently on skin areas that are not
regularly exposed to the sun but are exposed in the PUVA treatment: e.g.,
like the legs, where they do not commonly occur in the general public.
Curiously enough, many of the skin carcinomas taken from PUVA-treated patients
were found to have mutations in the p53 gene that pointed at UVB
radiation instead of the PUVA treatment as the direct cause (Nataraj et
al. 1997). This would indicate that UVB radiation may even have contributed
to the development of these clearly PUVA-related skin tumors. Clearly,
there are a number of factors that can amplify the risks from UV exposure,
and vice versa. Identifying high-risk populations will open up the possibility
for well-targeted mitigating strategies.