Skin cancer risk factors

Excess exposure to ultraviolet (UV) radiation is the main preventable risk factor for skin cancer.1,2 The sun is the principal source of natural UV radiation (Figure 4.1). Sunbeds produce artificial UV radiation.

Figure 4.1: Solar UV Radiation in Relation to the Earth and Ozone Layer

UVA and UVB are the two types of solar radiation which reach us on Earth. Both types are linked to skin cancer. UVB is predominantly responsible for burning, whilst UVA penetrates deeper into the skin and is linked with premature ageing.

It is also expected that climate change will cause more skin cancer cases in the future, as more UV radiation reaches us on Earth, and warmer temperatures encourage people to spend more time in direct sunlight.98

There is sufficient evidence that too much exposure to solar UV radiation is the main cause of both melanoma and NMSC in humans, according to the International Agency for Research on Cancer (IARC).3,4 It is estimated that around 11,100 (86%) melanoma cases in the UK in 2010 were linked to UV radiation exposure.1 Among NMSCs, an estimated 50-70% of squamous cell carcinomas (SCCs) and 50-90% of basal cell carcinomas (BCCs) in fair-skinned people are caused by UV radiation.5

Intermittent sun exposure and sunburn

Risk of melanoma is most strongly linked to intermittent exposure to high-intensity sunlight (for example from sunbathing, doing watersports or holidaying in a place where the sun is strong), a meta-analysis has shown.6 Intermittent sun exposure was associated with a 60% increased risk of melanoma, though this effect was smaller and not significant in studies of UK, US, Canadian or Australian populations.6

Intermittent exposure to high-intensity sunlight often results in sunburn,6 and a history of sunburn doubles the risk of melanoma.6,7 Having had 26 or more episodes of ‘painful’ or ‘severe’ sunburn during your lifetime increases the risk of melanoma by two to three times in women, a pooled analysis showed.8 Melanoma risk is increased regardless of whether sunburn occurred in childhood or adulthood.6,9

Sunburn, especially in childhood, or intermittent exposure to sunlight, also increases the risk of BCC.10,15 Sunburn and intermittent sunlight exposure is believed to have less of an effect on SCC risk.12,14,16

Exposure to UV radiation has increased in recent decades in the UK population, as people have increasingly sought a suntan by holidaying abroad.

Chronic sun exposure

Chronic or more continuous sunlight exposure, for example that received by people with outdoor occupations, did not appear to increase melanoma risk in a recent meta-analysis, though the review authors commented that occupational sun exposure still probably increases risk over no sun exposure at all.6

There is evidence that chronic sun exposure increases the risk of NMSC. People who work outdoors are at 43% higher risk of BCC,17 and 77% higher risk of SCC,18 and these effects are stronger in countries nearer the equator, two meta-analyses have found.17,18

section reviewed 23/07/12
section updated 23/07/12

There is sufficient evidence that use of sunbeds causes melanoma, IARC states.4 IARC also states there is limited evidence that sunbed use causes SCC.4 There is currently no IARC statement on sunbeds and BCC.

Use of a sunbed for the first time before age 35 increases the risk of melanoma by 59%, and use at any age increases melanoma risk by 20-25%, the most recent meta-analysis showed.19,103 Women using a sunbed once a month or more in their 30s increase their melanoma risk by 49%, and those doing so in their 40s face a 61% increased risk, one large study included in that analysis showed.20 Sunbed use is estimated to cause around one hundred deaths a year from melanoma in the UK.19,21

Sunbed use at any age increases the risk of SCC by 67%, and increases BCC risk by 29%, according to the most recent meta-analysis.101 Risk increases for both types of non-melanoma skin cancer in relation to sunbed use were also shown in an earlier meta-analysis.19 Exposure before age 25 appears to confer even greater risk increases, though in meta-analysis the effect was significant only for BCC (40% risk increase).101

Using a sunbed without ever burning appears to be no safer – it can increase the risk of melanoma and early-onset BCC (diagnosed in people under 40 years old) by more than 60%, recent case-control studies have shown.22,23

Prevalence of sunbed use

In 1999 a quarter of men and a third of women in Britain reported trying to get a tan in the previous six months. There were even higher rates amongst younger people (Figure 4.2). Overall, 2% of adults trying to get a tan did so using a sunbed or tanning machine only.24  In 2008 and 2009, 6% of 11-17-year-olds in England reported they had used a sunbed,25 and in 2008, 5% of under-25s in Northern Ireland reported currently using sunbeds.26

Use of sunbeds by under-18s is now banned across the UK, with legislation passed in Scotland in 2009, England and Wales in 2011, and Northern Ireland in May 2012.

Figure 4.2: Proportion of Adults that Tried to get a Tan in the Last Six Months, Great Britain, 1999

Download this chart XLS (55KB)

The possibility that younger people and those with high-risk skin types are at greatest risk of skin cancer due to sunbed use is widely recognised. In 2003 the International Commission on Non-Ionizing Radiation Protection (ICNIRP) and the World Health Organisation (WHO) recommended that certain categories of people should not use sunbeds (Table 4.1).27,28 ICNIRP also concluded that anyone using suntanning appliances is likely to raise their risk of skin cancer, eye damage, photodermatosis, photosensitivity and premature skin ageing.27 Cancer Research UK recommends that people do not use sunbeds.

Table 4.1: Categories of People who Should Not Use Sunbeds, as Recommended by ICNIRP and WHO

• Download this table (41KB)

section reviewed 23/07/12
section updated 18/01/13

People with light eyes, skin or hair, or with a phototype than means they sunburn easily or do not tan, have an increased risk of skin cancer.10,11,14,29,30 These factors in combination are used to define the skin type (Table 4.2). This classification is used in most studies exploring pigmentary characteristics and skin cancer risk.29

Table 4.2: Skin Types

Based on: Fitzpatrick T. Soleil et peau. J Med Esthet 1975;2:33-4.

• Download this table (42KB)

In comparison with people with skin phototype IV, those with skin phototype I are at more than double (2.27 times) the melanoma risk, phototype II at double (1.99 times) the risk, and phototype III at 35% increased risk, a recent meta-analysis reported.29

Table 4.3 gives incidence rates for malignant melanoma by major ethnic group in England.31

Table 4.3: Ranges of Age-Standardised Incidence Rates per 100,000 Population for Malignant Melanoma, 2002-2006 by Major Ethnic Group and Sex, England

• Download this table (41KB)

In comparison with dark-eyed people, those with blue/blue-grey eyes have a 57% higher melanoma risk, and those with green/grey/hazel eyes have a 51% increased risk.29

In comparison with dark-haired people, those with red/red-blonde hair were shown to be at up to triple the melanoma risk.29,32 Blondes are at double the risk, and people with light brown hair are at 46% increased risk.29

People with freckles were found to have around double (1.99 times) the risk of melanoma, versus people without freckles.29 People with freckles have increased melanoma risk, regardless of the number of moles they have.33

People with blue/green-blue/green-grey eyes are at increased risk of BCC.34 People with red and light-coloured hair are at increased risk of BCC and SCC.35-37

section reviewed 23/07/12
section updated 23/07/12

Meta-analyses show people with any unusually shaped or large moles (also called atypical naevi; these are usually larger than common naevi, with a more variegated appearance; poorly-defined border, and some areas slightly raised) have around four to ten times increased risk of melanoma,38,39 and the risk increases with the number of atypical moles.38 People with very high numbers (100+) of common moles on their bodies have nearly seven times the risk compared to people with very few (0-15 moles),38 and every additional common mole increases the risk of melanoma by around 2%.39

People with dysplastic mole syndrome (also known as Familial Atypical Multiple Mole-Melanoma Syndrome or FAMMM; characterised by multiple atypical moles that continue to appear in adulthood) and a family history of melanoma have a 500-fold increased risk of developing melanoma;40 however this is very rare and accounts for less than 5% of melanoma cases.38

Most moles are genetically determined (inherited), though sun exposure can increase the number of moles. Most moles appear during childhood.33,41 The emergence of moles in adolescents is under strong genetic control, a UK study of moles in twins concluded.42 Chronic sun exposure rather than number of sunburn episodes is the most important environmental factor determining mole development.33

section reviewed 23/07/12
section updated 23/07/12

The impact of sunscreen use on skin cancer risk remains unclear, due largely to methodological limitations.43,44

Sunscreen should be used together with clothing and shade to protect the skin from sun damage, and should never be used to spend longer in the sun. Research shows sunscreen users may counteract the protective effect of sunscreens by spending longer in the sun than non-users; applying their sunscreen incorrectly; or failing to use protective clothing.45-48

section reviewed 23/07/12
section updated 23/07/12

The only established benefit of exposure to solar UV radiation is the synthesis of vitamin D, which is vital for bone health. Higher circulating levels of vitamin D in the blood are associated with lower risk of colorectal cancer, although it is unclear whether this is a causal relationship.49-53

However, sunbathing, tanning or burning should not be necessary to make sufficient vitamin D to obtain health benefits. In fair-skinned people, the time taken to make enough vitamin D is short, and less than the time taken for skin to redden or burn.54 Once sufficient vitamin D is made, any extra is turned into inactive substances.55 So more sun exposure does not equate to greater health benefits, and excessive exposure to solar UV radiation is not a means of reducing the incidence or mortality of cancer.

section reviewed 23/07/12
section updated 23/07/12

People with a family history of melanoma have roughly double the risk of developing the disease, compared to people without a family history.56,57 A small percentage of melanoma cases (around 10%) are attributable to inherited risk.58,59

FAMMM is one of several rare hereditary syndromes associated with an increased risk of melanoma.60 FAMMM is associated with the hereditary susceptibility genes CDKN2A and CDK4. CDKN2A mutation carriers often have three or more family members with melanoma, or have multiple primary melanomas with no family history. CDKN2A mutation carriers living in Europe have a ∼58% risk of developing melanoma by age 80.61 Melanoma and NMSC have been linked to Li Fraumeni syndrome in some studies, although they are not among the ‘core’ cancers occurring in Li Fraumeni families.62

People with a family history of SCC have an increased risk of SCC.63 People with a family history of melanoma have an increased risk of BCC.64

section reviewed 23/07/12
section updated 23/07/12

Previous melanoma is associated with eight- to twelve-fold increased risk of a second melanoma.65,66 The effect is stronger for women and for invasive (versus in situ) melanoma.65,66 People with a previous melanoma and a parent with melanoma are at more than 30-fold risk of a second melanoma.67 Melanoma risk is higher among people with a previous diagnosis of various other cancers, including female breast cancer;68,69 non-Hodgkin lymphoma;70,71 renal cell carcinoma;72 certain childhood cancers;73 prostate cancer;68,74 thyroid cancer;68 and leukaemia.68 Generally the increase in risk was less than double. Often these associations are bi-directional,66 supporting shared genetic or environmental factors.

Previous SCC is associated with ten times higher risk of a second BCC or SCC, whilst previous BCC is associated with ten times higher risk of second BCC but a lower increase in second SCC risk.75-77 Previous melanoma is associated with three-fold increased risk of NMSC.78 People who have had NMSC are also at increased risk of other second primary cancers.79

section reviewed 23/07/12
section updated 23/07/12

Organ transplant recipients are at 29-fold increased NMSC risk, and two-fold increased melanoma risk, a meta-analysis shows.80-82 This may be related to the use of immunosuppressant drugs called azathioprine and cyclosporine, which IARC states are causes of NMSC.4

People with atopic dermatitis (the most common form of eczema) appear to have an increased risk of NMSC, but the association between atopic dermatitis and melanoma remains unclear.83,84 Contact allergy may reduce the risk of NMSC very slightly.85 A drug called methoxsalen is used in conjunction with exposure to UVA to treat eczema, and IARC classifies this as a cause of NMSC.4

Melanoma risk is apparently doubled in men with Parkinson’s disease, but there is no significant association for women, or for NMSC.86

Rheumatoid arthritis patients taking tumour necrosis factor inhibitors are at 45% higher risk of NMSC,87 but they do not have a higher risk than patients taking other disease-modifiying anti-rheumatic drugs.99 Their melanoma risk does not appear significantly altered.87

People with HIV or AIDS have been shown to have increased melanoma and NMSC risks;80 presently IARC states that there is limited evidence that HIV type 1 infection causes NMSC.4

section reviewed 23/07/12
section updated 23/07/12

Taller women appear to be at increased risk of melanoma and BCC. One study showed women taller than 5 feet 3 inches were at 28-64% higher BCC risk than shorter women; there was no effect of height for men.34  Melanoma risk increased by 32%-51% for every 10cm increase in height, according to recent large studies.88,100

Melanoma risk increases by 17% per 5kg/m2 increment in body mass index (BMI) for men, a meta-analysis reported.89 There does not appear to be a significant effect of obesity on melanoma risk for women.89,90 Obesity is associated with decreased NMSC risk, perhaps due to less UV exposure in larger people.91 Women with BMI lower than 25kg/m2 were at 26-43% higher BCC risk and 20-41% higher SCC risk than women with a higher BMI, US cohort studies have shown.34,91 For men in these analyses, the difference in BCC risk was only significant for people with BMI 30-34.99 versus BMI under 25, and there was no significant effect of BMI on SCC risk.34,91

There is sufficient evidence that X-radiation and gamma radiation (both types of ionising radiation) cause non-melanoma skin cancer, according to IARC.4 Radiotherapy for a previous cancer is estimated to have caused 17.9% of second primary melanoma cases in women and 2.8% of second primary melanoma cases in men in 2010.92 Exposure to cosmic radiation has been posited as an explanation for the higher rates of melanoma in airline staff, but recent evidence suggests that excessive UV exposure and sun-sensitive skin phenotypes are more likely causes.93,94

Some chemical exposures that can take place in certain occupations cause NMSC, IARC states.4 These include coal tar pitch, soot, mineral oils and shale oils. It has been estimated that around 7% of NMSCs in men and around 1% in women in Britain are due to occupational exposures (including solar radiation).95

People diagnosed with genital warts (associated with infection with HPV types 6 and 11) have a 30% increased risk of BCC.102

section reviewed 23/07/12
section updated 23/07/12

Use of oral contraceptives or hormone replacement therapy does not significantly impact on melanoma risk in women, a meta-analysis shows.96 Other reproductive factors (age at birth of first child, number of children) show small effects on melanoma risk, though these effects are largely explained by socio-economic factors.96

Unlike for many other cancers, smoking and alcohol consumption do not appear to increase melanoma or NMSC risk overall.4 However, smoking appeared to slightly increase SCC risk but not BCC risk, when NMSC subtypes were examined separately in a recent meta-analysis.97

section reviewed 23/07/12
section updated 23/07/12