Spot the Difference: Why There Are Two Ways to Develop Melanoma

Cutaneous melanoma is a cancer with a clear environmental culprit: ultraviolet (UV) radiation. In the melanoma capital of the world, Australia, 95% of melanomas are attributable to UV radiation. Globally, it’s 75% (Arnold et al. 2018).

However, even in sun-drenched Australia, many melanomas form on relatively sun-protected body sites, like the lower back, instead of more heavily damaged areas, like the face. Why is this? This blog discusses how both UV radiation and non-UV pathways can promote melanoma.

The connection between sunlight and melanoma was made in the 1950s, when Lancaster showed that the risk of melanoma in white populations was directly associated with the latitude at which they lived. Soon after, the skin, hair, and eye characteristics that increased susceptibility to melanoma were described — pale skin that burns easily and tans poorly, freckles, red or blond hair, and blue eyes (Rebecca et al. 2013). (People with skin of color can also develop melanomas, but it tends to be the types not associated with UV radiation, like acral melanoma on the sole of the foot.)

More recently, a UV-specific mutational signature was discovered — single base substitution 7 (SBS7), where C bases are replaced with T bases. Almost all cutaneous melanomas have a large number of SBS7 mutations.

Where the Sun Doesn’t Shine (Much)

However, there’s another easily spotted phenotype linked to melanoma risk: having many nevi (moles). The link between moles and melanoma was noted in the 1800s, when doctors noticed that families with multiple cases of melanoma also had many moles (Rebecca et al. 2013). Having more than 20 moles increases your risk of melanoma 10-fold (Duffy et al. 2019). These people also tend to get melanoma at a younger age than heavily sun-damaged people. Moles are typically most dense on the trunk, and people with many moles are more likely to have melanoma on the sun-protected trunk rather than the more chronically sun-exposed face or limbs (Caini et al. 2009).

Is the answer that melanomas in mole-y patients are not caused by UV exposure?

Well, no. These melanomas still carry the UV mutation signatures SBS7 (C>T), although at lower levels than in melanomas on chronically sun-exposed sites (Stark et al. 2018).

Two Paths Diverged in a Skin…

Eventually, Whiteman et al. proposed that there are two pathways to melanoma formation that both begin with UV exposure and then diverge (Whiteman et al. 1998, Whiteman et al. 2003). Once UV exposure in early life has transformed normal melanocytes into “initiated” melanocytes, their fate depends on a mix of further UV exposure and host factors.

People have many moles because their melanocytes have inherently high proliferative potential, so relatively low levels of further UV exposure can trigger malignant transformation. They also often have a strong family history of melanoma and have had moderate or intermittent UV exposure. Other people have melanocytes that are less inclined to proliferate, so they need a higher level of UV damage, accumulated over a lifetime, to make the leap to melanoma (Figure 1).

Fig. 1. The divergent pathways to melanoma. Chronic UV exposure, often working on a background of pale, easily-sunburned skin, results in head, neck, and limb melanomas in older people. Host factors like a high mole count, family history of melanoma, and moderate UV exposure predispose people to melanomas on the trunk at a younger age.

Melanomas from the heavily sun-exposed head, neck, and lower legs have p53 mutations, are associated with a tendency to sunburn, and are associated with having many solar keratoses, a marker of chronic UV exposure, but not with having many moles. Melanomas on the more protected trunk tend not to have p53 mutations, are associated with having 60+ moles, and are not associated with solar keratoses (Whiteman et al. 1998, 2003).  

More recent work on polygenic risk scores supports the divergent pathways hypothesis. Polygenic risk scores take into account the small but cumulative increases in melanoma risk caused by at least 85 low-penetrance, small effect gene variants. Melanoma patients with low polygenic risk scores have markers of high cumulative UV exposure, while those with high polygenic risk scores had high-level ambient UV exposure in early life — being born and raised in Australia — but few markers of high cumulative exposure (Olsen et al. 2020).

Double Trouble

Sometimes, both pathways are at play — and increase a person’s risk synergistically instead of additively. For example, Australians with more than 20 moles have a 10-fold increase in melanoma risk compared to people with fewer than five moles. People with two copies of MC1R R mutations, which promote the chronic UV damage pathway by causing red hair and sun-sensitive skin, have a four-fold increase in melanoma risk compared to those with wild-type alleles. People with both 20+ moles and two MC1R R alleles? In the high-UV environment of Australia, these people have a 25-fold increase in melanoma risk and an absolute melanoma risk of 19.3% for women, 23.3% for men (Figure 2). In other words, these people have a one in 5 chance of developing melanoma instead of the Australian average of 1 in 17 (men) or 1 in 24 (women) (Duffy et al. 2019).

Fig. 2. Published data provided courtesy of Katie Lee showing the synergistic effect of MC1R genotype and mole count on melanoma risk. When people have elements of both melanoma pathways at play, their risk increases precipitously (Duffy et al 2019).

Genetic contributions to each pathway are also tangled. Several known melanoma susceptibility alleles are pleomorphic, influencing more than one phenotype: for example, IRF4 influences both mole count and skin pigmentation (Duffy et al. 2010). CDKN2A, MC1R, and MTAP mutations seem to influence both pathways: they are found in people with early-onset (under age 40) melanoma, representing the low UV pathway, but also in people with multiple melanomas all in UV-damaged areas (McMeniman et al. 2020).

Two Pathways, One Takeaway

Recent research in other cancers, like lung cancers (Hill et al. 2023), support this general pathway of initiated cells later triggered into full malignancy by other mutagens, inflammatory agents, or host genomic instability. This might be why the risk of lung cancer drops after you quit smoking — the smoking-induced mutations remain, but you’ve removed the inflammatory element of regular smoking. Likewise, Australians with a history of multiple primary melanomas may be able to halve their risk of developing a subsequent melanoma by using sunscreen, hats, and other sun-protective practices (von Schuckmann et al. 2019).

This is all promising news for both cancer researchers and people at risk of cancers with a strong environmental component. People with high cancer risk can take immediate steps to reduce environmental inputs that might encourage their initiated cells to become fully malignant. Researchers become one step closer to untangling the interactions between environmental and host causes of cancer, to enable personalized preventative advice and treatment for patients.