Plain language summaries of GenoMEL published papers

GenoMEL research carried out to identify new inherited genes which increase the risk of melanoma

We all have inherited variants of genes which increase the risk of cancer. Most of these are common and are associated with an increased risk of melanoma (albeit a small increased risk). These are called low penetrance genes.Although each gene has a very small effect on risk, when we have many of such genes then our cancer risk is considerably increased. Some inherited genetic variants or mutations are more serious and these are associated with a very much increased risk. These are called high penetrance genes. They  are rare and they are associated with familial cancer. For melanoma a very small proportion of melanoma patients carry these genes, and the most common is a gene called CDKN2A. The protein this gene produces is called p16. Mutations in this gene have been known to increase the risk of melanoma since 1994.

One of the main aims of GenoMEL is to carry out research on these inherited genes. GenoMEL has worked:-

  1. To establish what having inherited genetic mutations means for family members in terms of their risk of cancer (most has been done for CDKN2A families as this is the most frequent mutation found)
  2. To find out if the inherited genetic mutations increase the risk of other cancers as well as melanoma
  3. To determine the biological effects of the mutations are as this tells us something about how melanoma is caused and in the end that might help us to find new treatments for melanoma
  4. To find new high penetrance (familial melanoma genes) causing the melanoma in families who do not have the CDKN2A mutation. This has proved a very difficult task but in very recent years (from 2014) new mutations have been found.
  5. To find low penetrance genes which increase the risk of melanoma. These genes are usually found using genome wide association studies (GWAS) and GenoMEL is carrying out a very large study indeed.

Two papers have in 2014 been published which show that some families with melanoma have inherited genetic mutations in a gene called POT1

POT1 loss-of-function variants predispose to familial melanoma.

Robles-Espinoza CD, et al. Nat Genet. 2014 Mar 30.

Some families have far more cases of melanoma than we would expect by chance alone. In roughly half of these families this increased risk is explained by changes in genes that control how cells divide (CDKN2A). However, the other 50% of these families do not carry any of the known mutations (harmful changes in a gene).

Several GenoMEL groups looked at genes in over a hundred families from the UK, the Netherlands and Australia that do not carry any known mutations. The work described in this paper was carried out  with Dr David Adams at theSanger Institute, and Dr Robles-Espinoza was a PhD student in his laboratory. We found mutations in a gene called ‘POT1’ that increased the risk of melanoma in a small number of these families. POT1 plays a role in protecting the ends of our chromosomes*, a little like the caps on the ends of shoelaces that stop them fraying. When the POT1 gene works correctly it protects the chromosomes from stress and damage, and also controls their length.

Comment: This is an important discovery. However, it will take more work to understand what carrying a mutation in the POT1 gene means for families in terms of risk of melanoma and other cancers. We expect that some time in the future genetic counsellors will start counselling families and testing for this mutation if patients choose this option. GenoMEL is continuing its research to provide the information to make this possible.

The scientific abstract is available on the PubMed website

Simultaneously a number of additional GenoMEL groups found the same finding so the two papers were published together and this is the reference for the second paper

Rare missense variants in POT1 predispose to familial cutaneous malignant melanoma

Shi J, Yang XR, Ballew B, Rotunno M, Calista D, Fargnoli MC, et al. . Nat Genet. 2014 May;46(5):482-6.

*A helpful definition of chromosomes can be found at


The most frequent high penetrance melanoma gene found to be mutated in melanoma families is CDKN2A and the protein produced by the gene is called p16. p16 binds to CDK4 to control how fast cells divide. In 1996 two families were shown to have inherited mutations in  the CDK4 gene which increased their risk of melanoma. Families with this mutation are now known to be very rare. GenoMEL groups looked at what having this sort of mutation means for melanoma families in the paper below by Puntervoll and colleagues.

Melanoma-prone families with CDK4 germline mutation: Phenotypic profile and associations with MC1R variants

Puntervoll HE et al. J Med Genet 2013;50:264-270

Families with many cases of melanoma tend to have harmful changes (mutations) in genes that control how rapidly cells divide. The most commonly affected gene is called CDKN2A, and the features of families with mutation in this gene have been well described. A much less common cause of inherited melanoma is mutations in a gene named CDK4. We wondered whether families with CDK4 mutations had different features to families with CDKN2A mutations.

We looked at all the families with known CDK4 mutations across the world. This included families from Norway, the UK, Latvia, France, Italy, Greece, the USA and Australia. We found that CDK4 families were similar to those with CDKN2A mutations including the age of onset, body location and type of melanoma, and number of melanomas.

Comment: This study showed that it is not possible to make a distinction between CDK4– and CDKN2A-mutated families before they are genetically tested. Therefore, when a person belonging to a melanoma-prone family seeks genetic counselling and testing, the CDK4 gene should always be examined if a CDKN2A mutation cannot be found.


The following papers reported the results of GenoMEL work carried out which might be of value to genetic counselors and therefore people seeking information about their risk of cancer

Features associated with germline CDKN2A mutations: a GenoMEL study of melanoma-prone families from three continents.
Goldstein et al. 2007 J Med Genetics 2007; 44: 99-106.

The genes that we inherit from our parents can affect our risk of developing certain diseases. Sometimes genes have faults (called mutations) that can increase our chances of developing a disease. Certain mutations in a gene called CDKN2A can increase your chances of developing melanoma. This paper describes how GenoMEL looked at the families (with 3 or more cases of melanoma) within its studies to see what having these CDKN2A mutations can mean. 385 families were involved across Europe, North America and Australia. Each family contained three or more relatives who had previously developed a melanoma.

Of the total number of families studied 39% had CDKN2A mutations. However, the percentage of families with a mutation varied by continent. Only 20% of the families in Australia contained a CDKN2A mutation, whilst in North America and Europe the figure was 45% and 57% respectively. We have speculated that these differences my have occurred because in very hot countries such as Australia, 3 people with melanoma might occur just because the family has red hair and has had a lot of sunshine rather than inheriting a rare higher risk gene.

In comparison with other families it is known that families with a CDKN2A mutation may have:

  • More cases of melanoma.
  • Melanomas diagnosed at an earlier age.
  • More family members being diagnosed with more than one melanoma.
  • More family members being diagnosed with cancer of the pancreas.

GenoMEL looked to see if these four features were related to CDKN2A mutations in the 385 families studied. It found that the features were related to the presence of a mutation, although the Australian families did not show an increased risk of pancreatic cancer. Also, the relative importance of these features varied by continent.

The strongest link between having a mutation in this particular gene in the family and having multiple family members with melanoma, early onset, and so forth, was in Europe followed by North America and then Australia. This is probably explained by the varying rates of melanoma on each continent and the role of the environment. Australia has the highest rate of melanoma of the three continents due to its very sunny climate and its population largely having lightly pigmented skin. Therefore, in comparison to families on the other continents, Australian families with several cases of melanoma were more likely to be due to these factors rather than possessing a CDKN2A mutation.

The way in which our genes, environment and life style interact influences our chances of developing melanoma. This paper adds to our understanding of this interaction, and GenoMEL is continuing its studies to learn more.

The original paper can found at the web site of the Journal of Medical Genetics –


High-risk Melanoma Susceptibility Genes and Pancreatic Cancer, Neural System Tumours, and Uveal Melanoma across GenoMEL
Goldstein et al. Cancer Research 2006: 66: (20) October 15, 2006.

GenoMEL has collected blood samples to look for genetic changes (mutations) that increase the risk of developing melanoma. We looked at particular genes, called CDKN2A and CDK4, using the samples given for research by melanoma families. GenoMEL also looked to see if having mutations in these genes increased the risk of pancreatic cancer, nervous system tumours or some rare melanomas of the eye (uveal melanoma).

This study included 2,137 participants from 466 families from around the world. These families had at least three people who had developed a melanoma.

Overall, 41% of families in this study had one of these mutations. Most of these mutations were in the CDKN2A gene, affecting a protein called p16. The CDKN2A gene is very unusual in that mutations can also affect another, quite different, protein called ARF. Although mutations affecting the ARF protein were found in some melanoma families, they were rare. Occasionally there are mutations that affect a different gene, called the CDK4 gene, but these were also uncommon. There were only five mutations in the CDK4 gene and seven mutations in the part of the CDKN2A gene that affects ARF.

There were large differences in the types and numbers of particular mutations found in different geographical areas.

Most families studied had an increased risk of melanoma. Mutations in the CDKN2A gene also increased the risk of pancreatic cancer in some families, but the level of that increase varied.  There was also variation between countries so that there was no definite increased risk of pancreatic cancer in Australian families.

The increased risk of pancreatic cancer is clearly of great concern to families and to GenoMEL, and we are hard at work determining how to predict which families are at increased risk. At the moment, it is not clear whether the increased risk in some families is due to the type of mutation or to interaction with environmental factors such as smoking.

There did not appear to be an increased risk of eye (uveal melanoma).  The results for nervous system tumours were less clear. There was a possible relationship between nervous system tumours and mutations altering ARF but more work is needed to understand this relationship.

This GenoMEL study provides the most detailed description of mutations in these particular high-risk genes in families with three or more melanoma cases to date.

The original paper can found at the web site of Cancer Research at:


GenoMEL has been using genome wide association studies (GWAS) to find inherited low penetrance genetic variation associated with an increased risk of melanoma: that is, common genes which increase risk just a little. Where people inherit several or many of these common genes then they are especially at risk. In any country, people who inherit more of these low penetrance melanoma genes are more likely to get melanoma if they sun bathe. Many of the genes are associated with pale skin, skin which burns easily, freckles and lots of moles.

The first paper was published in 2009, and a series of more recent papers have shown increased numbers of these low penetrance melanoma susceptibility genes.


Genome-wide association study identifies three loci associated with melanoma risk
Bishop et al. 2009 Nature Genetics 2009; 41, 920-925.

GenoMEL used the latest technology to look for genetic factors that increase the risk of melanoma. This study involved blood samples from over 10,000 people from across Europe and Australia. It looked for differences between two groups of people: those with a history of melanoma (cases) and those without (controls). The genetic make-up of each individual was examined for the presence of over 300,000 variations.

A number of clear genetic patterns emerged. It is already known that people with red hair, fair skin and a tendency to sunburn are most at risk of melanoma. Not surprisingly, the cases were found to be much more likely to be carrying the genes associated with red hair and freckles. However, the association was even stronger than expected.

Other genetic regions were also found to differ between cases and controls. These regions are not associated with skin colour but are associated with the number of moles a person has. It has been known for some time that people with many moles are at increased risk of melanoma.

Finally, the study showed that these same genes influenced the risk of melanoma across all the countries represented in the study and to approximately the same extent.

The original paper can be found on the Nature Genetics website.



GenoMEL has carried out research to find out what sort of sun exposure causes melanoma

In particular, is it holiday sun exposure that causes melanoma predominantly or is it both holiday sun and chronic sun exposure (such as one would have working outside)?


Sun exposure and melanoma risk at different latitudes: a pooled analysis of 5700 cases and 7216 controls

Chang et al., International Journal of Epidemiology 2009; 38: 814-30.

Although it is well known that the risk of melanoma is influenced by exposure to the sun, it is less clear exactly what aspects of sun exposure increase risk. We investigated this by combining data on different aspects of sun exposure obtained from 15 studies of melanoma patients (cases) compared with individuals without melanoma (controls). The studies were conducted in different parts of the world, some with much sunnier climates (lower latitude such as Australia and Hawaii) than others, so that we could see whether the findings were different in different climates.

At all latitudes, sunburn (especially in childhood) and recreational sun exposure, including sunbathing, were clearly related to an increased risk of developing melanoma, especially on parts of the body not usually exposed to the sun (such as arms, legs, back and trunk).

More continuous sun exposure (e.g. through working outdoors) showed some evidence of being related to increased risk of melanoma on the limbs -but only in hot countries.

——————————————————————————————————————————————————————————————————————————————————————————————————————————————–Relationship between sun exposure and melanoma risk for tumours in different body sites in a large case-control study in a temperate climate

Newton-Bishop et al., European Journal of Cancer 2011; 47: 732-41.

Although it is well known that the risk of melanoma is influenced by exposure to the sun, there is a lot still to understand about the patterns of sun exposure that are harmful. We conducted a study of 960 melanoma patients (cases) and 687 individuals without melanoma (controls) in Northern England, a region with low sun exposure, to investigate this.

A reported history of sunburn was associated with higher risk of melanoma. Individuals with features that are associated with a tendency to sunburn, such as fair skin, red or blonde hair and freckling, were also at greater risk.

Although people with more holiday sun exposure were more likely to report having been sunburnt, the amount of holiday sun exposure itself was not associated with increased risk of melanoma. This may be partly because sun-sensitive people (those people who are at increased risk of melanoma) reported less sun exposure than those who are not so sensitive. That is, that the failure to see a relationship between recreational sun exposure and risk which our previous studies have so clearly shown (Chang et al), might be because in a smaller study its difficult to “allow” for the effects of inherited variation in risk.

Average weekend sun exposure in the warmer months of the year was associated with a lower risk of melanoma, especially on the arms and legs.

Overall, it is clear that sunburn should be avoided, which means that those prone to burning must take particular care to avoid excessive sun exposure. However this study also suggests that in cooler climates moderate sun exposure without sunburn may be protective against melanoma. The reasons for this are not yet understood, but possible explanations are the beneficial effects of higher vitamin D levels or acclimatization to the effects of the sun.


Swimming pool

GenoMEL work performed on prevention of melanoma by avoidance of intense sun exposure and sunburn

Although health educational research designed to improve prevention of melanoma is not the main aim of the GenoMEL consortium, it is important to us that approaches to prevention are improved. GenoMEL therefore carried out some projects as part of its Network of Excellence research grant from the European Commission under Framework 6

Predictors of Sun Protection Behaviours and Severe Sunburn in an International Online Study
Bränström et al. Cancer Epidemiology, Biomarkers and Prevention 2010, 19: 2199-2210

Several things are clear about melanoma and its cause:-

  • The rate of new melanoma cases continues to increase in many countries.
  • Melanoma is largely caused by sun exposure on holidays or in recreational activities in people who are susceptible
  • The melanoma risk factor that is potentially at least easiest to change, is sun exposure.
  • Avoiding sunburn is essential in reducing the risk of melanoma.

We researched how people protect themselves from the sun. We also looked at how people’s behaviour related to different melanoma risk factors. We asked people whether they used shade and clothing to protect their skin from the sun, as well as about other behaviours such as wearing sunscreen.

A total of 8,178 individuals from across the world completed the survey. Half of them reported at least one severe sunburn during the previous 12 months. Perhaps surprisingly, even 27% of those with a previous melanoma also reported at least one severe sunburn during the previous 12 months. Use of protective clothing and shade, as well as avoiding midday sun exposure, were more strongly related to reduced risk of sunburn than sunscreen use. Unsurprisingly, those who considered sun protection to be difficult and impractical, and those who reported a positive attitude towards suntans, were less likely to protect themselves.


Despite public health messages about the importance of sun protection, many people, including some with a previous melanoma, are not protecting themselves effectively resulting in severe sunburn.

If we want to reduce the number of people getting sunburnt we have to do rather better at empowering people at risk to use adequate  sun protection.

——————————————————————————————————————————————————————————————————————————————————————————————————————————————Melanoma risk factors, perceived threat and intentional tanning: an international online survey
Branstrom et al. European Journal of Cancer Prevention 2010, 9:216-26

In this second piece of research we looked at why people seek a tan and how tanning is related to different melanoma risk factors. Members of the public in Europe, Australia, the USA and Israel were invited to complete an online survey. A total of 8,178 individuals completed the survey. Overall, 70% reported some degree of intentional tanning during the past year. Perhaps surprisingly, 38% of respondents who reported a previous diagnosis of melanoma had also intentionally tanned. Preference for a dark suntan seemed to be the strongest reason for intentional tanning even amongst those with a previous melanoma.


This research suggests that the desire to have a tan is so strong, that many people from different countries generally thought it was more important to tan than reduce their risk of developing melanoma.

The desire to have a tan therefore remains very strong despite the fact that as a result, melanoma continues to become more common.

Health educationalists have to find a way of explaining the risks and how to reduce those risks.

A daughter consortium of GenoMEL, called BioGenoMEL, works on understanding what factors affect survival from melanoma.

BioGenoMEL has worked to study how inherited genetic variation affects survival. That is, that BioGenoMEL is asking whether people inherit differences in how their bodies defend them against the melanoma. If we can understand this then we may be able to improve survival.

The first genes BioGenoMEL looked at were those we know increase the risk of melanoma. The most common low risk melanoma gene is MC1R, the red hair and freckle gene.



Inherited variants in the MC1R gene and survival from cutaneous melanoma: a BioGenoMEL study

Davies JR, Randerson-Moor J, Kukalizch K, Harland M, Kumar R, Madhusudan S, et al. Pigment Cell & Melanoma Research. 2012 25:384-94


Particular versions of a gene called MC1R cause freckles and red hair. We investigated whether such versions are also associated with survival in melanoma patients. We suspected they might be because the MC1R gene is known to interact with other genes involved in cancer.

We recruited melanoma cases, looked to see what type of MC1R genes they had, and monitored their survival over a period of years.  We found that the MC1R variants associated with red hair colour were also associated with better survival in melanoma patients.  We found support for our findings when we combined the data of nine other groups from Europe and the USA.

Comment: This study showed, to the best of our knowledge, the first evidence that inherited variation could affect survival in melanoma patients. Also, while red hair is a known risk factor for developing a melanoma it may indicate a higher chance of survival.


The scientific abstract is available on the PubMed website


Since BioGenoMEL published this study another research group confirmed the findings stengthening the significance of the results. This is the reference for the second paper.

Inherited variation at MC1R and ASIP and association with melanoma-specific survival.

Taylor et al Int J Cancer 2015: 136: 2659-67

High risk of tobacco-related cancers in CDKN2A mutation-positive melanoma families.


Helgadottir H et al. J Med Genet. 2014 Aug;51(8):545-52


In this study we investigated cancer risks in members of Swedish melanoma-prone families that are carriers of an inherited mutation in the tumor suppressor gene CDKN2A. The carriers all have an identical “founder” mutation in the same region of the gene. This means that they are all, going back many generations, related, through the same ancestor in which the mutation occurred many centuries ago.


The information on the cancers diagnosed among known carriers and their relatives was collected through the Swedish Cancer Registry, which has a near complete coverage of all cancers diagnosed in the country. The carriers of the CDKN2A mutation were found to have significantly increased risk to develop melanoma and other cancers. High risks were seen for cancers in the respiratory and upper digestive organs, particularly in the pancreas, mouth, throat, lungs, gullet and stomach. Smoking has previously been shown to increase the risk of these cancers in the general population and people who had inherited he genetic mutation were at least 15 times more likely to develop one of these smoking related cancers.


Through questionnaires and medical records, information was collected on gene carriers’ history of tobacco smoking. The majority of the carriers that were diagnosed with cancers in respiratory and upper digestive organs had a history of tobacco smoking. It appears that smoking increases the cancer forming effect of the mutation and increases the cancer risks in the CDKN2A mutation carriers.



The increased risk for tobacco smoking associated cancers was seen in carriers of the Swedish founder mutation in the CDKN2A gene. The association with tobacco smoking has also been seen among Dutch carriers of the Leiden founder mutation (Pojter TB et al. 2015). These two studies suggest that avoidance of smoking is especially important for carriers of these genetic mutations.


The Swedish founder mutation is less common outside Sweden and it is uncertain if carriers of other mutations in the CDKN2A gene have the same risk spectrum and more research is needed on this.  It is quite probable however that this is so and so, for now, we inform all carriers of CDKN2A mutations that smoking can further increase their risk of cancers and recommend total abstinence from tobacco smoking.


Open access full text of this article: