This entry is part 1 of 30 in the series Our milestones
We have been funding cutting-edge research for more than 100 years. As we’ve explained, cancer research is a careful and painstaking process. While it’s often relatively easy to see the impact of, say, a clinical trial for a new treatment, it generally takes years of research and testing even to get that far.
So, often it’s only with the benefit of hindsight that we can see how research into a tiny molecule or a specific gene ends up making a difference.
In this series of posts, we’ll be picking out some of the most important discoveries and breakthroughs that we’ve been involved in – from fundamental cancer biology to clinical trials and population studies – to explain how they’ve made a major impact on our mission to beat cancer.
So in no particular order, we turn first to a fascinating piece of gene research from 2002.
CHEK2 – a new breast cancer gene
This post in Our Milestones series is based on research published in May 2002.
Entitled “Low-penetrance susceptibility to breast cancer due to CHEK2*1100delC in noncarriers of BRCA1 or BRCA2 mutations” – it’s a great example of genetic research, showing how the investigation of a single family’s genes led to the identification of a gene fault that may be responsible for a significant minority of inherited breast cancers.
Who did the research?
The work was carried out by an international team of scientists, including Professor Doug Easton and his colleagues at the Cancer Research UK Genetic Epidemiology Group at the University of Cambridge, and the results were published in the journal Nature Genetics.
The work was funded by Cancer Research UK, the Dutch Cancer Society, and several other European cancer research organisations
Researchers carried out analysis of a family where several members had been diagnosed with breast cancer at a relatively early age. Such a ‘clustering’ of cancer cases points the finger at cancer risk being inherited in this family.
The idea of inherited cancer risk was certainly nothing new in 2002. Indeed, the clustering of breast cancer within families was noted even in Roman times, and documented more than 100 years ago by the French surgeon Pierre Broca. Modern science has now pinpointed the ‘gene’ as the unit of inheritance, and shown that faults in genes can be passed down from parents to their children, sometimes increasing their risk of disease.
The family in this study were chosen for investigation because it was already known that they did not possess faults in BRCA1 or BRCA2, two previously identified ‘high-risk’ breast cancer genes. Inheriting a faulty version of one of these genes increases a person’s risk of developing breast cancer. But they are rare in the population and only cause 2-5 per cent of all breast cancer cases.
Because members of the family did not have faulty BRCA genes yet still seemed to have increased rate of breast cancer, the researchers hoped to pinpoint other gene mutations that might account for their susceptibility to breast cancer. And, importantly, they wanted to work out whether any such faults might account for breast cancer in the wider population.
What did they do?
Using well-established laboratory techniques like DNA sequencing together with some hard-core statistical analysis, the researchers studied samples of genetic material taken from family members with and without the disease to try to identify genetic regions that could be linked to breast cancer. Their results pointed them in the direction of a particular region on chromosome 22 that seemed to be linked to the cancer in this family.
Because of earlier work, the team already knew that a gene called CHEK2 is found in this region. This was encouraging, as other researchers had also found that the protein encoded by CHEK2 is involved in stopping cells from dividing when DNA is damaged – an important mechanism used by the body to stop damaged cells from multiplying and potentially causing cancer.
It was therefore reasonable to suggest that mutations in CHEK2 could be linked with breast cancer.
The next step was to figure out whether CHEK2 was damaged in this family, and if so, whether this damage to CHEK2 might be the reason the family developed more cancers.
Gene faults – or mutations – can occur when a ‘typo’ occurs in the normal sequence of building blocks that a gene is made of.
Mutations, like typos, can have varying effects. Sometimes they go virtually unnoticed – but other mutations can have a dramatic impact on how the cell functions. To use our analogy, a typo that changed “onion” to “oniun” wouldn’t have much of an effect on a recipe for soup, but changing “tomato” to “potato” certainly would.
Others researchers had identified a specific fault which was known to stop the CHEK2 protein working properly. Maybe, thought the scientists, this known fault could be causing the cancers?
So, to test their idea, the researchers decided to give the whole family a genetic test to see if any of them were carrying this mutation.
What did they find?
Of the 17 people who had breast cancer in this family, 7 people had the CHEK2 mutation – more than you’d expect by chance. Clearly, CHEK2 was at least partly to blame for this family’s tragic history of cancer.
But did this apply to just this family? Or was this affecting other people too? The scientists needed more evidence to say for sure that faulty CHEK2 played a role in breast cancer in the general population.
And this is where the study really came into its own. Now that they had a hunch that CHEK2 mutation was possibly causing the cancers in this family, the scientists tested other, much larger numbers of people from the UK, the Netherlands, North America and Germany for the presence of the CHEK2 mutation.
They classified all these people into different groups, including:
- A healthy, ‘control’ group – that is, people who had not been diagnosed with cancer.
- People with inherited breast cancer, from families without faults in their BRCA genes (the ‘BRCA-negative’ group, like the original family studied).
- People with breast cancer who had not been selected based on their family history (called the ‘population-based group’). This was to get a snapshot of how many people with breast cancer in general, irrespective of family history, carry the CHEK2 fault.
By comparing how many people in each group had the CHEK2 mutation, the scientists reached some interesting conclusions.
Firstly, and importantly, around 5 of every 100 people from the BRCA-negative group carried the CHEK2 mutation, but only about 1 in every 100 people in the ‘no breast cancer’ control group had it. This was evidence that CHEK2 mutations increase breast cancer risk at least in certain people.
However, equally important, there was only a tiny difference in how many people had the CHEK2 mutation in the population-based breast cancer group versus the control group. This means that although the CHEK2 mutation increases breast cancer risk, it only increases it by a relatively small amount.
What’s the impact of this work?
So, this is an interesting study, scientifically speaking. But what does this mean to you and me, seven years on? The scientists sum up by calling the CHEK2 mutation a ‘low penetrance mutation’, which simply means that a person carrying the fault has a ’medium’ risk of developing breast cancer compared to the ‘high risk’ BRCA genes.
So knowing that someone has the CHEK2 mutation alone may not be very helpful. Why then have we highlighted this as an important paper?
Well, this is where we come back to a recurring blog theme. Cancer research happens through incremental discoveries and innovations, the impact of which may not be instantly obvious.
What makes this work so noteworthy is that, although they had been thought to exist, this was the first time ever that a medium-risk breast cancer gene was located. It therefore opened the door to an avalanche of genetic research looking for other such genes.
Simply put, CHEK2 lent substantial weight to an important theory (which has subsequently been accepted as fact by many cancer researchers) – that for many people, their breast cancer risk is determined my many ‘medium’ or even ‘low’ risk genes, all acting together.
Since 2002, researchers, including many funded by Cancer Research UK, have found several other low- and medium-risk breast cancer susceptibility genes. With each such discovery, and further research into how we can reduce our risk of cancer, the tantalising prospect of personalised breast cancer care becomes more real.
We’re not there yet, but tangible progress is being made, and will continue to be made with the valuable help of our supporters.
We’ve covered a couple of related topics about breast cancer screening and genetic tests for cancer in other posts.
Olly Childs, Senior Science Information Officer
Broca P. Triate des Tumeurs. Paris: Asselin (1866)
Chehab et al. (2000). Chk2/hCds1 functions as a DNA damage checkpoint in G(1) by stabilizing p53 Genes Dev , 14 (3), 278-288
Meijers-Heijboer, H. et al. (2002). Low-penetrance susceptibility to breast cancer due to CHEK2*1100delC in noncarriers of BRCA1 or BRCA2 mutations Nature Genetics, 31 (1), 55-59 DOI: 10.1038/ng879
Wu, X. (2000). Characterization of Tumor-associated Chk2 Mutations Journal of Biological Chemistry, 276 (4), 2971-2974 DOI: 10.1074/jbc.M009727200
- CHEKing genes for breast cancer clues
- Finding faults in the BRAF gene
- Discovering the p53 cancer protein
- Kaposi sarcoma and AIDS – unravelling a medical mystery
- Carboplatin and the “Calvert formula”
- EGFR – Wanna be starting something?
- Chasing down the APC bowel cancer gene
- A story of Myc and death
- Tracking down the BRCA1 gene
- Tracking down the BRCA2 gene
- The story of vismodegib and skin cancer
- Tamoxifen – the start of something big
- The story of temozolomide
- Counting copies: HER2 and the development of Herceptin
- The link between working with asbestos and mesothelioma – case closed?
- Changing the future of pancreatic cancer: The ESPAC trials
- Mustard gas – from the Great War to frontline chemotherapy
- Classifying leukaemia: one size doesn’t fit all
- Understanding how cells divide – the story of a Nobel prize
- The shape of all things – the year of crystallography
- Our milestones: A recipe for success – the birth of combination chemotherapy
- Our milestones: the discovery of the NRAS signalling gene
- Our milestones: Cisplatin – the story of a platinum-selling life-saver
- Our milestones: the birth of abiraterone for prostate cancer
- Our milestones: How anastrozole became a number one hit
- Our milestones: How a surprise discovery is helping improve cancer diagnosis
- Our milestones: How a combination of events led to a cure for choriocarcinoma
- Our milestones: A radical change that improved breast cancer surgery
- Our milestones: Nudging breast cancer radiotherapy in the right direction
- Our milestones: Europe’s search for an anti-cancer diet
Coco January 19, 2011
“This might sound scary” for gods sake can this condescending crap tone be dropped. Its like they are talking to a child about a hairy spider. Just give us the data clearly and drop the shock horror childish commentary in parts of the response to what was a clear, concise and well argued post.
Michael Regnier April 24, 2009
The final analysis of the stats, as presented, seems a bit backward to me, and it would be helpful to see the statistics turned around.
Olly, you say the CHEK mutation occurs in about 1% of the population, slightly higher in the population of people with all genotypes of breast cancer, and in about 5% of BRCA negative people with breast cancer. This doesn’t immediately yield conclusions about the influence of the CHEK gene on breast cancer, does it? Formally, I’d say it would suggest an association between BRCA negative breast cancer and the CHEK mutation, but not necessarily a causal one and not necessarily with CHEK contributing to the occurrence of breast cancer.
It may be really straightforward to calculate, but can you just show how this leads to numerical conclusions about the risk associated with having the CHEK mutation? Ie, how do you get from frequencies of CHEK mutations to an estimate of what percentage of people with the CHEK mutation get breast cancer and what percentage do not.
Oliver Childs April 27, 2009
Thanks for your insightful comments. You raise some interesting points. When writing about such work, it’s always a tricky balance between presenting the scientists’ conclusions and explaining how they got to them. And we are grateful for queries like this to help us better explain the important work of our researchers.
One thing you are alluding to is that correlation does not necessarily imply causation – this is a very important point. It is certainly true to say that the work did not ‘prove’ that the CHEK2 fault caused breast cancer in the 5.1% of BRCA-negative people with the disease who carried the mutation. But the fact that this finding was highly statistically significant does give a strong signal that this gene could have a role in the development of breast cancer, either directly or indirectly through interactions with other ‘stuff’ in the cell.
There was a great accompanying editorial in the Nature Genetics issue that this work was published in, which talks about the strengths and weaknesses of the study. A crucial point, and one that we talk about in the original post, is that the statistically significant result (showing correlation, not causation) is bolstered by the fact that CHEK2 is a ‘biologically plausible’ candidate gene for breast cancer – because it is involved in the cells response to DNA damage. We know that cells with faulty responses to DNA damage can become cancerous.
You also raise a query about the “numerical conclusions about the risk associated with having the CHEK mutation”. Again, an important question. So, around 1 in every 100 healthy people have the fault, but around 5 in every 100 people with BRCA-negative breast cancer carry the fault. This difference is statistically significant, but there are healthy people with the fault as well as people with breast cancer who carry the fault. It is certainly not a simple case of you carry the fault, you get breast cancer.
To work out the risk associated with carrying the fault, the authors used a statistical computer package called MENDEL to compare the presence of the mutation in the various groups of people. They found that women with the mutation are at around twice the risk of breast cancer. This might sound scary, but the average lifetime risk of developing breast cancer is 1 in 9. So the work indicates that this risk is raised to 2 in 9 in women carrying the CHEK2 fault. In the grand scheme of things, the increased risk is not huge.
So, to re-iterate what we say in the post, this work is interesting because it was the first to locate a medium risk breast cancer gene. And by ‘locate’ we mean the first to confirm with a high degree of statistical significance. But it’s only with the benefit of hindsight we can look back and assess the impact of such work – and since 2002, lots more research into CHEK2 and other medium-risk genes has been carried out and added weight to these early results.
The most exciting thing about this work is the prospect of being able to test for lots of such medium-risk genes and figure out how these all work together to increase risk of cancer. Doctors might then be able to tailor preventive measures and treatments according to this information. We’re not quite there, but we’re getting closer, in part thanks to work like this.