Friday February 27 2015
Some types of leukaemia take decades to form
BBC News reports that, according to researchers, “It is ‘almost inevitable’ that your blood will take the first steps towards leukaemia as you age”.
Researchers analysed the blood of 4,219 people, looking for DNA errors (mutations) linked to blood cancers (leukaemia).
The number of mutations in healthy older people without the disease was higher than expected. The research focused on 15 genetic hotspots of leukaemia-linked mutations and found them in 0.8% of individuals aged under 60, and 19.5% of those aged 90 or older.
The media quoted figures suggesting that more than 70% of people in their 90s would have some form of leukaemia-associated mutation. This was based on predictions of the prevalence of other mutations outside of the 15 tested.
The good news is that this form of age-related leukaemia is highly unlikely to kill you. The bad news is you are far more likely to die of something else before the mutations trigger the onset of leukaemia.
Still, as some are predicting that average lifespans will rise dramatically in the decades ahead, the results of this study could become more of an issue for future generations, and might also apply to other cancer types.
Where did the story come from?
The study was carried out by researchers from Wellcome Trust Sanger Institute, Cambridge (UK), and was funded by the Wellcome Trust, Leukaemia Lymphoma Research, the Kay Kendal Leukaemia Fund and the Spanish Ministerio de Economía y Competitividad Subprograma Ramón y Cajal.
The study was published in the peer-reviewed medical journal Cell Reports. It is open-access, so is free to read and download online.
Generally, the media reported the story accurately. The Independent quoted Dr Vassiliou, senior author of the study, reassuring readers that: “These mutations will be harmless for the majority of people, but for a few unlucky carriers, they will take the body on a journey towards leukaemia. We are now beginning to understand the major landmarks on that journey”.
One small gripe is with the choice of figures used. The main study result was that leukaemia-linked mutations were found in 0.8% of under-60s and 19.5% of over 90s. This was based on studying 15 leukaemia-linked hotspots.
Both the Independent and the BBC quoted figures suggesting that 20% of 50 to 60 year olds, and over 70% of over 90s, had dormant leukaemia-linked mutations. These much higher figures come from the study discussion and were not directly tested in the current research. They were figures based on assumptions about combining results from the 15 hotspots with other non-hotspot mutations from previous studies. We are not able to appraise the non-hotspot mutations studies, so we don’t know how accurate these figures are.
What kind of research was this?
This was a genetic study investigating how common small, leukaemia-linked DNA changes were in cancer-free adults.
Cancers, including leukaemia, develop through the combined action of mutations that are acquired over time. The researchers say that leukaemia-associated DNA mutations can occur without evidence of the disease. They wanted to find out how common these were in healthy people, and how common they were as people got older.
What did the research involve?
The researchers analysed DNA samples at 15 pre-defined leukaemia-associated mutation hotspots, using highly sensitive tests. They analysed the DNA of 4,219 people aged 17 and over.
The majority of DNA tests were on healthy people, but for comparison, they analysed the genes of a number of blood cells from people with myeloid leukaemia.
The production of the range of different types of mature white blood cells starts with a small number of stem cells. More specialised cells develop from these, like tree branches. The stem cells replicate themselves, producing clones. Some of these clones receive signals from the body, causing them to replicate and develop (differentiate) into more specialised white blood cells. Different signals produce different types of cells. The researchers were looking at what stage in the production process the mutations were occurring. If the mutations happened early in the differentiation process, they would be found in many downstream white cell types. If they occurred later, then they would be found in fewer cell types.
What were the basic results?
The main results were that age-related leukaemia-linked mutations were much more common than previously predicted.
Using only the 15 hotspots studied, they identified leukaemia-linked mutations in 0.8% of individuals under 60, rising to 19.5% of those aged 90 years and over. Coupling these estimates with other mutation rates from previous studies (outside of the 15 hotspots tested) they came up with much higher estimates. They predicted that more than 70% of people aged 90 or older would have some form of leukaemia-associated mutation. The 70% figure made it into the media coverage; the 19.5% was not mentioned.
On closer inspection, they found that mutations DNMT3A-R882 were most common and, although their prevalence increased with age, were found in individuals as young as 25. By contrast, mutations affecting spliceosome genes SF3B1 and SRSF2, closely associated with the myelodysplastic syndromes, were identified only in those aged over 70 years, with several individuals harbouring more than one mutation.
Myelodysplastic syndrome is an uncommon condition of unknown cause, that can lead to a drop in the number of healthy blood cells being produced. In some cases, it can progress into acute myeloid leukaemia.
Mutations in gene NPM1 were not seen in the group. This gene is thought to act as a “gatekeeper” to leukaemia. If it goes wrong, your risk of leukaemia rises considerably. As the group were symptom-free, it is not surprising that this gene was not affected in most people.
How did the researchers interpret the results?
The study group said: “individuals without overt features of a haematological [blood] disorder may harbor hemopoietic cell clones [blood stem cells] carrying leukemia-associated mutations” and that accumulating these mutations “is an almost inevitable consequence of aging in humans”
This study estimated that 0.8% of individuals under 60, and 19.5% of those aged 90 years and over, had leukaemia-linked mutations. These mutations caused no immediate harm and the people didn’t have leukaemia. The mutations were lurking in the background, but could have the potential to contribute to leukaemia in the future.
The research primarily focused on 15 genetic hotspots of leukaemia-linked mutations.
However, in their discussion, they predicted that more than 70% of people aged 90 or older would have some form of leukaemia-associated mutation. This formed the basis of their comment that these mutations seem an inevitable part of ageing. It is important to realise that this much higher estimate was not directly tested in the study. That is not to say it is not true, but we can’t confirm or refute it either way. Further research could confirm this prediction.
Scientists know that cancer is caused by the accumulation of genetic mutations over many years. This is why most cancers occur in older people, and the risk of cancer increases with age. What is surprising about this study is the relatively high prevalence of leukaemia-linked background mutations in healthy adults. The implication is that if people were to live a lot longer, say 150 years, they might expect to get leukaemia. In theory, this could also apply to some other types of cancer.
This is all largely theoretical. The impact on the average person is minimal, though if lifespans continue to increase, it could be a potential problem for your grandchildren.
It’s important to remember that we can all reduce our risk of cancer by making some simple changes to our lifestyle.
For example, healthy eating, taking regular exercise and stopping smoking will help to lower your risk.
Read more about how a healthy lifestyle can help to reduce your chances of developing cancer.
Analysis by Bazian. Edited by NHS Choices. Follow Behind the Headlines on Twitter. Join the Healthy Evidence forum.