Cancer Research UK on Google+ Cancer Research UK on Facebook Cancer Research UK on Twitter

Let's beat cancer sooner

What causes cancer? It’s a simple question to ask, and – in so many ways – one of the hardest to answer.

And last week, as 2015 slowly got into gear and many of us put the finishing touches to our new year’s resolutions, some eyebrow-raising headlines appeared claiming that scientists had ‘proved’ that the answer is ‘mainly bad luck’.

This story came from a research paper from scientists at Johns Hopkins University in the US, (press released here). And it appeared to contradict the message that many organisations have been trying to hammer home (including us): that although there are no guarantees, we can stack the odds of avoiding cancer in our favour if we embrace a healthy lifestyle.

(In fact, just a week ago we published new stats showing how, if the UK’s population had been healthier overall, an estimated 600,000 cases would have been avoided over the last five years.)

We weren’t the only ones with raised eyebrows – a whole host of blogs and opinion pieces appeared over the weekend, scrutinising the claims. We’ll look at some of these criticisms below, and discuss why the research DOESN’T mean that two-thirds of cases of cancer are ‘caused by bad luck’ (whatever that means).

But before we do, let’s have a very quick recap of what the researchers did, and what they found.

The root of the story

The researchers set out to answer a simple but fascinating question: why do cancers arise more often in some tissues (e.g. the bowel) than others (e.g the brain)?

To try to answer this, they looked at what was already known about cell division rates in a given tissue (specifically, specialised cells called ‘stem cells’, which renew and replenish the tissues in our bodies) and how this related to the overall risk of cancer in that tissue.

‘The Stats Guy’ blog summarised it thus:

[T]he authors … looked at the published literature on 31 different types of cancer (eg lung cancer, thyroid cancer, colorectal cancer, etc) and estimated 2 quantities for each type of cancer. They estimated the lifetime risk of getting the cancer, and how often stem cells divide in those tissues.

They found a very strong correlation between those two quantities: tissues in which stem cells divided frequently (eg the colon) were more likely to develop cancer than tissues in which stem cell division was less frequent (eg the brain).

In other words, they found a mathematical relationship between the rate stem cells divide in a tissue, and the rate of cancer in that tissue (although several blogs questioned the way the researchers had used the statistics).

According to the researchers, the maths could explain two-thirds of the variation between different tissues.

It’s an interesting finding, which casts new light on an old mystery. But as Professor David Spiegelhalter noted on his blog:

“[This] may be a fairly reasonable statement to make about population rates in different tissues, but of course says nothing about variation in risks between individuals, and certainly does not say that two-thirds of cases are just luck.” [emphasis ours]

And that’s the key criticism of the way this paper has been interpreted. The media coverage has inadvertently jumped from talking about cancer rates in different tissues to speculating about cancer rates in the population. (Although it’s worth noting that the authors themselves still support the fact that certain cancers can be prevented by lifestyle changes.)

Other criticisms

The study also had another notable weakness: it looked only looked at cancer types where there was hard data about the rates of stem cell division.

As a result, they missed two of the most common cancers – breast cancer (which is influenced by lifestyle factors like a woman’s reproductive health, her weight after the menopause, and how alcohol much she drinks) and prostate cancer (which is only weakly influenced by lifestyle).

So although their finding is intriguing, there’s a way to go before we can say for sure it applies to all cancer types.

So where does ‘luck’ come in?

Molecular machinery of a dividing cell.

Molecular machinery of a dividing cell.

Let’s recap. Every time a cell in your body divides to create two new ones – something that happens billions of times a day – there’s a very (very) small chance a mistake could creep into its DNA, and one of the resulting cells could begin the long deadly journey towards developing into cancer.

And, as these researchers have shown, the more often a cell divides, the more chance there is of something untoward eventually happening.

So, in this sense, there’s an element of ‘bad luck’ about cancer. And it’s true that the disease can affect any of us – from the most avid gym bunny to the most determined couch potato.

But we also know there are a whole host of things that affect the chances of DNA damage developing in a dividing cell.

The chemicals in tobacco smoke, for instance, can modify DNA molecules, resulting in errors when they’re copied. The same is true of ultraviolet light.

And, similarly, things that change the rates of cell division over our lives will also influence the chance of a ‘bad’ cell division. For example, fluctuating hormone levels through puberty, periods, pregnancy and the menopause, can affect cell division rates in the breasts, womb and ovaries. Obesity too can influence our hormone levels, and thus cancer risk.

And long-term (‘chronic’) inflammation, as caused by certain infections, asbestos, and a host of other external influences, is thought to speed up cell division in other tissues.

(This is not an exhaustive list, just a flavour of the different things that can influence the development of cancer).

So to ascribe a particular patient’s cancer to ‘bad luck’ is essentially impossible. It’s a combination of myriad influences, some of which we can control, others which we can’t. And it’s certainly not about ‘blame’, as we’ve pointed out before.

In his response to the media stories, Professor Spiegelhalter had this excellent analogy:

[Imagine] there are tickets in a bucket marked cancers of different types, and a lot of blank tickets (and some marked ‘run over by bus’ etc). Smoking means you might get 20 times as many ‘lung-cancer’ tickets, but you still may be lucky and not draw one: many smokers don’t get lung cancer.

So chance plays a very strong role, even in so-called preventable cancers. This leads to the apparently paradoxical observation that most lung cancers are ’caused’ by smoking, while all lung cancers are also a matter of bad luck.

What about the media coverage?

It’s often tempting to blame ‘the media’ for mangling science stories, but, as is often the case, there’s a more complicated story here (one that the excellent 2020 Science has dissected in full).

But this isn’t about pointing fingers. It just seems a huge shame that something that is, as far as these things go, established scientific fact – that cancer rates are affected by lifestyle – has been called into question.

For those that doubt the effect these stories can have, just read this tweet:

One of the gratifying things about this story was the speed with which other voices all over the Internet shone a spotlight on the story even before the day was out. So we’ll leave the final word to Adam over at The Stats Guy blog:

We often see medical research badly reported in the newspapers. Often it doesn’t matter very much. But here, I think real harm could be done. The message that comes across from the media is that cancer is just a matter of luck, so changing your lifestyle won’t make much difference anyway. We know that lifestyle is hugely important not only for cancer, but for many other diseases as well.

For the media to give the impression that lifestyle isn’t important, based on a misunderstanding of what the research shows, is highly irresponsible.

– Henry



Victor S January 14, 2015

Dan, you missed the point of the study that says that some cancers are only bad luck.

Dan Cooke January 13, 2015

I think a lot of the commentards here are missing the point. Whether you get cancer or not is a matter of chance (or “luck”) but you increase your chances of getting it (or ” bad luck”) with certain lifestyle choices. As the article says, you can smoke and not get lung cancer, and not smoke and get lung cancer. You are just 20 times more likely to get it if you smoke!!!

Sharada January 13, 2015

I do appreciate the comments made in this article, however I have struggled to understand why I had breast cancer at the age of 27. (I am now 42 – it has not returned thankfully). I didn’t have a particularly healthy lifestyle but no more unhealthy than my friends. There is no history of breast cancer in my family. After having read the article on risk factors, the only thing I can’t confirm is whether my breast tissue is more dense than usual. But still – I was 27 years old. I personally have two ideas as to why I might have got breast cancer. One – I’ve always had ‘hormonal’ problems, problematic periods and PMT etc. However, I’d had several hormone tests and none of them revealed any problems. Secondly – mental health. I’d had a very bad episode of depression for a couple of years prior to getting breast cancer. At one point, when I was feeling suicidal, I wished cancer upon myself. Of course this was dismissed by the medical team looking after me. However, I believe that negative thought and stress can play a huge part in physical health, I’m surprised that it hasn’t been mentioned as a factor in keeping a healthy lifestyle.

Victor S. January 13, 2015

Smoking wasn’t listed in the original paper at bad luck. You should edit the article and remove smoking from the argumentation.

Dr John Barrett January 11, 2015

To ascribe a disease to “bad luck” simply means we do not know the cause. As a medical student in the 1950s almost all the patients I saw with cancer had no known cause, and that was bad luck. Who knows what research over the next 60 years could reveal.

Louise Heffernan January 10, 2015

This is quite a difficult article to read if your husband has recently died from cancer. He was told it was ‘bad luck’ at the beginning of his treatment as he didn’t smoke or drink much and was only a little overweight. I spend my days wondering if we could have done anything differently. This article does not help me deal with my grief. He was unlucky in that the surgery and the radiotherapy actually completely removed the cancer in the throat, but unfortunately a few cancerous cells it seems had already travelled to the liver and were discovered a couple of years later….

Jo January 9, 2015

My husband was the healthiest fit person non smoker fruit loving surfing /cycle maniac u could meet at 59 he developed pancreatic cancer and given months to live 8 months later he was still surfing however 1year after being diagnosed he died his lifestyle was perfect he did everything right I get very annoyed when people suggest a bad lifestyle link to pancreatic cancer perhaps in this case it was bad luck/bad genes

Sue January 9, 2015

And what about children who get cancer! Is is bad luck for them? Cannot be lifestyle led can it? Having watched my grandson being treated, successfully thank god, for non-Hodgkin’s lymphoma, and other children who were not, try telling the parents that it was bad luck!

Alan Kay January 9, 2015

Whilst I agree that a healthy lifestyle is important I am very annoyed with the inference that people who are diagnosed with Bowel cancer do not lead a healthy lifestyle, your article tars everyone with the same brush. My wife was diagnosed with Bowel cancer in Sept, has never smoked we never eat processed foods, she exercised regularly and had no symptoms, it was detected through the screening programme,after having major surgery and now going through Chemotherapy a very difficult time is now being made worse by inferring that the cancer was more than likely caused by lifestyle your reporting needs to be more focused with a broader view.

Diane January 9, 2015

I am 58. I have always been slim. I have never drunk more than 2 units occasionally in a week. I avoid lactose. I have never smoked. I live backing into woods. I cycle, walk, sail. I have stage 4 breast cancer. Which part of my life style caused it?

Patricia Dagnall January 9, 2015

I tended to agree with the recent findings of cancer being just an unfortunate and unlucky [prtion for some of us. I had breast cancer 10 years ago, was 56, just had menopause, did not drink or smoke, weighed 9st 4 lbs, had 2 children reasonabley fit with a job and had a reasonable healthy diet.. Then a recurrence 3 years later, thankfully am fit and well now but it did become a real talking point in my family as to why the disease came to me. Thankyou for your information,,, to think about.

beverley January 9, 2015

unfortunately my cancer, apart from being not one of the high profile conditions, seems to have been just bad luck. I was a fit and healthy 49 year old who went to the gym, ate healthily and didn’t smoke started to suffer from a hoarse voice and after being told I had laryngitis was eventually diagnosed as thyroid cancer and had major surgery to remove both thyroid and neck glands. Even my GP had only heard of it once before so while the more high profile conditions may be helped by lifestyle what about the lesser known conditions.

Julian Desser January 9, 2015

A very interesting and highly informed article (as you would expect) which makes good effort to explain very complex facts to “laypersons”. However, I’m still no wiser as to why, for example, some people can smoke 40 cigarettes a day and not get lung cancer while lifelong abstainers (like my late grandmother) can fall victim. I suppose it’s “in the DNA” as they say. In other words, something we’re born with. Therefore, I believe we can shift the risk one way or another through lifestyle choices, but only marginally, which may make a difference, of course, in a few cases. (By the same token you can reduced the risk of being killed in a car crash by never venturing out of the house!) The one thing I did object to in the article was the in the phrase “. . . one of the resulting cells could begin the long deadly journey towards developing into cancer.” The word deadly is used emotively (perhaps to scare?) which is unfortunate as CRUK is always keen to point out that not all cancers are fatal.

yvonne oakley January 9, 2015

Hooray…the nation must be given accurate information and not distorted headlines taken from research papers, which can and in this case has led to confusion. Ultimately any inaccuracies can and possibly will result in more people suffering from cancer. it would be a tragedy if more lives are lost due to misleading media coverage.

Diane Lower January 9, 2015

Of course Life-style has an effect on whether people are more susceptible to cancer or not. For those with a parent who has suffered cancer the risk seems to me to be slightly higher. But hey why not try to improve your chances by being sensible about what you eat and drink.

April Ozet January 9, 2015

Very informative as always, thank you.

Georgia January 8, 2015

I did not choose breast cancer, it chose me. I’m the one with one very special (now smaller) right breast. The very day I was told that treatment worked and was cancer free, my arm was twisted by a lovely young girl to run the 2010 Race for Life – I could walk it as long as I was warring something pink aparently. Since then: I stopped smoking, lost a stone and a half in weight, and run 5 Races for Life, 2 Half Marathons and numerous other smaller races. I’m approaching my 60th birthday but I was never fitter or so good looking in my life! I know I may get Cancer again but in the mean time I want to run at least one full Marathon, get a new job, ware this little black dress, learn to drive…………….. the list is endless. And yes, this new zest for life was given to me by a lovely young girl desperate to sign me on to my first R4L five years ago this month. Will I go back to my unhealthy life style? NO WAY! Chances to get cancer again may or may not increase with unhealthy life style but living a healthy way is more fun. Honest

Jacqueline January 8, 2015

I am the epitome of how a person should run their life when it comes to health. I almost never drink (2 unit of alcohol a week is consider a lot for me), I don’t smoke (never have & never will), excercises at my local gym 4 times a week, sometimes play tennis or go for a run, eat fresh fruit & veg, nuts, fish, poultry etc every day( Macdonalds etc are swear words in our house), drink plenty of water, do enjoy dark chocolate and sometimes coffee. So when I was diagnosed with Breast Cancer in November 2013 my husband and I were totally shocked, stunned and pretty much lost for words, also as I have no history of it in my family. Therefore you cannot tell me that this is not bad luck. I do agree some cancers are a direct result of bad lifestyle choices and do believe we should always live as healthily as possible. Though in my case it was random and pure bad luck!!

Janice perkin January 8, 2015

However I do think when you go on about lifestyle choices you can make it feel that people are choosing to get cancer by being irresponsible in their choices and therefore deserve to get cancer. I did not choose to get breast cancer at 65 and go through 6 lots of chemo with some horrendous side effects at the time and continuing even now. Yes I had HRT for about13years after a hysterectomy at 45. Yes I do drink and yes I am overweight and no I don’t excercises regularly but have childminded for 4 years which keeps you pretty active! But I don’t deserve breast cancer in comparison with others which sometimes your research info suggests. I was unlucky that my cells didn’t behave!, they might have done but they didn’t. So I do get upset with some of your diagrams and they do have the result of making me feel bad about myself and I already have enough to contend with long term damage over the chemo and the side effects of taking the anastrazole for the next 5 years

Jo Hutt January 8, 2015

“Bad Luck” in scientific terms means we haven’t enough knowledge to explain fully what is going on. Genetics is a fast growing area for causal factors in cancer and other diseases. As a participant in such research, I have been told that whilst I have been tested for three known, genetic risks and found to be negative, it seems that there is a risk of some yet unknown genetic condition that may provide an explanation of the occurrence of three different cancers. Research presently being undertaken with cancer indicates this. “Lifestyle” is what “we know” at this moment in time as a potential or actual cause. This gives us general guidance about keeping healthy and subduing risks. Although science has some way to go, it has got to be worth doing something about lifestyle factors we can control for all sorts of physiological and psychological reasons as well as cancer prevention. Our A&E departments would be under far less pressure if we follow the excellent guidance given.

Kate Elliott January 8, 2015

Following on from my earlier post – I also think GPs who miss cancer diagnosis before it is too late for some are misguided by the fact that healthy people, young and old can and do develop cancer all the time, and too much importance is placed on lifestyle factors and not peoples’ symptoms. You hear about it all the time: people being sent away for months on end because, ‘oh you’re too young/healthy to have breast/colon/ovarian/cervical/bowel cancer. Fortunately, I had/have an excellent GP who gave me a proper MOT for university when I was 17, and didn’t just send me away with a condescending wave despite my being in excellent health, and discovered my enlarged spleen hidden beneath my muscular dance-induced stomach.

Koray Tascilar January 8, 2015

I believe this has been a very good case example of poor knowledge translation, that is how scientific information is to be conveyed to the public. There were two very tricky points here, first the word “stochastic” and the phrase “65% of the variability”.
Yes a lot of very well known risk factors -both modifiable and not- influence the risk of cancer in a lifetime but what the researchers had tried to point out could be summarized as “the basic risk that cancer is observed in a given tissue is an intrinsic property of that very tissue, and the magnitude of any extra risk that can be brought upon by an external risk factor is also an intrinsic property of that tissue” This practically means in an example that “yes by completely eliminating smoking in the society you can effect a drastic decrease in the number of lung cancers, but you still probably cannot make lung cancer as rare as cartilage cancer”, because tissue structures and functions necessitate that more cell divisions -hence more chance errors- have to happen in certain tissues than in others in order to maintain them.