‘Chocolate detects cancer’ headlines are misleading

No, chocolate can’t detect cancer. Sorry.

Much to the (misplaced) joy of chocoholics everywhere, the Daily Mail’s headline yesterday proclaimed “Chocolate and fizzy drinks could be used as cancer detectors because malignant tumours feed off sugar”.  Unfortunately, this is a gimmicky, inaccurate misrepresentation of what is actually a fascinating and important experimental research finding.

The story is based on a paper, published in the journal Nature Medicine, by scientists at University College London (UCL). They’re part of the UCL and KCL Comprehensive Cancer Imaging Centre, which is jointly by Cancer Research UK and the EPSRC. They’re developing a new scanning technique, GlucoCEST, which detects glucose – a type of sugar – taken up by cancer cells in mice.

The scientists injected animals with a solution of pure glucose dissolved in salty water, using a relatively small amount of the sugar. Scaling it up, the researchers suggest that in human trials (which are underway), volunteers would need to take in around 14 grammes of glucose – about the same as the amount found in half a standard size chocolate bar.

But – here’s the crucial point – people in the trial would need to fast beforehand, and would  drink glucose in the form of a controlled sugar solution rather than chomping a Mars Bar or glugging a can of cola.

Other news outlets managed to report the research – which, as we’ll see below, is extremely exciting – with rather more accurate headlines (for example, “New MRI technique uses spoonful of sugar to detect cancer”). So what’s this research all about, and why is it important?

What did the scientists actually do?

Researchers have known for many years that cancer cells produce energy and use sugar slightly differently from healthy cells.  Tumours tend to favour a relatively inefficient process called glycolysis to make energy from glucose, burning through the sugar at an impressive rate.

But healthy cells use glycolysis to produce smaller molecules that they then feed into a different cellular ‘engine’, which puts out more energy for a given amount of glucose than glycolysis alone. So healthy cells generally get much more bang for their glucose buck, so to speak.

PET scans use radioactive glucose to detect tumour activity

This difference between cancer cells and healthy tissue is already used to monitor the disease – for example, PET scans measure how tumours absorb small doses of radioactively-labelled glucose.

But giving people radioactivity, even in small amounts, carries a slight health risk, especially with repeated scans. And it means that they aren’t so suitable for people who are more vulnerable to the effects of radiation, such as pregnant women and children.

So the UCL team set about developing a scanning technique that could detect the differences in glucose use between tumours and healthy cells, without having to resort to using radioactivity.

They decided to use a technique called MRI scanning, or magnetic resonance imaging, which looks at how different molecules ‘wobble’ when they’re inside a strong magnetic field. It usually picks up signals from fat and water molecules, building up a picture of the tissue inside the body, and is already used in the NHS to detect and monitor cancer.

In this study, the researchers adapted their MRI scanner to look for certain characteristic signals of glucose, rather than fat or water, on the basis that bowel tumours should have higher levels of the sugar than healthy tissues.

The scientists discovered that their technique could accurately detect bowel tumours in mice just as well as other methods. And they also found that it was sensitive enough to tell the difference between different types of bowel tumour, as well as picking up areas of tumours with relatively low oxygen levels, and could also detect other glucose-like molecules that are involved in energy production.

The UCL team is now testing their technique, called GlucoCEST, with patients, although – as we’ve mentioned – these volunteers definitely won’t be getting a choccy bar in the scanner, but are being given a glucose solution to drink. They need to find out if it works in humans as well as it does in mice, and whether giving glucose to people to drink produces a good enough signal, or whether they’d need to inject it instead.

A slight diversion about cancer and sugar

There is a pervasive belief, particularly among alternative-medicine fans, that sugar “feeds cancer cells”. This is an unhelpful oversimplification of a highly complex area that researchers are only just starting to understand.

“Sugar” is a catch-all term, referring to a range of molecules including glucose and fructose –simple sugars found in plants – and sucrose, the white stuff in the bowl on your table, which is a made from glucose and fructose stuck together.  And sugars fall under the banner of carbohydrates (“carbs”) – molecules made from carbon, hydrogen and oxygen. This also includes starch, found in foods like bread and pasta, which is made of long chains of glucose molecules.

When we eat food containing carbohydrates – whether it’s a cake or a carrot – they get broken down in our digestive tract to release glucose and fructose, which get absorbed into the bloodstream to provide energy for us to live.

All our cells, cancerous or not, use glucose for energy. But because cancer cells are usually growing very fast, compared to healthy cells, they have a particularly high demand for this fuel. As we mentioned above, there’s also evidence that they use glucose and produce energy in a different way to healthy cells.  And it’s the high demand for glucose in tumours that’s the basis for the UCL team’s new technique.

Researchers around the world – including those funded by Cancer Research UK – are working hard to understand the differences in energy usage in cancers compared to healthy cells, and trying to exploit them to develop more effective treatments (including the infamous DCA). This field, known as “tumour metabolism”, is an extremely hot topic, but there aren’t yet any effective drugs targeting cancer metabolism that have been shown to work in large-scale clinical trials.

But all this doesn’t mean that “sugar” (i.e. sucrose in cakes, sweets and other sugary foods) specifically “feeds” cancer cells, as opposed to any other type of carbohydrate. Our body doesn’t pick and choose which cells get what fuel – it converts pretty much all the carbs we eat to glucose, fructose and other simple sugars, and they get taken up by tissues as and when they’re needed.

We recommend that cancer patients and the general public limit sugary foods as part of an overall healthy diet, but it’s not clear that eating sugary foods specifically “feeds” cancer cells.  It’s important that cancer patients discuss their diet with their doctor or specialist nurse before making any big changes – for example, some people may be advised to have a high-calorie diet during chemotherapy to help cope with the effects of treatment.

Why is this research exciting?

Although it still needs to be fully tested, if GlucoCEST works in patients it could potentially be very important. The technique could be rolled out relatively quickly using the existing MRI scanning technology available in the NHS. It’s also likely that it would be cheaper than PET scans, although there’s some variation in how much the two techniques cost to run. And it also avoids the need to use radioactive glucose, making it safer and suitable for all patients.

Figuring out how cancer is behaving in the body – knowing where it is, how it’s growing, and whether it’s responding to drugs – is vitally important. If it works in patients, and if it does become widely used, GlucoCEST could help doctors decide the best approach for treatment for an individual patient, and accurately monitor how their disease is responding to therapy. Right now, we simply don’t know whether it will work in patients, but we look forward to seeing the results of the team’s clinical study when they’re published.

So why are we annoyed about a headline?

It may seem like nit-picking or pedantry to complain about the coverage of this paper or, indeed, any of the other times we’ve criticised the way a particular science story is covered, such as the recent “HIV cures dying girl” episode. But we write these posts because we strongly feel that it’s important that we provide easily understandable, accurate and – hopefully – engaging and interesting coverage of the latest progress in cancer.

We’re living in a hugely exciting time for cancer research, with advances in understanding and technology accelerating year on year, bringing significant improvements in survival along with it. Yet it does a great disservice to patients and the public if the media, researchers and funders fail to find ways to talk about these impressive advances in a way that captures the imagination, yet maintains scientific accuracy.

Misleading, over-hyped or just plain wrong headlines help no-one. They lead to confusion about the current state of research, and reinforce unhelpful narratives about what “they” (i.e. scientists) think is “good for you”, “bad for you”, can “cure cancer” or whatever else.

Sadly, certain headlines about this particular paper are an inaccurate reflection of what is an important and potentially very useful piece of research, which should demand attention on its own merits, rather than resorting to gimmicks.

Kat

• This research has also been covered in detail on the NHS Choices blog.
• Images (brain scan and chocolate) from Wikimedia Commons

Reference:

Walker-Samuel S., et al. (2013). In vivo imaging of glucose uptake and metabolism in tumors., Nature medicine, PMID: 23832090