Image of the Week!

Image of the Week

Sugar makes cancer light-up in MRI scanners

07 July 2013

A new technique for detecting cancer by imaging the consumption of sugar with magnetic resonance imaging (MRI) has been unveiled by UCL scientists. The breakthrough could provide a safer and simpler alternative to standard radioactive techniques and enable radiologists to image tumours in greater detail.

The new technique, called ‘glucose chemical exchange saturation transfer’ (glucoCEST), is based on the fact that tumours consume much more glucose (a type of sugar) than normal, healthy tissues in order to sustain their growth.

Image 1: UCL scientists have developed a new technique for detecting the uptake of sugar in tumors, using magnetic resonance imaging (MRI). Image 2: Tumours use large quantities of glucose to sustain their growth. By injecting normal, unlabeled sugar, UCL scientists have developed a way to detect its accumulation in tumors using magnetic resonance imaging (MRI).

The researchers found that sensitising an MRI scanner to glucose uptake, caused tumours to appear bright on MRI images of mice.

Lead researcher Dr Simon Walker-Samuel, from the UCL Centre for Advanced Biomedical Imaging (CABI) said: “GlucoCEST uses radio waves to magnetically label glucose in the body. This can then be detected in tumours using conventional MRI techniques. The method uses an injection of normal sugar and could offer a cheap, safe alternative to existing methods for detecting tumours, which require the injection of radioactive material.”

Professor Mark Lythgoe, Director of CABI and a senior author on the study, said: “In principal, we can detect cancer using the same sugar content found in half a standard sized chocolate bar.

“MRI uses standard imaging technology available in many large hospitals,” he continued. “Our research reveals a useful and cost-effective method for imaging cancers. In the future, patients could potentially be scanned in local hospitals, rather than being sent for referrals to specialist medical centres.”

According to UCL’s Professor Xavier Golay, another senior author on the study: “Our cross disciplinary research could allow vulnerable patient groups such as pregnant women and young children to be scanned more regularly, without the risks associated with a dose of radiation.”

The study has been published in the journal Nature Medicine (PDF) and trials are now underway to detect glucose in human cancers.

Dr Walker-Samuel added: “We have developed a new state-of-the-art imaging technique to visualise and map the location of tumours that will hopefully enable us to assess the efficacy of novel cancer therapies.”

Image 3: Glucose uptake varies within tumors, as demonstrated using a new technique developed by scientists at University College London. 'Hot' regions at the edge of the tumor show increased uptake compared with 'cold' central regions, which could be used in the future to determine the best therapies to give to individual patients.

This work was supported by public and charitable funding from the Department of Health’s NIHR Biomedical Research Centres funding scheme, Cancer Research UK, Engineering and Physical Sciences Research Council (EPSRC) and the British Heart Foundation (BHF).

Q & A with Dr Walker-Samuel 

  • Why did you decide to do this research project and what prior work led up to this latest paper? 
    The avid consumption of glucose by tumours is known as the Warburg effect, and is a key discriminator of malignant from normal tissue. This effect is exploited in the clinic with positron emission tomography (PET) imaging, in which a radioactively-labelled glucose analogs is administered to patients in order to detect metastatic disease. However, we hypothesised that magnetic resonance imaging (MRI) scanners could be adapted to allow the uptake of unlabelled glucose, without the radiation dose associated with PET. This could allow vulnerable patient groups such as pregnant women or children to be scanned, or for longitudinal measurements to be performed with less risk to the patient.
  • Can you explain the methodology used in your paper? Why did you decide to adopt this approach? Conventional MRI images mostly map the signal from water and fat, but we have adapted our scanner to allow the signal from glucose to be amplified, via the water signal. We used a technique called chemical exchange saturation transfer (CEST), which uses the exchange of protons between glucose hydroxyl groups and bulk tissue water protons. By selectively saturating the magnetisation of protons in glucose hydroxyl groups, using radiofrequency pulses, the exchange of protons causes an associated reduction in the signal from water, which we can then map using a standard MRI acquisition. Using colorectal carcinoma xenograft models, we tested out whether this effect was sensitive enough to detect the accumulation of injected glucose in vivo, and to evaluate whether the measured glucose signal corresponded with other gold-standard measures of glucose uptake. We have called the technique 'glucoCEST'.
  • What were the most significant findings? How do they relate to what was already known about the subject? We found that glucoCEST could distinguish between two types of colorectal xenograft models (SW1222 and LS174T), and was also significantly correlated with the uptake of fluorodeoxyglucose (FDG), a glucose analog used in PET imaging. Moreover, we found that in one tumour type, glucose uptake was correlated with hypoxia (low tissue oxygenation). Interestingly, we also found that, alongside free glucose, the first four molecules in the glycolytic pathway could be detected with glucoCEST, along with amino acids produced via glutamine synthesis This could therefore allow glucoCEST to be used to probe various aspects of tumour glucose metabolism.
  • How do you plan to take this work forward? What are the implications for future research? We are currently evaluating glucoCEST in a number of patient groups in order to evaluate its sensitivity on clinical scanners. To mirror our experiments in mice, we are administering glucose via a sugary drink in the MRI scanner. If sufficiently sensitive, glucoCEST could complement FDG-PET, or provide a cost-effective and safer alternative, potentially with a greater spatial resolution. It could also be used to assess response to therapy or to characterise individual tumour pathophysiology for treatment stratification.

Notes for editors:

  1. Members of the media who would like more information, or to interview the researchers quoted,  please contact David Weston UCL Media Relations Office on tel: +44 (0)20 3108 3844, out of hours: +44 (0)7917 271 364, email:
  2. High resolution images are available from UCL Media Relations. Please credit UCL if used.
  3. The paper "Imaging glucose uptake and metabolism in tumors" is published online ahead of print in Nature Medicine, July 7th 2013.
  4. The UCL Centre for Advanced Biomedical Imaging is a new multidisciplinary research centre for experimental imaging. The Centre is built around a number of groups at UCL and brings together imaging technologies across UCL with specific applications in the biomedical sciences. Dr Simon Walker-Samuel and Professor Mark Lythgoe are affiliated to UCL Division of Medicine. Professor Xavier Golay is affiliated with the UCL Institute of Neurology.

Page last modified on 11 jul 13 19:10