A powerful lens into tumour metabolism

I first heard about Cancer Grand Challenges on a BBC Radio 4 programme back in 2015. After discovering that one of the problems was to map tumours at a molecular and cellular level, I started discussing with other leading MSI experts which imaging modalities we could combine to tackle this question. I also talked to leading cancer biologists to find out about the types of maps they needed to help improve their understanding of tumour behaviour. Together, we formulated the Rosetta programme, setting out to apply molecular imaging to explore the secrets of tumour metabolism.  

Over the past five years, we’ve successfully established a truly groundbreaking untargeted MSI-driven systems biology approach that can probe tumour metabolism in unprecedented detail. By performing measurements with a carefully compiled selection of techniques at many locations across a sample, we can achieve coverage at not only the required molecular breadth but also a range of length scales from subcellular to whole tissues. The easiest way to conceptualise our pipeline is by likening it to Google Earth: it generates a detailed atlas of tumour metabolism – from the organ level right down to the subcellular level – with the ability to zoom in at different degrees of magnification.  

The information that we are uncovering with this fundamentally new way of analysing cancer metabolism is already providing a much deeper understanding of tumour behaviour, prognosis and response to treatment. Looking ahead, we hope that the Rosetta pipeline will lead to the development of a novel suite of clinical decision-making tools that can help transform outcomes for people with cancer.  

During the first two years, we focused on developing and validating, optimising and combining the different components of the Rosetta pipeline – including matrix-assisted laser desorption/ionisation (MALDI), desorption electrospray ionisation (DESI) and secondary ion mass spectrometry (SIMS).  

We set out by distilling a list of around 200 highly informative metabolites – including amino acids, fatty acids, carbohydrates, vitamins and lipids – which we could use to measure our pipeline’s success in terms of our degree of coverage. But the sheer breadth of this list posed a huge analytical challenge. We needed to establish methods for a range of modalities that could monitor all these target molecules with vastly different chemical properties – including polar and non-polar, neutral, or positively or negatively charged entities with a wide range of molecular weights – in complex samples at the highest resolution, while maintaining the quality of each measurement. It was also absolutely crucial that we could review and compare data from in vitro models, including primary cell lines and patient-derived organoids, with measurements from ex vivo and in vivo samples, including whole animal organs and tissue biopsies.  

A major legacy of the Rosetta pipeline will be our development of new and innovative modalities, which we did in response to our early data when we didn't have the required sensitivity or molecular coverage. Typically, fresh frozen tissue samples give the best results for MSI, while other traditional preservation methods are less optimal or even incompatible. Developing new methods to enable the analysis of formalin-fixed paraffin-embedded material has also proven absolute dynamite – allowing us to turn our attention to beautifully curated and compiled repositories of samples from around the world that would otherwise have been impossible to analyse with our pipeline.  

It’s also taken a colossal and heroic effort by many people in the team to develop a world-class computational pipeline to enable the mining of these enormous unique datasets, which we have been improving, testing, validating and extending throughout the programme. Most recently, we’ve refined our methods to run unsupervised analyses, enabling the team to make unbiased interpretations of the data to identify novel points of interest that might otherwise have been missed.   

We started the project with a parallel launch of several programmes focusing on colorectal, breast and pancreatic cancers, as well as brain tumours, which have all reached different stages.   

In work led by Owen Sansom at the Cancer Research UK Beatson Institute, Glasgow, we identified the antiporter SLC7A5 as a potential new therapeutic target for KRAS-driven colorectal cancers by using genetically engineered mouse models. Challenging the long-held assumption that glutamine metabolism fuels the growth of cancers, we found that SLC7A5 ejects this amino acid from cancer cells and imports other metabolites from the surrounding tissue. The results uncovered a promising new targetable vulnerability for drug discovery: targeting SLC7A5 could help starve cancer cells of metabolites necessary for cancer growth while having little effect on healthy tissues, and could also help overcome drug resistance in some patients. But without our pipeline, we wouldn't have been able to detect glutamine in the tissue surrounding the cancer cells rather than inside them.   

Another study, which was led by George Poulogiannis at the Institute of Cancer Research, London, involved screening breast cancer cell lines, tumour samples and mouse models. We identified that disruptions to the PIK3CA kinase pathway cause an increase in the metabolite arachidonic acid, which helps fuel the growth of these breast tumours. We also found that drugs that interfere with the PIK3CA pathway are much more effective in slowing breast tumour growth in mice fed a diet without fatty acids than a standard diet. This is the first study showing how dietary fat restriction might play a major role in the response to treatment – and again, would not have been possible without our pipeline.  

Developing ways to translate the Rosetta pipeline into the clinic offers unique opportunities for enabling the delivery of precision medicine. For the first time, we can start to envision an analytical pipeline that can produce data that can be compared across all stages of the cancer patient journey – from screening and diagnostics to advanced therapies and follow-up monitoring (see figure – Cancer Care 2.0).   

The intelligent surgical knife (iKnife) – which was invented by Rosetta team member Zoltan Takats at Imperial College London – provides a powerful example of a diagnostic tool with the potential to deliver benefits at different points along the patient journey. The electrosurgical device uses rapid evaporative ionisation mass spectrometry (REIMS) and is designed to deliver real-time information about tumour metabolism that can help guide margin control during surgery. In our PIK3CA programme, our results provide a proof of principle that this method could also be used to stratify cancers into subtypes according to metabolic phenotype to help guide treatment selection. Specifically, we showed that iKnife can detect elevations in arachidonic acid in cell lines, which can be mapped back to disruptions in the PIK3CA pathway and correlate with a subtype of breast cancer. We are now exploring ways to detect changes in tumour metabolism with less-invasive approaches, such as the analysis of urine and stool samples or exhaled breath.  

Underpinning our team’s success is that we’ve united some of the most exceptional experts in the world in each of the four cancer types, tumour biology and metabolism, and each of our technologies. Although the journey has often been difficult, we’ve all thoroughly enjoyed working with one another. The long-term support for our extraordinary generation of early-career future leaders has also made a real difference – they’ve been spending time in each other’s laboratories and are the people truly driving the project.  

From a personal perspective, this is the most important research I've ever been involved in. Analytical chemists have often been relatively agnostic to the problems brought to us and have applied very technology-focused approaches to problem-solving. But since taking on this challenge, I've become completely fascinated with the field of metabolism. I think my laboratory will now work in this space forever.  

As told to Alison Halliday 

The Rosetta team is funded by Emerson Collective through Cancer Research UK.   

This article was originally included in our annual progress magazine Discover: A year of scientific creativity.