Making Medicine More Precise Using Metabolomics

The metabolome is exquisitely sensitive to our physiology, sitting at the interface of our genome and environment. Developing handheld devices that enable people to monitor themselves for changes in their metabolites that can predict future disease will empower them to take charge of their own healthcare, transforming medicine.

“Precision medicine is an approach to delivering care to an individual that considers their genes and environment to define the best therapeutic or preventative strategy for them,” explains Dr Clary Clish, director of the Metabolomics Platform at the Broad Institute of MIT and Harvard, USA. 

In terms of realising this vision, much attention is understandably focussed on our genomes – using genetic information to predict what conditions we may develop in the future, to diagnose our disease more precisely, and for selecting the right treatments.

But metabolomics, the characterisation and interpretation of the hundreds of thousands of small molecules in our bodies, also has a lot to offer.

“Genomics tells you what might happen. But what we really want to know is what’s actually happening, which is what metabolomics tells you,” says Dr David Wishart, a Professor of the Departments of Biological Sciences and Computing Science at the University of Alberta, Canada.

Metabolomics Screening Can Save Lives

In fact, metabolomics is already powering precision medicine, exemplified by the use of newborn blood spot test screening programmes for inborn errors of metabolism.

“A metabolomics read-out can not only diagnose one of these rare genetic conditions but also in many cases detects it early enough that something can be done - so it never develops into anything serious,” says Wishart.

In a similar way, metabolomics could also raise a red flag for more common conditions such as diabetes, cancer or Alzheimer’s – even up to 10 or 15 years before it is diagnosed.

“You just have to look at how successful metabolomics has been for newborn screening and how many lives it’s saved or made better – if it can have the same impact on common conditions, that would be exciting,” says Wishart.

The idea is to screen panels of metabolites, then interpret this information to inform personalised lifestyle or prophylactic medication strategies. While this isn’t a new approach - for example, a basic screening for lipids in the blood can already give doctors enough information to prescribe statins – metabolic profiling will offer a step-change in its sophistication.

The Search for Novel Metabolites

However, before this revolution can happen, a lot more research is needed to characterize metabolic changes that are predictive of different diseases before clinical symptoms appear. Scientists are trying to identify these by following people in longitudinal population studies over many years.

“If certain individuals develop a disease, we can go back to their samples and analyse them to look to derangements that existed way before they were diagnosed,” says Clish.

With recent improvements to technology, researchers can now measure hundreds of thousands of molecules in a sample all at once.

“While in the past, we would have looked at a more targeted or limited set of metabolites – but now we’re doing untargeted screens that are extremely broad – increasing the probability that we might find a novel metabolite that’s associated with a disease,” explains Clish.

Researchers aren’t simply looking for the presence or absence of a metabolite but need to go further, interrogating the data for more detailed information about how much of the metabolite is present or how elevated or depressed it is. The big challenge then lies in interpreting the data to work out how specific a change is for that particular metabolite in that disease.

Metabolite Analysis Provides Promising Results 

Metabolomics is already generating results that demonstrate its potential for spotting diseases early.

“In one recent study we identified a novel metabolite associated with liver fat – and it also turns out to be a strong predictor of type 2 diabetes up to twelve years before the onset of disease. In another study, which we haven’t published yet, we’ve found a promising novel set of indicators for coronary heart disease in women five years ahead of diagnosis,” says Clish.

“And a few years ago, we did a study of pancreatic cancer, where we found metabolic derangements up to five years before diagnosis. What was exciting about this was it then suggested that the disease is probably present sub-clinically for many years before it’s diagnosed,” he adds.

Quantification is Key 

It can take years after the discovery of a metabolic change to then translate it into a reliable diagnostic tool. But recent advances in technology are speeding up progress, along with the fundamental realization that quantification is essential for both reproducibility and for clinical application.

“Over the last decade or so, I think everything has improved – both with the sensitivity of the instrumentation and with the design of studies,” says Wishart.

“We used to look for qualitative measures – so whether something’s up or down, but we couldn’t compare across platforms or between labs. But now, with metabolomics moving more towards actual quantification, it will mean we can move things into clinical practice much more quickly,” he adds.

Benefits of ‘omics Integration 

But the biggest benefits may come from combining big datasets across different ‘omics.

“For precision medicine to work for a human being, it’s not just about their metabolome, their genome, their proteome or their microbiome - but about all of these. If we can integrate data across all ‘omics studies, we will be able to come up with new diagnostic, prognostic and therapeutic insight, which would certainly make medicine more precise, more holistic,” explains Wishart.

“But as a community, we continue to struggle to measure and record these ‘omics studies and then integrate and interpret them - I think that’s our biggest challenge,” he adds. “If we can overcome this, I think it will unlock tremendous insights for diseases that are already existing or about to appear – and that’s the essence of precision medicine.”

Personalized Healthcare: Giving the Power to the People

Looking into the future, researchers envision handheld devices that people can use to monitor themselves, enabling them to take control of their own healthcare.

“The idea of having some sort of micro-analyser – using a device like a smartphone – that people could use to monitor their own metabolic profiles at home on a regular basis, then you would really enable precision medicine – because everyone would know their own baseline and be able to spot any changes,” says Clish.

“And over recent years attention on the microbiome has increased – certainly in my lab, we’ve been looking at stool samples for metabolites associated with disease. So there’s even the concept of developing a ‘smart toilet’ – so that, combined with the smartphone – and you’ve got a pretty powerful set of measures,” says Clish.

“We’ve always had the desire to manage our own healthcare so moving the tools for precision medicine into the palm of your hand I think will transform medicine,” says Wishart. 

The Canaries of the Genome Take Centre Stage 

Metabolomics, which has previously often been in the shadows of other ‘omics, is now considered as central for reaping the full potential of precision medicine.

“Metabolites are the canaries of the genome, the sentinels of the immune system, they are the first responders,” enthuses Wishart. “We ignore them at our peril – so that’s why I think metabolomics is so important. It’s something we’ve neglected for too long and it’s time to rebalance that more carefully.”