Kirk Pappan, Translational Science Team Lead and Jonathan Kitley, Scientific Content Creator at Owlstone Medical explain why genes are just part of the precision medicine journey.
To date, most research around precision medicine has focused on what can be achieved through genetics and genomics. That makes logical sense as a starting point, as genes are the foundations of biological systems. However, when it comes to disease mechanisms, genes are only part of the story.
The complex relationships between individual genes, molecular pathways and disease phenotypes make it challenging to understand diseases, their causes, and potential treatments from genetics alone. With the maturation of other ‘omics’ fields, a continued focus solely on genetics may be leading us to an incomplete view of what can be achieved through a tailored approach to healthcare. Environment and experience, reflected in the human metabolome, offer more direct indicators of a patient’s current health at any given time.
The metabolome – the full complement of all biochemically relevant small molecules in the body – provides a comprehensive and current snapshot of changes within the body. As a close reflection of phenotype, the metabolome is much more directly and diversely affected by disease and offers a more applicable way to assess the presence, development and treatment of a wide range of illness. As such, metabolomics could realise the true potential of precision medicine by analysing the combined downstream impact of genetics, environment and disease.
This article will discuss the relative advantages of metabolomic analysis and the challenges that have been overcome to enable it to be applied in precision medicine. This will be illustrated through examples of non-invasive metabolomic analysis, such as breath research, for applications including asthma, liver disease, gut health, and cancer – with an emphasis on the additional information available through this approach.
What is metabolic precision medicine?
Genomics is currently the dominant force in precision medicine thanks to the relative ease and reliability of established methods for collecting and interpreting genetic information. Metabolomics offers a more diverse, albeit more complicated, source of information linked to the human body’s metabolic networks. Perturbations of these networks are the underlying mechanisms of disease, reflecting not just genomics but also environment, and changes to the metabolome can directly reflect real-time disease progression.
Genes are not typically changed by disease (except in certain cancers and viral infections), so they cannot usually reveal whether a patient is ill, instead suggesting a predisposition towards developing a particular illness at some point in their life (e.g., BCRA1suggesting an increased risk of breast cancer). In contrast to genes, the relative abundance of proteins and their locations often change in response to illnesses, however accessing suitable protein samples can be particularly challenging and is still insufficient to capture all forms of disease.
Metabolites are small and are readily circulated throughout the body in the bloodstream. Therefore, sampling these metabolites provides a consistent window into disease, including at its earliest stages. Metabolic biomarkers also offer high sensitivity as the effect of a change to a single gene or protein can be amplified to a difference of 100s of metabolite molecules.
Many diseases can have significant metabolic impact with little or no effect on the genetic or protein complement of a cell. In contrast, metabolism reflects how the body’s biological systems are interacting with environmental factors at any given time, and how this changes distinctly during illness. That means that analysis of metabolic biomarkers has utility not only as a sensitive means for detection and diagnosis, but also on disease progression and treatment monitoring.
Breath samples reflect metabolic activity
There are many ways to sample genetic, protein-based, and metabolic biomarkers that are already in clinical use. Tissue and liquid biopsies are both regularly offered for the diagnosis and treatment of a wide range of diseases. While not yet widely adopted, breath offers a significant number of unique advantages as a metabolic sampling medium. Exhaled breath is a rich source of volatile organic compounds (VOCs) that can originate either from within the body (endogenous VOCs) or from external sources such as diet, prescription drugs and environmental exposure (exogenous VOCs). Studies have identified over 1,000 VOCs that can be found on breath, many of which are products of metabolism.
Breath sampling is non-invasive and investigations using breath sampling with vulnerable groups have reported high levels of patient comfort, even when used with patients self-reporting shortness of breath1. Further, exhaled breath is also an effectively inexhaustible waste product that the human body constantly produces in large quantities, while the amount of material that can be collected through other methods is highly limited. Breath is therefore ideal both for the early detection of disease where the markers may be in very low abundance, and in applications where multiple longitudinal samples are needed such as to monitor disease progression and treatment response.
There is a growing level of international awareness that breath may hold the solution to many difficult healthcare challenges, and that is reflected by the huge growth in the number of breath research papers published annually, over recent decades (Figure 1).
Phenotyping disease through breath biopsy
Breath biopsy, another name for measurement of breath biomarkers, can be applied to precision medicine in several clinically relevant contexts. If breath biomarkers can be identified that predict both therapy response and treatment efficacy, it could lead to diagnostic breath tests that enable therapeutics to be provided to those most likely to respond. Additionally, longitudinal sampling can monitor treatment response, providing early indicators of recovery, resistance development and/or adverse effects. Overall, this has great potential to reduce the time and cost of finding the most effective treatments.
An excellent example of the potential of this approach can be seen in asthma. Asthma encompasses a broad spectrum of conditions that cause chronic inflammation in the airways of the lungs and affects up to 339 million people worldwide2. Currently, a lack of accurate stratifying diagnostics means that treatments are often prescribed via a ‘trial and error’ approach – leading to poor patient outcomes (such as limited disease control and increased chance of exacerbations) and wasteful healthcare spending. In 2019, a paper published in the American Journal of Respiratory and Critical Care Medicine, Schleich et al.demonstrated that breath analysis offers the possibility of a rapid, straightforward, and non-invasive method to stratify asthma patients and identify precision treatments for different disease endotypes3.
Schleich et al.conducted a study of 500 asthma patients that was able to identify four VOCs as key identifiers of asthma types. Participants had first been associated with asthma inflammatory endotypes based on induced sputum testing – a test that is time-consuming, complex and not suitable for young children, but that is currently the best method for asthma stratification. Subsequently, breath samples were analysed via GC-MS where it was found that patients with neutrophilic asthmas exhaled elevated levels of nonanal, 1-propanol and hexane relative to other phenotypes, while eosinophilic asthmas could be characterised by lower levels of hexane and 2-hexanone. The precision medicine model the study produced to phenotype patients based on these VOCs, achieved 70% accuracy, 45% sensitivity and 85% specificity with an area under receiver operating characteristic curve (AUROC) of 0.72.
Comparing VOC measurement to FeNO and blood eosinophil readings, both diagnostic methods that are currently more widely accepted, Schleich et al.found comparable accuracy (Figure 2). Combining all three readings enabled the team to achieve an AUROC of 0.87 (accuracy 76%, sensitivity 79%, specificity 78%) in their validation cohort. Notably, the breath collection procedure was also easier to perform and is generally preferable for patients. The effect of combining tests also demonstrates that the metabolic VOC biomarkers reflect factors that are orthogonal to other testing methods, so add diagnostic value both alone and in combination.
Assessing metabolic activity using probes
Probes are a way to boost the diagnostic efficacy of various testing methods.4PET scans and MRIs, for example, use a range of contrast agents to highlight image features that are relevant to specific diagnostic needs. Similarly, metabolic probes can be used to monitor the activity of target pathways to allow more sensitive and reliable measurement of key biological responses to disease. Isotope labelled probes have been investigated for a range of applications, for example 13 C-methacetin liver disease tests, which produce13 CO2 detectable on breath. Digestive substrates which trigger fermentation by the gut microbiome in the context of various gastrointestinal conditions can be investigated using hydrogen and methane breath tests.
Recently developed exogenous VOC (EVOC) Probes, take the use of probes one step further. EVOC Probes and their volatile products can be monitored on breath to assess the metabolic activity of target pathways and to help stage disease or reveal how a patient metabolises prescribed drugs. Further, different EVOC Probes can be selected to deliberately target different metabolic pathways to develop precision medicine tests. Breath biopsy combined with an EVOC-probe-tagged drug therefore could evaluate drug pharmacokinetics, to predict how a patient may respond to treatment. There is also evidence that these pathways may relate to drug-induced toxicity, providing advanced warning of potential side effects. In the future we may be able to develop an EVOC Probe that specifically reports on cancer.
An area where substantial progress is being made, towards the clinical application of EVOC Probes, is in liver disease. Factors such as poor diet, obesity, and type 2 diabetes are contributing to a rapid rise in the prevalence of liver diseases in many countries, affecting as many as 3 in 10 adults and an increasing number of children, as seen in a US study5 where 10% of children between the ages of two and 19 showed signs of non-alcoholic fatty liver disease (NAFLD).
Monitoring liver diseases, such as NAFLD and non-alcoholic steatohepatitis (NASH) as well as more advanced cirrhosis, is a significant clinical challenge. Current methods are often expensive and invasive but a 2020 publication from Ferrandino et al.demonstrated that limonene found on breath offered a potential non-invasive testing option6.
Limonene is a naturally occurring VOC that is not produced by the body but can be found in food and drinks such as orange juice and can be reliably and accurately detected on breath. The cirrhotic liver is less able to break limonene down, meaning that after consumption, higher concentrations of limonene will be released in the breath of a patient with liver disease. By using breath biopsy to study the exact levels of limonene on breath, Ferrandino et al.were able to produce a model that differentiated between healthy and cirrhotic individuals with 75% accuracy (73% sensitivity, 77% specificity).
Further study is needed but breath limonene appears to compare favourably to known blood indicators of liver function, currently collected through more invasive approaches. Limonene shows great potential, not just to diagnose liver disease, but more specifically to monitor and stage liver disease,as levels were shown to be especially pronounced in those with more severe disease (Child-Pugh classification) (Figure 3). Interestingly, limonene levels on breath can decrease dramatically following a successful liver transplant7, suggesting that this approach also holds great promise for monitoring response to therapy.
In the future, breath biomarkers could be used to develop companion diagnostics to screen patients for inclusion in clinical trials. Using precision medicine to select the patients most likely to respond to a particular therapy would enable smaller and more cost-effective trials. Proactively stratifying patients would also ensure greater prior knowledge of the patients included in each trial. As a result, clinical trials would have a better chance of success and be more likely to result in regulatory approval for therapies that are only effective for specific disease phenotypes.
When will we see breath tests for metabolic precision medicine in the clinic?
In 1970 Linus Pauling successfully demonstrated that there was more to exhaled breath than simply nitrogen, oxygen, carbon dioxide and water vapour. It has taken time, however, for the technologies and understanding of the many components of breath to develop to the stage where clinical applications have become a realistic possibility.
We now know exhaled breath to be an incredibly complex medium, with the VOC biomarkers relevant to metabolic precision medicine often occurring in low abundance (at the parts per billion and trillion levels). Reliable collection and accurate detection are therefore key to identifying suitable biomarkers for clinical testing, but for a long time a lack of standardisation of either of these processes across the breath sciences has limited progress.
In recent years experts in the field have worked together with the European Respiratory Society Task Force8 and the Breathe Free Project to better standardise breath collection to ensure comparable results across studies, and new reliable and reproducible technologies developed by this partnership, such as the ReCIVA Breath Sampler, are now leaders in the field.
At the same time, while many techniques have been applied to breath analysis, gas chromatography mass spectrometry (GC-MS) has emerged as the leading method for VOC biomarker discovery, offering superior sensitivity and the ability to specifically characterise individual compounds. The latest high-resolution GC-MS platforms excel in identification and quantification of biomarkers within the complexity of a breath sample – providing vital biological insight across the full range of VOCs on breath, which simply is not possible with many other approaches. While these are the established gold standards, the vital insights that these approaches are currently providing will ultimately help to develop more portable point of care solutions, capable of easily performing reproducible and reliable breath tests in the clinical setting.
Together, GC-MS and advances in sampling technologies that allow pre-concentration of VOCs and help to reduce the presence of environmental VOCs in breath samples are providing the solutions needed to enable the development of novel tests in areas of high clinical need. As a result, leaders in the field are predicting the increased prevalence of breath tests in the clinic to support metabolic precision medicine, within the next few years.
Figure 1
Figure 2: Receiver operating characteristic curve analyses of the sensitivity and specificity of volatile organic compounds (VOCs) (hexane and 2-hexanone), blood eosinophil levels, exhaled nitric oxide fraction (FeNO), and combined VOCs, blood eosinophils, and FeNO for the diagnosis of sputum eosinophilia in the replication cohort. AUC = area under the curve.
Figure 3: Boxplots demonstrating the levels of limonene on exhaled breath from patient’s in different groups (healthy, cirrhosis and cirrhosis with hepatocellular carcinoma (HCC)) (left). Boxplots of levels of exhaled limonene for cirrhosis patients by Child-Pugh classification of disease severity.References
- Holden, K.A., et al., Use of the ReCIVA device in breath sampling of patients with acute breathlessness: a feasibility study.ERJ Open Res., 2020. 6: 00119-2020
- The Global Asthma Report. The Global Asthma Network 2018. (http://www.globalasthmareport.org/index.html)
- Schleich, F.N., et al., Exhaled Volatile Organic Compounds are Able to Discriminate between Neutrophilic and Eosinophilic Asthma. Am J Respir Crit Care Med., 2019. 200(4): p 444-453.
- Djago, F., et al., Induced volatolomics of pathologies. Nature Reviews Chemistry., 2021. 11.
- Definition & Facts of NAFLD & NASH. National Institute of Diabetes and Digestive and Kidney Diseases. (https://www.niddk.nih.gov/health-information/liver-disease/nafld-nash/definition-facts)
- Ferrandino, G., et al., Breath Biopsy Assessment of Liver Disease Using an Exogenous Volatile Organic Compound—Toward Improved Detection of Liver Impairment.Clinical and Translational Gastroenterology, 2020. 11(9): p. e00239.
- Fernandez de Rio, R., et al., Volatile Biomarkers in Breath Associated With Liver Cirrhosis — Comparisons of Pre- and Post-liver Transplant Breath SamplesEBioMedicine, 2015. 2(9): p. 1243-1250.
- Horváth, I., et al., A European Respiratory Society technical standard: exhaled biomarkers in lung disease.Eur Respir J, 2017. 49: 1600965
About the authors
Kirk Pappan, Translational Science Team Lead at Owlstone Medical.Born and raised in Kansas, Pappan’s earliest influence in science and medicine came from helping his father, a large animal veterinarian, do his rounds on various horse and cattle farms. His interest in metabolism began with post-doctoral research into metabolic factors during the development of type 2 diabetes. Kirk has a decade of metabolomics and precision medicine experience, and over 30 peer-reviewed publications in the field.
Jonathan Kitley, Scientific Content Creator at Owlstone Medical.Kitley was fascinated by science from an early age and a voracious reader of books on every subject. Now at Owlstone Medical, alongside occasional writing for the Royal Society of Chemistry, he is passionate about ideas that have the power to change the world, for the benefit of all.
