What we do

Our research focuses on the role of volatile organic compounds (VOCs) as non-invasive markers of human disease. VOCs are small molecules that are produced by the body and released via the breath, urine, stool and skin. Variation in their levels is associated with a number of diseases, including cancer.

VOC analysis in breath has been the major focus of our research owing to universal acceptability amongst patients of all backgrounds. We have conducted a number of large clinical trials examining the performance of breath testing for the detection of oesophagogastric, colorectal and pancreatic cancers. To support the wider of adoption of breath testing in clinical practice we have developed quality-controlled protocols for breath collection and handling in clinical practice to ensure findings are accurate and reproducible. Parallel research is also being conducted to understand the origin and mechanisms of VOC release from healthy and cancer patients. Our laboratory is equipped with state-of-the-art mass spectrometry instruments that are able to reliably detect VOCs levels up to 1 part per billion (1:1000,000,000).

Why it is important

At an early stage many cancers, including those affecting the gastrointestinal tract, often have non-specific symptoms. For example, symptoms such are heartburn and indigestion are common to many benign (non-cancer) conditions, but in a minority of patients may signify early stomach or oesophageal cancer. When patients present to medical attention with such symptoms it can be difficult for healthcare professionals to decide if further testing is necessary. Many existing tests for diagnosing cancer are invasive and expensive, meaning that it is difficult to justify their use in the majority of patients who will not have cancer. Consequently, for many patients with early cancers, further investigation is delayed until symptoms become more apparent. At this point the tumour is more likely to be at an advanced, incurable, stage.

Our goal is to develop a novel breath test, for use in primary care patients with gastrointestinal symptoms who are at risk of cancer, to determine which patients would benefit from further definitive investigation. In achieving this we hope to improve cancer survival through earlier detection.

How it can benefit patients

Our research intends to make investigation of early cancer symptoms more accessible to patients. By detecting cancers earlier, we will be able to improve patient survival. For the NHS and other healthcare systems the test will provide an accurate, affordable and acceptable solution for early cancer detection.

Summary of current research

The groups research is divided between three core themes.

Clinical trials in disease detection through VOC analysis

We have conducted large scale clinical trials of breath analysis for the detection of gastroesophageal, colorectal and pancreatic cancers recruiting over 4000 patients. We also performed a large-scale trial examining the feasibility of different recruitment and engagement strategies for breath testing in primary care. We will start recruitment to major trials enrolling more than 10,000 symptomatic patients at risk of oesophagogastric (AROMA trial) and colorectal (COBRA II trial) cancers.

Studies to determine the biology and mechanisms of VOC production and release

Our group has undertaken extensive work to understand the molecular and genetic drivers of VOC production in cancer. We have also studied factors influencing the release of VOCs from different sites within the body in order to better understand optimal approaches to their detection. Our current research focuses on the role of the cancer associated microbiome in VOC production.

Methodologies for high throughput and quality-controlled VOC analysis

We maintain a strong focus on establishing accurate and reliable methods for VOC detection in clinical practice. We have developed and optimised methods for VOC analysis in breath, urine and other biofluids. In establishing those methods, we have sought to minimise sample losses and contamination that can occur at all points of the analytical pathway. We also developed a machine learning platform for analysis of VOC data that has overcome the challenges of large-scale data processing in breath analysis and enabled molecular network analysis. The next step will be to acquire ISO accreditation for our laboratory in preparation for large-scale breath testing in clinical practice.