Volatile hypoxia signatures in oesophageal adenocarcinoma

Clinical need: Every year in the UK, 9,200 patients are diagnosed with cancer of the gullet (oesophagus) but only 15 in 100 patients can expect to live beyond five years. Most patients do not respond to chemotherapy or radiation therapy and this lowers their chances of survival. More effective treatments are urgently needed to improve patient outcomes.

A lack of oxygen ('hypoxia') in tumours is a cause of treatment resistance. Promising new drugs have been developed which reduce hypoxia and could improve outcomes for patients. However, previous studies have shown that only certain patients with hypoxic tumours will respond. We currently have no simple method of measuring tumour hypoxia (i.e. a 'biomarker') to predict who will benefit or know if the drugs are working. This makes designing clinical trials challenging, hence no hypoxia-targeted therapies are approved for use in oesophageal cancer. A simple test that measures tumour hypoxia could accelerate clinical trials and make these new drugs available to patients.

Solution: Breath testing is safe, acceptable to patients and can be repeated easily. Breath tests are already used in law enforcement to detect blood alcohol levels and in healthcare to diagnose stomach infections (H. pylori) and asthma. At Imperial College London, we have also shown that volatile chemicals in the breath measured using mass spectrometry can accurately diagnose oesophageal cancer. Recent work has shown that some of these chemicals originate from cancer cells and can be influenced by hypoxia. This may mean that we could measure tumour hypoxia through a simple breath test.

Aim: The project aim is to develop a non-invasive breath test to predict and monitor the response to hypoxia-targeted therapy in oesophageal cancer.

Institutions: This research will be delivered through a strategic collaboration between Imperial College London and Manchester Cancer Research Centre (MCRC). The Volatile Organic Compound (VOC) laboratory at Imperial has recruited more than 3000 patients with cancers of the digestive tract for clinical trials of non-invasive diagnostics over the last 2 years. MCRC has extensive experience in tumour hypoxia research and delivering biomarker-driven trials of hypoxia-targeted therapy.

New findings: We have identified: (i) target genes that can be used to detect hypoxia from tumour samples (a hypoxia gene 'signature') and (ii) volatile chemicals that are produced by tumours in response to hypoxia that can be detected in the breath.

Project structure: Our aim is to fully characterize the volatile chemicals that are produced by hypoxic tumours and the mechanisms of their production. This will involve analysing volatile chemicals in: (i) breath and cancer tissue from patients retrieved during a camera test ('endoscopy') as part of their routine investigations, and (ii) cancer cells grown under hypoxic and normal conditions. The second aim is to determine how these chemicals change in response to therapy. This will involve analysing: (i) cells grown in a laboratory treated with hypoxia-targeting drugs, and (ii) the breath of oesophageal cancer patients in a clinical trial of a hypoxia-targeting drug.

Importance: This project aims to develop a non-invasive breath test to improve patient selection and monitoring of anticancer drugs targeting hypoxia. It is intended that this breath test will accelerate clinical trials to make more effective therapies available to patients. Successful development of a breath test for treatment monitoring would also be the first of its kind worldwide. This innovative approach could promote the concept of breath-based monitoring for a wide range of applications in healthcare.

Oesophageal adenocarcinoma (OAC) has poor survival rates (<15% at 5 years) and there is an urgent need to develop more effective therapies to improve clinical outcomes. Tumour hypoxia is druggable, but hypoxia stratification and monitoring are needed to advance clinical trials of hypoxia-modifying therapy (HMT). Mass spectrometric analysis of exhaled volatile organic compounds (VOCs) is a promising approach to diagnose OAC non-invasively. Our recent work has shown that exhaled VOCs reflect deregulated volatile carbonyl (VC) chemistry in OAC and can be influenced by hypoxia.

By comprehensively characterizing volatile signatures of hypoxia targeted at the carbonyl chemistry of OAC, we hypothesize that exhaled VOCs could be used to predict and monitor the response to HMT non-invasively. This will be investigated by developing a VOC signature of hypoxia using human and model systems with mechanistic understanding and examining the VOC signature in response to HMT.

The VC chemistry of OAC will be analysed using quantitative MS and validated methods in (i) breath and tumour samples stratified by a rationally selected hypoxia gene signature and (ii) cell lines under hypoxic and oxic conditions to evaluate kinetics. Genetically constrained cell models and stable isotope tracing will be used for mechanistic evaluation. Hypoxia-targeting drugs will be evaluated in OAC cell lines with serial VOC measurement. The influence of hypoxia and VCs on therapeutic targets will be assessed, focusing on the hypoxic DNA damage response. Finally, patients participating in an HMT trial will be recruited for a 'proof-of-principle' study of response prediction and therapeutic monitoring using breath analysis.

The proposed scientific work will develop for the first time the VOC signature of hypoxia in OAC. This innovative approach will provide the opportunity to use breath testing as a non-invasive tool to monitor the response to targeted cancer therapy.