An interview with Dr. Jane Hill, Dartmouth College, conducted by April Cashin-Garbutt, MA (Cantab)
When was it first discovered that patient breath could be used to diagnose disease?
The breath of patients has been used to diagnose disease since was first recorded, as far as we know, by the ancient Greek physician, Hippocrates, who wrote described "fetor hepaticus" (breath of the dead) back in ~400 BC, a reference to the pungent smell of thiols on the breath of patients with liver dysfunction.
The first, FDA-approved system using breath to diagnose an infectious agent was approved in 1996. It’s a urea-based breath test for Helicobacter pylori (the stomach ulcer bacterium).
Over the past couple of decades, the analytical technology and the clinical microbiology insight needed to create a breath test for infections has seen a considerable number of advances and there are now numerous research teams focused on evaluating patient breath as a diagnostic fluid for infectious diseases. Significant challenges still need to be overcome.
In what ways has our understanding advanced in this field over recent years and what advances in technology have been important to your research?
The collection of breath samples in a reliable manner is one of the ways we have increased our understanding over the last few years in this field. Components in the breath are affected by exercising, sitting down and standing up etc., and so it is vital to have a reliable baseline of breath measurement.
At Dartmouth College we use state-of-the-art analytical equipment to analyze breath samples, this includes secondary electrospray ionization mass spectrometry on a system we modified system from AB Sciex, which enables us to carry out essentially real-time analysis of volatile.
Another instrument that we use is a two-dimensional gas chromatography time-of-flight mass spectrometry system from LECO, the profiles of which are more chemically comprehensive. With these instruments, we are able to look for molecules in the breath comprehensively. Developing real-time sensitive detection instruments that can be translated into a clinical setting is a vital area of research that needs to be further developed.
The inclusion of medical doctors on breath research teams i.e., alongside analytical chemists, has also helped focus breath-based infectious disease research questions. Statisticians, engineers, and clinical microbiologists are also becoming increasingly more common members of breath research teams, in a bid to aid translation of research into the clinic.
How can translation from laboratory to clinical setting be achieved?
At this moment in time, very few proof-of-concept biomarkers end in successful translation into a clinical setting. I suggest that success is more likely when the underlying biochemical hypothesis is strong and the link to disease pathogenesis well known. Without a solid understanding of the organism and the host’s response during infection with the organism, as well as the clinical confounders, such as co-morbidities, or biochemical mimics, progress will founder.
Researchers need to be aware that there are several stages in the development of a breath test for infectious diseases. The process is lengthy, involving the iteration of results obtained via data discussions with chemists, medical doctors, clinical microbiologists, etc. Translation of research is achievable, however it is important that assumptions are minimized and always contextualized with input from by clinicians.
Cell culture systems can and should be used when considering a developing technology, which might, for example aim to measure volatile molecules particularly quickly or sensitively. In this experimental system, evaluating the technology is an essential proving ground.
Then, if the proof-of-concept works, a translation has to be done in conjunction with medical doctors and other people with insight into a clinical application. Although knowledge of the organism and the patient population may be well-known, other clinical confounders must be considered. These confounders are quite different to the more discrete and well-understood problems analytical chemists typically consider in a laboratory.
For example, a patient may have a co-morbidity factor, such as an underlying heart disease issue or diabetes, which can influence the molecules in the breath as well as the control patient populations recruited to the breath study. If the heterogeneity that occurs in a patient population is not accounted for, both in terms of the diseases that might appear similar to the one that you're trying to develop a diagnostic for, as well as those that might exist in the general population, your evaluation of putative breath biomarkers may not be effective. Prevention of this common error is usually obtained with the active involvement in the project, including the early planning stages, by clinical staff, such as physicians and nurses.
A common mistake when trying to translate laboratory research for clinical application for an infectious breath test is assuming that the culture volatile molecules grown in the lab will be exactly the same as the molecules detected in the human patient, which is highly unlikely. It is therefore vital that scientists remain hypothesis-driven and that both in vitro and in vivo phases.
What do you think the future holds for diagnosing infections through molecules in patient breath?
There are still some stages to overcome, getting FDA and MHRA approval is nontrivial, taking several years to make it through the approval process. However, over the next decade or so we will see more breath tests come onto the market. My team hopes to be a part of that wave.
Please can you outline your upcoming talk at Pittcon 2017?
My talk at Pittcon will cover the effective use of cell cultures to develop proof-of-concept datasets useful to the clinical context. I will also talk about the design of these experiments within the bigger context of the clinical relevance.
In particular, I will focus on the following three bacterial infection contexts:
Tuberculosis kills approximately one and a half million people each year making it the biggest infectious disease killer in the world each year. I will present work expanding from the lab bench to patient breath analysis.
- Pseudomonas aeruginosa
Pseudomonas aeruginosa is an opportunistic pathogen that particularly impacts vulnerable populations, like those with cystic fibrosis or chronic obstructive pulmonary disorder, creating a high level of morbidity and mortality for those patients. We have carried out a lot of bench and clinical work in this context and I will try and make those connections in my talk at Pittcon.
- Multidrug-resistant bacteria
I will share our early work on possible strategies to target the identification of multidrug-resistant bacteria.
What Pittcon Can do for You from AZoNetwork on Vimeo.
Where can readers find more information?
Pittcon® is a registered trademark of The Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, a Pennsylvania non-profit organization. Co-sponsored by the Spectroscopy Society of Pittsburgh and the Society for Analytical Chemists of Pittsburgh, Pittcon is the premier annual conference and exposition on laboratory science.
Proceeds from Pittcon fund science education and outreach at all levels, kindergarten through adult. Pittcon donates more than a million dollars a year to provide financial and administrative support for various science outreach activities including science equipment grants, research grants, scholarships and internships for students, awards to teachers and professors, and grants to public science centers, libraries and museums.
Visit pittcon.org for more information.