Despite significant advances in research, the exact disease mechanism(s) underpinning ME/CFS remain unclear, and no validated biomarkers for the diagnosis of the disease are available.
Recognising this, a team of researchers in Canada aimed to develop a method to detect changes in the liquid part of blood – blood plasma – which they hoped would enable them to differentiate between participants with ME/CFS and healthy, yet sedentary, controls.
The team collected blood samples from 115 people with ME/CFS and from 45 controls, these samples were taken at rest, and again 90 minutes after a ‘non-invasive stress test’ that was designed to induce the cardinal symptom of ME/CFS; post exertional malaise (PEM). Importantly, all participants provided their informed consent to participate in the study.
Samples were analysed using a relatively low cost tool called ‘label-free Raman spectroscopy’ (RS) which looks for molecular changes in blood and tissue samples. Building on this analysis, the team used machine learning (ML) models – computer models that can learn from patterns in data to make predictions or decisions – on the information from the RS analysis and assessed whether it was possible to differentiate between those with ME/CFS and healthy controls at each timepoint.
Results were complex but showed that there were indeed differences between those with ME/CFS and healthy controls, and that these differences could be seen both at rest and following the stress test.
Specifically, findings were consistent with disturbances in the biology of protein building blocks (amino acids), fat molecule (lipid) regulation, and energy production and utilisation (metabolism) in people with ME/CFS – all of which have been previously implicated in ME/CFS.
It is important to note that this study was a small exploratory study, known as a pilot study, and before firm conclusions can be drawn validation in larger and more diverse cohorts is required. Despite this, the researchers conclude:
“RS-ML may contribute to improved patient stratification, more objective assessment of disease activity, and a better understanding of the biological mechanisms underlying ME.”
