Biomarkers will reveal chronic fatigue

Chronic fatigue syndrome is a serious illness accompanied by persistent physical and mental fatigue, sleep disorders, headache, muscle pain, and dizziness. Despite the scale of the problem (in the USA, from 836,000 to 3.3 million people suffer from ME/CFS), objective diagnostic methods still did not exist.

“Some doctors still question the reality of this diagnosis due to the lack of biomarkers and attribute the symptoms solely to psychosomatics,” said Jerya Unutmaz, Professor of immunology and one of the authors of the study.

The team of scientists used the innovative BioMapAI AI tool developed by Dr. Ruoyun Xiong to analyze more than 250 biological and clinical parameters in 249 people (153 patients with ME/CFS and 96 healthy ones). The tool integrated data from intestinal metagenomics, plasma metabolome, immune profile, results of routine blood tests, and symptoms described by patients.

BioMapAI was able to distinguish patients with ME/CFS from healthy participants with 90% accuracy, with information about immune cells being the most accurate. The microbiome and metabolites were the best predictors of sleep, digestive tract, and emotional health problems.

“This is fundamentally important, because ME/CFS is characterized by many different symptoms that vary in intensity and duration,” explained microbiologist Julia Oh, co—author of the study. “For the first time, we have linked clinical manifestations to specific disorders at the level of microbiota, metabolites, and immune response.”

Among the main findings are a decrease in the level of butyric acid (butyrate), a beneficial fatty acid produced by the intestinal microbiota, and an increase in substances such as tryptophan and benzoate, indicating an imbalance of microflora. An increase in inflammatory activity was also found, especially from MAIT cells— immune cells that are closely related to the state of the intestinal microbiota.

“MAIT cells act as a bridge between the gut and the immune system as a whole,” explained Unutmaz. “Disorders in this system, along with butyrate deficiency and activation of tryptophan pathways, indicate a profound inflammatory imbalance.”

Interestingly, patients with a more recent diagnosis (who had been ill for less than four years) had fewer violations in these networks than patients with a long-term course of the disease (more than ten years), which may indicate the progressive nature of the disease.

Although the findings require further confirmation, scientists consider them a major step forward. Animal models are not able to reproduce the full range of symptoms of ME/CFS, so it is the multi-layered analysis in humans that gives a chance for accurate diagnosis and personalized treatment.

“The metabolome and the microbiome are dynamic. This means that we can influence them through diet, lifestyle, or targeted medications,” says Julia O.

The model also showed about 80% accuracy when tested on external datasets, which confirms the reliability of the identified biomarkers.

According to the researchers, their goal is to create a detailed map of the interaction of the immune system, microbiota, and metabolites. They intend to make the BioMapAI data available to the scientific community so that other researchers can use it in their work on ME/CFS, long‑covid and other chronic conditions.

“We want medicine to become really accurate. So that each patient’s complaint can be explained and linked to biological mechanisms,” concludes Julia O.