Machine learning was used to evaluate the blood microbiome of cancer patients. Researchers wanted to see if the plasma microbiome reflected the microbial population in the stomach in patients with systemic lupus erythematosus (SLE) and healthy controls (HCs).

Microbial 16S ribosomal RNA sequencing was used to examine the microbiome makeup of paired plasma and feces samples from female SLE patients and HCs.

Both HCs and SLE patients had lower microbial alpha diversity in stool compared to plasma and different plasma and gut beta diversity. There was no change in gut microbial diversity; however, plasma alpha diversity was lower in SLE patients than HCs. In both groups, the prevalent bacteria differed between plasma and feces. Although the leading genus bacteria in plasma and feces were comparable in SLE and HC patients, several were significantly distinct.

The plasma microbiome had a separate community and more variability than the gut microbiome, indicating that most circulating microbiome may come from places other than the gastrointestinal tract (e.g., oral or skin). The lower plasma but not gut alpha diversity in SLE patients compared to HCs suggests that the plasma microbiome in SLE is changed, which may be essential for systemic immunological perturbations and SLE disease pathogenesis.