Autoimmune Cytopenias

Machine-Learning Model for Resistance/Relapse Prediction in Immune Thrombocytopenia Using Gut Microbiota and Function Signatures

At the 63rd annual meeting of the American Society of Hematology (ASH), a multicenter team of investigators used a metagenomic sequencing technique (abstract 18) developed through machine learning to evaluate in immune thrombocytopenia (ITP) the relationship between gut microbiota and relapse within 3 months of therapy with corticosteroids, as well as the likelihood of benefit from thrombopoietin receptor agonists (TPO-RAs) in subsequent therapy. As the authors note, the gut microbiome has emerged as a useful tool for assessment and prediction of immunomodulatory therapy in autoimmune diseases. Using deep shotgun metagenomic sequencing, they analyzed 75 fecal samples (60 before and 15 after corticosteroid therapy) from 60 patients with newly diagnosed ITP and 41 samples from with persistent/chronic ITP before and after treatment with TPO-RAs, including eltrombopag and avatrombopag. They found that while microbiome diversity and composition following corticosteroid therapy changed significantly from baseline, the gut microbiota of ITP patients in remission was similar to that of the [control], "which implies that a shift in the gut microbiome could represent a return to homeostasis," they concluded. Compared with nonresponders to treatment with TPO-RAs, responders had increased baseline levels of Ruminococcaceae, Clostridiaceae, and Bacteroides, they reported, adding that their prediction model based on the gut microbiome for response to TPO-RA "was robust across the cohorts and showed 89.5% and 79.2% prediction accuracy for persistent/chronic ITP patients in the training and validation sets, respectively."

ASH 2021 Annual Meeting and Exposition