Specific chemokines may improve the diagnosis and bleeding risk assessment in primary immune thrombocytopenia (ITP), according to a study published in the British Journal of Haematology. The authors conducted a prospective analysis to identify chemokines as potential biomarkers in patients with ITP. The researchers recruited 60 patients with ITP and 17 patients with other forms of thrombocytopenia (non-ITP), measuring plasma chemokine concentrations using a Luminex-based assay. The researchers also compared a subgroup of 12 patients with ITP who had bleeding episodes (ITP-B) to 33 without bleeding episodes (ITP-NB). Machine learning algorithms identified CCL20, interleukin-2, CCL26, CCL25, and CXCL1 as potential biomarkers for diagnosing ITP, while CCL21, CXCL8, CXCL10, CCL8, CCL3, and CCL15 were linked to bleeding risk in ITP. These findings were confirmed with enzyme-linked immunosorbent assays in a validation cohort of 43 patients with ITP and 19 patients with non-ITP. Overall, the study highlights specific chemokines as promising biomarkers for both diagnosing ITP and assessing bleeding risk.
The post Machine Learning Supports ITP Bleeding Assessment first appeared on Physician's Weekly.