Biomarkers for predicting response to corticosteroid therapy for immune thrombocytopenic purpura.

Abstract

Patients with immune thrombocytopenic purpura (ITP) often receive corticosteroids as a first-line treatment strategy. The ability to predict the therapeutic response to corticosteroids before initiating treatment would reduce the risk of adverse events, but biomarkers of this parameter have not yet been established. Here, in a single-centre, retrospective, cohort study of 127 ITP patients who received corticosteroids as first-line treatment, we compared several characteristics and test results between those patients with a favourable response to corticosteroids (responder cohort, n= 68) and those with a poor response to corticosteroids (non-responder cohort, n= 59) to identify potential biomarkers that were predictive of corticosteroid response. We extracted six factors as indicative of poor response to corticosteroid therapy for ITP: old age (≥81 years) (odds ratio [OR], 2.44; p= 0.02); low platelet count (<9 × 109 /L) (OR, 2.25; p= 0.02); high level of platelet-associated IgG (≥445 ng/107 cells) (OR, 3.95; p< 0.01), high platelet distribution width (≥ 14.0g/dL) (OR, 2.00; p= 0.03), high lymphocyte-to-monocyte ratio (≥ 3.52) (OR, 1.40, p= 0.04), and low megakaryocyte count in bone marrow (< 85.5/μl) (OR, 1.72; p= 0.04). Thus, our present data support the fact that these six factors are useful biomarkers for predicting corticosteroid response in patients with ITP.