Predicting development of pediatric chronic immune thrombocytopenia at disease onset using a statistical risk model

Area of Science:

• Pediatric Hematology
• Immunology
• Biostatistics

Background:

• Immune thrombocytopenia (ITP) in children has an unpredictable course, with up to 30% developing chronic disease.
• Limited known predictors for chronic ITP complicate early clinical management and patient counseling.
• The complex pathophysiology of immune dysregulation in ITP contributes to disease chronicity.

Purpose of the Study:

• To develop and validate a statistical model for predicting the development of chronic immune thrombocytopenia (ITP) in children.
• To identify key clinical and laboratory variables associated with ITP chronicity.
• To provide a tool for early risk stratification in pediatric ITP.

Main Methods:

• Retrospective development of a multivariable logistic regression model using a training cohort of 611 children with ITP.
• Validation of the model using two external cohorts totaling 161 children.
• Inclusion of variables such as age, sex, immunoglobulin levels (IgG, IgA, IgM), presenting platelet and lymphocyte counts, secondary cause, and DAT positivity.

Main Results:

• The developed model effectively predicted the development of chronic ITP.
• Key predictors identified include age, sex, IgG, IgA, IgM, presenting platelet count, presenting lymphocyte count, secondary cause at diagnosis, and DAT positivity.
• External validation confirmed the model's consistent discriminative performance and clinical utility.

Conclusions:

• A validated multivariable model can predict chronic ITP in children at the time of diagnosis.
• This predictive tool can assist hematologists in counseling families and tailoring early treatment strategies.
• An online tool (https://opal.shinyapps.io/citp-rm/) is available for clinical use.