Immunoglobulin Levels As Predictors of the Development of Chronic Disease in Pediatric Immune Thrombocytopenia

Background: Immune thrombocytopenia (ITP) is the most common acquired bleeding disorder in children. Although ~75% of these patients will go on to have spontaneous resolution of disease, up to 25% will develop chronic ITP, with significant and prolonged bleeding symptomatology and impaired quality of life. At this time, there is no definitive method to predict the development of chronic disease at the time of ITP diagnosis.

Aims: We aim to identify clinical biomarkers predictive of the development of chronic ITP at the time of diagnosis among pediatric ITP patients.

Methods: The clinical records of 280 pediatric ITP patients enrolled in a biological banking study at a large pediatric tertiary care referral center from July 1, 2015 to July 1, 2019 were reviewed in accordance with IRB-approved protocols. ITP diagnosis and chronicity of disease (vs. spontaneous resolution within 1 year of diagnosis) was confirmed by a pediatric hematologist, as defined by current international expert committee standards (Neunert et al, Blood, 2011;117(16):4190-207; Provan et al, Blood, 2010;115(2):168-186). Patients with non-classical ITP were excluded (1 child with drug-induced ITP in the setting of acyclovir therapy, and 1 neonate with passive ITP in the setting of maternal lupus); and patients who were lost to follow-up or had ongoing disease of <1 year duration were subsequently excluded. Patient characteristics including age, gender, ethnicity, presence of concurrent autoimmune disease, and time to resolution were collected. Pertinent biomarkers including immunoglobulin levels (IgG, IgA, and IgM), anti-nuclear antibody (ANA), C-reactive protein (CRP), and immature platelet fraction (IPF) which were obtained at the time of diagnosis were documented, if if obtained prior to the administration of any ITP-directed therapy, and within 2 weeks of diagnosis.

Chi-squared test or Fisher's exact test was utilized to compare nonparametric categorical data. Mann-Whitney U test was used to compare nonparametric continuous data. A multivariate backwards logistic regression model was conducted to determine independent associations with the development of chronic ITP. Statistical analyses were

performed using SPSS Statistics 26 (IBM, Armonk, New York). A Bonferroni correction was applied to correct for multiple comparisons. A p value < 0.05 was defined as statistically significant.

Results: We identified 251 pediatric ITP patients with sufficient length of follow-up to define acute (<1 year) vs. chronic (>1 year) disease within our 4-year study period. This included 132 patients with acute ITP (53%) and 119 patients who went on to develop chronic ITP (47%). Mean age of diagnosis among those with acute ITP was 5.5 years, while mean age of diagnosis among those with chronic ITP was 7.7 years. Within the entire cohort, 126 (50%) were male, and 124 (49%) were of Hispanic ethnicity. Fifty-eight had ITP attributable to systemic autoimmune disease - 17% within the Acute ITP group, and 30% within the Chronic ITP group. Among all patients (n=251), 188 with evaluable biomarkers (immunoglobulin levels, ANA, CRP, or IPF) at the time of diagnosis were identified. These included 174 patients with Ig levels at diagnosis, 43 patients with ANA titers at diagnosis, 25 patients with CRP levels at diagnosis, and 78 patients with IPF at diagnosis.

By univariate analysis, we found that abnormalities in immunoglobulin levels at diagnosis were significantly associated with the development of chronic ITP. Low IgG levels (p = 0.015) were more prevalent in chronic (10.3%) vs. acute ITP (1.9%). High IgG levels (p = 0.015) were more prevalent in chronic (6.7%) vs. acute ITP (1.9%). High IgM levels (p = 0.03) were more prevalent in chronic (26.5%) vs. acute ITP (13.3%). Of note, the presence of Evans syndrome (p = 0.0005) and other systemic autoimmune disease (p = 0.011) were also significantly associated with the development of chronic ITP. Subsequent multivariate analysis identified low IgG level at diagnosis to be independently predictive of chronic ITP (p = 0.043, with an odds ratio of 5.7).

Conclusion: Although further study is needed within a larger international pediatric ITP cohort, the findings from this institutional study indicate that biomarkers obtained in routine clinical care at the time of ITP diagnosis, specifically immunoglobulin levels, can be utilized to help predict development of chronic disease among pediatric ITP patients.