Anti-Platelet Antibodies Are Predominantly IgM in Childhood Immune Thrombocytopenia

Introduction

Platelet clearance in childhood immune thrombocytopenia (ITP) is thought to occur through antibodies binding to platelet glycoproteins such as GPIIb/IIIa, Ib/IX and V. Most children recover spontaneously, but ~10% remain thrombocytopenic beyond 12 months (chronic ITP).

Anti-platelet antibodies of IgG subclass have been shown in adult ITP patients. In children, limited data is available. Van Leeuwen et al. (Scand J Haematol 1981) suggested that children with chronic ITP might have class-switched platelet IgG antibodies, with superior effector functions, whereas anti-platelet IgM antibodies may be found in spontaneously recovering patients. 30% of children with ITP do not respond to intravenous immunoglobulin (IVIG) therapy, and a lower response rate to IVIG has been described in adult ITP patients with anti-platelet GPIb/IX antibodies. This association is yet unknown in children.

In the present work we investigated the prevalence of autoantibodies in childhood ITP and their effect on spontaneous recovery, development of chronic disease, and resistance to IVIG therapy.

Methods

We evaluated the presence and glycoprotein specificity of platelet autoantibodies in 200 children with newly diagnosed ITP who participated in a multicenter randomized controlled trial. In this study, children aged 3 months to 16 years were allocated in a 1:1 ratio to careful observation or a single dose of IVIG, and followed clinically for 1 year. Blood samples obtained at diagnosis were analyzed by indirect platelet monoclonal antibody-specific immobilization of platelet antigens (MAIPA, for IgG and IgM) at a central reference laboratory. To correlate autoantibody levels with clinical outcomes, we investigated the association between platelet antibody profiles and spontaneous resolution of thrombocytopenia (observation arm) and response to IVIG (treatment arm).

Results

Platelet antibodies were evaluated by incubation of patient serum with donor platelets, as the low sample volumes of children made it impossible to directly detect immunoglobulin binding to platelets of a thrombocytopenic child. At diagnosis, platelet autoantibodies of IgM subclass were found in 52% (95% confidence interval [CI], 44-60%) of patients. Glycoprotein-specific IgM antibodies to GPIIb/IIIa were most frequent (47%; 95% CI, 40-55%), followed by GPIb/IX (31%; 95% CI, 24-38%) and GPV (29%; 95% CI, 22-36%). In turn, glycoprotein-specific IgG antibodies were present in only 8% (95% CI, 4-12%) of children. Only 3 patients were positive for IgM and IgG autoantibodies at the same time. IgM anti-GPIIb/IIIa antibodies were more prevalent in young children, and correlated with lymphocytosis.

There was no association between the presence of platelet autoantibodies and resolution of thrombocytopenia 1 week after diagnosis, both for patients that were allocated to observation only (P=0.45 for IgM; P=1.0 for IgG) or treated with IVIG (P=0.94 for IgM; P=0.46 for IgG). The rate of no, partial and complete response were also similar amongst antibody-positive and -negative individuals. Buchanan bleeding scores were comparable amongst patients with and without antibodies.

Finally, we could not detect differences in autoantibody levels or isotype between patients who remained thrombocytopenic beyond 12 months (chronic ITP), and patients who recovered. During the 1-year follow-up period, we found no evidence of class switching from IgM to IgG, also not in children who developed chronic ITP.

Conclusions

Our data show that in contrast to adult ITP, platelet antibodies in newly diagnosed childhood ITP are predominantly of IgM subclass. The spontaneous recovery and response to IVIG was not different between children with or without antibodies.