Association between Platelet Autoantibody Specificity and Response to Intravenous Immunoglobulin G in the Treatment of Patients with Immune Thrombocytopenia.

Background: Recent studies in murine model of immune thrombocytopenia (ITP) suggest that autoantibodies against GPIba may cause thrombocytopenia through Fc-independent pathway and ITP mediated by anti-GPIb may be less responsive to intravenous immunoglobulin G (IVIG) treatment (Nieswandt B, et al. Blood 2000; Webster ML, et al. Blood 2006). The objective of this study was to investigate the potential association between platelet autoantibody specificity and response to intravenous immunoglobulin G (IVIG) treatment in patients with ITP.

Methods: We retrospectively reviewed the clinical history of ITP patients who had platelet autoantibody test and received IVIG for treatment of thrombocytopenia. ITP was diagnosed by usual clinical criteria. All platelet autoantibody tests were performed by the Blood Center of Wisconsin (Milwaukee, WI) using ELISA, which detects antibodies reactive with platelet GPIIb/IIIa, GPIb/IX, and GPIa/IIa. A response was defined as a platelet count of > 50 x 10⁹/L with a minimum increment of 30 x 10⁹/L.

Results: We found 17 patients who received IVIG and had platelet autoantibody test performed. The median age was 58 years (range, 20–83) and 10 (59%) were males. ITP was considered idiopathic in 12 patients. In the remaining 5 patients, the following hematologic conditions were present: chronic lymphocytic leukemia (2), Hodgkin lymphoma (1), myelodysplastic syndrome (1), and non-Hodgkin lymphoma (1). Antibodies reactive with GPIIb/IIIa, GPIb/IX, and GPIa/IIa were detected in 13, 10, and 8 patients, respectively. In 3 patients, none of these antibodies were detected. Among the 14 patients with antibodies, 9 (64%) had antibodies against more than 1 glycoprotein. The median pre-treatment platelet count was 9 x 10⁹/L (range, 1–68). Overall, 10 (59%) patients responded to IVIG. A response occurred in 7 of 7 patients without anti-GPIb/IX but in only 3 of 10 patients with anti-GPIb/IX. The difference in the response rates between the 2 groups was statistically significant (P= .0098).

Conclusions: Our clinical results support the murine model findings that ITP mediated by anti-GPIb may be less responsive to IVIG treatment. However, because of the small number of patients investigated and the retrospective nature of the study, our findings must be confirmed by other groups in a larger patient population.