Double jeopardy

Patients treated with the widely used “RGD-mimetic” platelet inhibitors sometimes experience severe but self-limited thrombocytopenia after exposure to one of these agents. Since thrombocytopenia occurs within a few hours, it is generally thought that the responsible drug-dependent antibodies affect only circulating platelets.

In this issue of Blood, Greinacher and colleagues describe the case of a 67-year-old man who experienced severe thrombocytopenia and profuse bleeding after a second exposure to the platelet function inhibitor eptifibatide.¹  An eptifibatide-dependent, platelet-reactive antibody specific for the platelet glycoprotein GPIIb/IIIa complex was detected in the patient's serum, providing a likely explanation for the acute drop in platelet levels. Unlike most patients with eptifibatide-induced immune thrombocytopenia, however, the platelet count remained profoundly low (< 5000/uL) for 4 days. A normal count was not achieved until more than a week later. Four days after the acute episode, the overall numbers of megakaryocytes in the marrow was reduced, and they had a “young” morphology. This suggested that the drug-dependent antibody might have injured mature megakaryocytes in addition to causing destruction of circulating platelets, and provided a likely explanation for the 4- to 6-day delay before platelet levels began to recover. To test this possibility, CD34⁺ cord blood stem cells were cultured under conditions favoring differentiation into megakaryocytes and were then treated with patient Immunoglobulin G (IgG) in the presence and absence of eptifibatide. Loss of cell viability (trypan blue exclusion) was significantly greater under these conditions than after treatment with drug alone or IgG alone. Cytologic studies showed that drug-dependent cytotoxicity preferentially affected megakaryocytes that had a high surface density of GPIIb/IIIa, presumably a population of relative mature cells. From these findings, the authors conclude that prolonged thrombocytopenia observed in this patient was caused by the cytoxic effect of a GPIIb/IIIa-specific, drug-dependent antibody on mature megakaryocytes.

The issue of whether platelet-specific antibodies can damage megakaryocytes and impair platelet production has a long and interesting history. Almost a century ago, Frank observed that, although normal or increased numbers megakarocytes were present in the bone marrow of patients with “essential (idiopathic, autoimmune) thrombocytopenic purpura” (ITP), many of the cells were small, agranular, and devoid of free platelets clustered about their periphery.²  From these findings, he concluded that low platelet counts in this condition were the result of insufficient platelet production by defective megakaryocytes. However, a contemporary, George Minot, concluded from his examinations of bone marrow that, in such cases, platelets were destroyed “as fast as they were formed.”³  Thereafter followed a lively and long controversy over which of these mechanisms was the main cause of thrombocytopenia in ITP. In the very first issue of Blood, published in 1946, William Dameshek summarized his extensive study of marrow morphology in patients with ITP and agreed with Frank that “… the fundamental defect leading to thrombocytopenia is a dysfunction of the megakaryocytes … ”⁴  In the subsequent 20 years, it was shown that ITP is associated with platelet-specific autoantibodies.^5,6  Moreover, technical advances made it possible to measure the lifespan of transfused and autologous platelets, and studies using the new tools showed that platelet survival is markedly shortened in almost all cases of ITP.⁷  Accordingly, Baldini concluded in an authoritative review published in 1966 that “… the old hypothesis … of a toxic depression of megakaryocytes and their activity resulting in thrombocytopenia (in ITP) has (now) been disproved … ”⁸ 

The idea that platelet-reactive antibodies might be cytotoxic for megakaryocytes would not die, however. In the subsequent 4 decades, in vivo and in vitro studies provided evidence that antibodies, and perhaps cellular immune mechanisms as well, can in fact act on megakaryocytes to suppress platelet production.^9,10  Patients with immune thrombocytopenia appear to be heterogeneous in respect to which of the 2 mechanisms predominates, making it likely that platelet antibodies differ from patient to patient in their ability to adversely affect megakaryocyte viability and maturation. The antibody studied by Greinacher et al provides a particularly striking example of an immunoglobulin that not only caused acute destruction of peripheral platelets, but also depleted the bone marrow of mature megakaryocytes expressing GPIIb/IIIa. This causes more severe and prolonged thrombocytopenia than is usually the case in patients sensitive to the platelet inhibitors eptifibatide or tirofiban.¹¹  Antibodies associated with this condition appear to recognize a restricted domain in the vicinity of the RGD recognition site of GPIIb/IIIa. Studies to determine whether other immunoglobulins of this type are selectively cytotoxic to megakaryocytes could be rewarding.