Abstract 2408

Poster Board II-385

Drug-induced immune thrombocytopenia (DITP) is a recognized complication of many different medications. Often, drug-dependent, platelet-reactive antibodies (DDAbs) can be detected in such patients but, not uncommonly, they appear to be absent. An important reason for this is that a metabolite of the primary drug can cause sensitization; in such cases, a DDAb can be detected only if the appropriate metabolite is used in testing. Since most drugs have multiple metabolites that are not readily available, serologic identification of a metabolite-specific DDAb can be a difficult task. Animal models have been of limited value for studying immune destruction of human blood cells because of their rapid destruction by xenoantibodies. We have shown that immune clearance of human platelets by human antibodies can be studied in the NOD/SCID mouse, which lacks immunoglobulins, including xenoantibodies (Thromb Haemost. 2007; 5 Suppl 1:305-9) and have now begun studies to determine whether this model can be used to identify metabolite-specific antibodies in patients with apparent DITP in whom a DDAb cannot be detected.

In preliminary studies, we showed that rapid clearance of circulating human platelets results when an animal previously infused with a DDAb specific for quinine or sulfamethoxazole is challenged by intraperitoneal (IP) injection of the appropriate drug. We then infused mice with platelets and a DDAb from a patient previously shown (Blood 109:3608, 2007) to be sensitive to acetaminophen glucuronide, a major metabolite of acetaminophen, and challenged the animal with acetaminophen. Platelet survival at 24 hours averaged 19% in animals given acetaminophen vs 80% in those given buffer alone (p =0.01). Sensitivity to the glucuronide metabolite of the NSAID naproxen can also cause severe DITP (Blood 97:3486, 2001). In a second study, we infused serum from a patient who experienced acute thrombocytopenia while taking naproxen but had no detectable naproxen-specific antibodies and challenged the animals with naproxen or buffer. Platelet survival at 24 hours averaged 26% in animals given naproxen and 111% in those given buffer only (p =0.01). Similar findings were made with serum from a second patient studied only once to date.

The findings show that NOD/scid mice produce metabolites of acetaminophen and naproxen for which human antibodies that cause DITP are specific and that these metabolites accumulate in the animal's circulation at levels high enough to promote platelet destruction by metabolite-specific DDAbs. Although drug metabolism can be species-specific, certain metabolic pathways, including those involved in glucuronide conjugation, are shared by humans and mice. Thus, the NOD/scid model may prove useful for documenting the presence of metabolite-specific antibodies in patients with cytopenias caused by sensitivity to drug metabolites and for studying other aspects of drug sensitivity.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.