Eltrombopag treatment of a dog with idiopathic aplastic pancytopenia

Abstract Idiopathic aplastic pancytopenia is an uncommon disease in dogs which results in pancytopenia and for which an immune‐mediated etiology is suspected. A small number of affected dogs reported in the veterinary literature have responded to immunosuppressive medication but the prognosis generally is considered poor with a reported mortality rate of 80%. Reported response rates to immunosuppression alone in affected people are low with overall and complete responses of 65 and 10%, respectively. With the addition of eltrombopag, an orally available thrombopoietin receptor agonist, reported overall and complete response rates in people increase to 94 and 58%, respectively. Herein, we report the use of eltrombopag in a dog with idiopathic aplastic pancytopenia. Eltrombopag was started after no response was seen to treatment with prednisolone and cyclosporine. Complete remission was achieved after the addition of eltrombopag and was sustained after stopping the medication.


| INTRODUCTION
Idiopathic aplastic pancytopenia is characterized by cytopenias in multiple cell lines with hypoplasia of the bone marrow and replacement with adipose tissue with or without fibrosis and for which an inciting cause cannot be identified. 1 Aplastic pancytopenia can be caused by infectious agents, drugs, and toxins but a diagnosis of idiopathic aplastic pancytopenia accounts for <1% of bone marrow samples evaluated in dogs. 1 In the absence of a known inciting agent, the term idiopathic is used but increasing evidence suggests an immunemediated pathogenesis. 1 In the veterinary literature, only a few cases of idiopathic aplastic pancytopenia have been reported, with mortality rates of 66%-80% and with most animals dying or being euthanized within 2-3 weeks of diagnosis. [1][2][3] Given the suspicion of an immunemediated etiology in dogs, trial treatment with immunosuppressive drugs has been recommended and a small number of reported cases have responded to immunosuppressive treatment. [1][2][3][4][5] In human patients, immunosuppression with cyclosporine is considered the cornerstone of treatment for aplastic pancytopenia, but overall and complete response rates are low at 65 and 10%, respectively. 6 In human patients with aplastic pancytopenia, the addition of the thrombopoietin receptor agonist eltrombopag to immunosuppressive treatment leads to improved outcomes with overall and complete response rates of 94 and 58%, respectively. 7

| CASE DESCRIPTION
A 1-year-old male neutered Border Collie was referred for investigation and treatment of pancytopenia. The dog was presented to the primary care veterinarian with a 2-day history of lethargy and anorexia. There was no known access to toxins and the dog had never traveled outside of the United Kingdom. The dog was found to be pyrexic and a CBC showed pancytopenia. At the time of presentation to the referral hospital, the dog was quiet and pyrexic (39.5 C). Petechial hemorrhages were visible on the ventral abdominal skin and the oral mucosa. Mucous membranes were slightly pale.

| DISCUSSION
Eltrombopag is an orally available thrombopoietin receptor agonist which was first approved for use in human patients with refractory idiopathic thrombocytopenia purpura. 8 After initiation of its use in the clinical setting, eltrombopag was found to have bi-lineage and trilineage bone marrow stimulatory effects and, in recent years, several studies have shown clinically relevant responses to eltrombopag in humans with aplastic pancytopenia. 6,7,9,10 The mechanism by which this thrombopoietin receptor agonist stimulates multi-lineage recovery is not completely understood but is thought to be related to stimulation and proliferation of primitive hematopoietic stem cells and other progenitor cells that may express the thrombopoietin receptor. 11 Aside from patients with refractory disease, eltrombopag has been used alongside immunosuppression in treatment-naive human patients with severe aplastic pancytopenia and has been shown to accelerate cell count recovery. One study described the use of eltrombopag for a set period of 10 weeks in treatment-naive human patients and found a response rate of 87% at 3 months after starting treatment and response rates were sustained at 6 months, 3 months after discontinuing eltrombopag. 6 The duration of eltrombopag treatment used in our case (2 months) was based on this study and, as also was found in this study of human patients, clinical remission in the dog reported here was sustained after discontinuing the eltrombopag.
The times to initial and complete hematological response in this dog were 30 and 44 days, respectively. This response is similar to a study in humans investigating the combined use of eltrombopag and cyclosporine in treatment-naive patients with pancytopenia, which found a median response time of 35 days. 6 Multiple studies have shown sustained responses when eltrombopag treatment is discontinued, which has led to the suggestion that starting eltrombopag early in treatment-naive patients may be warranted because of its proposed ability to salvage and cause expansion of residual hematopoietic stem cells. 6,10 Ours is the first report of the use of eltrombopag in a dog. Idiopathic aplastic pancytopenia in dogs generally carries a poor prognosis with a reported 21-day mortality rate of 80%. 1 The high mortality rate reported in dogs likely reflects poor response to immunosuppressive treatment, similar to the poor response seen when immunosuppressive treatment is used alone in human patients. The decision to start treatment with eltrombopag in this case was made when no response to immunosuppressive treatment was seen and given the relatively high response rates to eltrombopag reported in humans. In this case, the dog responded to treatment with immunosuppressive medication and eltrombopag with sustained remission after discontinuing the eltrombopag with 10 months follow-up. Aside from polyuria, polydipsia, and polyphagia, which resolved after discontinuation of the corticosteroids, no adverse effects were observed. The dosage used in this case was chosen based on the smallest available tablet size and the starting dosage used in children. The main limitation of this case report is that we cannot prove that the eltrombopag was responsible for the response seen, although we feel that this is likely the case. It is possible that the response seen was a response to the immunosuppressive medications in combination with supporting the dog for a sufficiently long period of time. We however feel that this is unlikely given that no response was seen initially when immunosuppressive medications were administered alone and because the time until response seen in this case after starting treatment with eltrombopag closely resembles the median response time reported in human patients. Eltrombopag may be useful for treating aplastic pancytopenia in dogs, but further studies are required to confirm or refute this hypothesis. Although the drug currently is available in many countries worldwide, its cost may make it prohibitive for use in some cases. If used for only a relatively short period of time (eg, 2 months in treatment-naive patients), the potential for eltrombopag to result in substantially higher sustained response rates when used along with immunosuppressive medications may offset the financial cost of the drug. The purpose of our report is to increase awareness of this drug with the aim of stimulating further investigation into its use in dogs with aplastic pancytopenia, a disease that when treated with the previously described medications (eg, immunosuppressive agents, granulocyte colony stimulating factor, erythropoietin), 12 typically carries a poor prognosis.

CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
Authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.