Hotline

Etanercept Update on Recent "Dear Doctor" Safety Letter

Etanercept (soluble TNF p75 receptor fusion protein) was FDA approved for use in rheumatoid arthritis (RA) in Nov. 1998 and has since been given to over 80,000 patients, with an estimated drug exposure of over 60,000 patient?years. The manufacturer recently announced changes to their package insert based on post?marketing surveillance that has shown sporadic reports of hematologic and neurologic side effects in etanercept treated patients. The following summary data were available at the time of the recent "Dear Doctor" letter to the medical community.

Hematologic effects. There have been 10 reported cases of pancytopenia, including 3 cases of aplastic anemia and five of these resulted in death. The mean age was 56.4 years (range 37?78 years) six over the age of 60 years. Several possible explanations were observed. Two patients had a previous history of cytopenias; 4 patients died with cytopenias during serious infections; 9 patients were receiving potentially myelotoxic drugs (i.e., azathioprine, 6?mercaptopurine, cyclophosphamide, methotrexate, leflunomide, lisinopril); and one patient developed pancytopenia while receiving etanercept and cephalexin. Overall, 5 patients had multiple medical problems, 3 patients had extraarticular manifestations and 6 patients were receiving prednisone and combination therapies at the time of cytopenic event. Cytopenias developed 2 ?15 weeks (median 4 weeks) after initiating etanercept and two reporting physicians felt the cytopenia was due to etanercept. There are no historic controls to compare these findings and the rate of aplastic anemia or pancytopenia in RA is unknown. International and French Aplastic Anemia Study Groups have concluded the relative risk of aplastic anemia in RA to be 6.8-7.6.1 The incidence rate of aplastic anemia in the general population varies between 2.2 and 8.5 cases per million/yr. In this etanercept-treated RA population the prevalence rate of cytopenias is at least 0.0125%. It is unknown if this rate was influenced by the therapeutic inhibition of TNF in RA. TNF is not known to be involved in normal hematopoesis. Monitoring of blood cell counts are not recommended by the manufacturer, but patients with features of blood dyscrasias (e.g., fever, pallor, bleeding, sore throat) should be evaluated and withholding therapy should be considered.

Demyelinating Disorders. Eleven cases with neurologic disturbances have also been reported. These cases were either de novo cases (2 patients) of multiple sclerosis (MS), exacerbation of pre-existing MS (4 patients), 2 cases each with encephalopathy or myelitis and two with optic neuritis and one with ischemic optic neuropathy. The expected rate of MS in the RA population is unknown. The association is felt to be exceedingly uncommon and far less common than the occasional association of MS with lupus or Sjögren's syndrome. The prevalence is estimated to be between 250,000 and 350,000 or 93?130 cases per 100,000 population. By extrapolation, between 72-104 cases of MS should have occurred among etanercept-treated RA patients. Patients reported with demyelinating disorders differ from those with cytopenias (above) in that the patients were younger (mean 44.2 years; range 37?56 years), were more likely to have psoriatic arthritis or seronegative or poorly defined RA, and were less likely to have multiple comorbidities or be treated with combination therapy (4/11 patients were on MTX). Although TNF has been implicated a possible factor in the pathogenesis of demyelinating disease in humans (MS) and animal models (EAE), the medical literature also supports the plausibility of a TNF inhibitor causing these findings.2,3 One report of 2 MS patients who were treated with infliximab failed to show any clinical changes but was associated with worsening of MRI changes over a 6 month period. Another study of lenercept (a p55 TNF receptor) in MS patients failed to show improvement and instead showed increase rates of MS flares when compared to placebo. Thus, it appears that demyelinating disorders may be drug class (TNF inhibitors) effect rather than a drug specific (etanercept) effect. Rheumatologists should therefore exercise caution in prescribing a TNF inhibitor to patients with MS or a history of optic neuritis or other demylenating disorders.

November 2000
John J. Cush, MD; Robert Spiera, MD
Hotline Editors

1. Blood 81 (6): 1471-78, 1993
2. Neurology 47: 1531-34, 1996
3. Neurology 53: 457-65, 1999

Hotline is provided by the ACR Communications and Marketing Committee as a service to members. This Hotline reflects the views of the author(s) and does not represent a position statement of the College.