The rate of infectious events that required hospitalization and/or parenteral antibiotic treatment was 2.3 per 100 patient-years of exposure.
This finding offers clinicians a number to consider when they compare serious infectious risks between methotrexate and other treatments, including biologic agents, Dr. Jennifer Powers said during a poster discussion session at the annual meeting of the American Academy of Dermatology.
Dr. Daniel E. Furst noted in an interview that the issue of infection in patients on methotrexate is one physicians often forget about, but it is significant because methotrexate’s infection risk should be a factor when physicians consider whether to initiate biologic therapy.
Unfortunately, the research does not address whether the infection risk is even higher in patients who receive both methotrexate and a biologic agent than in patients on one or the other as monotherapy. But the findings still are of interest, said Dr. Furst, who is the Carl M. Pearson Professor of Rheumatology at the University of California, Los Angeles.
In contrast, there was no statistically significant higher risk for a serious infectious event (SIE) associated with methotrexate in a meta-analysis of five psoriasis trials (2.2 serious infectious events per 100 patient-years) or a meta-analysis of five psoriatic arthritis studies (0.9 events per 100 patient-years).
The small number of eligible studies in psoriasis and psoriatic arthritis might explain the lack of statistical significance, said Dr. Powers, a first-year resident at Grand Rapids (Michigan) Medical Education and Research Center. It could also be that patients with rheumatoid arthritis generally are sicker. The Food and Drug Administration first approved methotrexate in 1953 and granted a new indication to treat rheumatoid arthritis in 1988. These approvals predate the era when the agency required more stringent reporting of infectious adverse events, Dr. Powers said.
She and her colleague, Dr. Richard W. Martin, conducted a literature search for randomized, placebo-controlled studies published from January 2005 through May 2009 in Embase Biomedical Answers, the National Library of Medicine’s Medline database, CINAHL (Cumulative Index to Nursing and Allied Health Literature), and the Cochrane Library. Patients in each study received oral methotrexate (7.5-30 mg/wk) for at least 12 weeks.
Previous observational studies reported conflicting results. “It was interesting to me when going over the literature [to find] how controversial infection with methotrexate is and how little consensus there is,” Dr. Powers said. “These are rare cases, but they are out there.”
“There are not great data about [SIEs] in methotrexate,” Dr. Powers said. “So there was a hole in the literature we wanted to fill.”
Randomized, controlled studies were included in the meta-analyses only if they met objective quality criteria. For example, participants had to be adults with clearly defined disease, and they could not be taking more than 10-mg prednisone. Also, studies were excluded if 20% or more of patients were lost to follow-up.
Biologic agents have transformed psoriasis treatment, but they are associated with increased serious infectious events and significant expense, which revives interest in the comparative efficacy of standard vs. new therapies, the authors wrote.
“This is really an important counterpoint when we are discussing SIE risks for methotrexate. ... It is generally accepted that methotrexate has lower risk of SIE than biologics,” Dr. Powers said. For example, the product labeling for ustekinumab (Stelara) notes that serious infections have occurred. In addition, the risk of serious infections that can lead to hospitalization and death are included on black box warnings for etanercept (Enbrel), infliximab (Remicade), and adalimumab (Humira).
A meeting attendee said the findings would be more robust if the methotrexate trials in the meta-analyses went beyond 12 weeks. Dr. Powers agreed, adding that these results, “are definitely limited by the fact that there are trials out there that do not follow patients for a longer time.”
None of the investigators had any relevant financial disclosures.