Researchers at the School of Medicine and the Veterans Affairs (VA) Medical Center have analyzed a national database of VA patients to investigate the effects of the rheumatoid arthritis drug leflunomide in the first years after its approval.
The study reveals that high doses of leflunomide may cause adverse side effects that lead some patients to stop taking the drug.
The Food and Drug Administration (FDA) approved the use of leflunomide for rheumatoid arthritis in October 1998 and recommended starting treatment with high doses to rapidly trigger the drug’s benefits.
But the findings, which appear in a recent issue of the journal Arthritis Care & Research, may prompt physicians to consider slightly modifying the standard leflunomide treatment regimen.
“Our data suggest that by starting on a lower dose initially, patients tolerate the drug better,” said Seth A. Eisen, M.D., professor of medicine and of psychiatry and a VA staff physician. “The disadvantage is that it may take a little bit longer for patients to improve clinically; the advantage is that patients may be more likely to continue treatment.”
Eisen and Chuck Siva, M.D., a rheumatology fellow, were principal investigators of the study, which was conducted in collaboration with several other VA medical centers across the country.
Rheumatoid arthritis, one of the most crippling forms of arthritis, afflicts about 2 million Americans with joint pain and inflammation. It is a chronic condition linked to immune system malfunctions. Leflunomide is one of about a dozen drugs available for treatment.
In addition to suggesting an alternative approach to initial treatment, the new findings also help put to rest lingering concerns about whether leflunomide’s toxicity was adequately assessed in phase III FDA clinical trials that ended in 1998.
“As far as we could tell, there were no surprises in terms of toxicity,” Eisen said. “Sometimes it takes a lot more patients than the 3,300 we studied to pick up rare adverse outcomes, but I think our findings are reassuring to the larger community of patients and clinicians.”
Eisen emphasized that because the study is a review of already existent data, no control and intervention groups could be established and assessment of side effects could not be standardized.
“In a subset of the patients for whom the medication was discontinued, we examined their medical records in an attempt to determine the reason the medication was stopped,” Eisen said. “We were particularly looking for evidence of severe toxicity and did not find any. But still, it’s a matter of trying to interpret clinical notes.”
Although researchers were aware of the disadvantages of this type of study, there also are a number of potential advantages. Only relatively small investments of money and time were required, and the patient population at the VA offered a chance to study portions of the general population that typically do not enroll in clinical trials.
Eisen explained that clinical trials often enroll more women and younger patients from higher socio-economic groups. The sex bias is particularly prevalent in clinical trials for treatments for rheumatological diseases, which afflict women more often than men.
“VA medical data complements information obtained from other sources because VA patients are predominantly male and older and include a higher proportion of African-Americans and individuals from lower socio-economic groups than general studies do,” Eisen said.
Eisen suggested that researchers should consider conducting studies like this on other drugs already approved for clinical use.
“Post-approval follow-up is very important because it may demonstrate problems, sometimes decades later, that weren’t appreciated,” Eisen said, citing the example of premarin, a treatment for post-menopausal symptoms. Risks recently linked to the drug were identified several decades after it was approved for clinical use.
“Because post-approval studies tend to require large numbers of people in order to evaluate the less-common adverse outcomes, they are typically very, very expensive to do,” Eisen said. “Studying large databases like the VA’s is relatively inexpensive, and it is very feasible to collect and analyze important information.”