A new study on the adverse effects caused by nine biologics used to treat rheumatoid arthritis and other conditions will help clinicians make more informed choices.

"The data provides some guidance regarding the safety of different biologic drugs, but these are not head-to-head trials," says study author Jasvinder Singh, MD, MPH, associate professor in the division of immunology and rheumatology at UAB.

"The novelty of the study is the approach, because we combined all the trials across all conditions," Singh says. The data covered 50,010 patients in 163 randomized trials for any use, including treating rheumatoid arthritis, malignancy, neurological disease and any disease except HIVH. Though they found only one study on HIVH, it was excluded because of the unique autoimmune status caused by the disease itself.

The trials went as far back as 1965 and up to January 2010. "Biologics have only become available recently," Singh says. "But as with all Cochrane systematic reviews, a sensitive search strategy was used to find all the studies for these medications."

Using a trained full-time Cochran librarian for the searches, the review pulled from randomized controlled trials (RCT) and controlled clinical trials (CCT) that reported on any adverse events, including the seven main adverse events cited in the review. Duration of trials lasted anywhere from 6 months to 2 years.

"Some outcomes were pretty much what we expected," Singh says. In three of the adverse event options, the biologics were associated with significantly higher odds than control treatment: total adverse events, withdrawals due to adverse events, and TB reactivation.

In the other four adverse events, however, biologics were not statistically significantly different from control treatment. Those included the rate of serious infections, lymphoma, congestive heart failure, and serious adverse events, such as death and hospitalization over 24 hours.

"A lack of difference doesn't mean that a difference doesn't exist," Singh says. "It just means we don't have enough patients and adverse events to better distinguish their risk in comparison to a control."

"We realize that all the trials are not the same," Singh says. "Patients in one study for a given drug may be different than in the study for the next drug, but that is why we compared them each to their own control group. The ideal would be a blind randomized study of all nine biologics, but that is not likely to happen."

The study's Indirect comparisons, however, revealed that abatacept and anakinra were associated with a significantly lower risk of serious adverse events compared to most other biologics. Adalimumab and infliximab, however, were associated with more adverse events.

For serious infections, abatacept rated a significantly lower chance than infliximab and tocilizumab. Certolizumab pegol was associated with significantly higher odds compared to etanercept, adalimumab, abatacept, anakinra, golimumab, infliximab, and rituximab. However, the study warns that the overall numbers were relatively small.

For withdrawals due to adverse events, Infliximab showed a significantly higher risk, and abatacept, adalimumab, etanercept and golimumab were significantly less likely than infliximab to cause this adverse event.

Singh was surprised to find so many differences in the head-to-head indirect comparisons. "Some findings were a little surprising, but we know the biologics have different mechanics of action, so we expected the adverse profile to be different. But we cannot say the differences are related to the mechanism of action, the trials, or in the patient populations. That is a limitation of these comparisons," he says.

The researchers adjusted for dosage and other variances, including older drugs having had more studies completed. "But some studies for more recent drugs have been larger than older studies on drugs that came before. But I won't say that the number is then equal," Singh says.

"We have to do statistical maneuvering, but we're very aware that we can't fix all the differences between the trials," he says. "So we do not try to over interpret those results."

Singh advises that physicians use these results as a guide when starting biologics. If the biggest concern with the patient is infection, then the results showing abatacept with a lower associated serious infection rate — compared to control or indirectly to other biologics — might help make a more informed choice.

"We know the limitations of these comparisons, and we don't pretend that all the trials are the same," Singh says. "But physicians are having to make these comparisons when choosing among these nine drugs, and we can give you more information on which to base that decision."