Many With JIA Can Stop TNF Inhibitors

Janis C. Kelly

July 26, 2011

July 26, 2011 — A retrospective chart review study of 171 patients with juvenile idiopathic arthritis (JIA) treated with tumor necrosis factor (TNF) blockers found that half maintained inactive disease for 6 months after discontinuation and 33% were still in clinical remission at 12 months after discontinuation. Lead author Kevin Baszis, MD, from Washington University School of Medicine in St. Louis, Missouri, told Medscape Medical News that "it is important to remember that a proportion of these patients remained on [disease-modifying antirheumatic drugs], such as methotrexate."

His group's paper was published online June 23 in Arthritis & Rheumatism.

"I think our findings support the idea that more understanding is needed into the biological differences between patients with JIA. For instance, if we could determine, before stopping TNF inhibitors, which subset of patients would fall into the 33% that remain in clinical remission, this would be ideal. This may involve checking cytokines and other biologic markers of disease activity," Dr. Baszis said.

The study goals were to estimate the length of time to flare and the likelihood of clinical remission after discontinuation of TNF blockers. Median patient observation was 59.7 months.

The researchers found no significant association between time to flare after stopping TNF blockers and time from diagnosis to beginning of TNF blockers, duration of therapy after the onset of inactive disease, and total duration of TNF blockers prior to discontinuation.

Dr. Baszis said that "for clinicians, I think these findings show that it is worthwhile to try stopping anti-TNF therapy in patients with JIA who have had inactive disease for some time (median, 6 months in our study). Although the majority (67%) still will flare within 1 year, if we are able to spare one third of children the cost and side effects of treatment for 1 year, that is beneficial."

Michael Henrickson, MD, MPH, clinical director of the division of rheumatology at Cincinnati Children's Hospital Medical Center in Ohio, is not sure the data support this approach. Dr. Henrickson, who reviewed the study for Medscape Medical News, said: "This article has a number of epidemiologic design problems that weaken its conclusions and call into question whether it has validity."

Dr. Henrickson said that "Baszis et al use a retrospective (observational) study, reviewing 12 years of medical records at a single center. This study design is rapidly falling out of favor because of the numerous data collection and potential biases that exist." Dr. Henrickson explained that biases are insufficiently discussed in the paper and should have included incidence–prevalence bias, self-selection bias, nonresponder bias, and observer bias.

"Retrospective studies, done well, are essentially embedded ("nested") case–control studies in a cohort study. This infers that internal comparisons are made to controls. However, Baszis' study has no specific controls," Dr. Henrickson pointed out. He also questioned the use of the hazard ratio for statistical analysis, because it assumes a constant proportion of hazard in comparison groups, and this study has 6 different subgroups.

"Conclusions made in a retrospective study can only suggest exploring ideas (hypotheses). This type of study design generates hypotheses, but it should not make any conclusions. I feel the paper reaches too far, without outlining its very specific limitations. Given all these potential errors, I am not convinced by the paper's data or conclusions," Dr. Henrickson said. "The issues overlooked are the numerous potential biases, flaws in the study design, lack of controls, and the mixed use of prevalent and incident cases."

Dr. Henrickson also drew a different conclusion from the study: "The most credible finding, not emphasized by the authors, is that biologics, specifically TNF inhibitors, are not disease-modifying drugs. If 50% of patients previously treated with TNF inhibitors flared by 6 months, it is reasonable to conclude these medications do not offer an appreciable benefit in reducing risk of overall disease activity or recurrence," he said.

Dr. Baszis and Dr. Henrickson have disclosed no relevant financial relationships.

Arthritis Rheum. Published online June 23, 2011. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: