Meta-analysis of anti-TNF agents in RA treatment trials suggests increased risk of infection, malignancies

A meta-analysis of 9 randomized, placebo-controlled studies of rheumatoid arthritis (RA) patients treated with the anti-tumor necrosis factor (anti-TNF) agents infliximab and adalimumab suggests an increased risk of malignancies dependent on dose and an increased risk of serious infections.

A meta-analysis of 9 randomized, placebo-controlled studies of rheumatoid arthritis (RA) patients treated with the anti-tumor necrosis factor (anti-TNF) agents infliximab and adalimumab suggests an increased risk of malignancies dependent on dose and an increased risk of serious infections.

"Our analysis contributed to the findings that challenge the previously presumed safety profile of anti-TNF therapy," stated the authors of the study, which was published in the Journal of the American Medical Association (JAMA). "The detected increase of 2 serious adverse events [malignancies and serious infections] has to be interpreted in the light of the high effectiveness of anti-TNF therapy in patients with RA and the lack of therapeutic alternatives in cases with high disease activity irresponsive to traditional DMARD therapy."

The researchers extracted randomized, placebo-controlled studies of at least 12 weeks in duration from the Cochrane Library, EMBASE, and MEDLINE from inception through December 2005, and searched electronic abstracts from the European League Against Rheumatism and the American College of Rheumatology dating to 1996.

Researchers identified documented adverse events, including serious infections (defined as requiring antimicrobial therapy or hospitalization) or malignancy. To detect potential differences in higher doses of anti-TNF therapies, pooled estimates also were evaluated, and patients were divided into high-dose (≥6 mg/kg of infliximab every 8 weeks or 40 mg of adalimumab every other week) and low-dose (≤3 mg/kg of infliximab every 4 weeks, or 20 mg of adalimumab weekly) groups.

Sensitivity analyses were used to exclude trials with a moderate or high risk of bias. Malignancies reported within 6 weeks of the beginning of the study and nonmelanoma skin malignancies also were omitted using sensitivity analyses. Of 144 potentially relevant publications, 9 were included in the meta-analysis.

Overall, 5,014 patients received anti-TNF therapy or control treatment. Published data reported 24 malignancies in 3,493 patients who received ≥1 dose of anti-TNF therapy (0.8%) and 2 malignancies in 1,512 controls (0.2%). Safety data reported to FDA from these trials included 37 malignancies in treatment groups and 3 discrepancies. The final numbers used in the meta-analysis included 29 malignancies in the treatment group and 3 malignancies in the placebo group.

Compared with patients receiving placebo, patients who received anti-TNF therapies had a higher risk of developing malignancies (pooled OR=3.3; 95% CI, 1.2–9.1). Malignancy estimates remained statistically significant when all statistical tests were applied, and patients treated with higher doses of anti-TNF therapies had a increased risk of developing malignancies compared with those treated with lower doses (OR=3.4; 95% CI, 1.4–8.2). For patients treated with anti-TNF therapies, the number needed to harm was 154 (95% CI, 91–500) for 1 additional malignancy within a 6- to 12-month treatment period.

Serious infections were reported in 126 of the anti-TNF-treated patients versus 26 control patients. Two additional infections were reported to FDA and were included in this analysis. The risk of serious infection in anti-TNF-treated RA patients greater than in patients receiving placebo (pooled OR=2.0; 95% CI, 1.3–3.1). Estimates remained statistically significant when different statistical tests were applied. Stratified analysis according to dose did not show statistical significance (P=.07). For serious infections, the number needed to harm was 59 (95% CI, 39–125) within a 3- to 12-month treatment period.

Assessment of validity revealed potential bias sources in the studies, including many trials with reported higher numbers of withdrawals (placebo ineffectiveness), more infusion-site reactions or allergic skin reactions leading to partial unmasking of study treatments, and lack of reporting on patients lost to follow-up.

Study limitations included: the statistical tests being based on limited data; wide confidence intervals; the need for inferences about certain patient subgroups to be cautiously made because of the mixed population; and study estimates being made for trials with varying durations. The studies analyzed also had many confounding variables and high numbers of patient drop-outs. The definition of serious infections also was broad.

The authors suggested that screening for subclinical malignancies in patients being evaluated for anti-TNF therapy, followed by surveillance of those patients, may represent future strategies to improve the safety of anti-TNF therapy treatment. However, these proposed approaches have not yet been fully evaluated.

SOURCE Bongartz T, Sutton AJ, Sweeting MJ. Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies. JAMA. 2006;295:2275–2285.