Cancer in patients with rheumatic diseases exposed to TNF antagonists

Objective To describe the risk of cancer in patients exposed to tumor necrosis factor (TNF) antagonists. Methods The following 2 clinical cohorts were studied: (1) BIOBADASER 2.0: a registry of patients suffering from rheumatic diseases exposed to TNF antagonists (2531 rheumatoid arthritis (RA), 1488 spondyloarthropathies, and 675 other rheumatic conditions); and (2) EMECAR: a cohort of 789 RA patients not exposed to TNF antagonists. Cancer incidence rates (IR) per 1000 patient-years and incidence rate ratios (IRR) were calculated for BIOBADASER 2.0 and EMECAR patients. The IR over time in BIOBADASER 2.0 patients was analyzed by joinpoint regression. The IRR was estimated to compare cancer rates in exposed versus nonexposed RA patients. Standardized incidence and mortality ratios (SIR, SMR) were also estimated. Risk factors for cancer in patients exposed to TNF antagonists were investigated by generalized linear models. Results The SMR for cancer in BIODASER 2.0 was 0.67 (95% CI: 0.51-0.86), and the SIR was 0.1 (95% CI 0.03-0.23). The IR in RA patients exposed to TNF antagonists was 5.8 (95% CI: 4.4-7.6), and the adjusted IRR was 0.48 (95% CI: 0.09-2.45). The IR in patients with previous cancer was 26.4 (95% CI: 4.1-171.5). Age, chronic obstructive pulmonary disease, and steroids were associated with a higher risk of developing cancer. The IR decreased after the first 4 months of exposure, without statistical significance. Conclusion Overall cancer and mortality rates in patients with rheumatic diseases exposed to TNF antagonists are no higher than in the background Spanish population. However special attention should be paid to elderly patients, those with previous cancers, and patients treated with steroids.