Abstract
OBJECTIVE: To investigate the risk of cancer recurrence in patients with rheumatoid arthritis (RA) and a prior solid cancer in remission treated with biological disease-modifying antirheumatic drugs (bDMARDs) compared with those who received only conventional synthetic DMARDs (csDMARDs). METHODS: Nationwide registry-based cohort study of Danish patients with RA and one of the six algorithm-specific solid cancers in remission (breast, colorectal, melanoma, bladder, endometrial and lung) who initiated treatment with a bDMARD or a csDMARD. Three bDMARD exposure groups were defined according to the type(s) of bDMARD initiated: (1) any bDMARD, (2) tumour necrosis factor inhibitors (TNFi) and (3) rituximab. Patients were identified in Danish Rheumatology Quality Register and followed for cancer recurrence from 2002 to 2021 using validated cancer-specific recurrence algorithms. Inverse probability of treatment weighting and weighted cause-specific Cox models were used to calculate HRs for cancer recurrence in each bDMARD exposure group compared with a csDMARD-treated group. RESULTS: Among 720 unique patients with RA and an algorithm-specific solid cancer in remission, 170 any bDMARD, 81 TNFi, 99 rituximab and 651 csDMARD initiators were identified. No statistically significant increased HRs for cancer recurrence were found with any type of bDMARD 0.92 (95% CI 0.38 to 1.73), TNFi 1.10 (95% CI 0.21 to 3.16) or rituximab 0.94 (95% CI 0.32 to 2.11). Also, no increased HRs were shown for breast cancer recurrence specifically. CONCLUSION: No indications of increased cancer recurrence risk were found for bDMARDs as used in clinical practice in patients with RA and a solid cancer in remission when compared with csDMARD treatment.