Risk of first and second primary keratinocyte cancers in relation to treatment of rheumatoid arthritis with JAKi, TNFi, and non-TNFi bDMARDs—a Swedish nationwide study

ABSTRACTObjectives This study aims to assess the risks of primary and second primary keratinocyte cancers (KCs) in patients with rheumatoid arthritis (RA) and in relation to treatment with biologic disease-modifying antirheumatic drugs (bDMARDs) or targeted synthetic disease-modifying antirheumatic drugs. Methods Nationwide cohort study of patients treated with Janus kinase inhibitors (JAKi), tumour necrosis factor inhibitor (TNFi), or non-TNFi bDMARDs, using data from the Swedish Rheumatology Quality Register linked to other registers including the National Cancer Register, 2012 through 2023. Adjusted hazard ratios (HRs) were estimated via Cox regression using TNFi as reference. Results We identified 21,756 unique patients with RA. Based on 155 incident KC with JAKi, 458 with non-TNFi and 766 with TNFi, the HR for JAKi vs TNFi was 1.39 (1.16-1.68), corresponding to 1 extra KC case per every 244 patients per year. For non-TNFi vs TNFi, the HR was 0.96 (0.86-1.08). By subtype, the HR for JAKi vs TNFi was 1.41 (1.13-1.75) for basal cell carcinoma and 1.49 (1.09-2.05) for squamous cell carcinoma (SCC). For abatacept vs etanercept, the HR for SCC was 1.48 (1.11-1.97). The HR for a second primary KC was 1.31 (0.94-1.82) for JAKi and 0.94 (0.75-1.17) for non-TNFi bDMARD vs TNFi. Conclusions Patients treated with JAKi have an elevated risk of KC compared with patients treated with TNFi. Although the class of non-TNFi bDMARDs is not associated with increased KC risk, we repeated a drug-specific signal of increased risk for SCC with abatacept.