BACKGROUND: Topical tacrolimus (TAC) is indicated for the treatment of moderate to severe atopic dermatitis (AD), and topical pimecrolimus (PIM) for the treatment of mild to moderate AD. Data on the risk of lymphoma associated with use of these medications are inconclusive.
OBJECTIVES: To estimate the incidence rate ratio (IRR) of lymphoma, including cutaneous T-cell lymphoma (CTCL), in children and adults comparing new users of TAC and PIM with users of moderate- to high-potency topical corticosteroids (TCS), and users of TCS with general population untreated subjects.
METHODS: Cohort study in the PHARMO Database Network (Netherlands), the Danish and Swedish national registers, and the Clinical Practice Research Datalink (United Kingdom), with RTI-HS acting as coordinating/pooled analysis center. New users of TAC and PIM were frequency matched to users of TCS on twentiles of propensity scores; users of TCS were individually matched to untreated subjects on age, sex, region, and calendar year. We estimated IRRs and 95% confidence intervals (CI) using Mantel-Haenszel methods.
RESULTS: We included (a) 19,948 children and 66,127 adults treated with TAC matched with 79,700 children and 264,482 adults treated with TCS; (b) 23,840 children and 37,417 adults treated with PIM matched with 90,268 children and 149,671 adults treated with TCS; and (c) 79,040 children and 257,074 adults untreated with any study medication. For use of TAC vs. TCS, the adjusted IRR (95% CI) for any lymphoma was 3.74 (1.0014.06) in children and 1.27 (0.941.71) in adults; by lymphoma type, the highest IRR was 1.76 (0.813.79) for CTCL in adults. For PIM vs. TCS, the IRR for any lymphoma was 1.07 (0.254.60) in children and 1.03 (0.711.51) in adults; by lymphoma type, the highest IRR was 1.31 (0.335.14) for CTCL in adults. The IRR for adults treated with TCS vs. untreated subjects was 1.49 (1.191.87) for any lymphoma and 10.66 (2.6043.75) for CTCL.
CONCLUSIONS: These results suggest an association between TAC and PIM and the risk of lymphoma. Residual confounding by AD severity, reverse causation (particularly for CTCL), and surveillance bias cannot be ruled out.