Dupilumab versus tralokinumab in atopic dermatitis: A propensity score adjusted comparison from BioDay

Dupilumab and tralokinumab for atopic dermatitis (AD) target the type 2 axis through different mechanisms of action, which may lead to variation in effectiveness and safety. Head-to-head trials, however, are lacking.

To compare the real-world effectiveness and safety of dupilumab and tralokinumab in AD.

This prospective cohort study enrolled biologic-/Janus kinase inhibitor-naïve AD patients (≥12 years) from the BioDay registry who initiated dupilumab or tralokinumab between November 2021 and September 2024. Visits were scheduled at baseline, 4 weeks and every 3 months up to 52 weeks. Effectiveness outcomes included Eczema Area and Severity Index (EASI), weekly mean pruritus Numeric Rating Scale (NRS), treat-to-target thresholds (EASI ≤ 7; NRS-pruritus ≤ 4, with patients discontinuing treatment considered non-responders) and drug survival. Adverse events (AEs) were assessed at each visit. Inverse probability of treatment weighting (IPTW) was used to balance treatment groups.

In total, 750 patients were included (643 dupilumab; 107 tralokinumab). After IPTW, baseline characteristics were well balanced. During follow-up, dupilumab patients had lower EASI scores than tralokinumab patients, although differences were not consistently statistically significant (p = 0.10). NRS-pruritus scores were significantly lower with dupilumab at all visits (p < 0.0001), mean differences did not exceed the 2-point clinical relevance threshold. The probability of achieving EASI ≤ 7 and NRS-pruritus ≤ 4 was higher with dupilumab (both p < 0.0001), with risk differences of 34.7% and 40.2% at 52 weeks, respectively. After 52 weeks, dupilumab drug survival was 92.6% vs. 70.6% for tralokinumab. Ocular surface disease incidence was similar (HR 1.0, 95% CI 0.6-1.6, p = 0.94) between treatments, leading to discontinuation of dupilumab in n = 23 (3.4/100 PY) and tralokinumab in n = 5 (5.4/100 PY).

In this real-world comparison, dupilumab provided superior effectiveness compared with tralokinumab. In responders continuing treatment, EASI and NRS-pruritus differences were small. More substantial differences were observed when treatment targets EASI ≤ 7 and NRS-pruritus ≤ 4, and discontinuation rates were taken into account.