Tralokinumab versus Dupilumab for the Treatment of Moderate-to-Severe Atopic Dermatitis with Hand Involvement: a Matching-Adjusted Indirect Comparison

Introduction Tralokinumab is a high-affinity interleukin (IL)-13 inhibitor while dupilumab targets the IL-4/IL-13 pathway via inhibition of the IL-4 receptor α. Both are approved for the treatment of moderate-to-severe atopic dermatitis (AD), including on the hands where AD often presents an additional burden and treatment challenges. As no head-to-head studies have been conducted, the 16-week efficacy of tralokinumab and dupilumab were compared in patients with AD with hand involvement in a matching-adjusted indirect comparison (MAIC) based on two phase 3 trials, ADHAND and LIBERTY-AD-HAFT. Methods In the ADHAND trial (NCT05958407), adults with moderate-to-severe AD with hand involvement were randomized 2:1 to double-blind treatment with tralokinumab 300 mg or placebo every 2 weeks for 16 weeks. In LIBERTY-AD-HAFT (NCT04417894), adults and adolescents (≥12 years) with moderate-to-severe AD with hand and/or foot involvement were randomized 1:1 to dupilumab 300 mg (200 mg in adolescents with body weight <60 kg) or placebo every 2 weeks for 16 weeks. An anchored MAIC was conducted using individual patient data (IPD) from ADHAND and aggregate published data from LIBERTY-AD-HAFT, with placebo as the common anchor. IPD from ADHAND were weighted to match age, sex, race, and baseline Hand Eczema Severity Index (HECSI) score in LIBERTY-AD-HAFT. Endpoints compared at week 16 were the proportions of patients with Investigator’s Global Assessment (IGA) scores of 0/1 (IGA-Atopic Hand Eczema in ADHAND and Hand and Foot-IGA in LIBERTY-AD-HAFT), HECSI-75 and HECSI-90, percentage reduction in HECSI and ability to perform regular daily activities other than work using the Percent Routine Activity Impairment domain of the Work Productivity and Activity Index. Reductions in itch and pain (assessed with 11-point numerical rating scales) were also compared, as were proportions of patients with ≥4-point improvement in itch. There were minor differences between trials in the scales used for IGA, itch and pain. Results LIBERTY-AD-HAFT included 133 patients (dupilumab, n=67, placebo, n=66) while ADHAND included 235 patients (tralokinumab, n=156; placebo, n=79). The effective sample size after matching was 99 (tralokinumab, n=76, placebo, n=23). Anchor-adjusted relative treatment differences significantly favored tralokinumab compared with dupilumab for the proportions of patients achieving HECSI-90 (17.5 [95% CI: 0.8, 34.2]; p=0.040) and decrease in Percent Routine Activity Impairment (16.0 [95% CI: 4.8, 27.1]; p=0.005). Relative differences for other endpoints favored tralokinumab over dupilumab but were not significantly different between treatments: proportions of patients achieving IGA score 0/1 (8.4 [95% CI: −11.3, 28.1]; p=0.403) or HECSI-75 (10.7 [95% CI: −13.3, 34.7]; p=0.383), percent reduction in HECSI (20.1 [95% CI: 41.3, −1.2]; p=0.064), percent change in itch (9.8 [95% CI: −7.7, 27.4]; p=0.273), change in pain (0.8 [95% CI: −0.4, 1.9]; p=0.213) and proportions of patients with ≥4-point improvement in itch (2.4 [95% CI: −18.9, 23.6]; p=0.827). Conclusion In this MAIC, tralokinumab resulted in significantly more patients achieving HECSI-90, indicating a deeper response, and a significantly greater improvement in ability to perform regular daily activities versus dupilumab at 16 weeks. Other endpoints were numerically in favor of tralokinumab.