Short-term efficacy of biologics targeting the IL-17/IL-23 axis in scalp, nail, and palmoplantar psoriasis: a network meta-analysis of randomized controlled trials.

BACKGROUND: Scalp, nail, and palmoplantar psoriasis are termed "difficult-to-treat sites" owing to their unique anatomical features and therapeutic resistance, substantially impairing patient quality of life. Although anti-IL-17 and anti-IL-23 biologics are widely used, head-to-head comparative evidence for achieving high-level lesion clearance at these specific sites remains limited. METHODS: Following PRISMA-NMA guidelines, we systematically searched PubMed, Embase, and other databases from inception through November 2025 for randomized controlled trials (RCTs). Primary outcomes were defined as complete or near-complete clearance. Frequentist network meta-analysis was performed to calculate odds ratios (ORs), with treatment rankings derived from surface under the cumulative ranking curve (SUCRA) values. RESULTS: Twenty-four RCTs involving 5,946 patients and eight biologics plus placebo were included. For palmoplantar psoriasis, secukinumab ranked highest (SUCRA = 79.7%), followed by bimekizumab (72.2%) and ustekinumab (69.7%). For nail psoriasis, bimekizumab ranked first (78.9%), followed by ixekizumab (77.2%) and brodalumab (67.5%); however, local inconsistency for the ixekizumab-placebo comparison and low-certainty evidence warrant caution. For scalp psoriasis, brodalumab (87.7%) and ixekizumab (86.9%) ranked highest, followed by bimekizumab (69.0%) and guselkumab (65.1%); the brodalumab estimate relied on a single contributing study. CONCLUSION: During induction-phase follow-up, IL-17 inhibitors tended to rank highly for complete or near-complete clearance at difficult-to-treat psoriasis sites. Bimekizumab and ixekizumab showed consistently favorable rankings across sites, but treatment selection should consider certainty of evidence, sensitivity analyses, long-term response, and patient-level factors. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO, identifier CRD420251271255.