Selective IL-23 Inhibition in Conventional Treatment-Refractory Pyoderma Gangrenosum: A Multicenter, Retrospective Study

Background: The therapeutic management of pyoderma gangrenosum (PG) is challenging. Conventional therapies, including systemic corticosteroids and cyclosporine, are limited by incomplete responses, frequent relapses, and cumulative toxicity. Emerging evidence supports a pathogenic role of the interleukin (IL)-23/IL-17 axis in PG, suggesting selective IL-23 inhibition as an appealing therapeutic approach. This study assessed the effectiveness, safety, steroid-sparing potential, and predictors of response to selective IL-23 inhibitors in conventional treatment-refractory PG. Methods: This multicenter retrospective study included adult patients with refractory PG or intolerant to conventional systemic therapies, treated with selective IL-23 inhibitors (guselkumab, risankizumab, or tildrakizumab). Clinical assessments were performed at baseline and during follow-up (1, 3, 6, 12 months, and last observation), evaluating ulcer area, number of lesions, ulcer depth, border wound bed characteristics, Investigator's Global Assessment for PG (IGAPg), and pain intensity assessed by the Numerical Rating Scale (NRS). Longitudinal changes were analyzed using non-parametric tests for repeated measures. Results: Selective IL-23 inhibition in 18 patients was associated with a progressive reduction in total ulcer area, significant from 1 month onward (p = 0.0069; all subsequent p ≤ 0.0003). The number of active ulcers did not change at 1 month (p = 0.41) but significantly decreased from Month 3 onward (p = 0.004 to < 0.001). Ulcer depth remained unchanged at 1 month (p = 0.41) and significantly improved from Month 3 onward (p = 0.01–0.046). Border/perilesional inflammation and wound bed characteristics showed significant improvement at all follow-up visits (p = 0.0008–0.034 and p = 0.013–0.046, respectively). IGAPg score showed a significant and progressive improvement over time (all p ≤ 0.027). Pain intensity showed a significant reduction during follow-up (p = 0.004 to < 0.001). A significant steroid-sparing effect was observed, with a progressive reduction in median daily prednisone-equivalent dose during follow-up (p = 0.002–0.014). Guselkumab and risankizumab showed comparable efficacy (p = 0.182). In univariate analysis, baseline ulcer depth was the only predictor of ulcer area reduction (p = 0.039). No unexpected safety signals were observed. Conclusions: Selective IL-23 inhibition appears to be a well-tolerated therapeutic option for refractory PG, providing meaningful clinical improvement and a relevant steroid-sparing effect. These findings support further controlled studies to define the role of IL-23 inhibitors within PG treatment algorithms.