IL-12/IL-23 Blockade in Psoriasis Outcomes

Introduction

What is the relationship between IL-12/IL-23 blockade and clinical outcomes in plaque psoriasis and psoriatic arthritis?

Evidence Summary

The evidence base here is substantial and spans nearly two decades of RCTs, meta-analyses, and real-world data. The story differs meaningfully between the two indications, so it's worth taking them in turn.

Plaque psoriasis: strong, consistent evidence across multiple agents

The p40-targeting antibody ustekinumab established proof of concept in the PHOENIX 1 and PHOENIX 2 trials, reporting PASI 75 responses of 66–76% at week 12 and 85% by week 24, versus 3–4% for placebo. The ACCEPT trial then showed ustekinumab outperformed high-dose etanercept — PASI 75 of 67.5% and 73.8% (45mg and 90mg, respectively) versus 56.8% for etanercept. Responses were maintained at 148 weeks in open-label extensions, which is a clinically meaningful durability signal.

The selective p19 inhibitors (guselkumab, risankizumab, tildrakizumab), which block IL-23 without affecting IL-12, then raised the bar considerably. A meta-analysis of 13 RCTs pooling 5,155 patients found PASI 75 response rates roughly 11-fold higher than placebo, and head-to-head data showed these agents outperformed both adalimumab and ustekinumab. Among p19 inhibitors, risankizumab consistently ranked highest in PASI 90 scores in network analyses, while guselkumab showed the strongest PASI 90 response as primary endpoint in its phase 3 trials (above 70% at week 16). Tildrakizumab achieves PASI 75 of 62–74% at 200mg but lags the other two on deeper responses. A non-clinical comparative study found risankizumab and guselkumab have 5-fold higher IL-23 binding affinity than tildrakizumab and ustekinumab, with preclinical data suggesting this translates to clinical rank order.

Safety across all IL-23 inhibitors in psoriasis is reassuring. Phase 3 and long-term extension data show no increased risk of serious infections, malignancies, cardiovascular events, opportunistic infections, inflammatory bowel disease, or TB reactivation — unlike IL-17 inhibitors, which carry candidiasis and IBD signals. The most common adverse events are upper respiratory tract infections and nasopharyngitis. A 2024 review found selective IL-23 inhibitors had the most favorable long-term risk-benefit profile among all psoriasis biologics.

Psoriatic arthritis: effective for joint outcomes but with important caveats

Ustekinumab's PsA evidence is the most robust. PSUMMIT 1 (615 patients, anti-TNF naive, published in The Lancet) found ACR20 at week 24 in 42.4% (45mg) and 49.5% (90mg) versus 22.8% for placebo, with sustained responses at 52 weeks including improvements in enthesitis, dactylitis, and HAQ-DI. PSUMMIT 2 extended this to anti-TNF-experienced patients, showing similar ACR20 advantages (44% combined vs 20% placebo), with reduced efficacy in patients who had failed two or more prior anti-TNF agents — ACR20 of 13–39% versus 59–73% in anti-TNF naive patients. Integrated radiographic data from both PSUMMIT trials showed ustekinumab also slows structural joint damage.

For the selective p19 inhibitors in PsA, a 2025 meta-analysis (6 RCTs, primarily guselkumab, risankizumab, and tildrakizumab) found significant improvements across all joint and skin outcomes: ACR20 RR of 1.86, ACR70 RR of 3.06, PASI90 RR of 5.98, and meaningful improvements in MDA, enthesitis, and dactylitis. A separate 2023 meta-analysis of 6 RCTs (2,971 patients) confirmed ACR20 RR of 1.74 with a favorable safety profile; guselkumab showed the strongest ACR20 association in another systematic review.

The most mechanistically interesting finding comes from a Annals of the Rheumatic Diseases tissue-profiling study: synovial IL-23 expression in PsA is highly heterogeneous — unlike skin, where it is uniformly high. Skin responses consistently favor ustekinumab over anti-TNF, while joint responses in that study favored anti-TNF. IL-23 transcriptomic expression in the synovia correlated with high-grade synovitis that wasn't distinguishable by conventional clinical measures, suggesting synovial molecular profiling might eventually identify which PsA patients will respond best to IL-23 blockade.

Axial PsA

One notable gap: axial PsA remains poorly covered by this mechanism. A PSUMMIT subgroup analysis in patients with physician-diagnosed spondylitis showed BASDAI improvements with ustekinumab (BASDAI20/50/70: 54.1%/27.9%/14.4% vs 26.2%/13.1%/0% for placebo), but dedicated spondyloarthritis trials with IL-23 blockade have shown only modest effects on axial disease compared to IL-17 inhibitors, suggesting the IL-23 pathway may be less central to axial involvement than to peripheral arthritis and skin.

Conclusion

These results come from a focused initial search, so the picture here represents a reasonable orientation to the field rather than a systematic appraisal — a deeper search with extended thinking could capture comparative effectiveness data and more recent long-term extension results.