Protein Signatures Track PsA Activity, With Vinod Chandran, MD, PhD

Circulating protein signatures were associated with disease activity across several domains of psoriatic arthritis (PsA) and showed changes following treatment with deucravacitinib in a pooled post hoc analysis of 2 phase 3 trials.¹ Changes in the signatures also differed between patients who achieved an American College of Rheumatology 20% response (ACR20) and those who did not.

Vinod Chandran, MBBS, MD, DM, PhD, presented the findings at the American College of Rheumatology (ACR) Convergence 2026 in Orlando, Florida. Chandran is a professor of medicine at the University of Toronto and director of the Gladman Krembil Psoriatic Arthritis Program at the Schroeder Arthritis Institute, University Health Network. In an interview with HCPLive, he discussed the potential for protein signatures to inform future biomarker development and the need for further validation.

Why PsA Needs Objective Biomarkers of Disease Activity

“Psoriatic arthritis is a heterogeneous disease, and there are many domains that are affected, including skin, synovitis of peripheral joints, enthesitis, dactylitis, and even spinal disease, and the response rates with different drugs are not uniform,” Chandran said. “Now that we have many drugs, we like to choose the right drug at the right time, and we don’t have any tools to select the right drug for that particular patient.”

PsA encompasses multiple clinical domains, and treatment responses may vary across manifestations. Although clinical assessments measure disease activity and treatment response, validated molecular biomarkers capable of guiding individualized treatment selection remain an unmet need.¹

Deucravacitinib (Sotyktu), an oral selective tyrosine kinase 2 (TYK2) inhibitor, demonstrated efficacy in the phase 3 POETYK PsA-2 trial. At week 16, ACR20 response rates were 54.2% with deucravacitinib versus 39.4% with placebo (P = .0002).²

Previous research from the POETYK program examined associations between disease activity and 48 selected PsA- and immune response-related biomarkers.³ The current analysis expanded the investigation to approximately 3000 proteins to identify composite signatures associated with disease activity across individual PsA domains.¹

How POETYK PsA Protein Signatures Were Developed

Investigators analyzed plasma samples from 1225 patients enrolled in the phase 3 POETYK PsA-1 and PsA-2 trials. Samples were collected at baseline, week 16, and week 52, with protein measurements performed using the Olink Explore 3072 platform.¹

The investigators used a 3-step analytical framework to develop disease-tracking signatures. Proteins associated with clinical endpoints were identified through correlation analyses using a false discovery rate-adjusted significance threshold of P <.05. The top 10% of proteins within equally sized groups were then evaluated using least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation.

The resulting weights were normalized and combined into composite weighted z scores. Investigators assessed treatment-associated changes using mixed-effects models.¹

The analysis evaluated signatures associated with several measures of disease activity and patient-reported outcomes, including:

• Psoriatic Arthritis Disease Activity Score (PASDAS) and Disease Activity Index for Psoriatic Arthritis (DAPSA)
• Tender and swollen joint counts
• Leeds Enthesitis Index (LEI) and Spondyloarthritis Research Consortium of Canada (SPARCC) Enthesitis Index
• Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-Fatigue)
• Short Form-36 (SF-36) physical and mental component summary scores¹

Deucravacitinib Attenuated Protein Signatures at Week 16

Investigators identified a 9-protein signature associated with PASDAS, while signatures for the other clinical outcomes contained 7 to 10 proteins each. The signatures included both overlapping and distinct protein profiles across disease activity measures and clinical domains.¹

According to the investigators, the profiles reflected biological processes involving systemic inflammation, innate and myeloid activation and recruitment, tissue injury, matrix remodeling, and counter-regulatory responses.

At week 16, deucravacitinib was associated with statistically significant attenuation of protein signature scores compared with placebo across PASDAS, tender and swollen joint counts, LEI, SPARCC, and patient-reported outcomes.¹

Changes in signature scores also differed according to clinical response. Patients who achieved ACR20 at week 16 demonstrated significantly greater attenuation across disease activity and functional outcome signatures than nonresponders. The molecular separation between responders and nonresponders was greater in the deucravacitinib group than in the placebo group.¹

These findings support an association between changes in circulating protein signatures and clinical disease activity during treatment. However, the observed differences between responders and nonresponders do not establish whether the signatures can prospectively predict treatment response.

Chandran also discussed an exploratory assessment of whether baseline signature scores could anticipate subsequent clinical response. Results from this assessment were not detailed in the published ACR abstract.

Independent Validation Needed Before Clinical Application

Although the analysis identified protein signatures associated with disease activity and treatment response, the findings remain exploratory.

“The most important thing is replication in an independent cohort, which we have not yet done,” Chandran said. “So we consider this as a signal, a hypothesis generation.”

The investigators concluded that the circulating protein signatures tracked longitudinal changes in PsA disease activity and individual clinical domains.¹ However, the analysis was conducted using data from the deucravacitinib clinical development program, and the signatures have not been independently validated as tools for clinical decision-making.

The week 16 comparisons provide evidence of treatment-associated changes relative to placebo, but they do not establish whether the signatures can identify the most appropriate treatment for an individual patient before therapy begins. Although samples were also collected at week 52, the placebo-controlled molecular comparisons reported in the abstract focused on week 16.

Chandran emphasized the distinction between identifying biological signals associated with response and developing biomarkers capable of informing treatment selection among available therapies. Further studies would need to evaluate the signatures in independent patient populations and across different treatment classes.

Editor’s Note: Chandran has reported relationships with Bristol Myers Squibb, AbbVie, Eli Lilly, Johnson & Johnson, Novartis, UCB, and other companies in ACR disclosures. The POETYK PsA clinical development program was sponsored by Bristol Myers Squibb.

References

1. Chandran V, Li S, FitzGerald O, et al. Identification of Protein Signatures Tracking Changes of Disease Activity of PsA Disease Domains and Improvement With Deucravacitinib Treatment in Phase 3 Clinical Studies. Abstract 0879. Presented at: ACR Convergence 2026; November 6-11, 2026; Orlando, FL. Arthritis Rheumatol. 2026;78(suppl 9). https://acrabstracts.org/abstract/identification-of-protein-signatures-tracking-changes-of-disease-activity-of-psa-disease-domains-and-improvement-with-deucravacitinib-treatment-in-phase-3-clinical-studies/

2. Mease PJ, Chandran V, Armstrong AW, et al. Deucravacitinib in Active Psoriatic Arthritis: Efficacy and Safety Up to 52 Weeks From the Randomized, Double-Blind, Phase 3 POETYK PsA-2 Trial. Arthritis Rheumatol. Published online July 20, 2026. doi:10.1002/art.70280

3. Chandran V, FitzGerald O, Mease P, et al. Effect of Deucravacitinib Treatment on Disease Activity–Associated Plasma Biomarkers in Patients With Active Psoriatic Arthritis: Results From 2 Phase 3 Studies. Abstract 0590. Presented at: ACR Convergence 2025; October 24-29, 2025; Chicago, IL. Arthritis Rheumatol. 2025;77(suppl 9). https://acrabstracts.org/abstract/effect-of-deucravacitinib-treatment-on-disease-activity-associated-plasma-biomarkers-in-patients-with-active-psoriatic-arthritis-results-from-2-phase-3-studies/