0847: Development of an IFN 5-Gene Signature Score to Identify IFN-high and IFN-low Subsets and as a Pharmacodynamic Biomarker for Deucravacitinib Treatment in a Phase 2 Trial in Patients with Systemic Lupus Erythematosus

Chun Wu¹, Yanhua Hu¹, Mary K. Crow², Amit Saxena³, Cristina Arriens⁴, Coburn Hobar¹, Adrian Coles⁵ and Ian M. Catlett⁶, ¹Bristol Myers Squibb, Princeton, NJ, ²Hospital for Special Surgery, New York, NY, ³NYU Langone, New York, NY, ⁴Oklahoma Medical Research Foundation and University of Oklahoma Health Sciences Center, Department of Arthritis & Clinical Immunology, Oklahoma City, OK, ⁵Bristol Myers Squibb, Lawrenceville, NJ, ⁶Bristol Myers Squibb, Pennington, NJ

Background/Purpose: Elevated IFN activity is observed in a subset of lupus subjects suggesting that those with higher levels of IRGs may benefit from interferon targeted therapies, however, to date responses by this dichotomization have been inconsistent. Tyrosine kinase 2 (TYK2) mediates cytokine pathways (eg, type I IFN, IL-12 and IL-23) linked with SLE pathogenesis. Deucravacitinib is a first-in-class, oral, selective, allosteric TYK2 inhibitor and was found to be efficacious in a phase 2 SLE trial.¹ We developed a customized IFN 5-gene signature score, assessed the pharmacodynamic effects of deucravacitinib on the IFN score, and evaluated the score's association with SLE disease activity and clinical response in the phase 2 trial (NCT03252587).

Methods: Patients were randomized equally to placebo or deucravacitinib (3 mg twice daily [BID], 6 mg BID, or 12 mg once daily [QD]). DxTerity chemical ligation-dependent probe amplification was used to measure 51 immune system–related genes from whole blood. IFN genes were selected based on distribution, correlations, hierarchical clustering, and consistency of k-means clusters. Serum proteins, blood cell subsets, and antibodies were measured by immunoassays and flow cytometry. Systemic Lupus Erythematosus Responder Index-4 (SRI[4]) and British Isles Assessment Group–based Composite Lupus Assessment (BICLA) were measured at weeks 32 and 48.

Results: An IFN 5-gene (MX1, HERC5, IFIT1, RSAD2, and EIF2AK2) signature score was identified and used to classify patients into IFN-high or IFN-low subgroups (Figure 1). Higher baseline score was associated with higher baseline SLEDAI and Cutaneous Lupus Erythematosus Disease Area and Severity Index scores, higher IFN activity biomarker (eg, IFN𝛼, IFN𝜆, B-cell activating factor, C-X-C motif chemokine ligand 10) and anti–double-stranded DNA levels, and lower complement and lymphocyte counts. Deucravacitinib reduced the IFN score from weeks 4 through 44 by > 50%. Patients with high IFN scores had numerically higher SRI(4) response rates at week 32 and BICLA response rates at week 48 in the 3-mg BID and 12-mg QD dose groups, but not in the 6-mg BID group, compared with patients with low IFN score (Figure 2).

Conclusion: These data support the IFN 5-gene signature score as a biomarker to classify patients with SLE into IFN-high or IFN-low subgroups; however, clinical response by IFN score was inconsistently improved. IFN-regulated gene expression performs well as a pharmacodynamic biomarker to confirm deucravacitinib mechanism of action and to aid in phase 3 dose selection.

Reference:
1. Morand E, et al. Arthritis Rheumatol 2023;75:242–252.






C. Wu: Bristol Myers Squibb, 3; Y. Hu: BMS, 3, 12, BMS stock holder; M. Crow: AMPEL BioSolutions, 2, AstraZeneca, 2, Bristol Myers Squibb, 2, Eli Lilly, 2, Gilead Sciences, 5, GlaxoSmithKlein(GSK), 2; A. Saxena: AbbVie/Abbott, 1, AstraZeneca, 1, GlaxoSmithKlein(GSK), 1; C. Arriens: AstraZeneca, 1, 5, 6, Aurinia, 6, Bristol-Myers Squibb, 1, 5, Cabaletta, 1, GSK, 1, Kezar, 1, UCB, 1; C. Hobar: Bristol-Myers Squibb(BMS), 3; A. Coles: Bristol-Myers Squibb(BMS), 3; I. Catlett: Bristol Myers Squibb, 3, 8.