Background: Upadacitinib (UPA), an oral JAK1 selective inhibitor, showed greater efficacy compared to adalimumab (ADA) in patients with active rheumatoid arthritis (RA) despite treatment with methotrexate (MTX) in the SELECT-COMPARE phase 3 study1. The regulatory immune networks affected by JAK1 inhibition compared with TNF-blockade have not been explored previously in a head-to-head trial.
Objectives: To infer the relative immunological pathway modulation of UPA compared with ADA in patients with RA via the evaluation of a pre-defined set of plasma proteins associated with inflammation.
Methods: Patients from the SELECT-COMPARE studies were randomly selected (PBO, n=100; UPA 15 mg QD, n=100; ADA 40 mg EOW, N = 100). The levels of 184 inflammation related proteins were analyzed using the Olink® platform; change from baseline in protein levels were expressed as Log2 Fold Change; a Repeated Measure Mixed Linear Model identified proteins differentially modulated by UPA and ADA compared to PBO, and between Responders (R defined as achieving Low Disease Activity [LDA] based on CDAI [≤ 10] at week 12) and Non Responders (NR defined as not achieving LDA at week 12) for the UPA and ADA groups. Pathway analysis were performed with Ingenuity® Pathway Analysis (Qiagen Inc.); selection criteria: mean |Log2 FC| ≥ 0.1 AND a FDR ≤ 0.05 → n = 88 out of 184 proteins; the top 10% pathways based on Z score for the ADA and UPA groups, and each visit (Week 2 and Week 8) were selected for comparison.
Results: Both UPA and ADA inhibited protein biomarkers (pBM) associated with Neutrophil / Macrophage biology. However, UPA preferentially inhibited pBM associated with T cells and ADA preferentially inhibited pBM associated with M1 or ‘inflammatory’ Macrophages. The pathways implicated in silico by these pBM changes in response to UPA and ADA tended to be similar except for T cell activity related pathways that were preferentially modulated by UPA.
In the ADA group, clinical response was mainly associated with lower levels of pBMs such as IL6, TNFRSF1A, MMP10, IL2RA, PLAUR, CCL2, TNFRSF10C, and SERPINE1, suggesting that control of these pathways may be important for response to ADA in RA.
In the UPA treated group, clinical response was mainly associated with slightly higher levels of the pBM IL17A, IL17C, CCL11, CCL20, and TIMP4.
Analysis of the reciprocal changes in the above showed that IL6, TNFRSF1A, IL2RA, NPPB, and SERPINE1 were downregulated similarly in UPA R and NR patients. By comparison, IL17A was modestly upregulated similarly in ADA R and NR patients; CCL20 was downregulated similarly in ADA R and NR patients; and IL17C, IL22RA1, TIMP4, and CCL11 were not modulated by ADA. Finally, taken together, among the 184 inflammation-related pBM tested, none were associated with clinical response for both ADA and UPA.
Conclusion: We detected common but also discrete alterations in pBMs upon exposure to ADA and UPA overall and further distinctions when we examined pathway changes associated with achievement of a low disease clinical status achieved by each patient population. Whereas both drugs exhibit inhibition of macrophages and granulocyte associated pathways, ADA appears to preferentially affect M1 macrophages and UPA appears to preferentially affect T cells. This modulatory pattern by UPA is consistent with its broad cytokine receptor inhibition profile compared with highly specific TNF inhibition and could account at least in part, for the greater efficacy of UPA over ADA in the SELECT-COMPARE study.
References: Fleischmann, R., et al. (2019). Arthritis Rheumatol 71(11): 1788-1800.
Disclosure of Interests: Thierry Sornasse Shareholder of: AbbVie Inc, Employee of: AbbVie Inc, In-Ho Song Shareholder of: AbbVie Inc., Employee of: AbbVie Inc., Timothy Radstake Shareholder of: AbbVie Inc, Employee of: AbbVie Inc, Iain McInnes Grant/research support from: Bristol-Myers Squibb, Celgene, Eli Lilly and Company, Janssen, and UCB, Consultant of: AbbVie, Bristol-Myers Squibb, Celgene, Eli Lilly and Company, Gilead, Janssen, Novartis, Pfizer, and UCB
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.