Background/Purpose: Janus kinase inhibitors (JAKi) are effective in rheumatoid arthritis (RA); however, concerns regarding safety, particularly the risk of infections and malignancies, have been raised. This study aimed to assess the impact of JAKi on immune cell subsets in RA patients.

Methods: This prospective study included RA patients initiating JAKi therapy (tofacitinib, baricitinib, upadacitinib, and filgotinib), diagnosed according to ACR/EULAR 2010 criteria. Control groups consisted of healthy donors and RA patients starting biological disease-modifying antirheumatic drugs (RA-bDMARDs). Peripheral blood mononuclear cells were analyzed using multiparametric flow cytometry to characterize the immunophenotype of various immune cell subsets.

Results: We analyzed 78 RA patients treated with JAKi (Baricitinib: 46.2%, Filgotinib: 24.4%, Tofacitinib: 17.9%, Upadacitinib: 11.5%), 20 RA patients treated with bDMARDs (tocilizumab: n=11, abatacept: n=9), and 20 healthy controls (Table).RA-JAKi patients showed a significant decrease in cytotoxic NK Dim cells (vs RA-bDMARD) and NK cells expressing Nkp30 (vs RA-bDMARD and healthy controls). The proportion of effector memory (EM) CD4⁺ T cells was higher in RA-JAKi (20.47 [14.77–30.69]) vs RA-bDMARD (12.12 [7.62–17.05]) and healthy controls (10.79 [7.16–16.25]), while central memory (CM) T cells were reduced (38.40 [29.26–45.66] vs RA-bDMARD: 48.32 [43.82–58.8] and healthy controls: 48.87 [42.07–56.52]). Total CD4⁺ T-helper cells were significantly increased in RA-JAKi (1.50 [0.66–3.13]) vs healthy (0.38 [0.14–1.55]) and RA-bDMARD (0.33 [0.14–1.18]) (p=0.005 and p=0.003). Th17 and Th1⁺17 cells were reduced in RA-JAKi vs both controls (Figure 1).The proportion of double-negative B cells differed significantly between JAKi patients and controls, with switched memory B cells elevated in RA-JAKi (20.25 [11.18–29.08] vs healthy: 12.24 [8.31–18.82], p=0.018). CD21⁻ low B cells were higher in RA-JAKi (4.44 [3.05–7.97] vs RA-bDMARD: 3.16 [1.42–4.75], p=0.045) (Figure 2).

Conclusion: JAKi differentially impact innate immune cells compared to biological DMARDs. The observed reduction in cytotoxic activated NK cells and intermediate monocytes may contribute to increased susceptibility to viral infections and potential malignancies. Alterations in T and B cell subsets suggest an exhausted immune phenotype, emphasizing the need for further functional studies to fully elucidate the long-term immunomodulatory effects of JAKi.

Table. Clinical characteristics of the patients included in the study.

Figure 1. Differentiation status and percentage of T-Helper subsets in the three groups studied. A) Dot-plots depicting the percentage of T-helper according to their differentiation status categorized as Effector Memory T helper (TH EM, CD3+ CD4+ CD62L- CD45RA-), Central Memory T helper (TH CM, CD3+ CD4+ CD62L+ CD45RA-) and Effector memory re-expressing RA T helper cells (TH TEMRA, CD3+ CD4+ CD62L- CD45RA+) among healthy (green), rheumatoid arthritis control group (blue) and jakinib group (pink). B) Dot-plots depicting the percentage of T-helper subsets categorized as T helper 1 (TH1, CD3+ CD4+ CD45RA- CXCR3+ CCR6-), T helper 17 (TH17, CD3+ CD4+ CD45RA- CXCR3- CCR6+), T helper 1-17 (TH1-17, CD3+ CD4+ CD45RA- CXCR3+ CCR6+) and T helper 2 ( TH2, CD3+ CD4+ CD45RA- CXCR3- CCR6- CD294+) among healthy (green), rheumatoid arthritis control group (blue) and jakinib group (pink).*p < 0.05, **p < 0.01 and ***p < 0.001.

Figure 2. Differentiation status of B cells. Dot-plots depicting the percentage of B cells according to their differentiation status, categorized as Naïve (CD19+ IgD+ CD27-) Switched memory B cells (CD19+ IgD- CD27+), Unswitched memory B cells (CD19+ IgD+ CD27+) and Double negative B cells (CD19+ CD27-IgD-) among healthy (green), rheumatoid arthritis control group (blue) and jakinib group (pink).*p < 0.05, **p < 0.01 and ***p < 0.001.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/changes-in-nk-cells-and-th-cell-phenotype-in-rheumatoid-arthritis-patients-treated-with-janus-kinase-inhibitors-implications-for-adverse-effects/