Non-selective Janus-Kinase-Inhibitors (ns-JAKi) significantly impair signal-transducer-and-activator-of-transcription (STAT) type 1 (STAT1) signaling of varicella-zoster-virus-(VZV)-specific cellular memory responses

Background: Janus-Kinase-inhibitors (JAKi) have been approved for several indications in autoimmune diseases. Clinical studies showed that non-selective JAKi (ns-JAKi) are associated with an increased risk of herpes zoster (HZ) by reactivation of varicella-zoster-virus (VZV), whereas selective JAK-1 inhibitors (JAK-1i) have a lower risk for HZ.

Goals: In order to understand the molecular mechanisms behind the different HZ risk between ns-JAK und JAK-1i, the study aims to analyze transcriptome patterns and VZV-specific cellular responses induced by administration of JAKi in vitro on lymphocytes obtained from patients with rheumatoid arthritis (RA) and healthy controls (HC).

Methods: Lymphocytes were stimulated by VZV-antigens in vitro under different concentrations with JAK-1i or ns-JAKi. ELISpot and immune-phenotyping by flow-cytometry were performed to screen for cellular reactivity. mRNA sequencing techniques were used for transcriptome analysis. Western-blot-analysis was performed to determine the phosphorylated signal-transducer-and-activator-of-transcription-(STAT) proteins to estimate functional JAK activity.

Results: JAKi-treated lymphocytes showed significantly lower spot-forming-units (SFU) in a concentration-dependent manner compared to untreated controls and a reduced expression of activation markers (CD69). Relative intensity of phosphorylated STAT1 (pSTAT1) was reduced by 81% by JAK-1i and by 94% by ns-JAKi using plasma-concentration-equivalent doses. JAKi-untreated VZV-stimulated conditions induced expression of STAT1 2.4-fold, STAT2 2.8-fold, STAT3 1.2-fold and STAT4 1.7-fold. Interferon-response-factor-7 (IRF7) was induced 3.6-fold. After treatment with JAK-1i, STAT1 was still induced 1.5-fold, but significantly reduced by 82% by ns-JAKi.

Summary: The data demonstrated that the impact on VZV-specific cellular responses was significantly higher in ns-JAKi-treated than in JAK-1i-treated lymphocytes. This may result in less severe impairment of VZV-induced STAT-signaling by selective JAK-1i compared to ns-JAKi and may explain the lower HZ incidence in JAK-1i compared to ns-JAKi-treated patients.