Background/Purpose: In autoimmunity, T follicular helper cells (TfH) are considered drivers of autoantibody production, and T helper 17 (Th17) cells are implicated in tissue-specific inflammation. In juvenile dermatomyositis patients, the Ueno lab showed that the ratio of TfH cells expressing markers of Th17 cells was increased relative to Th1, and this Th17/TfH ratio correlated with disease activity. Although patients with rheumatoid arthritis (RA) feature joint-specific inflammation and autoantibody production, TfH cell subsets have not been well-studied in context of disease activity or therapy. 

Methods: We identified seropositive RA patients receiving stable TNF inhibitor (TNFi) therapy (n = 26) or non-biologic DMARDs (n = 35), who were enrolled in the RA Comparative Effectiveness Research (RACER) registry at University of Pittsburgh. Treatment groups were matched for age, gender, ethnicity, concomitant DMARD and steroid use. 8 healthy controls were also enrolled. A second cohort of 6 subjects were identified who showed a good response following initiation of TNFi. PBMC were isolated and CD4+CD45RO+ T cell subsets analyzed by flow cytometry using chemokine receptors as functional subset markers. 

Results: There were no significant differences in blood memory cell frequency between RA and healthy controls. Total Th1 (CXCR3+) and TfH (CXCR5+) cells did not differ between RA treatment groups or controls. However, the proportion of TfH cells that co-expressed the Th17 marker CCR6 (termed Th17/TfH) was increased compared to healthy controls (p = 0.044, t-test). Surprisingly, Th17/TfH cell negatively correlated with disease activity in the TNFi group, as assessed by CDAI (Spearman r = -0.72, p = 0.0001) and DAS28crp (r = -0.6844, p = 0.0001). However, no correlation between disease activity and Th17/TfH cells was observed in patients receiving non-biologic DMARD therapy (r = 0.175, p = 0.25 for CDAI; r = 0.059, p = 0.745 for DAS28crp). While CXCR3-expressing Th1/TfH cells showed a trend towards being decreased in RA subjects, Th1/TfH cells did not correlate with disease activity in either treatment group. Analysis of longitudinal samples from TNFi subjects showed strong correlations between visits for Th17/TfH cell frequencies (r = 0.818, p = 0.006). To confirm these cross-sectional results, we analyzed a second longitundinal cohort (n = 6) of subjects initiating TNFi therapy and showing a subsequent good clinical response. Indeed, Th17/TfH cell frequencies remained stable after TNFi therapy (r = 0.943, p = 0.017). 

Conclusion: Th17/TfH cells correlate with disease activity in RA subjects receiving TNFi, but not DMARDS alone. Th17/TfH frequencies remained remarkably stable over time for each individual. These data indicate that Th17/TfH cells are not directly affected by TNFi, but rather that proportions of these cells serve as a predictor of future disease severity for patients receiving TNFi therapy. Hence different levels of Th17/TfH cell contribution to RA pathogenesis in individuals could be one factor underpinning the clinical problem of heterogeneity in RA therapy responses.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/th17tfh-cell-predict-disease-severity-in-rheumatoid-arthritis-patients-receiving-tnf-inhibitor-therapy/