Synovial, Skin Gene Expression Differences May Explain PsA Treatment Responses

Bianca Nogrady

December 15, 2020

Differences in gene expression between the skin and synovial tissues of individuals with psoriatic arthritis could explain why treatments targeting proinflammatory mechanisms don't improve joint symptoms in some patients.

A paper published in Annals of the Rheumatic Diseases presents the results of an observational, open-label study involving 27 patients with active psoriatic arthritis, 18 of whom were treated with anti–tumor necrosis factor (anti-TNF) therapies and 9 with the monoclonal antibody ustekinumab (Stelara). This drug targets the axis of proinflammatory cytokine interleukin-23 and effector cytokine IL-12, which are believed to play an important role both in skin and nail psoriasis, and psoriatic arthritis.

However, while anti–IL-23 antibodies seem to work well to address skin manifestations of psoriasis, they tend to improve joint symptoms only in selected patients.

"The lack of a clear mechanism to explain such divergent responses prompted this study," said Alessandra Nerviani, lead author of the study, from the Barts and The London School of Medicine & Dentistry.

Participants also had biopsies taken from the synovium – in particular, from joints that were clinically and ultrasonographically active – and from lesional and nonlesional skin for gene expression analysis.

In terms of treatment response, the ustekinumab-treated group showed significantly higher scores for erythrocyte sedimentation rate, joint pain, and disease activity, compared with the anti–TNF-treated group. Psoriasis Area and Severity Index scores were similar in both treatment arms, but significantly more patients in the anti-TNF group met the EULAR Disease Activity Score for response (70.6% vs. 22.2%).

The gene expression analysis, which assessed the activity of 80 genes related to inflammation in 14 patient samples from synovial tissue, lesional skin, and nonlesional skin, found that patterns of expression in the synovium clustered away from those from skin.

This was particularly the case when it came to genes related to drug targets. The targets for anti-TNF showed similar levels of expression in both skin and synovial tissue. However, the targets for ustekinumab – namely interleukin (IL)–23A, IL-23R and, IL-12B – showed higher levels of expression in lesional skin than in nonlesional skin and synovial tissue.

"Interestingly, we observed that, while some patients did express IL-23 cytokines/receptor in both skin and joint, others had discordant expression, that is, active IL-23 pathway in the lesional skin but not in the synovium," the authors wrote.

When researchers then stratified patients according to how much synovial inflammation they had, they found that patients who had higher scores also showed higher expression of genes for IL-12B and IL-23R, but not IL-23A, despite showing no other major clinical differences.

The authors also looked at the protein expression levels for IL-23p40, IL-23p19 and IL-23R, and found that while the percentage of cells positive for these proteins was significantly higher in lesional, compared with nonlesional skin, it was also higher in the synovium among patients with more inflammation.

"Except for the LIKERT patient score, we did not detect other significant correlations between IL-23 axis expression and clinical parameters at baseline, suggesting that patients with comparable disease severity may have, in fact, heterogeneous histopathological features and expression of drug targets within the diseased synovium," they wrote.

More Selective Expression of IL-23 in Synovium

Commenting on the findings, the authors highlighted that the expression of targets for anti-TNF was much more homogeneous across skin and synovial tissue, but the IL-23A/IL-12B/IL-23R genes generally showed higher levels of expression in lesional skin. compared with either nonlesional skin or synovium. However, even within the synovium, expression of these genes varied enormously, from levels similar to those seen in paired lesional skin to levels well below those.

"It is plausible to speculate that an overall higher presence of IL-23 in the psoriatic skin supports the concept of a generally better response in terms of skin manifestations, including almost complete clearance of psoriatic lesions," Nerviani said in an interview. "While, on the other hand, the more selective expression of IL-23 in the synovium, namely in histologically more inflamed synovium characterized by immune cells infiltration, may explain the overall more modest success to meet stringent response criteria in the joints."

Of particular significance was the observation that IL-12B and IL-23R transcription levels were higher in patients with higher levels of synovial tissue inflammation.

"We confirmed that IL-23 axis expression relates to the synovial histopathology not only in PsA at different stages of the disease, including early treatment-naive patients, but also in the early phase of RA, investigated as disease control," they wrote.

Nerviani said the results could inform a more tailored "precision medicine" approach to treating patients with psoriatic arthritis.

"While randomized synovial biopsy–driven clinical trials are now a reality in rheumatoid arthritis, in psoriatic arthritis, these kinds of studies have not been performed yet but may become actual in the future," she said. "An in-depth characterization of the synovial tissue represents the first essential step towards addressing current unmet clinical needs and, potentially, changing our practice."

However, she stressed that the study was not powered to test the correlation between the expression level of these pathways in disease tissue and clinical response to treatment.

"Further dedicated clinical trials should be designed to look at the relationship between synovial pathology and molecular characteristics, and response to targeted treatment to address this question," Nerviani said.

The study was supported by the Queen Mary University of London and the Fondazione Ceschina, and in part by grants from Versus Arthritis. No conflicts of interest were declared.

SOURCE: Nerviani A et al. Ann Rheum Dis. 2020 Nov 26. doi: 10.1136/annrheumdis-2020-218186.

This article originally appeared on, part of the Medscape Professional Network.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: