Adeline Ruyssen-Witrand

Background: As periodontitis is a recognized risk factor for rheumatoid arthritis (RA), it should be adequately treated before the initiation of biologic disease-modifying antirheumatic drugs (bDMARDs), along with any concomitant acute oral infection. bDMARDs remains a concern for patients, and can pose challenges for dentists, particularly regarding the management of invasive procedures. Our aim was to assess adherence to oral health recommendations, and perioperative management of invasive dental procedures in RA patients with bDMARDs. Methods: A French online cross-sectional survey was conducted within the ART e-cohort. A semi-structured 77-item questionnaire was developed by a multidisciplinary team (rheumatologists, dentists, expert patients). Validated oral health scores (Oral Health Impact Profile [OHIP-14] and Periodontal Screening Score [PESS]) were calculated from responses and combined with routine cohort data (RAPID-3, EQ-5D-5 L). Results: Of 543 e-cohort participants, 279 completed the survey (response rate 51.4%; 71% female; 70% aged ≥ 50 years). RA duration exceeded 5 years in 63.7%. Most participants (90%) were receiving bDMARD, including 67% on anti-TNF. Moderate-to-severe RA activity was common (53% with RAPID-3 ≥ 7). The mean EQ-5D-5 L index was 0.605. Based on a PESS > 5 41.6% of respondents were identified at risk of severe periodontitis. The risk increased with age, smoking, alcohol consumption, and RA severity, but was not associated with treatment type. Only 16.5% of participants fully adhered to French professional oral health recommendations, including twice-daily tooth brushing, interdental cleaning and annual dental check-ups). One in four patients had undergone invasive dental procedures within 2 years; 68.5% continued bDMARD therapy. Post-operative complications were rare, all after extractions without antibiotic cover. Conclusions: In the present cohort, RA patients showed low adherence to oral health recommendations and high risk of severe periodontitis. Postoperative complications while continuing bDMARDs were infrequent.

Background Methotrexate is the anchor first-line disease-modifying treatment for rheumatoid arthritis, yet many patients do not reach remission and strategies to potentiate the early treatment response are needed. Metformin has immunometabolic and anti-inflammatory properties, including attenuation of neutrophil extracellular trap formation and reactive oxygen species production, that could enhance the response to methotrexate. We aimed to determine whether adding metformin to methotrexate improves disease activity at 6 months in methotrexate-naïve patients with active rheumatoid arthritis.Methods METorMET2 was an investigator-initiated, multicentre, double-blind, placebo-controlled, parallel-group, phase 2 superiority trial done in 12 rheumatology centres in France. Eligible participants were adults (aged ≥18 years) with rheumatoid arthritis fulfilling the 2010 ACR/EULAR criteria, active disease (DAS28 >3·2), and no methotrexate or metformin in the previous 6 months; participants were not selected for body weight or metabolic status. Participants were randomly assigned (1:1, stratified by centre) to methotrexate plus metformin (titrated to a target of 1500 mg/day) or methotrexate plus matching placebo. Methotrexate was started at 10 mg/week and increased to 15 mg/week (up to 25 mg/week if needed). Participants, investigators, outcome assessors, and statisticians were masked to allocation. The primary outcome was DAS28-CRP at month 6, analysed as the adjusted between-group difference under a treatment-policy estimand with multiple imputation for missing data, in all randomly assigned eligible patients. The trial is registered with ClinicalTrials.gov (NCT04196868) and is complete.Findings Between December 2020, and June 2024, 128 patients were randomly assigned; 126 (65 metformin, 61 placebo) were eligible and analysed. Mean age was 57·9 years (SD 12·7), 89 (70·6%) of 126 participants were women, and median rheumatoid arthritis duration was 1 month (IQR 0·0–2·0). At month 6, mean DAS28-CRP was 2·8 (SD 1·1) with metformin and 2·9 (SD 1·3) with placebo; the adjusted between-group difference was −0·02 (95% CI −0·44 to 0·40; p=0·92). Results were consistent for the secondary estimand (−0·02, −0·46 to 0·41; p=0·92) and across all sensitivity analyses, all excluding the prespecified clinically relevant difference of 0·6 points. Remission was reached by 29 (48·3%) of 60 patients in the metformin group versus 30 (54·5%) of 55 in the placebo group. Serious adverse events occurred in 2 (3·1%) of 65 patients in the metformin group and 3 (4·9%) of 61 in the placebo group; gastrointestinal events were more frequent with metformin, and no deaths occurred.Interpretation Adding metformin 1500 mg/day to first-line methotrexate did not improve disease activity at 6 months in methotrexate-naïve patients with active rheumatoid arthritis and is not supported as an adjunct to methotrexate in this setting. Whether metabolism-targeting strategies benefit rheumatoid arthritis populations selected for metabolic dysfunction remains to be established.

Background/Objective: Adherence to treatment in axial spondyloarthritis (axSpA) remains poorly documented, with estimates ranging from 28.2% to 70.6%. Psychological factors may influence adherence. This study aimed to assess adherence to biological disease-modifying antirheumatic drugs (bDMARDs) and evaluate the impact of catastrophizing, fibromyalgia, anxiety, and depression on adherence. Methods: We conducted a 1-year multicenter cross-sectional study, including consecutive axSpA patients fulfilling ASAS criteria. Participants completed a self-administered questionnaire covering demographic data, disease characteristics, and activity scores. Adherence was assessed using the Girerd scale (GS); psychological assessments included the Pain Catastrophizing Scale (PCS), Fibromyalgia Rapid Screening Tool (FIRST), and Hospital Anxiety and Depression Scale (HADS). Results: Among 500 patients, 49.2% showed good adherence (GS=0). Prevalence of catastrophizing, fibromyalgia, anxiety, and depression was 17.2%, 21.4%, 25.6%, and 11.0%, respectively. Good adherence was significantly associated with older age (mean 53.3 vs. 45.9 y; p p p =0.003). Poor adherence was linked to anxiety (OR: 1.53; 95% CI: 1.02-2.30; p =0.041) and catastrophizing (OR: 1.61; 95% CI: 1.00-2.58; P =0.049). In multivariate analysis, younger age (OR: 0.96; 95% CI: 0.95-0.98; p p =0.031) remained significantly associated with poor adherence. Conclusions: About half of axSpA patients had suboptimal adherence to bDMARDs. Younger age, anxiety, and catastrophizing were associated with poor adherence and should be considered in patient management.

Objective: The previously developed FLARE questionnaire assesses the multidimensional aspects of a flare of psoriatic arthritis (PsA). We aimed to further validate this questionnaire. Methods: Reanalysis of data from the ReFLAP study (ClinicalTrials.gov: NCT03119805 ), a longitudinal observational study carried out in 14 countries. Demographic, clinical, and patient-reported data were collected at baseline and 1 follow-up visit. The FLARE questionnaire (range 0-10) was completed at both timepoints (mean interval 4.5 months). Using a patient anchor question for perceived flare (yes/no), the optimal cutoff for defining a flare was obtained using receiver-operating characteristic curve analysis. Based on this cutoff, patients were compared for flare (yes/no) using clinical and patient-reported data, including composite measures of disease activity. The magnitude of score differences in patients experiencing a new flare at second visit were compared with independent samples t -scores. Results: Of 147 patients, 61 (41%) reported a flare at baseline; the optimal score cutoff from the FLARE questionnaire was 4 (sensitivity 0.86, specificity 0.76) and, using this cutoff, patients in flare had significantly worse clinical and patient-reported outcomes. Of those patients reporting a flare, the most frequently affirmed items from the questionnaire referred to pain, mobility, frustration, and fatigue. Those with a new flare at the second visit had patient-reported and clinical change scores significantly worse than those not reporting a flare. Conclusion: This study has provided further data on the FLARE questionnaire performance and validity. The questionnaire is ready for use in clinical trials and longitudinal observational studies.

Lymphotoxin-alpha (LTα3) is a soluble cytokine of the TNF superfamily. Its role in inflammation and arthritis is not well known. Macrophages are important in K/BxN Serum-Transfer Arthritis (STA) and rheumatoid arthritis (RA). Anti-TNF monoclonal antibodies as well as etanercept (ETA), a soluble TNF receptor II that also neutralizes LTα3, are efficient in the treatment of RA. Objectives: To evaluate the role of LTα3 in macrophage phenotypes and in arthritis. Methods: Macrophages were cultured in the presence of recombinant LTα3, and their phenotypes were studied. The clinical effect of blocking LTα3 in STA was evaluated, as well as the effect of switching from anti-TNF monoclonal antibodies to etanercept in the “ROC” register of RA patients. Results: We showed that recombinant LTα3 was capable of directing mouse and human macrophages towards a pro-inflammatory “M1” phenotype. In K/BxN STA, ETA decreased clinical score and joint swelling. Anti-LTα3 reduced arthritis only in TNF-KO mice, indicating that the effect of LTα3 was visible in the absence of TNF. The “ROC” register indicated that switching anti-TNF mAb to ETA did not induce clinical and biological improvement in RA. Conclusion: We show a pro-inflammatory role for LTα3 in murine and human macrophages. The neutralization of both TNF and LTα3 is not beneficial in the treatment of RA.

(1) Background: TNF inhibitors (TNFis) have revolutionized the treatment of rheumatoid arthritis (RA). However, 30–40% of RA patients do not respond adequately to those biologics. In addition to neutralizing soluble TNF, TNFis have the ability to bind the transmembrane form of TNF, tmTNF. Importantly, tmTNF can act itself as a receptor that induces “Reverse Signaling” (RS) in cells. We previously showed that certolizumab, a Fab’ TNFi, activates RS in human primary monocytes, at least in part through the transcription factor Nrf2 that is known to regulate the expression of genes involved in anti-inflammatory response and oxidative stress. (2) Methods: Here, we have developed an assay for the prediction of clinical response of RA patients to TNF inhibitors. This assay is based on mRNA quantitation of CD36 activation and of six genes induced by Nrf2 following tmTNF RS in fresh monocytes. (3) Results: We could predict the response to anti-TNF monoclonal antibodies (mAbs) with 93.3% accuracy. However, our method was not suitable for the prediction of the response to TNF soluble receptor etanercept. (4) Conclusions: We have developed a rather simple, short-term test that can be standardized. Predicting the response to TNF mAbs will help physicians offer the best available treatment and provide patients with personalized medicine.

Objectives: This post hoc analysis evaluated change from baseline (Δ) in weight/body mass index (BMI) and association with disease activity or lipid changes in tofacitinib‐treated patients with rheumatoid arthritis (RA).Methods: Data up to month 12 were pooled from eight phase 3 and 3b/4 studies of patients with RA receiving tofacitinib 5 or 10 mg twice daily or tofacitinib 11 mg modified‐release once daily (alone or combined with conventional synthetic disease‐modifying antirheumatic drugs), or placebo. Assessments included Δweight/BMI and the proportion of patients with weight gain ≥5%, at months 3, 6, and 12. Correlations between ∆weight/∆BMI and baseline/∆Disease Activity Score in 28 joints, erythrocyte sedimentation rate (DAS28‐4[ESR]), baseline C‐reactive protein (CRP), and ∆lipids were assessed. Statistical analysis included a longitudinal linear mixed model for repeated measures.Results: The analysis included 5,335 patients (tofacitinib 5 mg twice daily [n = 2,349], 10 mg twice daily [n = 1,611], 11 mg once daily [n = 694], and placebo [n = 681]). Increases in least squares mean Δweight and ΔBMI were significantly greater (P < 0.05) at months 3 and 6 with all tofacitinib doses versus placebo; increases continued to month 12. Significantly greater (at least P < 0.05) proportions of tofacitinib‐treated patients (all doses) had weight gain ≥5% at months 3 and 6 versus placebo. There were weak correlations between weight/BMI changes with tofacitinib and DAS28‐4(ESR), baseline CRP, or lipid changes.Conclusion: Patients receiving tofacitinib experienced weight and BMI changes (primarily increases) over time, with weak correlations with disease activity or lipids.

Objective: To explore thresholds for the Psoriatic Arthritis (PsA) Impact of Disease questionnaire (PsAID12) score against disease activity measures in an observational setting, in patients with PsA. Methods: The baseline data from the ReFlaP observational, prospective, multicentre and international study were used (NCT03119805). Cutoffs for PsAID12 were determined against disease activity scores, defining disease impact states (i.e. remission, low impact, moderate impact and high impact). Statistics used to assess the optimal cutoff point included Youden’s index and the 75th percentile method, with external anchors (i.e. Disease Activity index for Psoriatic Arthritis [DAPSA], very low disease activity [VLDA]/minimal disease activity [MDA] and single questions for both patients and physicians) serving as gold standards. The diagnostic performance of these cutoffs was evaluated using receiver operating characteristic (ROC) curve analyses. Results: A total of 410 patients were analysed. Mean (s.d.) PsAID12 score was 3.4 (2.5). The prevalence of remission varied between 12.4% and 36.1%, while low disease activity ranged from 37.8% to 59.8%. PsAID12 performed well against external anchors, with high areas under the ROC curves ranging from 0.75 to 0.94. Using the DAPSA as external anchor, the proposed PsAID12 cutoffs were 3.1 to <4.8 for moderate impact and ≥4.8 for high impact. Compared with composite scores, patient and physician opinions performed less stringently. Conclusion: This study established cutoffs for PsAID12 in a clinical practice observational population, corresponding to remission and varying levels of disease impact. However, these proposed cutoffs need further validation, and an expert consensus is essential to confirm the most accurate thresholds for future use.

Objectives To compare two strategies—a hydrocortisone replacement strategy and a prednisone tapering strategy—for their success in glucocorticoid discontinuation in patients with rheumatoid arthritis (RA) with low disease activity (LDA). Methods The Strategies for glucocorticoid TApering in Rheumatoid arthritis (STAR) study was a double-blind, double-placebo randomised controlled trial including patients with RA receiving a stable dose of glucocorticoid 5 mg/day for ≥3 months and were in LDA for ≥3 months. Patients were randomly assigned in a 1:1 ratio to either replace prednisone with 20 mg/day of hydrocortisone for 3 months, then reduce to 10 mg/day for 3 months before discontinuation or to taper prednisone by 1 mg/day every month until complete discontinuation, contingent on maintaining LDA. The primary outcome was the percentage of patients achieving glucocorticoid discontinuation at 12 months. Other secondary outcomes were proportion of flares, need for additional glucocorticoid use, disease activity, patient-reported outcomes and the results of adrenocorticotropic hormone (ACTH) stimulation tests. Results Of the 102 patients randomised in the trial (mean age 62.4 years, 70.6% females), 53 had hydrocortisone replacement and 49 tapered prednisone. At 12 months, 29 patients (55%) in the hydrocortisone replacement group and 23 patients (47%) in the prednisone tapering group achieved glucocorticoid discontinuation (p=0.4). No difference was observed between groups in the secondary outcomes. No cases of acute adrenal insufficiency were observed; however, 17 patients still had an abnormal ACTH stimulation test at 12 months, with no differences between arms. Conclusion A hydrocortisone replacement strategy was not superior to a prednisone tapering strategy for achieving glucocorticoid discontinuation success in patients with RA in LDA. Trial registration number NCT02997605.