POS1000 MORE THAN 30 % OF WOMEN WITH SPONDYLOARTHRITIS HAVE AN UNFAVORABLE PREGNANCY OUTCOME MOST FREQUENTLY DUE TO SMALL FOR GESTATIONAL AGE: ANALYSIS OF THE PROSPECTIVE GR2 COHORT

Background

Spondyloarthritis (SpA) is one of the most common chronic inflammatory diseases and regularly affects women of childbearing age1. However, there is limited knowledge about the impact of the disease and its treatment on pregnancy.

Objectives

The aim of the study was to determine the factors associated with adverse pregnancy outcome in women with SpA.

Methods

All SpA patients (diagnosis according to the Rheumatologist) included in the national multicenter GR2 cohort from 2015 to June 2021 were included in the analysis. Patients could be included either with a pregnancy wish (i.e., preconceptional period) or because of a clinical pregnancy (<12 weeks of gestation). The main endpoint was favorable pregnancy outcome, a composite outcome defined as a live birth at term ≥ 37 gestation weeks of a healthy newborn with a weight greater than the 10th percentile. We performed a multilevel logistic regression model, in which we considered patient and center random effects (patient random effect for some women included in the cohort two times). Disease activity was defined by a BASDAI score ≥ 4 at least once during pregnancy. We used a multiple imputation to address missing data among the explanatory variables. Results are presented as an odds ratio (OR) with confidence interval (CI).

Results

Among the 207 pregnancies in women with SpA included in the GR2 cohort, 126 were retained for analysis of obstetrical outcome. Of these, 29 (23.0%), 14 (11.1%), 69 (54.8%) were exposed to corticosteroid, NSAID and biologics at least once during pregnancy, respectively. An active disease at least once during pregnancy was found in 47 (37.3%) pregnancies. A live birth was found in 116 (92.1%) women, including 110 (87.3%) full-term births. Early miscarriages and stillbirths were observed in 7 (0.06%) and 3 (0.02%) women, respectively. A caesarean section was performed in 20 (17.2%) cases.

A favorable pregnancy outcome was found in 80 (63.5%) of the women. Unfavorable pregnancy outcome was most frequently due to small for gestational age, observed in 22 (19%) pregnancies. The multivariate model adjusted for age, BMI, nulliparity, active disease during pregnancy, smoking, and exposure to NSAIDs and corticosteroids during pregnancy found an association between unfavorable pregnancy outcome with nulliparity (OR 2.63 95% CI [1.01-6.81] p = 0.05).

Conclusion

This study provides original results on pregnancy in women with SpA. It found a favorable pregnancy outcome in 63.5% of women. Unfavorable pregnancy outcome was most frequently due to small for gestational age, which should lead to a coordinated management with obstetricians for the follow-up of pregnancy in women with SpA.

References

[1]Van den Brandt S. Arthritis Res Ther. 2017;19(1):64.

Table 1.

Multilevel logistic regression model: factors associated with unfavorable pregnancy outcome in women with SpA.

Univariate analysesMultivariate analysesCrude OR 95% CIpAdjusted OR 95% CIpAge1.01 [0.92-1.10]0.8591.05 [0.95-1.17]0.297BMI0.99 [0.91-1.07]0.7960.99 [0.90-1.08]0.747Nulliparity2.16 [0.94-4.94]0.0712.63 [1.01-6.81]0.049Smoking0.84 [0.23-3.03]0.8050.84 [0.22-3.21]0.805Disease activity*0.98 [0.40-2.43]0.9641.15 [0.43-3.07]0.778Corticosteroids**1.09 [0.45-2.65]0.8761.15 [0.51-2.71]0.902NSAIDs**0.65 [0.18-2.33]0.1960.67 [0.18-2.56]0.565

* BASDAI score ≥ 4 at least once during pregnancy.

** Use at least once during pregnancy

Acknowledgements

The GR2 Cohort is supported by the French Society of Rheumatology, the French Internal Medicine Society, and unrestricted grants from UCB.

Disclosure of Interests

SABRINA HAMROUN: None declared, Marion Couderc: None declared, Rene-Marc Flipo: None declared, Jérémie SELLAM: None declared, Christophe Richez Speakers bureau: CR has received consulting/speaker’s fees from Abbvie, Amgen, Astra Zeneca, Biogen, BMS, Celltrion, Eli Lilly, Galapagos, GSK, MSD, Novartis, and Pfizer, all unrelated to this manuscript., Rakiba Belkhir: None declared, Laure Gossec: None declared, Hubert MAROTTE: None declared, Emmanuelle Dernis: None declared, Aline Frazier-Mironer: None declared, Elisabeth Gervais: None declared, Cédric Lukas: None declared, Valerie Devauchelle-Pensec: None declared, Laëtitia Dunogeant: None declared, Alban Deroux: None declared, Nathalie Costedoat-Chalumeau: None declared, Anna Moltó: None declared

AB1131 IDENTIFICATION OF FACTORS ASSOCIATED WITH THE OCCURRENCE OF SEVERE FORMS OF COVID-19 INFECTION IN PATIENTS WITH AUTOIMMUNE/INFLAMMATORY RHEUMATIC DISEASES

Background

Patients with autoimmune/inflammatory rheumatic diseases (AIRD) were suspected to be an at-risk population of severe COVID-19. However, whether this higher risk is linked to the disease or to its treatment is difficult to determine.

Objectives

To identify, among AIRD patients, factors associated with occurrence of moderate-to-severe COVID19 infection and to evaluate if having an AIRD was associated with an increased risk of severe form of COVID19 infection (defined by hospitalization in ICU or death), compared to general population.

Methods

Data source: The “Entrepôt des Données de Santé (EDS)” collect data from electronic health records of all patients hospitalized or followed in the AP-HP (39 hospitals in Paris area, France). The French RMD COVID19 cohort is a national multi-center cohort that included patients with confirmed AIRD and diagnosed with COVID-19. All AIRD patients diagnosed with COVID-19 before September 2020 from both cohorts were included.

-We Identified factors associated with severe COVID-19 was made in a combined analysis of the 2 cohorts.

-Then, we compared COVID-19 infection severity in the EDS-COVID database in AIRD patients and controls, by a propensity score (PS)-matched case-control (1:4) study

Results

Among 1213 patients (334 in EDS and 879 in RMD cohort), 195 (16.1%) experienced a severe COVID19. In multivariate analysis, greater age, history of interstitial lung disease, arterial hypertension, obesity, sarcoidosis, vasculitis, auto-inflammatory disease and treatment with corticosteroids or rituximab were associated with severe COVID-19 (Table 1).

Table 1.

AIRD patient’s characteristics associated with severity of COVID-19

Patients with mild or moderate infectionPatients with severe infectionOR ajustés 95%CIp-value(N = 1018)(N = 195)Patients characteristics Age55.9 (16.7%)70.3 (14.3%)1.05 [1.03;1.07]<0.001 Gender: Female695 (68.3%)105 (54.1%)0.59 [0.38;0.94]0.025 Interstitial pneumonia38 (3.7%)20 (10.3%)2.94 [1.34;6.34]0.008 Obesity143 (17.8%)38 (27.7%)2.09 [1.26;3.43]0.004 Hypertension268 (26.3%)114 (58.5%)1.81 [1.13;2.89]0.013Underlying Disease: Chronic inflammatory arthritis618 (60.8%)72 (36.9%)Ref. Auto-inflammatory disease29 (2.9%)5 (2.6%)3.91 [1.2;11.32]0.025 Other29 (2.9%)4 (2.1%)0.35 [0.06;1.41]0.15 Connectivitis190 (18.7%)34 (17.4%)1.13 [0.62;2.01]0.69 Sarcoidosis40 (3.9%)24 (12.3%)5.19 [2.15;12.3]<0.001 Vasculitis111 (10.9%)56 (28.7%)1.8 [1.02;3.16]0.044Treatments Corticosteroid318 (31.2%)117 (60.0%)2.47 [1.58;3.87]<0.001 Leflunomide44 (4.3%)2 (1.0%)0.13 [0;0.97]0.045 Rituximab37 (3.7%)22 (11.5%)4.05 [1.96;8.27]<0.001Not significant in multivariate analysisCOPD, Asthma, Coronary heart diseases, stroke, diabetes, smoking, cancer, non-steroidal anti-inflammatory drugs, colchicine, hydroxychloroquine, methotrexate, salazopyrine, mycophenolate mofetil, azathioprine, intravenous immunoglobulins, anti-TNFα, anti-IL1, -IL6, -IL17, Abatacept, JAK inhibitor

Among 35741 COVID-19 patients in EDS, 316 with AIRD were compared to 1264 PS-matched controls. Severe form occurred in 118 (37,3%) AIRD cases and 384 (30.4%) controls (Adjusted OR (aOR) for severe form= 1.43 [1.1;1.9], p=0,01). In analysis restricted to rheumatoid arthritis (RA) and spondylarthritis (SpA), no increased risk of severe form (aOR=1.11 [0.68;1.81]) form or death (aOR=1.00 [0.55;1.81]) was observed.

Conclusion

In this multicenter study we confirmed that AIRD patients treated with rituximab or corticosteroids were at increased risk of severe COVID-19, as were those with vasculitis, auto-inflammatory disease, and sarcoidosis. Also, when compared to controls from the same cohort of hospitalized patients, AIRD patients had, overall, an increased risk of severe COVID-19, increased risk not observed in an analysis restricted to patients with RA or SpA.

Acknowledgements

FAI2R /SFR/SNFMI/SOFREMIP/CRI/IMIDIATE consortium and contributors

Patricia Martel

All clinicians/physicians implicated in COVID-19 patient care in APHP hospital and generated EDS patient data

Disclosure of Interests

None declared

POS0499 A NEW SERUM ASSAY MEASURING SYNOVIAL TURNOVER IN RHEUMATOID ARTHRITIS

Background

Rheumatoid arthritis (RA) is characterized by synovial tissue (ST) inflammation leading to pain, joint destruction, impaired mobility and ultimately increased morbidity and mortality. These various components of disease etiology can be monitored by pain scores and imaging technologies. Noninvasive assessment of joint tissues metabolism can be performed by measuring serum metabolites of tissue matrix turnover. Specific biochemical markers (BM) of bone and cartilage have been developed, but there is still a lack of a sensitive index of ST metabolism abnormalities, although there are early and key drivers of joint destruction.

Objectives

To develop a sensitive ELISA-based serum BM of ST turnover (SynHelix) and evaluate its performance in patients with RA.

Methods

We identified a 11 amino-acid sequence (HELIX-III) within the triple helical portion of type III collagen, a main component of ST matrix which is released during ST turnover and can be measured in the serum. A highly specific rabbit polyclonal antibody raised against the synthetic HELIX-III peptide was produced to develop a competitive ELISA. Specificity of the antibody was evaluated by competitive inhibitions with homologous sequences of type I, IV and XI collagens which are also present in ST, but in minor quantities for the two later. Immunohistochemistry analysis of human ST obtained during hip surgery was performed to investigate in situ localisation of the HELIX-III peptide. ELISA was then used to quantify serum HELIX-III fragments in two samples of patients with low (n= 40, median DAS 28: 2.1) and moderate (n=11, median DAS28: 3.5) RA activity participating in clinical trials. Levels in RA subjects were compared to sex and age-matched healthy controls randomly selected from epidemiological cohorts (n=70).

Results

The ELISA for SynHelix demonstrated adequate analytical performances with intra and interassay variations below 10 and 15%, respectively; analytical and functional limits of quantification of 0.21 and 2.5 ng/ml, respectively and dilution recovery of human serum ranging from 80 to 108%. Competitive inhibition experiments demonstrated that the antibody does not recognize HELIX-III peptide shortened or elongated by one amino-acid, indicating that immunogenicity is dependent on the presence of a neopitope resulting from the cleavage of the collagen molecule. The antibody does not recognize the homologous sequence of type I collagen, but shows significant immunoreactivity with the homologous sequences of type IV (alpha 5) and type XI (alpha 2) collagens with however a lower affinity than for type III collagen (a 2.2 and 4-fold higher concentration, respectively, is needed to displace 50% of the immune signal). Immunohistochemistry of ST tissue from RA subjects showed increased staining in the interstitial tissue and around vessels which are rich in type IV collagen. Median serum levels of SynHelix were significantly higher in patients with low (p=0027, +17%) and moderate RA activity (+220%, p=00004 vs healthy controls; +164% p=0.16 vs low RA) compared with those in age-matched controls, although CRP did not discriminate RA patients with low and moderate activity (2.2 mg/L for both groups, p=0.90). In RA subjects, serum SynHelix correlates modestly with CRP (r=0.59, p <0.0001), but not with DAS28.

Conclusion

The new SynHelix ELISA measures precisely circulating degradation fragments of Helix-III peptide-containing collagens. Serum SynHelix levels are already increased in patients with low activity RA, values correlating modestly the degree of systemic inflammation. Larger longitudinal studies are needed to further evaluate the value of SynHelix to predict disease outcome in RA.

Disclosure of Interests

Patrick Garnero: None declared, Evelyne Gineyts: None declared, Jean-Charles Rousseau: None declared, Hubert MAROTTE: None declared, Roland Chapurlat Consultant of: BMS, abbvie, Lilly, Galapagos, fresenius-Kiabi, novartis.

POS0695 EFFECT OF METHOTREXATE AND FOLIC ACID CO-ADMINISTRATION IN ARTHRITIS

Background

Methotrexate (MTX) is the recommended first-line treatment for rheumatoid arthritis. Adjuvant-induced arthritis (AIA) rat is a robust model with a high prevalence of arthritis used to investigate arthritis. MTX reduces inflammation, but associated with adverse events, such as gastrointestinal, hepatic, and hematology toxicity (1). To reduce these side effects, folic acid (FA) is administrated at distance to MTX with no defined recommendation for its dosing (5-25mg/week) or time point of administration (2) (1-3 days after MTX application). Whether the complicated therapeutic regimen with MTX once a week and FA at another time point affects compliance is an open question. MTX is metabolized in polyglutamates derivates (MTX-PG), which is a biomarker of MTX efficacy as its half-life (1-4 weeks) is longer than MTX (4h) (3).

Objectives

The aim of this study was to assess efficacy and tolerance of co-administration of MTX and FA compared to MTX with FA applied one day after MTX in the AIA.

Methods

Female Lewis rat were randomly divided in three groups and received an injection of Mycobacterium butyricum defining day (D) 0 to induce arthritis. An historic AIA group was used as control. Treatment began on D9, one day before arthritis onset in this model. The first group rats were treated with MTX only (n=13), the second group received MTX and FA at the same day (n=14), and the third group received FA one day after MTX administration (n=14). MTX was administrated intraperitoneally (IP) at 1 mg/kg every 3 days (4) and FA was delivered IP at 0.17 mg/kg. Arthritic index (AI) and ankle circumference (AC) were monitored to assess arthritis. Microcomputed tomography of the ankle was performed to assess bone loss. Moreover, complete blood count, transaminases, and MTX-PG were assessed.

Results

Arthritis developed at D10 in all groups. AI and AC were similar in MTX groups at the various time points. At D17, arthritis severity was lower in MTX groups (AI (mean and standard deviation): 1.4 ± 1.6; AC: 35 ± 7 mm) compared to AIA historical group (AI: 3.3 ± 0.6; AC: 42 ± 4 mm). Bone erosion and bone loss parameters were similar in all groups. Cortical porosity was around 0.40% ± 0.15 and bone volume / total volume was around 0.22% ± 0.13. MTX-PG1 was found at similar levels in MTX groups and correlated negatively with AI in MTX alone or MTX and FA at the same day groups (p<0.05 and p<0.01, respectively). Finally, white and red blood cells, platelets, hemoglobin, mean corpuscular volume, transaminases, and creatinine were found at a similar level in MTX groups.

Conclusion

Co-administration of MTX with FA on the same day is effective compared to FA application one day after MTX. MTX metabolism was not affected, as demonstrated by the MTX-PG concentrations. The biological tolerance between the protocols was comparable. Thus, co-administration of MTX and FA seems to be possible and may be more convenient to the patients and improve compliance at the end.

References

[1]Albrecht K, Müller-Ladner U. Side effects and management of side effects of methotrexate in rheumatoid arthritis. Clin Exp Rheumatol 2010;28:S95-101.

[2]Gaujoux-Viala C, et al. Recommendations of the French Society for Rheumatology for managing rheumatoid arthritis. Joint Bone Spine 2014;81:287–297.

[3]Angelis-Stoforidis P, et al. Methotrexate polyglutamate levels in circulating erythrocytes and polymorphs correlate with clinical efficacy in rheumatoid arthritis. Clin Exp Rheumatol 1999;17:313–320.

[4]Le Goff B, et al. A combination of methotrexate and zoledronic acid prevents bone erosions and systemic bone mass loss in collagen induced arthritis. Arthritis Res Ther 2009;11:R185.

Acknowledgements

We thank Ghislaine Roux, Diane Denis and Valentine Berrucas for their support in animal experiments. We thank Xavier Delavenne for MTX-PG dosage. We thank Nadia Boutahar for transaminases and creatinine dosage.

Disclosure of Interests

Elisa Dalix Grant/research support from: GEBRO, Mathieu Maalouf Grant/research support from: GEBRO, Sylvie Peyroche Grant/research support from: GEBRO, Arnaud Vanden-Bossche Grant/research support from: GEBRO, Charles-Antoine Arthaud Grant/research support from: GEBRO, Sophie Hodin Grant/research support from: GEBRO, Hubert MAROTTE Grant/research support from: GEBRO, Nordic Pharma, Rüdiger Müller Grant/research support from: GEBRO, Nordic Pharma

OP0254 FACTORS ASSOCIATED WITH THE SEVERITY OF COVID-19 INFECTION IN PATIENTS WITH SPONDYLOARTHRITIS: RESULTS OF THE FRENCH RMD COVID-19 COHORT

Background

To our knowledge, no published work has described precisely the severity and evolution of SARS-CoV-2 infection in patients with spondyloarthritis (SpA). Data on COVID-19 from cohorts of patients with immune-mediated inflammatory diseases concern small samples of SpA.

Objectives

Our objective was to describe the severity and course of COVID-19 in a large cohort of patients with SpA, including axial SpA (axSpA) and psoriatic arthritis (PsA), and to identify factors associated with severe forms.

Methods

Patients: individuals with Spondyloarthritis (SpA) from the French RMD COVID-19 cohort (observational, national, multicenter cohort) with a diagnosis of COVID-19 (clinical, PCR, CT or serology) were included.

Data collected: demographics, type of SpA, comorbidities, treatments, severity of COVID-19. Severity of COVID-19 was graded according to care needed: mild = outpatient care; moderate = non-intensive hospital treatment; severe = intensive care unit admission or death; severe = moderate or severe.

Statistical analyses: Logistic regression models were used to identify factors associated with these severe forms. All variables with p <0.20 in the univariate analysis were proposed in the multivariate model. Treatment variables (non-steroidal anti-inflammatory drugs (NSAIDs), methotrexate (MTX), sulfasalazine (SLZ), TNF inhibitors (TNFi), IL-17 inhibitors (IL-17i) and IL-23p19/p40 inhibitors (IL-23p19/p40i)) were included in the models, even if p≥0.20.

Results

Between March 2020 and April 2021, 626 SpAs reported COVID-19 with a mild course in 508 cases (81.1%), moderate in 93 cases (14.8%), and severe in 25 cases (3.9%), including 6 deaths.

The cohort analyzed included 349 women (55.8%), mean age 49.3 ± 14.1 years, mean BMI 27.1 ± 5.4 with 403 axSpA (64.4%), 187 PsA (29.9%) and 36 other SpA, duration of disease 11.3 ± 9.8 years; 352 (56.2%) had at least one comorbidity, of which obesity (23.6%), hypertension (15.5%), and smoking (10.4%) were the most frequent. Among them, 104 were treated with NSAIDs (16.6%), 186 with conventional synthetic disease-modifying antirheumatic drugs (DMARDs) including 156 MTX, and 460 (73.5%) with biological DMARDs (379 TNFi, 57 IL-17i, 15 IL-23p19/p40i, 9 others).

The following variables were associated with severe COVID-19 outcomes: age, body mass index, chronic obstructive lung disease, cardiovascular disease, diabetes, hypertension, interstitial lung disease, renal failure, and corticosteroids intake.

The factors independently associated with severe COVID-19 outcomes were corticosteroid intake (3.15 [CI95%: 1.46-6.76], p 0.004), and age (OR=1.06 [CI95%: 1.04-1.08], p <0.001] while anti-TNF (OR=0.26 [CI95%: 0.09-0.78], p=0.01]) was protective. NSAIDs intake (OR=0.97 [CI95%: 0.48-1.98]), SLZ (OR=7.9 [CI95%: 0.60-103]), or anti-IL17 (OR=0.37 [CI95%: 0.10-1.31]) was not associated with infection severity.

Conclusion

The course of COVID-19 was mild for the majority of SpA patients (81.1%). Corticosteroid intake was associated with more severe COVID-19 outcomes, whereas TNFi were found to be protective.

Disclosure of Interests

None declared

POS0621 MORE THAN 40% OF WOMEN WITH RHEUMATOID ARTHRITIS HAVE A TIME-TO-CONCEPTION LONGER THAN 1 YEAR: ANALYSIS OF THE PROSPECTIVE GR2 COHORT

Background

Rheumatoid arthritis (RA) is one of the most common chronic inflammatory diseases and regularly affects women of childbearing age1. However, there is limited knowledge about the impact of the disease and its treatment on fertility.

Objectives

The aim of the study was to determine factors associated with time-to-conception in women with RA.

Methods

All RA patients (diagnosis according to the Rheumatologist) included in the national multicenter GR2 cohort from 2015 to June 2021 were included in the analysis. Patients could be included either with a pregnancy wish (i.e., preconceptional period) or because of a clinical pregnancy (<12 weeks of gestation): for this analysis, only patients included preconceptionally were included. The main endpoint was time-to-conception, and the secondary endpoints were the number of subfertile patients (i.e., time-to-conception >12 months or non-achievement of pregnancy), as well as the number exposed to csDMARDs and biologics in the preconception period. We performed survival analyses, using a Cox model including a random effect for the center to account for heterogeneity of practices among participating centers. We used a multiple imputation to address missing data among the explanatory variables. Results are presented as a hazard ratio (HR) with confidence interval (CI) to assess associations between the factors studied and time-to-conception.

Results

Among the 167 patients with RA included in the GR2 cohort, 78 were selected for the main analysis of time-to-conception. Of these, 40 (51.3%) had a clinical pregnancy during follow-up. Subfertility was observed in 33 (42.3%) women and median time-to-conception was 19.1 months; mean preconception DAS28-CRP score was 2.3 (+/- 1.2).

Patients were treated during the preconceptional period with NSAIDs, corticosteroids, csDMARDs and biotherapy in 10 (12.8%), 35 (44.9%), 24 (30.8%), and 32 (41.0%) cases, respectively. The multivariate model adjusted for age, BMI, DAS28-CRP, disease duration, ACPA positivity, and exposure to corticosteroids and biologics in the preconception period found an association between increased preconception delay and age (HR (per year) 1. 12 95% CI [1.04-1.16] p = 0.01) as well as disease duration (HR (per year) 1.06 95% CI [1.02-1.15] p = 0.03).

Conclusion

This study provides original results on fertility in women with RA. It found a median time-to-conception of 19.1 months, with a subfertility rate of 42.3%, which is significantly higher than the general population2. In this context, it seems essential to discuss this topic from the beginning of the disease in women of childbearing age.

References

[1]Van den Brandt S. Arthritis Res Ther. 2017;19(1):64.

[2]Junul S. Hum Reprod. 1999;14(5):1250-4.

Table 1.

Survival analyses (Cox model): factors associated with time-to-conception in women with RA.

Univariate analysesMultivariate analysesCrude HR 95% CIpAdjusted HR 95% CIpAge1.11 [1.04-1.18]0.0021.12 [1.04-1.16]0.015BMI1.06 [0.99-1.16]0.1031.08 [0.99-1.16]0.062ACPA positivity1.75 [0.90-3.39]0.1071.44 [0.65-2.86]0.310Disease duration1.03 [0.98-1.08]0.2671.06 [1.02-1.15]0.032DAS28-CRP score1.08 [0.81-1.45]0.5921.08 [0.92-1.32]0.170Corticosteroids0.91 [0.51-1.65]0.7690.86 [0.42-1.68]0.620Biologics1.52 [0.82-2.81]0.1891.30 [0.62-2.78]0.630Figure 1.

Cumulative incidence curves for pregnancies in women with RA.

Acknowledgements

The GR2 Cohort is supported by the French Society of Rheumatology, the French Internal Medicine Society, and unrestricted grants from UCB.

Disclosure of Interests

SABRINA HAMROUN: None declared, Marion Couderc: None declared, Laure Gossec: None declared, Rene-Marc Flipo: None declared, Hubert MAROTTE: None declared, Christophe Richez Speakers bureau: CR has received consulting/speaker’s fees from Abbvie, Amgen, Astra Zeneca, Biogen, BMS, Celltrion, Eli Lilly, Galapagos, GSK, MSD, Novartis, and Pfizer, all unrelated to this manuscript., Aline Frazier-Mironer: None declared, Jérémie SELLAM: None declared, Elisabeth Gervais: None declared, Valerie Devauchelle-Pensec: None declared, Alban Deroux: None declared, Rakiba Belkhir: None declared, AZEDDINE DELLAL: None declared, Laëtitia Dunogeant: None declared, Cédric Lukas: None declared, Emmanuel Chatelus: None declared, Nathalie Costedoat-Chalumeau: None declared, Anna Moltó: None declared

OP0127 UNFAVORABLE PREGNANCY OUTCOME IS SIGNIFICANTLY ASSOCIATED WITH CORTICOSTEROID EXPOSURE DURING PREGNANCY IN WOMEN WITH RHEUMATOID ARTHRITIS: ANALYSIS OF THE PROSPECTIVE GR2 COHORT

Background

Rheumatoid arthritis (RA) is one of the most common chronic inflammatory diseases and regularly affects women of childbearing age1. However, there is limited knowledge about the impact of the disease and its treatment on pregnancy.

Objectives

The aim of the study was to determine the factors associated with adverse pregnancy outcome in women with RA.

Methods

All RA patients (diagnosis according to the Rheumatologist) included in the national multicenter GR2 cohort from 2015 to June 2021 were included in the analysis. Patients could be included either with a pregnancy wish (i.e., preconceptional period) or because of a clinical pregnancy (<12 weeks of gestation). The main endpoint was favorable pregnancy outcome, a composite outcome defined as a live birth at term ≥ 37 gestation weeks of a healthy newborn with a weight greater than the 10th percentile. Disease activity was defined by a DAS28-CRP score > 3.2 at least once during pregnancy. We performed a multilevel logistic regression model, in which we considered patient and center random effects (patient random effect for some women included in the cohort two times). We used a multiple imputation procedure to address missing data among the explanatory variables. Results are presented as an odds ratio (OR) with confidence interval (CI).

Results

Among the 167 pregnancies in women with RA included in the GR2 cohort, 92 were retained for analysis of obstetrical outcome. Of these, 43 (46.2%), 8 (7.9%), 40 (43.5%) were exposed to corticosteroid, NSAID and biologics at least once during pregnancy, respectively. A moderate or severe disease activity at least once during pregnancy was found in 20 (21.8%) pregnancies. A live birth was found in 83 (90.2%) women, including 69 (83.1%) full-term births. Early miscarriages were observed in 9 (0.1%) women. A caesarean section was performed in 22 (23.9%) cases.

A favorable pregnancy outcome was found in 52 (56.5%) of the women. Unfavorable pregnancy outcome was mainly due to prematurity and small for gestational age, observed in 14 (16.9%) and 17 (20.5%), respectively. The multivariate model adjusted for age, BMI, nulliparity, active disease during pregnancy, smoking, and exposure to biologics and corticosteroids during pregnancy found an association between an unfavorable pregnancy outcome and nulliparity (OR 6.2 95% CI [2.1-17.8] p = 0.002), age (OR (per year) 1.1 95% CI [1.0-1.3] p = 0.02) and exposition to corticosteroids during pregnancy (OR 3.2 95% CI [1.1-9.6] p = 0.04).

Conclusion

This study provides original results on pregnancy in women with RA. It found a favorable pregnancy outcome in 56.5% of women. Unfavorable pregnancy outcome was associated with age, nulliparity and corticosteroids use during pregnancy, which argues for their careful use during pregnancy.

References

[1]Van den Brandt S. Arthritis Res Ther. 2017;19(1):64.

Table 1.

Multilevel logistic regression model: factors associated with unfavorable pregnancy outcome in women with RA.

Univariate analysesMultivariate analysesCrude OR 95% CIpAdjusted OR 95% CIpAge1.09 [1.01-1.19]0.0361.14 [1.02-1.28]0.019BMI0.93 [0.83-1.04]0.1960.91 [0.77-1.08]0.204Nulliparity4.18 [1.66-10.53]0.0036.16 [2.13-17.76]0.002Smoking1.08 [0.29-3.36]0.9961.65 [0.37-7.22]0.141Disease activity*1.06 [0.40-2.81]0.9110.98 [0.21-2.28]0.753Corticosteroids**2.45 [1.05-5.68]0.0393.22 [1.09-9.57]0.038Biologics**1.05 [0.11-3.54]0.5892.02 [0.70-4.12]0.194

* Moderate or severe disease activity at least once during pregnancy.

** Use at least once during pregnancy

Acknowledgements

The GR2 Cohort is supported by the French Society of Rheumatology, the French Internal Medicine Society, and unrestricted grants from UCB.

Disclosure of Interests

SABRINA HAMROUN: None declared, Marion Couderc: None declared, Rene-Marc Flipo: None declared, Laure Gossec: None declared, Christophe Richez Speakers bureau: CR has received consulting/speaker’s fees from Abbvie, Amgen, Astra Zeneca, Biogen, BMS, Celltrion, Eli Lilly, Galapagos, GSK, MSD, Novartis, and Pfizer, all unrelated to this manuscript., Rakiba Belkhir: None declared, Aline Frazier-Mironer: None declared, Valerie Devauchelle-Pensec: None declared, Hubert MAROTTE: None declared, Jérémie SELLAM: None declared, Elisabeth Gervais: None declared, Alban Deroux: None declared, Cédric Lukas: None declared, Emmanuelle Dernis: None declared, Emmanuel Chatelus: None declared, Nathalie Costedoat-Chalumeau: None declared, Anna Moltó: None declared

OP0066 IMPACT OF DIAGNOSIS AND TREATMENT OF TROPHERYMA WHIPPLEI INFECTION IN PATIENTS WITH PRE-EXISTING CHRONIC INFLAMMATORY RHEUMATIC DISEASES: DATA FROM THE NATIONAL Tw-IRD REGISTRY

Background

Tropheryma whipplei (Tw) infection is a rare condition, characterized by inflammatory joint symptoms in more than 75% of the cases, which can lead the physician to diagnose chronic inflammatory rheumatic diseases (IRD) and to initiate DMARDs. DMARDs are often ineffective and may reveal digestive signs, systemic manifestations or involvement of other organs. We hypothesized that treatment of Tw infection has a favorable impact on rheumatologic and extra-rheumatologic manifestations attributed to IRD.

Objectives

To validate this hypothesis, we initiated a registry with the objectives to describe the characteristics of IRD and their treatments, the diagnostic and therapeutic modalities of Tw infections and the impact of the treatment of Tw infection on the evolution of IRD and associated DMARDs.

Methods

We initiated a French National register including adult patients with pre-existing IRD, treated with DMARDs, later diagnosed with Tw infection. Cases were identified through a call for observation via the “Club Rhumatismes et inflammations” website. We collected clinical and biological data about the characteristics of IRD and their treatments, the diagnostic and therapeutic modalities of Tw infections, and the impact of the treatment of Tw infection on the evolution of IRD and associated DMARDs.

Results

Seventy-three IRD patients were included. Mean age at diagnosis was 49 years (SD +/- 10.9), with 78% of men, median IRD duration was 79 months (IQR 36; 140), including rheumatoid arthritis (31 cases), spondyloarthritis (14 cases), psoriatic arthritis (6 cases) and other IRDs (22 cases). All IRD patients were treated with DMARDs, with no therapeutic response in 51% of the cases, worsening of rheumatologic symptoms in 34% of the cases, and occurrence of extra-articular manifestations in 27% of the cases. Screening for Tw infection mainly involved saliva and stool PCR, while diagnostic modalities involved organ specific PCR and biopsies, in particular duodenal biopsies (PCR positive in 87% of cases and histology in only 38% of cases). At the time of Tw infection diagnosis, mean age was 58 years (SD +/- 10.1), all patients had joint involvement, 33% axial involvement, 11% entheseal involvement, 84% extra-articular manifestations, 93% elevated CRP, 86% hypoalbuminemia and 67% anemia. Tw infection treatment modalities (median follow-up of 22 months) mainly involved a combination of doxycycline (95%) and hydroxychloroquine (96%), with complete recovery in 79% of the cases and Tw-related deaths in 2 cases. At the same time, Tw infection treatment was associated with IRD remission in 93% of cases, with a median time to remission of 2 months (IQR 1; 4.25), leading to DMARD withdrawal in 94% of cases and corticosteroid therapy withdrawal in 65% of cases.

Conclusion

A Tw infection should be considered in IRD patients with peripheral joint involvement and inadequate response to DMARDs, particularly in the presence of extra-articular manifestations, elevated CRP and hypoalbuminemia. In such patients, positive results of screening and diagnostic tests for Tw infection may lead to the initiation of Tw infection treatment which is associated with complete recovery of Tw infection and rapid remission of the IRD, allowing DMARD and corticosteroid therapy withdrawal in most the cases.

References

[1]Marth T. Tropheryma whipplei infection and Whipple’s disease. The Lancet Infectious Diseases 2016;16(3):e13–22.

[2]Puéchal X. Whipple’s arthritis. Joint Bone Spine 2016;83(6):631–5.

Acknowledgements

Club Rhumatismes et Inflammations.

Disclosure of Interests

None declared.

Reference microarchitectural values measured by HR-pQCT in a Franco-Swiss cohort of young adult women

Bone microarchitecture assessed by high-resolution peripheral quantitative computed tomography varies across populations of different origin. The study presents a reference dataset of microarchitectural parameters in a homogeneous group of participants aged within 22-27 range determined by a discriminant analysis of a larger cross-sectional cohort of 339 women.

INTRODUCTION

High-resolution peripheral quantitative computed tomography (HR-pQCT) non-invasively measures three-dimensional bone microarchitectural parameters and volumetric bone mineral density. Previous studies established normative reference HR-pQCT datasets for several populations, but there were few data assessed in a reference group of young women with Caucasian ethnicity living in Western Europe. It is important to obtain different specific reference dataset for a valid interpretation of cortical and trabecular microarchitecture data. The aim of our study was to find the population with the most optimal bone status in order to establish a descriptive reference HR-pQCT dataset in a young and healthy normal-weight female cohort living in a European area including Geneva, Switzerland, Lyon and Saint-Etienne, France.

METHODS

We constituted a cross-sectional cohort of 339 women aged 19-41 years with a BMI > 18 and < 30 kg/m². All participants had HR-pQCT measurements at both non-dominant distal radius and tibia sites.

RESULTS

We observed that microarchitectural parameters begin to decline before the age of 30 years. Based on a discriminant analysis, the optimal bone profile in this population was observed between the age range of 22 to 27 years. Consequently, we considered 43 participants aged 22-27 years to establish a reference dataset with median values and percentiles.

CONCLUSION

This is the first study providing reference values of HR-pQCT measurements considering specific age bounds in a Franco-Swiss female cohort at the distal radius and tibia sites.

L’infection a SARS-CoV-2 pourrait-elle induire des rhumatismes inflammatoires chroniques ? Une enquête nationale française

Introduction

Les cas publiés de manifestations inflammatoires non microcristallines après COVID-19 sont exceptionnels. L’objectif de cette étude est de décrire les caractéristiques clinicobiologiques, morphologiques et de suivi des manifestations rhumatologiques observées suite à une infection récente à SARS-CoV-2.

Matériels et méthodes

Cette étude est une série française ambispective, multicentrique, de patients présentant une infection à SARS-CoV-2, associée à la survenue de manifestations rhumatologiques. Elle provient des premiers résultats d’un registre national (étude du CRI) de patients atteints du SARS-CoV-2. Du 7 avril 2020 au 24 juin 2021, nous avons recueilli les données clinicobiologiques, morphologiques, et la prise en charge de ces manifestations.

Résultats

Cette étude comprend 15 patients ayant présenté une infection à SARS-CoV-2 avec haute suspicion clinique et cas contact confirmé (n = 2), confirmée par PCR nasale (n = 8), trachéale (n = 2), test salivaire (n = 1) ou sérologie (n = 2). L’âge moyen est de 55,1 ans (19–72) ; 53,3 % des patients sont de sexe féminin. Les comorbidités principales sont une hypertension artérielle (40 %), un diabète de type 2 (20 %), une obésité (20 %). Aucun patient ne présente d’antécédent de rhumatisme inflammatoire chronique sous traitement de fond, maladie auto-immune ou déficit immunitaire. Deux patients présentent un antécédent de psoriasis dans l’enfance. Le délai moyen d’apparition des manifestations rhumatologiques est de 32,9 jours après les premiers symptômes d’infection à SARS-CoV-2 et de 23,9 jours après confirmation diagnostique : arthralgies (100 %), myalgies (47 %), au moins une synovite (86 %). L’atteinte est majoritairement bilatérale (67 %), asymétrique (54 %), poly- ou oligo-articulaire (73 %). Trois patients présentent des signes extra-articulaires évoquant une maladie inflammatoire systémique (syndrome néphrotique, engelures, syndrome de Raynaud, papules de Gottron, télangiectasies). Un syndrome inflammatoire biologique est présent chez 80 % des patients. Un patient présente une positivité du facteur rhumatoïde à taux significatif, alors que les anticorps anti-CCP ne sont présents chez aucun d’entre eux. Les anticorps antinucléaires sont positifs à taux significatif (seuil de 1/160) chez 20 % des patients. Les radiographies ne montrent pas d’atteinte structurale. Dix patients ont bénéficié d’une évaluation échographique : sept présentent au moins une synovite, dont quatre avec hyperhémie Doppler. Une seule ponction articulaire de genou a été réalisée, retrouvant un liquide inflammatoire à prédominance de polynucléaires neutrophiles, sans cristaux. Nous disposons de données de suivi pour 12 patients. Le délai de suivi moyen par rapport au j0 des symptômes de l’infection à SARS-CoV-2 est de 15 mois (8–18). L’évolution a été favorable sous AINS ou corticothérapie chez 6 patients, et après abstention thérapeutique chez 2 patients. Parmi les 12 patients suivis, 6 ont présenté une rechute articulaire. Quatre patients ont fait l’objet d’un traitement de fond : csDMARDS (méthotrexate n = 1, sulfasalazine n = 1), bDMARDS (infliximab n = 1, adalimumab n = 1).

Conclusion

Notre étude rapporte plusieurs cas de manifestations rhumatologiques inflammatoires, non érosives, après une infection à SARS-CoV-2, après un suivi moyen de 15 mois. Elles restent exceptionnelles. L’hypothèse d’une arthrite de type réactionnelle ne peut être écartée.

Facteurs associés à la sévérité de l’infection COVID-19 chez les patients atteints de spondyloarthrite : résultats de la French RMD COVID-19 cohorte

Introduction

Il n’y a pas, à notre connaissance, de publication décrivant précisément la sévérité et l’évolution de l’infection à SARS-CoV-2 dans la spondyloarthrite (SpA). Les données sur la COVID-19 issues des cohortes de patients avec maladies inflammatoires à médiation immunitaire concernent de faibles effectifs de SpA.

Notre objectif était de décrire la sévérité et l’évolution de la COVID-19 dans une large cohorte de patients atteints de SpA (SpA axiale et rhumatisme psoriasique) et d’identifier les facteurs associés aux formes sévères.

Patients et méthodes

Patients : spondyloarthrites (SpA) de la French RMD COVID-19 cohort (cohorte observationnelle, nationale, multicentrique) avec un diagnostic de COVID-19 (clinique, PCR, scanner ou sérologie). Données collectées : démographiques, type de SpA, comorbidités, traitements, gravité de la COVID-19. La gravité de la COVID-19 a été classée en fonction des soins nécessaires : bénin = soins ambulatoires ; modéré = traitement hospitalier non intensif ; sévère = admission en unité de soins intensifs ou décès ; grave = modéré ou sévère.

Analyses statistiques : des modèles de régression logistique ont été utilisés pour identifier les facteurs associés à ces formes graves. Toutes les variables avec p < 0,20 en analyse univariée ont été proposées dans le modèle multivarié. Les variables de traitement (AINS, méthotrexate [MTX], sulfasalazine [SLZ], anti-TNF et anti-IL17) étaient incluses dans les modèles, même si p ≥ 0,20.

Résultats

Entre mars 2020 et avril 2021, 626 SpA ont déclaré une COVID-19 dont l’évolution avait été bénigne dans 508 cas (81,1 %), modérée dans 93 cas (14,8 %) et sévère dans 25 cas (3,9 %), dont 6 décès.

La cohorte analysée comprenait 349 femmes (55,8 %), âge moyen 49,3 ± 14,1 ans, IMC moyen 27,1 ± 5,4 avec 403 SpA axiale (64,4 %), 187 RPso (29,9 %) et 36 autres SpA, durée de la maladie 11,3 ± 9,8 ans ; 352 (56,2 %) avaient au moins une comorbidité dont l’obésité (23,6 %), l’hypertension artérielle (15,5 %) et le tabagisme (10,4 %) étaient les plus fréquentes. Parmi eux, 104 étaient traités par AINS (16,6 %), 186 par csDMARD dont 156 méthotrexate, et 460 (73,5) % par biomédicaments (379 anti-TNF, 57 anti-IL17 : 57, 15 anti-IL12/23, 9 autres).

Les facteurs indépendamment associés à une COVID-19 grave étaient la corticothérapie (OR = 2,83 [IC95 % : 1,41–5,66]) et l’âge (OR = 1,07 [1,05–1,09]) alors le genre féminin (OR = 0,64 [0,41–0,99]) et les anti-IL17 (OR = 0,51 [0,32–0,81]) avaient un caractère protecteur. Un traitement par AINS (OR = 0,91 [IC95 % : 0,47–1,77]), par sulfasalazine (OR = 6,81 [0,59–77,41]) ou par anti-TNF (OR = 0,67 [0,33–1,35]) n’était pas associé à la gravité de l’infection.

Conclusion

L’évolution de la COVID-19 a été bénigne pour la majorité des patients atteints de SpA (81,1 %). La corticothérapie était associée à des infections plus graves alors que les anti-IL17 avaient un caractère protecteur.

OP0284 OUTCOME OF COVID-19 IN PATIENTS WITH RHEUMATIC AND INFLAMMATORY DISEASES TREATED WITH RITUXIMAB: DATA FROM DE FRENCH RMD COVID-19 COHORT

Background:

Various observations have suggested that the course of the COVID-19 infection may be less favorable in patients with inflammatory rheumatic and musculoskeletal diseases (iRMD) receiving rituximab (RTX).

Objectives:

To investigate whether treatment with RTX is associated with severe infection and death.

Methods:

We performed an observational, multicenter, French national cohort study querying the French RMD COVID-19 cohort, including highly suspected/confirmed iRMD-COVID-19 patients. The primary endpoint was to assess the severity rate of COVID-19. Severe disease was defined by hospitalization in intensive care unit or death. The secondary objectives were to analyze death rate and length of hospital stay. Two control groups were considered for comparison with RTX treated patients: a first group including all non-RTX treated iRMD patients and a second consisting on RTX untreated iRMD patients with diseases for which RTX is a recognized therapeutic option. Adjusting on potential confounding factors was performed by using inverse probability of treatment weighting (IPTW) propensity score method.

Results:

We collected a total of 1090 records. Patients were mainly females (67.3%, 734/1090) with a mean age of 55.2±16.4 years, and 51.1% (557/1090) were over the age of 55. Almost 70% of the population had at least one comorbidity (756/1090). A total of 63 patients were treated with RTX, mainly for rheumatoid arthritis (RA) (31/63, 49.2%). RTX treated patients were more likely to be males, with older age, higher prevalence of comorbidities and corticosteroid use. The control population consisted on 1027 non-RTX treated iRMD patients, and 495 RTX untreated iRMD patients with diseases for which RTX is a recognized therapeutic option.

Of the 1,090 patients, 137 developed severe disease (12.6%). After adjusting on potential confounding factors (age, sex, arterial hypertension, diabetes, smoking status, body mass index, interstitial lung disease, cardiovascular diseases, cancer, corticosteroid use, chronic renal failure and the underlying disease), severe disease was confirmed to be observed more frequently in patients receiving RTX compared to all RTX untreated iRMD patients (effect size, ES 3.26, 95% confidence interval, CI 1.66 to 6.40, p<0.001) and the subgroup of untreated RTX patients with diseases eligible for RTX therapy (ES 2.62, 95% CI 1.34 to 5.09, p=0.005). Patients who developed a severe disease had a more recent rituximab infusion compared to patients with mild or moderate disease. Indeed, the time between the last infusion of rituximab and the first symptoms of COVID-19 was significantly shorter in patients who developed a severe form of COVID-19 (Figure 1).

Figure 1.

Distribution (Tukey’s box plot) of Lag time between last infusion of Rituximab according to disease severity. P-Values for comparison between disease severity with Kruskal Wallis test are reported; P-Value<0.001 for either post-hoc comparison of severe disease group with moderate or mild disease group (calculated using Dunn’s test).

Eighty-nine patients in our cohort died, resulting in an overall death rate of 8.2%. Death rate was numerically higher in RTX treated patients (13/63, 20.6%) compared to all RTX untreated iRMDs patients (76/1027, 7.4%) and the subgroup of untreated RTX patients with diseases eligible for RTX therapy (49/495, 9.9%). After considering the previously described confounding factors, the risk of death was not significantly increased in patients treated with RTX compared to all RTX untreated iRMDs patients (ES 1.32, 95% CI 0.55 to 3.19, p=0.53) (Table 2) and the subgroup of untreated RTX patients with diseases eligible for RTX therapy (ES 1.48, 95% CI 0.68 to 3.20, p=0.32). In line with a more severe COVID-19 disease, the length of hospital stay was markedly longer in patients treated with RTX compared to both untreated RTX patient groups.

Conclusion:

RTX therapy is associated with a more severe COVID-19 infection. RTX will have to be applied with particular caution in patients with iRMDs.

Acknowledgements:

Muriel Herasse played a major role in collecting the missing data of the cohort.

We thank Julien Labreuche (biostatistician, CHU-Lille) for the help in the statistical analysis.

Disclosure of Interests:

None declared

POS0068 HIGH LEVELS OF PORPHYROMONAS GINGIVALIS AND PREVOTELLA INTERMEDIA ANTIBODIES IN CHILDREN WITH JUVENILE IDIOPATHIC ARTHRITIS

Background:

Idiopathic juvenile arthritis (JIA) is a heterogeneous group of pathologies whose origin remains unknown at present (1). They are characterised by a systemic inflammatory and joint disease affecting children under 16 years of age. The current classification groups the different forms of JIA into 7 distinct entities (systemic forms, polyarticular forms with or without rheumatoid factors, oligoarticular forms, inflammatory arthritis associated with enthesopathies (ERA), arthritis associated with psoriasis and unclassifiable arthritis). Exact etiology of JIA is still unknown. To date, the various hypotheses put forward on the occurrence of JIAs integrate the genetic and environmental framework.

The link between periodontal disease and rheumatoid arthritis (RA) is largely reported. Recently, Porphyromonas gingivalis (P. gingivalis) infection explained the occurrence of arthritis in rodent and in RA (2). Several studies mention the beneficial effect of P. gingivalis treatment on disease activity.

Currently, there are very few studies on the prevalence of P. gingivalis in patients with JIA and the possible involvement of the germ in the development of inflammatory joint diseases in the pediatric population(3)(4).

Objectives:

The objective of our study is to determine presence of high IgG antibodies against P. gingivalis and Prevotella Intermedia in a cohort of patients with JIA compared to a control population and to determine variation of level according to sub-classes of JIA.

Methods:

Sera were obtained from 101 patients satisfying the ILAR classification criteria for JIA and in 25 patients with two other dysimmune disorders (type 1 diabetes and juvenile inflammatory bowel disease). Level of IgG antibodies against P. gingivalis and Prevotella Intermedia were obtained by homemade ELISA already used previously (5).

Results:

In the JIA group, major children were oligarthritis (47.5%), polyarthritis represents 31.7% of JIAs, ERA and systemic forms of JIA are respectively 9 and 11%. For the control group, 10 (40%) children had diabetes and 15 (60%) had IBD.

Levels of anti-P. gingivalis anti-Prevotella Intermedia antibodies were higher in AJI group compared at control groups (P<0.01, P<0.05). Theses difference are mainly related to oligoarthritis and ERA subsets for both P. gingivalis and Prevotella Intermedia.

Figure 1.

Relative titer of antibodies to P. gingivalis and anti Prevotella intermedia. *: P<0.05; **: P<0.01; ***: P<0.001. P. gingivalis (control vs oligoarthritis p= 0.0032. control vs ERA p= 0.0092). Prevotella intermedia (control vs oligoarthritis p= 0.0194. control vs ERA p= 0.0039).

Conclusion:

We confirmed high level of anti-P. gingivalis and anti-Prevotella intermedia antibodies in JIA compared to other inflammatory disorders. For the first time, we observed that this high level was mainly in oligoarthritis and ERA. Further investigations are required to investigate involvement of oral dysbiosis in AJI pathogenesis. As observed in RA, it could be a new way to integrate in JIA therapy management.

References:

[1]Thatayatikom A, De Leucio A. Juvenile Idiopathic Arthritis (JIA). StatPearls Publishing; 2020

[2]Cheng Z, Meade J, Mankia K, Emery P, Devine DA. Periodontal disease and periodontal bacteria as triggers for rheumatoid arthritis. Best Pract Res Clin Rheumatol. 2017;31(1):19–30.

[3]Romero-Sánchez C, Malagón C, Vargas C, Fernanda Torres M, Moreno LC, Rodríguez C, et al. Porphyromonas Gingivalis and IgG1 and IgG2 Subclass Antibodies in Patients with Juvenile Idiopathic Arthritis. J Dent Child Chic Ill. 2017 May 15;84(2):72–9.

[4]Lange L, Thiele GM, McCracken C, Wang G, Ponder LA, Angeles-Han ST, et al. Symptoms of periodontitis and antibody responses to Porphyromonas gingivalis in juvenile idiopathic arthritis. Pediatr Rheumatol Online J. 2016 Feb 9

[5]Rinaudo-Gaujous M, Blasco-Baque V, Miossec P, Gaudin P, Farge P, Roblin X, et al. Infliximab Induced a Dissociated Response of Severe Periodontal Biomarkers in Rheumatoid Arthritis Patients. J Clin Med. 2019 May 26;8(5).

Disclosure of Interests:

None declared.

POS0592 BIOSIMILAR INFLIXIMAB THERAPY IN RHEUMATOID ARTHRITIS, ANKYLOSING SPONDYLITIS AND PSORIATIC ARTHRITIS: A LONG-TERM FOLLOW-UP STUDY ON INFLIXIMAB-NAÏVE PATIENTS AND PATIENTS SWITCHED FROM THE ORIGINATOR TO THE BIOSIMILAR CT-P13

Objectives:

The ReFLECT study was carried out to investigate real life use of CT-P13, the first monoclonal antibody biosimilar to infliximab originator.

Methods:

ReFLECT is a multicentre, prospective, observational study conducted in France to determine the characteristics of patients (pts) receiving CT-P13, its effectiveness and safety in a real-life setting. Eligible were both pts who had been switched from infliximab originator (IFXS) and infliximab-naïve pts started on CT-P13 (IFXN). Intermediate results in pts with rheumatic diseases using descriptive statistical analyses from inclusion to a 24-month-follow-up period are described here.

Results:

Among the 1370 adult pts included between October 2016 and April 2019, data were analysed for 142 pts with rheumatoid arthritis (RA; 23.9% males; mean age: 61.5±10.9 years; 80 IFXN/61 IFXS), 411 with ankylosing spondylitis (AS; 58.9% males; 48.1±13.1 years; 228 IFXN/179 IFXS), and 96 with psoriatic arthritis (PsA; 41.7% males; 53.4±14.1 years; 55 IFXN/40 IFXS) after a median of between 8.7 and 12.9 years since diagnosis. At inclusion, 66.9%, 24.3% and 50% had one concomitant treatment, mainly methotrexate. At the time of the first administration of CT-P13, disease had been active in 94.2%, 76.6% and 83.3% of the IFXN pts compared with 45.3%, 24.0% and 32.0% of the IFXS pts. After 24 months of follow-up, both CRP levels and disease activity remained stable compared to baseline in the IFXS pts: median ΔCRP/RAS = 1, ΔCRP/ASS = 0, ΔCRP/PsAS = 0 and median ΔDAS28/RAS = 0, ΔBASDAI/ASS = 0, ΔDAS28/PsAS = 1. As expected, improvement was observed in the IFXN pts: median ΔCRP/RAN = -2, ΔCRP/ASN = -2, ΔCRP/PsAN = 0 and median ΔDAS28/RAN= -1, ΔBASDAI/ASN = -2, ΔDAS28/PsAN = -1. In naïve pts, CT-P13 brought both CRP and disease activity down to levels comparable to those seen in patients switched from the IFX originator: median [Q1;Q3] CRP/RAN = 5 [2;9] vs CRP/RAS = 5 [2;8]; CRP/ASN = 3 [2;9] vs CRP/ASS = 3 [2;5]; CRP/PsAN = 2 [1;4] vs CRP/PsAS = 2 [1;3] and median [Q1;Q3] DAS28/RAN = 3 [3;4] vs DAS28/RAS = 2 [2;3]; BASDAI/ASN = 3 [1;4] vs BASDAI/ASS = 2 [1;3]; DAS28/PsAN = 3 [2;3] vs DAS28/PsAS = 3 [2;3], (Figure 1). At 24 months, 62.7%, 74.9% and 75.0% remained on treatment with CT-P13 in RA, AS and PsA pts respectively. The reasons for treatment withdrawing and safety results are reported in Table 1.

Figure 1.

Disease activity at the time of first administration of CT-P13 and after 24 months

Table 1.

Reasons for treatment withdrawing and safety results

Reasons for treatment withdrawingIFX pts, n (%)RAN, 37RAs, 15ASN,64ASs, 38PsAN,11PsAs, 12Missing100001Remission02 (13.3)2 (3.1)000Therapeutic failure *25 (69.4)8 (53.3)41 (64.1)23 (60.5)5 (45.5)6 (54.5)Intolerance5 (13.9)1 (6.7)5 (7.8)2 (5.3)1 (9.1)1 (9.1)Other6 (16.7)4 (26.7)16 (25.0)13 (34.2)5 (45.5)4 (36.4)Safety data, n (%)RA, 142AS, 411PsA, 96Pts with ≥ 1 adverse event (AE)67 (47.2)222 (54.0)48 (50.0)Pts with ≥ 1 AE related to CT-P1333 (23.2)84 (20.4)19 (19.8)Pts with ≥ 1 serious AE19 (13.4)42 (10.2)9 (9.4)Pts with ≥ 1 allergic infusion reaction †4 (2.8)11 (2.7)1 (1.0)Pts with ≥ 1 infection ‡10 (7.0)18 (4.4)3 (3.1)

* Encompassing primary failure in IFXN pts and loss of efficacy in IFXN and IFXS pts

† Including acute and delayed hypersensitivity reactions

‡ Including severe infections, tuberculosis, opportunistic infections, hepatitis

Conclusion:

Year 2 follow-up data indicate that CT-P13 effectively induced improvement in disease activity in pts with RA, AS and PsA receiving infliximab for the first time and maintained stable disease activity in pts switched from IFX originator to CT-P13. This real-life study did not highlight any new safety concerns.

Disclosure of Interests:

Bruno Fautrel Speakers bureau: Lilly, SOBI, Roche-Chugaï, Consultant of: AbbVie, Biogen, BMS, Boehringer Ingelheim, Celgene, Lilly, Janssen, Medac, MSD France, Nordic Pharma, Novartis, Pfizer, Roche-Chugaï, Sanofi Aventis, SOBI and UCB, Grant/research support from: AbbVie, Lilly, MSD, Pfizer, Maryse Assing Employee of: Pfizer, Nadir Mammar Employee of: Pfizer, Hubert MAROTTE Speakers bureau: Sanofi-Aventis, Paid instructor for: Sanofi-Aventis, Consultant of: AbbVie, Biogen, Bristol Myers Squibb, Gilead, Lilly France, MSD, Medac, Nordic Pharma, Novartis, Pfizer, Sanofi-Aventis, UCB, Grant/research support from: Bristol Myers Squibb, Lilly France, MSD, Novartis, Nordic Pharma, Pfizer, Sanofi Aventis

AB0277 ASSOCIATION BETWEEN INTRA-ERYTHROCYTE METHOTREXATE POLYGLUTAMATE CONCENTRATION AND CLINICAL RESPONSE IN RHEUMATOID ARTHRITIS IN PATIENTS TREATED WITH METHOTREXATE INJECTABLE SUBCUTANEOUSLY

Background:

Methotrexate (MTX) is the first-line treatment for rheumatoid arthritis (RA). Due to its short half-life, blood MTX dosage is not performed in current practice. Erythrocyte MTX-polyglutamate (MTX-PG), which penetrates into the red blood cells, would be correlated with the area under the MTX curve and would be more accessible for dosing. When treatment is initiated, its concentration correlates with efficacy and therapeutic adherence.

Objectives:

To determine the interest of erythrocyte MTX-PG dosage in case of failure of 1st line MTX treatment.

Methods:

In this single-centre cross-sectional study, RA patients presenting for consultation at the Saint-Etienne University Hospital, with a stable dose of MTX for more than 3 months at least 15 mg/week subcutaneously with either clinical remission (DAS28<2.6) or active disease (DAS28>3.2) were included between July 2nd 2018 and May 28th 2020. In order to assess therapeutic compliance, the patient completed the compliance questionnaire of Rheumatology (CQR) questionnaire. The determination of erythrocytic MTX-PG was performed on a 5 mL blood sample by liquid chromatography method for the determination of the different PG forms.

Results:

Sixty patients were included, 34 in the active RA group and 26 in the RA group in remission. One patient withdrew his consent. Only 16% of patients were observed with a CQR score > 80%. Patients in remission were leaner with a longer duration of disease and MTX treatment. The sex ratio, RA status, creatinine clearance and MTX dose was not different in both groups. The CQR was better in the remission group than in the active RA group. However, total MTX-PG was not different in the two groups. The same results were observed for the different forms of MTX-PG 1, 2, 3 or 4. In contrast, in the remission PR group, total MTX-PG or MTX-PG2, 3, 4 and 5 correlated inversely with BMI while MTX-PG3 correlated positively with BMI in the active PR group. In the active PR group, MTX-PG5 correlated with MTX dose.

Conclusion:

MTX-PG dosage is not a biomarker of good response to MTX in our study. However, compliance is a key factor to be considered in RA with active disease in adapting patient management.

References:

[1]Pasma A, Boer E den, Spijker A van ’t, Timman R, Bemt B van den, Busschbach JJV, et al. Nonadherence to disease modifying antirheumatic drugs in the first year after diagnosis: comparing three adherence measures in early arthritis patients. Rheumatology 2016;55:1812–1819.

[2]Haandel L van, Becker ML, Leeder JS, Williams TD, Stobaugh JF. A novel high-performance liquid chromatography/mass spectrometry method for improved selective and sensitive measurement of methotrexate polyglutamation status in human red blood cells. Rapid Commun Mass Spectrom RCM 2009;23:3693–3702.

[3]Dervieux T. Pharmacogenetic and metabolite measurements are associated with clinical status in patients with rheumatoid arthritis treated with methotrexate: results of a multicentred cross sectional observational study. Ann Rheum Dis 2005;64:1180–1185.

Disclosure of Interests:

Marion POUDRET: None declared, Myriam Norman: None declared, Sophie Hodin: None declared, Adamah Stanislas AMOUZOUGAN: None declared, Karima BOUSSOUALIM: None declared, Astrid Coassy: None declared, Tiphany Neel: None declared, THIERRY THOMAS: None declared, Xavier DELAVENNE: None declared, Hubert MAROTTE Speakers bureau: Pfizer, Nordic, Paid instructor for: Amgen, Consultant of: Novartis, Grant/research support from: Nordic

AB0074 NO DIFFERENCE BETWEEN THE SEROLOGIES OF DENTAL GERMS AND THE PHENOTYPES OF SPONDYLOARTHRITIS WITHIN THE DESIR COHORT

Background:

Porphyromonas gingivalis (PG) and Prevotella Intermedia (PI) are two oral pathogens involved in the pathophysiology of chronic periodontopathies. Several studies have determined a role of these periodontal pathologies in the pathophysiology of rheumatoid arthritis (1) while the link with spondyloarthritis is not clearly established with currently contradictory studies (2,3,4).

Objectives:

The purpose of this work is to investigate a link between spondylarthritis and the presence of chronic periodontopathy evaluated by PG and IP serologies.

Methods:

The positivity and quantity of anti-PG and anti-PI antibodies were determined by ELISA method in patients from the DESIR cohort with one of the spondyloarthritis phenotypes compared to patients with common low back pain (n=50) (population control). Patients with spondyloarthritis were classified according to the following phenotypes (diagnosis chosen at 3 years in the DESIR cohort): axial spondylarthritis (n=126), psoriatic rheumatism (n=101), spondyloarthritis associated with chronic inflammatory bowel disease (n=36), undifferentiated spondylitis (n=241), reactive arthritis (n=2), SAPHO (acronym for Synovite, Acne, Pustulose, Hyperostosis and Osteitis. Correlations between patients’ serological status, and smoking status, body mass index and age were sought.

Results:

According to the general characteristics, the control group was significantly older than the axial spondylarthritis (p<0.001), and the psoriatic rheumatism group had a higher body mass index than the control population (p<0.05). The positivity and concentration of anti-PG and anti-PI antibodies were similar between different groups of spondyloarthritis compared to the control group. However, the subgroups “reactive arthritis” and “SAPHO” could not be analyzed due to insufficient staffing. Smoking status and body mass index were not related to antibody concentrations, however there was a statistical correlation between anti-PG antibody concentrations and age. After adjusting on age, there was no difference between the axial spondylarthritis and control groups.

Conclusion:

Our results therefore suggest the absence of a link between periodontal germs involved in chronic periodontopathies and spondyloarthritis, provided that the two phenotypes involving germs in their pathophysiology could not be analyzed. Our results from a population of beginner spondylitis are therefore not in favour of the involvement of oral flora in the pathophysiology of spondyloarthritis, as is the case in rheumatoid arthritis (1). The associations found previously could therefore be favoured by a systemic inflammatory phenomenon. In conclusion, our study suggests no link between chronic periodontopathy and the occurrence of spondyloarthritis. However, the effect of chronic periodontopathy on the evolution of spondylarthritis remains to be explored.

References:

[1]Detert J, Pischon N, Burmester GR, Buttgereit F. The association between rheumatoid arthritis and periodontal disease. Arthritis Res Ther. 2010;12(5):218.

[2]Ratz T, Dean LE, Atzeni F, Reeks C, Macfarlane GJ, Macfarlane TV. A possible link between ankylosing spondylitis and periodontitis: a systematic review and meta-analysis. Rheumatology. 2015;54(3):500-10.

[3]Pischon N, Pischon T, Gülmez E, Kröger J, Purucker P, Kleber B-M, et al. Periodontal disease in patients with ankylosing spondylitis. Ann Rheum Dis. janv 2010;69(01):34-8.

[4]Sezer U, Erciyas K, Pehlivan Y, Üstün K, Tarakçioğlu M, Şenyurt SZ, et al. Serum cytokine levels and periodontal parameters in ankylosing spondylitis: Ankylosing spondylitis and periodontal diseases. J Periodontal Res. juin 2012;47(3):396-401.

Disclosure of Interests:

None declared

SAT0115 COGNITIVE IMPAIRMENT WAS FREQUENT IN PATIENTS WITH RHEUMATOID ARTHRITIS STARTING A BIOLOGIC, WITH SIMILAR RATES OF INHIBITION OR OVERSTIMULATION: AN ANALYSIS OF 84 PATIENTS FROM THE SARIPRO STUDY

Background:

While cognitive impairment is an issue for patients with rheumatoid arthritis (RA), there are few data available on its frequency and possible link with other outcomes in RA.

Objectives:

To assess cognitive impairment in RA and its association with RA and patients’ characteristics.

Methods:

The SariPRO study (NCT 03449758) was a French multicenter study assessing the effects of sarilumab 200 mg on patient-reported outcomes in patients with moderately to severely active RA with an inadequate response or intolerance to conventional synthetic or biologic DMARDs. This report focuses on baseline data.

The main outcome of this analysis was cognitive impairment evaluated by the cognitive sub-score of the patient-reported multidimensional assessment of mood disorders (MAThyS) scale. This sub-score is scored between 0 (i.e. feeling that thoughts occur slower) and 40 (i.e. racing thoughts) where 20 is the best state. This sub-score was analyzed by tertiles where the lowest tertile indicates general inhibition, the second tertile includes normal states, and the highest tertile indicates general excitation. In addition to the MAThyS total score and sub-scores (Cognition, Emotion, Psychomotricity, Motivation and Sense perception), age, gender, duration of RA, methotrexate use, antibody status (rheumatoid factor/ACPA positivity), fatigue (FACIT-F), anxiety /depression (HADS), as well as patient global assessment (PGA) were collected. Cognitive impairment was defined as inhibition (first tertile) and excitation (third tertile). The association between cognitive inhibition and patients’ characteristics (demographic, psychological and disease activity) was estimated by univariate and multivariate logistic regression in an exploratory analysis. There was no imputation of missing data.

Results:

In all 84 patients were included, characteristics at baseline were as expected for an RA population initiating a biologic: mean (SD) age: 59.1 (12.3) years, 75.0% female, disease duration 10.0 (10.3) years, rheumatoid factor positivity 76.1%, ACPA positivity 81.3%, and DAS28 5.0 (1.1). The mean (SD) MAThyS cognition score was 18.2 (4.9). In the exploratory multivariate analysis, factors associated to cognitive inhibition were depression (HADS depression score ≥ 8, odds ratio, OR=3.15 [95% confidence interval, 1.16; 8.59], p=0. 025), emotion inhibition (lower tertile of the MATHYS emotion regulation: OR=4.76 [1.54; 14.28]; p=0.007) and low motivation (lower tertile of motivation: OR=4.17 [1.54; 11.11]; p=0.005).

Conclusion:

Cognitive impairment was frequent in this population of patients with active RA, with similar rates of cognitive inhibition and cognitive excitation. The results suggest that there may be an association between cognitive inhibition, depression, emotion dysregulation and absence of motivation. Unexpectedly, this exploratory analysis did not show an association between cognition impairment and demographic characteristics or disease activity.

References:

[1]Study was sponsored by Sanofi Genzyme

Disclosure of Interests:

Hubert MAROTTE Grant/research support from: Bristol Myers Sqibb, Lilly France, MSD, Novartis, Nordic Pharma, Pfizer, SanofiAventis, Consultant of: AbbVie, Amgen, Bristol Myers Sqibb, Lilly France, MSD, Novartis, Nordic Pharma, Pfizer, SanofiAventis, Paid instructor for: Sanofi-Aventis, Speakers bureau: Sanofi-Aventis, Florence E Lévy-Weil Employee of: Sanofi Genzyme employee, René-Marc Flipo Consultant of: Johnson and Johnson, MSD France, Novartis, Sanofi, Speakers bureau: Johnson and Johnson, MSD France, Novartis, Sanofi, Thierry Schaeverbeke: None declared, Eric Fakra Consultant of: Janssen, Lundbeck, Otsuka, Sanofi, Laure Gossec Grant/research support from: Lilly, Mylan, Pfizer, Sandoz, Consultant of: AbbVie, Amgen, Biogen, Celgene, Janssen, Lilly, Novartis, Pfizer, Sandoz, Sanofi-Aventis, UCB

FRI0095 SARILUMAB IMPROVED PATIENT-PERCEIVED IMPACT OF RHEUMATOID ARTHRITIS WHATEVER THE BASELINE DISEASE ACTIVITY: FIRST RESULTS FROM AN INTERVENTIONAL NON CONTROLLED STUDY: SARIPRO, IN MODERATE AND SEVERE RHEUMATOID ARTHRITIS PATIENTS

Background:

Sarilumab, an anti-IL-6R antibody, is approved for the treatment of moderate to severe RA and shown efficacy on disease activity and patient-reported outcomes (PROs). Detailed analyses of drug efficacy from the patient point of view is important. SariPRO is a pragmatic interventional study close to the daily practice.

Objectives:

To assess the effectiveness of sarilumab on several PROs using the RAID (Rheumatoid Arthritis Impact of Disease) score.

Methods:

The SariPRO study (NCT 03449758) was a French multicenter interventional study assessing the effects of sarilumab 200 mg on PROs in patients with moderately to severely active RA with an inadequate response or intolerance to conventional synthetic or biologic DMARDs. The primary endpoint was change in total RAID score from baseline to week 24 (RAID ranges 0-10 where 10 is maximal impact). Changes from baseline for RAID, DAS28-ESR and CDAI according to baseline disease activity were analyzed as secondary outcomes. Safety was assessed by monitoring adverse events (AE). All statistical analyses were descriptive, 95% CI was given when appropriate.

Results:

84 patients were included in 31 centers and 62 were evaluable and analyzed for effectiveness. They had similar characteristics to the 84 patients at baseline and were as expected for an RA population initiating a biologic: mean (SD) age: 59.9 (12.4) years, 71.0% female, disease duration 9.7 (10.3) years, rheumatoid factor positivity 82.5%, ACPA positivity 86.4%, and DAS28=4.9 (11). Total RAID score decreased significantly from 5.7 (2.0) at baseline to 3.3 (2.5) at W24; mean change was -2.4 [95% CI: -3.0; -1.8]. Furthermore, this improvement was noted both for highly and less active patients at baseline: for patients with DAS28-ESR < 5.1 (n=31), mean change was -1.56 [-2.28; -0.83] and for patients with DAS28-ESR≥5.1 (n=27), mean change was -1.98 [-2.91; -1.05]. Changes in DAS28-ESR and CDAI were significant (-2.8 [-3.2; -2.4] and -15.2 [-18.5; -11.8], respectively). AEs were consistent with the safety profile of anti-IL-6R antibodies and with results from RCTs (data not shown).

Conclusion:

In this real world study, treatment with sarilumab during 24 weeks in RA patients led to an improvement in the total RAID score irrespective of baseline levels of disease activity. This is the first time RAID score is used as the primary endpoint in a study.

References:

[1]Study was sponsored by Sanofi Genzyme

Disclosure of Interests:

Laure Gossec Grant/research support from: Lilly, Mylan, Pfizer, Sandoz, Consultant of: AbbVie, Amgen, Biogen, Celgene, Janssen, Lilly, Novartis, Pfizer, Sandoz, Sanofi-Aventis, UCB, René-Marc Flipo Consultant of: Johnson and Johnson, MSD France, Novartis, Sanofi, Speakers bureau: Johnson and Johnson, MSD France, Novartis, Sanofi, Thierry Schaeverbeke: None declared, Christine Albert: None declared, Athan Baillet Consultant of: Athan BAILLET has received honorarium fees from Abbvie for his participation as the coordinator of the systematic literature review, marie-Christophe Boissier: None declared, Cyrille Confavreux: None declared, Gregoire CORMIER: None declared, Emmanuelle Dernis Speakers bureau: Lilly, Novartis, Elisabeth Gervais Solau: None declared, Sophie Godot: None declared, Jacques-Eric Gottenberg Grant/research support from: BMS, Pfizer, Consultant of: BMS, Sanofi-Genzyme, UCB, Speakers bureau: Abbvie, Eli Lilly and Co., Roche, Sanofi-Genzyme, UCB, Philippe Goupille Grant/research support from: AbbVie, Amgen, Biogen, BMS, Celgene, Chugai, Lilly, Janssen, Medac, MSD France, Nordic Pharma, Novartis, Pfizer, Sanofi and UCB, Consultant of: AbbVie, Amgen, Biogen, BMS, Celgene, Chugai, Lilly, Janssen, Medac, MSD France, Nordic Pharma, Novartis, Pfizer, Sanofi and UCB, Speakers bureau: AbbVie, Amgen, Biogen, BMS, Celgene, Chugai, Lilly, Janssen, Medac, MSD France, Nordic Pharma, Novartis, Pfizer, Sanofi and UCB, Slim Lassoued: None declared, Thierry Lequerre: None declared, Frederic Lioté Consultant of: CME: Nordic Pharma, Christian Marcelli: None declared, Yves Maugars: None declared, Minh Nguyen: None declared, Aleth Perdriger: None declared, Yves-Marie Pers: None declared, Edouard Pertuiset: None declared, Lucile Poiroux: None declared, Carole Rosenberg: None declared, Christian Roux: None declared, Adeline Ruyssen-Witrand Grant/research support from: Abbvie, Pfizer, Consultant of: Abbvie, BMS, Lilly, Mylan, Novartis, Pfizer, Sandoz, Sanofi-Genzyme, Martin SOUBRIER: None declared, Pascale Vergne-Salle: None declared, Charles Zarnitsky: None declared, Eric Fakra Consultant of: Janssen, Lundbeck, Otsuka, Sanofi, Hubert MAROTTE Grant/research support from: Bristol Myers Sqibb, Lilly France, MSD, Novartis, Nordic Pharma, Pfizer, SanofiAventis, Consultant of: AbbVie, Amgen, Bristol Myers Sqibb, Lilly France, MSD, Novartis, Nordic Pharma, Pfizer, SanofiAventis, Paid instructor for: Sanofi-Aventis, Speakers bureau: Sanofi-Aventis, Florence E Lévy-Weil Employee of: Sanofi Genzyme employee

Azathioprine dose reduction in inflammatory bowel disease patients on combination therapy: an open-label, prospective and randomised clinical trial

BACKGROUND

Infliximab (IFX) combined with azathioprine (AZA) is more effective than IFX monotherapy in inflammatory bowel disease (IBD).

AIM

To identify the AZA optimal dose that is required for efficacy when receiving combination therapy.

METHODS

Patients with IBD in durable remission on combination therapy were enrolled in a 1-year, open-label, prospective trial after randomisation into three groups: AZA steady (2-2.5 mg/kg/day, n=28) vs AZA down (dose was halved 1-1.25 mg/kg/day, n=27) vs AZA stopped (n=26). Primary endpoint was failure defined as occurrence of a clinical relapse and/or any change in IBD therapy.

RESULTS

Eighty-one patients were included. Five (17.9%), 3 (11.1%), and 8 (30.8%) patients experienced failure at 1 year in groups AZA steady, AZA down and AZA stopped, respectively (P=.1 across the groups). The median trough levels of IFX at inclusion were close to those measured at the end of follow-up in group AZA steady (3.65 vs 3.45 μg/mL, P=.9) and in group AZA down (3.95 vs 3.60 μg/mL, P=.5), whereas these levels dropped from 4.25 to 2.15 μg/mL (P=.02) in group AZA stopped. Four (14.3%), four (14.8%) and 11 (42.3%) patients experienced an unfavourable evolution of IFX pharmacokinetics in groups AZA steady, AZA down and AZA stopped, respectively. A threshold of 6-TGN <105 pmoles/8.10⁸ RBC was associated with an unfavourable evolution of IFX pharmacokinetics.

CONCLUSIONS

Under combination therapy, AZA dose reduction, but not withdrawal, appears to be as effective as continuation of AZA at full dose.

New-onset vitiligo and progression of pre-existing vitiligo during treatment with biological agents in chronic inflammatory diseases

BACKGROUND

The development of vitiligo during treatment with biological agents is an unusual event and only a few isolated cases have been reported.

OBJECTIVES

To describe the clinical characteristics and evolution of patients developing new-onset vitiligo following initiation of a biological agent for chronic inflammatory disease; and also to report the clinical course of pre-existing vitiligo under biological therapy.

METHODS

This nationwide multicentre, retrospective study, carried out between July 2013 and January 2015, describes the characteristics of a large series of 18 patients (psoriasis N = 8, inflammatory rheumatic diseases N = 8, ulcerative colitis N = 1, uveitis N = 1) who developed new-onset vitiligo while receiving a biological agent.

RESULTS

TNFα inhibitors were the most common biological agent involved (13/18) while anti-IL-12/23 and anti-IL-17 agents or abatacept were less common (4/18 and 1/18 respectively). Mean duration of biological agent exposure before vitiligo onset was 13.9 ± 16.5 months. Outcome was favourable for most patients (15/17) while maintaining the biological agent. Data were also collected for 18 patients (psoriasis N = 5, inflammatory rheumatic diseases N = 10, inflammatory bowel diseases N = 2, SAPHO N = 1) who had pre-existing vitiligo when treatment with a biological agent started (TNFα inhibitors N = 15, ustekinumab N = 1, rituximab N = 1, tocilizumab N = 1). Vitiligo progressed in seven patients and was stable or improved in eight cases.

CONCLUSION

Vitiligo may thus emerge and/or progress during treatment with various biological agents, mainly TNFα inhibitors and could be a new paradoxical skin reaction. De novo vitiligo displays a favourable outcome when maintaining the biological agent, whereas the prognosis seems worse in cases of pre-existing vitiligo.

Spectacular improvement in vitamin D status in elderly osteoporotic women: 8-year analysis of an osteoporotic population treated in a dedicated fracture liaison service

In a population of postmenopausal women with a fragility fracture, we found a drastic reduction in the proportion of women with severe (<25 nmol/L) and moderate (25 to 75 nmol/L) hypovitaminosis D, especially from 2009 onwards. These results show that supplementation has been very widely integrated into current practice.

INTRODUCTION

Vitamin D (25(OH)D) is essential for bone health. In institutionalised osteoporotic women, it reduces the risk of fragility fractures. Numerous articles suggesting the possibility of extraosseous effects have generated a growing number of publications and recommendations on more widespread administration, to limit the risks of moderate or severe hypovitaminosis D. We assessed the impact on clinical practice of these recommendations concerning 25(OH)D supplementation in elderly at-risk populations.

METHODS

A total of 1486 postmenopausal osteoporotic women were seen in the context of a fracture liaison service (i.e. a rheumatology consultation following a peripheral fragility fracture), between May 2005 and December 2012. Of these, 1107 had a 25(OH)D assay (femur, n = 520; humerus, n = 207; wrist, n = 380).

RESULTS

The average age of the total population was 76.7 ± 9.9 years, while for women with an available 25(OH)D assay, the average age was 75.1 ± 11.8 years. The average 25(OH)D (nmol/L) level was similar for the three fracture sites: femur, 30 ± 36.2; humerus, 27.5 ± 24; and wrist, 31 ± 26. A drastic reduction in the proportion of women with severe (<25 nmol/L) and moderate (25 to 75 nmol/L) hypovitaminosis D was observed, especially from 2009 onwards, with a mean prevalence of 69 and 30 % respectively before that year and 35 and 52 % thereafter. Conversely, the proportion of women with 25(OH)D at the threshold value of 75 nmol/L increased from 1.2 to 24 %. Overall, mean serum 25(OH)D levels were significantly higher when comparing the two periods 2005-2008 and 2009-1012 (17.6 ± 14.6 and 48.4 ± 39.2 nmol/L, respectively; p < 0.0001).

CONCLUSION

These results show that supplementation has been very widely integrated into current practice. We can expect it to yield beneficial effects in osseous and extraosseous terms in osteoporotic women, particularly the very elderly.

Combination of C-reactive protein, infliximab trough levels, and stable but not transient antibodies to infliximab are associated with loss of response to infliximab in inflammatory bowel disease

BACKGROUND

Antibodies to infliximab [ATI] and trough levels to infliximab [TRI] are associated with loss of response in inflammatory bowel diseases [IBD]. The best way to predict loss of response [LOR] to infliximab [IFX] is unknown.

METHODS

We conducted a prospective observational cohort study enrolling all IBD patients who were in clinical remission at Week 14 after IFX treatment initiation. TRI, ATI and C-reactive protein [CRP] level were measured at Week 22 [T1] and thereafter at every other IFX infusion. Loss of clinical response was defined by a flare requiring therapeutic change [IFX dose intensification, initiation of another drug class, and/or surgery].

RESULTS

A total of 93 patients [59 Crohn's disease, mean duration of follow-up 17.2 months] were included; 32 patients [34.4%] lost clinical response during follow-up. Cumulative probability of LOR was 50% at 20 months. Mean TRI at T1 was significantly lower in IBD patients with stable ATI as compared with those with transient ATI or without ATI [0.052, 3.34 ,and 4.29 µg/ml, respectively; p = 0.001 between no ATI vs stable ATI, and p = 0.005 between stable and transient ATI] [p = 0.0001]. Three independent factors were predictive of LOR after Cox proportional hazards modelling: TRI > 5.5 µg/ml (hazard ratio [HR]: 0.21; 95% confidence interval [CI]: 0.05-0.89;p = 0.034) at T1, CRP > 5mg/l [HR: 2.5; 95% CI: 1.16-5.26; p = 0.019] at T1, and stable ATI defined by two consecutive ATI > 20ng/ml [HR: 3.77; 95% CI: 1.45-10.0; p = 0.007]. Transient ATI did not influence LOR.

CONCLUSIONS

LOR can be predicted based on a combination of CRP, TRI and stable ATI with a high degree of accuracy.