Background: There are limited data regarding therapy discontinuation in patients with giant cell arteritis (GCA) followed for a long period of time (>5 years). Objectives: We aimed to evaluate the long-term rate and factors associated with therapy discontinuation in a real-life GCA cohort. Methods: Patient and treatment data were collected at 3 different time points (baseline, 2 and 5 years later) in a multicenter, prospective GCA cohort. The rates of therapy discontinuation and its predictors were identified by uni- and multi- variable logistic regression analyses Results: 89 patients with data available for all time points of evaluation were included in the study; 75% were women with a mean age of 71.6±8.1 years and a median disease duration from diagnosis to last evaluation of 7.2 years. All patients had been treated initially with glucocorticoids (GCs, 100%, median daily dose: 40mg prednisolone) while during follow-up, 34% (30/89) had received cs-DMARDs and/or b-DMARDs (17%, 15/89), respectively. 40.4% (36/89) of patients had ≥1 disease relapses during this period. At the last follow-up visit, 48% (43/89) of patients managed to discontinue all therapies and 54% GCs (48/89), respectively. By uni-variable logistic regression analysis, factors associated with therapy discontinuation included the absence of relapses (OR=2.76; 95%CI: 1.13-6.74) and the lack of treatment with GC-sparing agents (b- and/or cs-DMARDs, OR=4.69; 95%CI: 1.76-12.50). By multi-variable logistic regression analysis, patients who did not require a GC-sparing agent (b- and/or cs-DMARDs) had a higher likelihood to discontinue all therapies (OR=5.15; 95%CI: 1.56.-17.01). Conclusion: In this real-life GCA cohort, approximately 40% of patients relapsed while only half were able to discontinue all therapies, ~7 years after diagnosis. The absence of relapses and requirement for GC-sparing agents were associated with a higher likelihood of drug discontinuation. Whether earlier introduction of GC-sparing agents could improve the chance for drug discontinuation by preventing relapses remains to be studied. REFERENCES: NIL. Acknowledgements: Supported in part by the Greek Rheumatology Society and Professional Association of Rheumatologists (ERE-EPERE) and the Special Account for Research Grants (S.A.R.G.), National and Kapodistrian University of Athens, Athens, Greece (DV #12085, 12086). Disclosure of Interests: None declared.

Background: There are limited real-life data regarding relapses and serious adverse event (SAEs) rates in patients with microscopic polyangiitis (ΜPA) and granulomatosis with polyangiitis (GPA) receiving rituximab (RTX) for remission-maintenance. Objectives: To estimate the incidence of relapses and SAEs as well the factors associated with the time to relapse in MPA/GPA patients under RTX-maintenance. Methods: Retrospective study of newly diagnosed/relapsing GPA/MPA patients who received RTX-maintenance (≥1 RTX-cycle with ≥6 months follow-up) after complete-remission (BVASv3=0 plus prednisolone ≤7.5mg/day) with induction regimens had been achieved. Relapses (BVASv3>0) were defined as major/minor, while SAEs included serious infections (SI), COVID-19-associated hospitalizations, deaths, cardiovascular events (CVE), malignancies and hypogammaglobulinemia. Incidence rates (IR) and relapse-free survival by Kaplan-Meier method were estimated while Cox-regression analysis was performed to investigate for factors associated with the time to first relapse. Results: Overall, 101 patients were included; GPA: 69%, 48% females, 53% newly diagnosed, median age: 63 years, C-/PR3-ANCA positive: 52% and p-/MPO-ANCA positive: 41%. During follow-up (294.5 patient-years, median [IQR] 3 [2-6] RTX-cycles), we identified 30 relapses in 24 patients (24%, IR 10.2/100 patient-years, 57% major). Up to the 3rd year from the first RTX-maintenance course, ~25% of patients had relapsed, while ~15% had experienced a major relapse (Figure 1). By Cox-regression analysis, renal involvement (adjusted-Hazard-Ratio-HR [95% Confidence-Interval]: 0.23 [0.06-0.78], p=0.019) and a higher number of RTX-administered cycles (adjusted-HR [95% CI]: 0.20 [0.09-0.44], p<0.001) were independently associated with a lower risk of relapses. We also recorded 17 SIs in 14 patients (14%, IR 5.8/100 patient-years), 11 COVID-19-associated hospitalizations (IR 3.7/100 patient-years), 4 malignancies (IR 1.4/100 patient-years), 6 CVE (IR 2/100 patient-years) and 10 deaths (10%, IR 3.4/100 patient-years). Among patients with available IgG data (n=53), 41 (77%) had normal IgG levels at baseline, of which 8 (20%) developed hypogammaglobulinemia. The rate of SIs did not differ significantly between those who developed hypogammaglobulinaemia (2/8, 25%) and those who did not (6/33, 18%, p=0.642). Conclusion: In this real-life study, relapses occurred in approximately one quarter of GPA/MPA patients under RTX-maintenance. The time to first relapse was longer in those with renal involvement and those who received >3 RTX cycles while a high rate of COVID19-associated hospitalizations was observed. Whether or not, longer duration of RTX maintenance therapy is required in these patients, needs to be examined in a randomized controlled trial. REFERENCES: [1] Delestre F et al. Rituximab as maintenance therapy for ANCA-associated vasculitides: pooled analysis and long-term outcome of 277 patients included in the MAINRITSAN trials. Ann Rheum Dis. 2023.[2] Smith RM et al. Rituximab versus azathioprine for maintenance of remission for patients with ANCA-associated vasculitis and relapsing disease: an international randomized controlled trial. Ann Rheum Dis. 2023;82(7):937-44. Acknowledgements: This work was supported in part by the Special Account for Research Grants, National and Kapodistrian University of Athens, Athens, Greece (DV #12085, 12086) and is conducted as a Master Thesis in Epidemiology – Research Methodology in biomedical sciences, clinical practice and public health (CG), Department of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens. This study has been accepted to be granted by EULAR with a Research Voucher for publication support. Disclosure of Interests: None declared.Figure 1a) Relapse-free survival and b) major-relapse free survival in the total cohort during RTX-maintenance. Relapse free survival according to c) renal involvement and d) number of RTX cycles received (≤3 vs. >3).

Background Despite significant advances in the diagnosis and therapy of giant cell arteritis (GCA), relapses during the disease course are common. The exact predictors for their occurrence at diagnosis or during the disease course have not been defined. Objectives To evaluate the frequency and predictors of relapses in a real-life GCA cohort. Methods Data were derived from an ongoing, multicenter, prospective cohort study of patients with GCA. Patients were evaluated at baseline and approximately 2 years later. Data regarding their demographics, clinical characteristics, treatment patterns, co-morbidities and relapses were retrospectively and prospectively collected and analyzed. Predictors of relapse were examined by logistic regression analyses. Results At the 1st cross-sectional evaluation, 254 patients with GCA were included; 68% were women with a mean age of 72.3 years at diagnosis and a mean disease duration of 2 years. Diagnosis had been established by temporal artery biopsy in 67% (n=171) and by ultrasound of temporal arteries in 21% (n=53) of patients. At the 2nd evaluation (median follow-up: 2.02 years), data were available for 199 patients (78%). All patients had been treated initially with glucocorticoids (GCs, mean prednisolone dose = 40 mg/day) and during the disease course additionally with non-biologic (39%, n=59) and/or biologic (16%, n=25) agents. At the 2nd evaluation time point, 73% of patients were still on therapy with GCs (60%) and/or GC-sparing agents (29%: non-biologics: 16%, biologics: 13%). During the prospective follow-up phase of 2 years, 27% (n=43) of patients had at least one disease relapse. Relapsing patients had shorter disease duration at 1st evaluation (1.4 vs. 4 years, p=0.015), more frequent large vessel involvement at diagnosis (21% vs. 8%, p=0.019) and cardiovascular events (CDV) during follow-up (14% vs. 4.2%, p=0.032), and had received a higher GC daily dose at diagnosis (mean 46.2 vs. 39.9 mg, p=0.01) compared to the non-relapsing ones. By multivariable logistic regression analysis, large vessel involvement at diagnosis [odds ratio (OR) (95% confidence intervals) = 4.22 (1.14, 15.58)], a cardiovascular event (CVD) during follow-up [OR = 4.60 (1.11, 19.13)] and a higher GC daily dose at diagnosis [OR = 1.04 (1.00, 1.08)] were associated with disease relapses. Conclusion In this large, real-life, long -term GCA cohort, despite continued treatment with GCs and/or GC-sparing agents, relapses occurred in 27% of patients. Factors associated with relapses included large vessel involvement and high GC dose at diagnosis as well as cardiovascular events during follow up. These are novel findings that could be useful in the design of the appropriate monitoring and therapeutic strategies in these patients. Acknowledgements Supported in part by the Greek Rheumatology Society and Professional Association of Rheumatologists (ERE-EPERE) and the Special Account for Research Grants (S.A.R.G.), National and Kapodistrian University of Athens, Athens, Greece (DV #12085, 12086). Disclosure of Interests None Declared.

Background AAVs are a group of rheumatic diseases with excess morbidity and mortality (~3-fold higher compared to the general population). Long-term studies looking at mortality trends in contemporary patient cohorts are limited. Objectives To investigate the overall long-term survival and all-cause mortality in a contemporary AAV patient cohort. Methods Multicenter cohort study of patients registered and prospectively followed in the Greek ANCA Registry. Results Data for 165 patients (989.38 patient-years of follow up) with a diagnosis of AAV (GPA n=95, 58%, MPA n=54, 33%, EGPA n=16, 9%) were analyzed (January 1, 1998 - January 10, 2022). 53% of patients were female, with a mean age of 65 (±16.4) years; the majority (97%) had generalized disease and were ANCA positive (76%). The mean follow-up since diagnosis was 5.9 (±5.1) years. At the end of follow-up, the overall mortality rate was 20% (33/165), whereas the cumulative mortality rates at 5 and 10 years were 24% and 26% respectively. Overall cumulative survival at 5 years was worse in patients with MPA (57%) compared to GPA (81%) and EGPA (92%), (p<0.001). There was no difference in long-term survival among those treated with different induction regimens including cyclophosphamide (CYC, n of deaths=24/83, 28.9%), rituximab (RTX, n=4/40, 10%) or the CYC+RTX combination (n=3/16, 18.7%). Furthermore, there was no difference in survival between relapsing (≥1 relapses) and non-relapsing (n=76) patients (Figure 1). Cumulative survival was worse in patients who initially presented with lung (66% vs. 90% at 5 years, p=0.007), kidney (56% vs. 96% at 5 years, p<0.001) and simultaneous lung and kidney (39% vs. 93% at 5 years, p<0.001) involvement. Among the 33 registered deaths, the most frequent causes were infections (52%), followed by cardiovascular events (24%), disease flares (14%) and malignancies (10%). Figure 1. Conclusion In a contemporary multi-center AAV cohort, the cumulative mortality rates at 5 and 10 years were 24% and 26% respectively. Overall survival was worse in patients with MPA as well as those with combined lung and kidney involvement at baseline while there were no survival differences according to the initial induction regimen. Infection was the most common cause of death. These findings emphasize the unmet needs for better, less toxic therapies. Acknowledgements Supported in part by the Greek Rheumatology Society and Professional Association of Rheumatologists (ERE-EPERE) and the Special Account for Research Grants (S.A.R.G.), National and Kapodistrian University of Athens, Athens, Greece (DV #12085, 12086). Disclosure of Interests None declared

OBJECTIVES:Systemic lupus erythematosus (SLE) patients show variably increased risk for pregnancy complications. We analysed pregnancy outcomes (foetal and maternal), patterns of disease activity and use of medications in a contemporary Caucasian SLE population.METHODS:Prospective observational study, involving hospital units and private rheumatologists in Greece, of incident pregnancies (period 2015-2018) in women with SLE. Clinical and obstetrical monitoring was performed at regular intervals up to 9 months post-partum. Regression and mixed model analyses were used to determine predictors for adverse foetal outcomes and flares.RESULTS:We monitored 82 pregnancies in 64 SLE patients. Foetal loss, prematurity and small for gestational age neonate occurred at 15.8%, 34.1% and 8.5%, respectively; 53.7% of pregnancies were complicated with at least one adverse outcome. Patients with antiphospholipid antibodies (aPL) had increased risk (odds ratio [OR] 5.67, p=0.015), whereas those at low disease activity at pregnancy onset were protected (OR 0.20, p=0.024) against foetal complications. Persistent activity and glucocorticoid intake during pregnancy also predicted poor foetal outcomes. SLE patients experienced an average 1.08 mild/moderate and 0.27 severe flares. The latter occurred more frequently post-partum, in patients with alopecia (OR 8.92, p=0.003), hypocomplementaemia (OR 10.34, p=0.038) and nephritis (OR 7.32, p=0.052). Lupusactivity post-labour was paralleled by decreased use of hydroxychloroquine, glucocorticoids and azathioprine.CONCLUSIONS:In SLE women, foetal complications are common especially in the presence of aPL and increased activity, which corroborates the importance of pregnancy planning and tight disease control at pregnancy onset. Flares, mostly mild or moderate, can occur both during and after pregnancy.

Background: Despite the increased incidence of influenza infection in rheumatoid arthritis (RA) patients, vaccination coverage has been shown to be suboptimal. Prospective data regarding the current rate and predictors of influenza vaccination adherence in RA patients are limited. Objectives: To calculate the current rate and predictors of influenza vaccination in a real-life, prospective, longitudinal RA cohort. Methods: Data regarding demographics, disease characteristics, treatments and co-morbidities from a multi-center, longitudinal cohort of Greek RA patients were collected at baseline and ~ 3 years later. Disease and patient characteristics were compared between patients with at least one influenza vaccine administration and non-vaccinated ones, during the 3 year follow-up period. Results: From a cohort of 1,569 RA patients, 1,406 with available vaccination data at baseline and 3 years later (mean interval: 2.9 years) were included; (women: 80.4%, mean age: 61.8 years, mean disease duration: 9.7 years, RF and/or anti-CCP positive: 50.4%, mean DAS-28 = 3.33, mean HAQ: 0.44, bDMARD use: 44.8%). At baseline, 54.2% of patients reported influenza vaccination in the past (31.8% during the previous season), while during the 3 year follow-up period, 81% had ≥1 influenza vaccinations (p=<0.001). Patients who received ≥1 influenza vaccine were older (63.5 vs. 54.7 years, p<0.001), were more likely to be seropositive (59.2% vs. 45.2%, p<0.001), had higher HAQ (0.46 vs. 0.36, p=0.02) and BMI (27.7 vs. 26.9, p=0.02) at baseline, more likely to be treated with bDMARDs (46.8% vs. 36.4%, p<0.001) and more likely to have chronic lung disease (9.7% vs. 5.3%, p=0.02), dyslipidemia (36.4% vs. 24.2%, p<0.001), hypertension (46.1% vs. 29.2%, p<0.001) and to report vaccination against influenza the previous season before baseline evaluation (34.9% vs. 18.2%, p<0.001). By multivariate analysis, history of influenza vaccination during the last season before baseline (OR=1.87, CI: 1.27-2.74, p=0.001), bDMARD treatment (OR=1.51, CI: 1.07-2.13, p=0.018) and age (OR=1.05, CI: 1.04-1.06, p<0.001) were independent predictors of influenza vaccination. Conclusion: In this ongoing, longitudinal, prospective, real-life RA cohort study, a significant increase in the influenza vaccination coverage was noted (from 53% to 81%). Influenza vaccination was independently associated with recent history of influenza vaccination, older age, and bDMARD treatment. Acknowledgments: Supported by grants from the Greek Rheumatology Society and Professional Association of Rheumatologists. Disclosure of Interests: Konstantinos Thomas: None declared, Argyro Lazarini: None declared, Evripidis Kaltsonoudis: None declared, Alexandros Drosos: None declared, ARGYRO REPA: None declared, Prodromos Sidiropoulos: None declared, Kalliopi Fragkiadaki: None declared, Maria Tektonidou Grant/research support from: AbbVie, MSD, Novartis and Pfizer, Consultant of: AbbVie, MSD, Novartis and Pfizer, Petros Sfikakis Grant/research support from: Grant/research support from Abvie, Novartis, MSD, Actelion, Amgen, Pfizer, Janssen Pharmaceutical, UCB, Panagiota Tsatsani: None declared, Sousana Gazi: None declared, Pelagia Katsimbri: None declared, Dimitrios Boumpas: None declared, Evangelia Argyriou: None declared, Kyriaki Boki: None declared, Gerasimos Evangelatos: None declared, Alexios Iliopoulos: None declared, Konstantina Karagianni: None declared, Lazaros Sakkas: None declared, Konstantinos Melissaropoulos: None declared, Panagiotis Georgiou: None declared, Eleftheria Grika: None declared, PANAYIOTIS VLACHOYIANNOPOULOS: None declared, Theodoros Dimitroulas: None declared, Alexandros Garyfallos Grant/research support from: MSD, Aenorasis SA, Speakers bureau: MSD, Novartis, gsk, Constantinos Georganas: None declared, Periklis Vounotrypidis: None declared, Konstantinos Ntelis: None declared, Maria Areti: None declared, George D Kitas: None declared, Dimitrios Vassilopoulos: None declared

Background: Comorbidities in rheumatic diseases (RDs) have been associated with increased morbidity and mortality. Evidence on prevalence of comorbidities in antiphospholipid syndrome (APS) and its difference from high comorbidity burden RDs is limited. Objectives: To compare the prevalence of common comorbidities between APS [primary (PAPS) and Systemic lupus erythematosus (SLE)-APS] and Rheumatoid arthritis (RA) patients. Methods: 326 APS patients from the Greek registry (237 women, mean age 48.7±13.4 years, 161 PAPS) were matched 1:2 for age and sex with 652 RA patients from Greek RA Registry. Prevalence of cardiovascular (CV) risk factors, stroke, coronary artery disease (CAD), osteoporosis, diabetes mellitus (DM), Chronic obstructive pulmonary disease (COPD), depression and neoplasms were compared between APS and RA using logistic regression analysis. Results: Regarding CV burden, hyperlipidemia and obesity (ΒMI≥30) were comparable while hypertension, smoking, CAD and stroke were more prevalent in APS compared to RA patients (Table 1). Osteoporosis and depression were more frequent in APS while DM, COPD and neoplasms were comparable between two groups. Comparison of APS subgroups to 1:2 matched RA patients revealed that smoking and stroke were more prevalent in PAPS and SLE-APS vs RA. Hypertension, CAD and osteoporosis were more prevalent only in SLE-APS vs. RA while DM was less prevalent in PAPS vs. RA patients. Table 1. Comparison of comorbidities between Antiphospholipid syndrome (APS) vs. matched Rheumatoid Arthritis (RA) patients and between primary APS (PAPS) or Systemic Lupus Erythematosus-APS (SLE-APS) vs matched RA patients APS RA OR* PAPS RA OR SLE-APS RA OR n (%) 326 652 161 322 165 330 Hypertension 97 (29.8) 136 (21) 1.61 (1.19-2.18) 40 (25) 75 (23.3) 1.09 (0.70-1.69) 57 (34.6) 61 (18.5) 2.33 (1.52-3.56) Smoking 175 (53.7) 264 (40.5) 1.70 (1.30-2.22) 87 (54) 142 (44) 1.49 (1.02-2.18) 88 (53.3) 122 (37) 1.95 (1.33-2.85) Hyperlipidemia 79 (24.2) 135 (20.7) 1.23 (0.89-1.68) 40 (24.8) 62 (19.3) 1.39 (0.88-2.18) 39 (23.6) 73 (22) 1.09 (0.70-1.70) Obesity 48 (20.5) 105 (19.5) 1.06 (0.73-1.56) 20 (17) 51 (19) 0.86 (0.49-1.52) 28 (24) 54 (19.7) 1.28 (0.76-2.15) Stroke ± 66 (20.3) 9 (1.4) 13.8 (6.5-29.1) 36 (22.4) 4 (1.2) 19.9 (6.6-59.9) 30 (18.2) 5 (1.5) 7.8 (2.7-22.6) Coronary disease ± 16 (4.9) 13 (2) 3.14 (1.17-8.45) 2 (1.2) 7 (2.2) 0.46 (0.04-4.77) 14 (8.5) 6 (1.8) 10.9 (2.7-44.3) Osteoporosis × 66 (20.3) 92 (14) 1.45 (1.01-2.06) 19 (11.8) 42 (13) 0.96 (0.54-1.73) 47 (28.5) 50 (15) 1.91 (1.20-3.05) Diabetes × 18 (5.5) 58 (9) 0.58 (0.33-1.01) 5 (3) 29 (9) 0.34 (0.13-0.89) 13 (8) 29 (9) 0.88 (0.44-1.79) COPD ≠ 11 (3.4) 14 (2.2) 1.26 (0.56-2.84) 3 (1.9) 6 (2) 0.96 (0.23-4.0) 8 (5) 8 (2.4) 1.28 (0.44-3.72) Depression # 53 (16.3) 66 (10) 1.70 (1.15-2.53) 23 (14) 30 (9.3) 1.69 (0.93-3.05) 30 (18.2) 36 (10.9) 1.65 (0.96-2.84) Neoplasms ˅ 14 (4.3) 27 (4.1) 1.05 (0.54-2.06) 5 (3) 12 (3.7) 0.84 (0.28-2.52) 9 (5.5) 15 (4.6) 1.31 (0.55-3.1) * OR: Odds ratio, crude or adjusted for: ± age, sex, smoking, hypertension, hyperlipidemia, BMI, corticosteroid (Cs) duration × Cs duration ≠ smoking, Cs duration # sex, disease duration, Cs duration ˅ age, disease duration Conclusion: Comorbidity burden in APS (PAPS and SLE-APS) is comparable or even higher to that in RA, entailing a high level of diligence for CV risk prevention, awareness for depression and corticosteroid exposure minimization. Disclosure of Interests: Stylianos Panopoulos: None declared, Konstantinos Thomas: None declared, Georgios Georgiopoulos: None declared, Dimitrios Boumpas Grant/research support from: Unrestricted grant support from various pharmaceutical companies, Christina Katsiari: None declared, George Bertsias Grant/research support from: GSK, Consultant of: Novartis, Alexandros Drosos: None declared, Kyriaki Boki: None declared, Theodoros Dimitroulas: None declared, Alexandros Garyfallos Grant/research support from: MSD, Aenorasis SA, Speakers bureau: MSD, Novartis, gsk, Charalambos Papagoras: None declared, PELAGIA KATSIMPRI: None declared, Apostolos Tziortziotis: None declared, Christina Adamichou: None declared, Evripidis Kaltsonoudis: None declared, Evangelia Argyriou: None declared, GEORGIOS VOSVOTEKAS Grant/research support from: MSD, Janssen, Consultant of: MSD, Novartis, Roche, UCB pharma, Bristol-Myers Squibb, AbbVie, Speakers bureau: UCB pharma, Menarini, Bristol-Myers Squibb, MSD, Petros Sfikakis Grant/research support from: Grant/research support from Abvie, Novartis, MSD, Actelion, Amgen, Pfizer, Janssen Pharmaceutical, UCB, Dimitrios Vassilopoulos: None declared, Maria Tektonidou Grant/research support from: AbbVie, MSD, Novartis and Pfizer, Consultant of: AbbVie, MSD, Novartis and Pfizer

Background: Pregnancy in women with SLE Systematic Lupus Erythematosus (SLE) has been related with adverse events both in the mother and the foetus. 1 Many studies have reported relapse of the disease during the pregnancy and post-labour, while others have not confirmed this finding. 2 To this end, most of these results originate from retrospective studies with patients of diverge ethnicities. Objectives: To record the Greek experience with pregnancies in mothers with SLE and their outcomes, as well as the course of the disease during first year post labor. Methods: This is a prospective, multicentre, observation study lasting three years. Women diagnosed with SLE who became pregnant consented to be monitored by their treating Rheumatologist. A structured questionnaire is used for monitoring at the beginning of pregnancy (positive pregnancy test) and at least every 3 months thereafter, depending on the course of the disease and pregnancy, until one year after childbirth. Results: A total 64 women and 81 pregnancies were recorded (1.27 pregnancies per patient). Patient’s age at conception was 32.8 ± 5.9 years (mean ± standard deviation). Thirteen patients (20.3%) had past history of nephritis. Regarding pregnancy outcomes, 62 (76.5%) pregnancies ended in live births, miscarriages during 1 st , 2 nd and 3 rd trimester occurred in 13 (16%). Six pregnancies were lost to followup. Prematurity occurred in 28 live births (45.1% in total), 26-32w (3.2%), 32-36w (22.5%), <37w (19.3%). No cases of preeclampsia occurred. Mean age of birth 36.9 weeks and mean birth weight 2750gr. The majority (72.5%) of deliveries were performed by caesarean section. In terms of disease activity, most of the women had mild disease at conception, (SLEDAI-2K: 2.67±2.69) that declined during 1 st /2 nd pregnancy trimester (SLEDAI-2K:1.91±2.09, 1.70±2.22)) but increased during the 1 st and 2 nd trimester post labor (SLEDAI-2K: 2.47±4.29 and 2.52±3.2). Conclusion: This is the first Greek inception cohort with prospective monitoring of pregnant SLE patients. Adverse outcomes occur with prematurity being the most frequent. In our cohort disease activity tends to increase during 1 st and 2 nd trimester post-labor without serious relapses. Vigilant monitoring during pregnancy and post-labour is advised. References: [1] Bundhun PK, Soogund MZ, Huang F. Impact of systemic lupus erythematosus on maternal and fetal outcomes following pregnancy: A meta-analysis of studies published between years 2001-2016. J Autoimmun 2017;79:17-27. [https://doi. org/10.1016/j.jaut.2017.02.009] [PMID: 28256367] [2] Wei S, Lai K, Yang Z, Zeng K. Systemic lupus erythematosus and risk of preterm birth: a systematic review and meta-analysis of observational studies. Lupus 2017;26:563-71. [https://doi. org/10.1177/0961203316686704] [PMID: 28121241] Acknowledgments: Hellenic Rheumatology Association Disclosure of Interests: Stella Ntali: None declared, Lina Pantazi: None declared, Kyriaki Boki: None declared, Dionysis Nikolopoulos: None declared, Antonis Fanouriakis: None declared, Despoina Dimopoulou: None declared, Ioannis Kallitsakis Grant/research support from: MSD, Speakers bureau: Genesis pharma, Bristol-Myers Squibb, CHARALAMPOS PAPAGORAS: None declared, Vasiliki Dania: None declared, Evgenia Emmanouilidou: None declared, George Bertsias Grant/research support from: GSK, Consultant of: Novartis

Background There are limited literature data regarding the characteristics of rheumatoid arthritis (RA) patients treated with biologic DMARD (bDMARD) monotherapy. Objectives To evaluate the disease and treatment characteristics of RA patients treated with bDMARD monotherapy. Methods Multicenter, cross-sectional RA epidemiological study in Greece (06/2015–05/2016, ERE RA Study Group). Demographics, disease characteristics, treatment and co-morbidity data were collected via a web-based platform Results 1036 RA patients treated with bDMARDs were identified during the one year recruitment period (female: 82%, mean age: 61.5±13 years, mean disease duration: 12.5±8.9 years, mean DAS28-ESR: 3.4±3.3). Among them, 26% (n=273) were receiving bDMARDs as monotherapy and 8% (n=23) of them had never tried conventional synthetic DMARDs (csDMARDs) before; The latter group (n=23) compared to the csDMARD-exposed (92%, n=250) group, had more often co-morbidities [cardiovascular disease (22% vs. 8%, p=0.029), chronic hepatitis (13% vs. 3%, p=0.02), hypertension (57% vs. 38%, p=0.08), COPD (13% vs. 4.4%, p=0.07)] or were active smokers (41% vs. 14%, p=0.001). csDMARD discontinuation was mainly due to adverse events (AEs, 58%) followed by inadequate response (IR, 44%). Compared to the cs- and b-DMARD combination therapy group, monotherapy treated patients were more frequently seropositive (64% vs. 57%, p=0.003), had lower DAS28-ESR (3.2 vs 3.5, p=0.009) and were more likely to have discontinued csDMARDs for AEs (58% vs. 21%, p<0.001) or IR (44% vs. 27%, p<0.001). Tocilizumab (22.3% vs. 12.7%, p<0.001) and rituximab (19.8% vs. 13.5%, p=0.013) were utilized more often, whereas adalimumab less often (8.8% vs. 14.8%, p=0.012) as monotherapy compared to as part of combination therapy. Conclusions In our large RA cohort, one out of four bDMARD treated patients, were receiving bDMARDs as monotherapy. Co-morbidities rather than RA characteristics influence the initial decision for bDMARD monotherapy whereas among those starting combination cs- and b-DMARD therapy, the majority discontinue csDMARDs due to AEs. Acknowledgements Grant support from the Hellenic Rheumatology Society and Professional Association of Rheumatologists. Disclosure of Interest None declared

OBJECTIVESTo evaluate the long-term safety of rituximab (RTX) in rheumatoid arthritis (RA) patients in daily clinical practice.METHODSThis was a multicentre (17 Greek Rheumatology sites), prospective, long-term, pharmacovigilance study of patients with moderate to severe RA and an inadequate response or intolerance to ≥1 anti-tumour necrosis factor (TNF) agents. Adverse events (AEs) were recorded and collected prospectively every 2-6 months.RESULTS234 patients (mean age: 59±12.5, 79.5% women, mean DAS28: 5.35±1.32) were included and followed for 27.7 months (median). The overall AEs, serious AE (SAEs) and serious infection (SIEs) rate were 48.36, 6.68 and 2.53/100 patient-years, respectively. Three cases of hepatitis B virus (HBV) reactivation were recorded (two in chronic and one in past HBV infection). Withdrawals due to AEs (5.6%) occurred more frequently during the first cycles of RTX therapy while repeated RTX cycles were not associated with an increased risk of AEs. There were 3 deaths with an incidence rate of 0.69/100 patient-years. Age ≥65 years was associated with a higher incidence rate ratio of AEs and SAEs as compared to <65 years (1.53, p=0.002 and 2.88, p=0.005, respectively). Drug retention rate during 434.28 patient-years of follow-up was 57.3%. Factors associated with drug discontinuation by multivariate analysis included age, baseline swollen joint count and no use of concomitant methotrexate therapy.CONCLUSIONSLong-term RTX therapy in a real-life RA cohort, did not reveal any new safety issues. Advanced age was associated with increased risk of AEs and premature drug discontinuation.

OBJECTIVES To assess in daily practice in patients with rheumatoid arthritis (RA) the effect of treatment with first tumour necrosis factor-α inhibitor (TNFi) in quality of life (Qol), disease activity and depict possible baseline predictors for gains in Qol. METHODS Patients followed prospectively by the Hellenic Registry of Biologic Therapies were analysed. Demographics were recorded at baseline, while RA-related characteristics at baseline and every 6 months. Paired t-tests were used to detect divergences between patient-reported (Health Assessment Questionnaire (HAQ), EuroQol (EQ-5D)) and clinical tools (Disease Activity Score-28 joints (DAS28)). Clinical versus self-reported outcomes were examined via cross-tabulation analysis. Multiple regression analysis was performed for identifying baseline predictors of improvements in QALYs. RESULTS We analysed 255 patients (age (mean±SD) 57.1±13.0, disease duration 9.2±9.1 years, prior non-biologic disease-modifying anti-rheumatic drugs 2.3±1.2). Baseline EQ-5D, HAQ and DAS28 were 0.36 (0.28), 1.01 (0.72) and 5.9 (1.3), respectively, and were all significantly improved after 12 months (0.77 (0.35), 0.50 (0.66), 3.9 (1.5), respectively, p<0.05 for all). 90% of patients who improved from high to a lower DAS28 status (low-remission or moderate) had clinically important improvement in Qol (phi-coefficient=0.531,p<0.05). Independent predictors of gains in Qol were lower baseline HAQ, VAS global and younger age (adjusted R2=0.27). CONCLUSIONS In daily practice TNFi improve both disease activity and Qol for the first 12 months of therapy. 90% of patients who improved from high to a lower DAS28 status had clinically important improvement in Qol. Younger patients starting with lower HAQ and VAS global are more likely to benefit.

Background Existing guidelines advocate aggressive management of dyslipidemia in rheumatoid arthritis (RA) patients according to cardiovascular disease (CVD) risk scores generated for the general population (SCORE). More specific RA scores such as the ERS-RA score have not been extensively studied. Objectives To evaluate the use of lipid-lowering agents for CVD prevention (primary/secondary) in RA patients according to different CVD risk scores (SCORE, ERS-RA). Methods Prospective, multicenter (12 hospitals, 6 private offices), cross-sectional, epidemiological study in Greece (06/2015–01/2016, RA Study Group). Demographics, disease characteristics, treatment and comorbidities were collected and SCORE/ERS-RA were calculated. Results Among 1475 patients, 178 (12%) had CVD (44% coronary artery disease, 44% peripheral vascular disease, 27% stroke) and 43% were not receiving any hypolipidemic therapy. From those on therapy, only 12% achieved an LDL-cholesterol (LDL-C) level <70 mg/dl while 48% had LDL-C<100 mg/dl. 859 patients without CVD, diabetes or hypolipidemic therapy were further analyzed (79% women, mean age 59.5 yrs, mean disease duration 3.4 yrs, mean DAS28-ESR 3.4, mean HAQ 0.45). Complete data for SCORE calculation were available in 225/859 patients (26%). When stratified according to CVD risk, 31.5%, 66%, 2% and 0.5% of patients had a SCORE of <1% (low risk), ≥1 to <5% (moderate), ≥5 to <10% (high) and ≥10% (very high), respectively. In moderate risk (1–5%) patients, 78% had an LDL-C>100 mg/dl while among high or very high risk (≥5%) patients, 100% had an LDL-C>70 mg/dl (target groups for drug therapy). Among 996 patients (40–75 yrs old), without CVD events, there were 99 diabetic patients (10%) and less than half (48%) were on hypolipidemic therapy. For the rest, the ERS-RA score was calculated (619/897, 69% with available data); 50% of them (308/619) had a 10-year risk of ≥7.5% (therapeutic cutoff) but only 40% of patients in this group were being treated. Conclusions A substantial proportion of RA patients with established CVD or diabetes do not receive lipid-lowering therapy or they do not achieve therapeutic targets. Among those without CVD or diabetes, more than half belong to moderate to high CVD risk groups and are not receiving appropriate hypolipidemic therapy. Acknowledgement Supported by grants from the Hellenic Rheumatology Society and Professional Association of Rheumatologists. Disclosure of Interest None declared

Background About one third of RA patients do not initially respond to treatment with an anti-TNF agent whereas a similar rate demonstrates lack of efficacy over time. Rituximab/Mabthera administration (temporary B-lymphocyte depletion) is one of the therapeutic options for them. Objectives The LAUNCH prospective study aimed at the evaluation of long-term efficacy and safety data following rituximab administration in standard clinical practice Methods 17 Rheumatology sites in Greece enrolled 234 adult patients (63.0±12.4 years, 79.5% women) with severe RA and an inadequate response or non-tolerance to anti-TNF treatment. Rituximab (1gr) was administered IV on days 1 and 14 of each cycle, repeated every 6-12 months, for up to 7 cycles. Of these patients 41.2% and 56.2% had received one, or more, anti-TNF agent(s), respectively. Adverse events, DAS28, and the quality of life evaluation indices (Euroqol) were collected every 2 to 6 months for 5 years according to each site’s standard clinical practice Results During 496 patient/years, 28 adverse events /100 pt-yrs (including9.9 serious adverse events and 7.7 serious infectious per 100pt-yrs, respectively) were observed. Of the total number of adverse events a 46.7% was not related to rituximab. The mean number of adverse events per patient remained stable during repeated treatment cycles. Disease activity at baseline (mean±SD DAS28 of 5.36±1.40) was significantly reduced in cycles 1,2, 3, 4, 5, and 6 by 1.34, 2.12, 2.25, 2,56, 2.42 and 2.79, respectively (p<0.01). Compared to baseline, significant improvement in quality of life was also observed in all cycles (p<0.01) Conclusions Rituximab administration in clinical practice for up to 5 years demonstrated an acceptable safety profile which was maintained over time. Likewise, maintenance and/or improvement of efficacy with repeated treatment cycles in patients with severe RA not responding to anti-TNF were evident Disclosure of Interest L. Settas: None Declared, A. Andrianakos: None Declared, S. Aslanidis: None Declared, P. Boura: None Declared, M. Katsounaros: None Declared, D. Vassilopoulos: None Declared, P. Athanassiou: None Declared, K. Tempos: None Declared, G. Skarantavos: None Declared, C. Antoniadis : None Declared, L. Sakkas: None Declared, A. Andonopoulos: None Declared, V. Galanopoulou: None Declared, F. Solioti: None Declared, K. Boki: None Declared, E. Vritzali Employee of: Roche Hellas SA, P. Sfikakis: None Declared

Objective: Primary Sjogren's syndrome is an autoimmune disease with glandular and extraglandular manifestations. The aim of our study was the incidence of pulmonary hypertension in patients with primary Sjogren's syndrome (pSS) and the relation with clinical and laboratory findings. Methods: A total of 107 (104 females-3 males, aged 56.5 ± 12.5 years) consecutive patients with pSS underwent complete echocardiographic study. Pulmonary artery systolic pressure (PASP) was calculated by continuous wave Doppler echocardiography as the peak systolic pressure gradient across the tricuspid valve plus the estimated right atrial pressure. Values above 40 mmHg was considered as pulmonary hypertension. Results: Sixteen (15%) patients had pulmonary hypertension which was mild in 13 and in 3 moderate degree. In univariate analysis the PASP values were correlated with hypocomplementemia, cryoglobulinemia, interstitial nephritis, lung disease and easy fatigability whereas in multivariate analysis the PASP values were significantly associated only with hypocomplementemia (p = 0.026), cryoglobulinemia (p = 0.028) and easy fatigability (p = 0.02). Conclusion: In patients with pSS mild pulmonary hypertension is not uncommon whereas severe is rare. Hypocomplementemia, cryoglobulinemia and easy fatigability were strong predictors of PASP. Follow-up studies must be done to examine whether pulmonary hypertension evolves and whether have an overall effect on disease prognosis.

OBJECTIVE:To measure aortic stiffness and global left ventricular (LV) function in patients with ankylosing spondylitis (AS) and no clinical evidence of heart disease. METHODS:Fifty-seven consecutive patients with AS (54 males, three females, mean age 41.78+/-10.02 years) without clinical evidence of cardiac involvement and 78 healthy subjects (73 males, five females, mean age 39.92+/-9.11 years) underwent complete echocardiographic study. Aortic stiffness was determined non-invasively by aortic distensibility (AoD) and the global LV function was evaluated by the myocardial performance index (the Tei index). RESULTS:AoD in patients with AS [(2.21+/-0.24)x10(-6) cm(2) dyn(-1)] was decreased compared to controls [(2.58+/-0.19) )x10(-6) cm(2) dyn(-1), p<0.01], confirming that aortic stiffness is increased in AS. The LV Tei index was significantly increased in the patient group compared to the control group (0.392+/-0.031 vs. 0.370+/-0.034, p<0.01). The ejection fraction (EF) did not differ between the two groups (p>0.05). In multivariate linear regression analysis, AoD was significantly associated with the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and LV isovolumic relaxation time (IVRT) whereas the LV Tei index was associated with BASDAI and the LV mass index. CONCLUSIONS:Patients with AS and no clinical evidence of cardiac disease have increased stiffness of the aorta and decreased global myocardial performance and both of these abnormal measurements correlate with disease activity. The abnormal Tei index may reflect an early manifestation of cardiac dysfunction in these patients.

OBJECTIVE:The 3E (Evidence, Experts and Exchange) Initiative is a multi-national effort that involves a large number of experts and practicing rheumatologists addressing specific questions relevant to everyday clinical practice, concerning the management of Ankylosing Spondylitis. Within this multinational group, the Hellenic working group, addressed specific issues complementary to the international ones, and formulated evidence-based recommendations, in order to improve everyday clinical practice for patients with Ankylosing Spondylitis.METHODS:A scientific committee of rheumatologists specializing in AS formulated a set of 7 questions in three domains: diagnosis, monitoring and treatment. Literature search in MedLine for papers published up to August 2006 was conducted. The evidence to support each proposition was evaluated and scored. To avoid any conflict of interest with the sponsor issues related to the use of biologics were not discussed. After extensive discussion among 50 rheumatologists and one Delphi round of votes, the final recommendations were formulated.RESULTS:A literature search resulted in a total of 320 relevant papers of which 29 were evaluated. A total of seven recommendations were formulated: two concerning diagnosis (role of HLA-B27 and MRI) and prognosis, one concerning monitoring for extra-articular manifestations and four concerning treatment (analgesics, disease modifying agents and physical therapy) were made. The level of evidence and the strength of recommendation were reported. The compiled agreement among experts ranged from 90% up to 100%.CONCLUSION:Recommendations for the management of AS were developed using an evidence-based approach followed by physicians' consensus with high level of agreement. These are complementary to existing ones, and address specific domains of everyday clinical practice.

Positive experience with mycophenolate mofetil (MMF) in the field of solid organ transplantation has made the drug attractive for the treatment of several autoimmune diseases. A rare case of acute hepatitis complicating the use of MMF for antinuclear cytoplasmic antibody (ANCA)-positive vasculitis is presented. Toxic hepatitis in our patient was thought to be causally related to the use of MMF based on a temporal relationship, negative serology for acute viral infection, negative antibody markers, and exclusion of drugs or other potential hepatotoxic agents.