Objectives: To investigate how the use of pharmacologic concomitant medications (CMs) is planned and reported in rheumatology randomized controlled trials (RCTs). Methods: We searched PubMed for RCTs on rheumatic diseases published in leading medical journals in the past 2 years. CMs potentially influencing primary outcome(s) were identified by a panel of experts and included in the analysis. Information on trial characteristics, how the use of CMs was planned and conducted throughout the trial were extracted. Data were summarized as number (%) for qualitative, and median (range) for continuous variables. Results: The 109 RCTs were mainly international (65%), industry-funded (81%) studies, including 267 CMs potentially influencing primary outcome(s). Forty-one RCT5 (38%) did not provide any data about the dosages of CMs allowed. Information on whether the intake of a permitted CM could have been modified during the study was missing in 24 (22%) or incompletely reported in 23 (21%) RCTs. As regards CMs use throughout the study, the baseline number of patients on CMs per arm and the mean baseline dosage was lacking in 20 (18%) and 57 (52%) trials, respectively. Ninety-five percent of RCTs did not provide any information on cumulative/mean exposure to CMs. Two of 109 trials described how many patients had modified CM dosage for reasons permitted by the protocol, and 3 reported the number of patients violating protocol because of an incorrect CM intake. Conclusions: The use of CMs potentially impacting on primary outcome is poorly reported in rheumatology RCTs. (C) 2019 Elsevier Inc. All rights reserved.

BackgroundThe biomarkers CXCL13 and sICAM1 have been associated with outcomes in patients with RA treated with TCZ.ObjectivesTo determine the association of CXCL13 and sICAM1 with response to TCZ and disease activity in early RA and DMARD-IR patients.MethodsPatient subsets from the FUNCTION (early RA) and LITHE (DMARD-IR) clinical trials were selected based on baseline and Week 24 sample availability; serum CXCL13 and sICAM1 levels were measured. Correlations between CXCL13 and sICAM1 levels and DAS28-ESR at baseline, and between change in CXCL13 and sICAM1 levels and change in DAS28-ESR at Week 24, were determined. Changes in CXCL13 and sICAM1 levels from baseline to Week 24 were compared between treatment arms using Welch t test. The effect of treatment, baseline DAS28-ESR and baseline CXCL13 and sICAM1 levels on the likelihood of DAS28-ESR remission and ACR50 response at Week 24 was determined via logistic regression. DAS28-ESR remission and ACR50 rates were compared against CXCL13 and sICAM1 status (high vs low based on median values) within each trial arm using a Cochran-Mantel-Haenszel test.ResultsOverall, 458 of 872 patients from FUNCTION (TCZ +MTX, n=60; TCZ monotherapy [TCZ-mono], n=157; placebo [PBO]+MTX, n=141) and 287 of 791 patients from LITHE (TCZ +MTX, n=137; PBO+MTX, n=150) were included. In these patient subsets, mean disease duration in FUNCTION was significantly shorter than in LITHE (0.45 vs 8.65 years). At baseline, correlation of serum CXCL13 levels with DAS28-ESR was moderate in the early RA population and weak in the DMARD-IR population (table 1). Correlation between baseline serum sICAM1 levels and DAS28-ESR was low in both populations. Serum levels of CXCL13 decreased significantly at Week 24 in all treatment arms in both populations, with greater reductions in the TCZ +MTX and TCZ-mono arms; sICAM1 levels decreased significantly at Week 24 in the TCZ-mono arm in patients with early RA and the TCZ +MTX arms in both populations but not in the PBO+MTX arms. Change in CXCL13 levels correlated moderately with change in DAS28-ESR at Week 24 in both populations (table 1). Change in sICAM1 levels correlated moderately with change in DAS28-ESR at Week 24 in the DMARD-IR population but weakly in the early RA population. Although the treatment arm had a significant effect on the likelihood of DAS28-ESR remission and achievement of ACR50, the effect of baseline levels of CXCL13 and sICAM1 were not significant. DAS28-ESR remission and ACR50 response rates at Week 24 within each treatment arm of the early RA and DMARD-IR populations were not significantly different between patients with high vs low baseline CXCL13 and sICAM1 levels. CXCL13. C-X-C motif chemokine ligand 13; DAS28-ESR, Disease Activity Score in 28 joints per erythrocyte sedimentation rate; DMARD-IR, inadequate response to disease-modifying antirheumatic drugs; RA, rheumatoid arthritis; sICAM1, soluble intercellular adhesion molecule 1. * Correlation between change in CXCL13 and sICAM1 levels from baseline to Week 24 and change in DAS28-ESR from baseline to Week 24; all patients combined.ConclusionsThe association of baseline CXCL13 levels with RA disease activity was stronger in the early RA population than in the DMARD-IR population. Changes in CXCL13 and sICAM1 correlated significantly with changes in DAS28-ESR at Week 24. However, baseline levels of CXCL13 and sICAM1 did not predict response to TCZ at Week 24, suggesting that although these biomarkers are associated with disease activity, they do not predict response to TCZ in all RA populations.AcknowledgementsFunded by F. Hoffmann-La Roche Ltd. and Genentech, Inc.Disclosure of InterestT. Sornasse Employee of: Genentech, Inc., C. Gabay Grant/research support from: Roche, Pfizer, AB2 Bio, Consultant for: Roche, Pfizer, AbbVie, Novartis, Sanofi, M. Townsend Employee of: Genentech, Inc., R. Laubender Employee of: Roche Diagnostics, J. Wang Employee of: Roche Diagnostics, K. Tuckwell Employee of: Genentech, Inc.

Introduction Macrophage activation syndrome (MAS) is a severe condition, which can appear as a complication of inflammatory rheumatic diseases such as systemic juvenile idiopathic arthritis (sJIA) and adult onset Still’s disease (AOSD), a viral infection, or a malignancy. Interleukin (IL)−18 is a pro-inflammatory cytokine of the IL-1 family, known as a strong interferon (IFN)-γ inducer, that is naturally inhibited by IL-18 binding protein (IL-18BP). High levels of unbound biologically active IL-18 have been described in patients with sJIA, AOSD, and MAS, suggesting that IL-18 is involved in the pathogenesis of these diseases. Objectives To examine the effect of excessive IL-18 signalling in a mouse model of MAS induced by repetitive toll like receptor (TLR)9 stimulation, and explore the consequences of IL-18 or IFN-γ blockade on MAS manifestations. Methods MAS was induced by repeated intraperitoneal CpG injections in IL-18BP deficient (IL-18BP-/-) mice and in wild type (WT) littermates. Anti-IL-18 receptor (IL-18R) or anti-IFN-γ monoclonal antibodies were administered prior to CpG injections. Clinical and biological manifestations of MAS were studied. Plasma levels of free IL-18 were measured by ELISA. Expression levels of IFN-γ and downstream IFN-γ-induced effectors (CXCL9, CIITA) were explored by ELISA or Luminex in plasma, and by RT-qPCR in spleen and liver. Results Naïve IL-18BP-/- mice had no spontaneous phenotype. After repeated CpG injections, IL-18BP-/- mice displayed significantly more severe MAS phenotype than their WT littermates, including more pronounced weight loss, splenomegaly, anaemia, thrombocytopenia, hyperferritinemia and hepatitis. This phenotype was associated with elevated plasma levels of unbound IL-18 and the presence of bone marrow hemophagocytes in IL-18BP-/- mice only. In addition, IL-18BP-/- mice displayed higher plasma levels of IFN-γ and CXCL9, as well as increased Ifnγ, Cxcl9 and CIIta mRNA expression in the spleen and liver. IL-18 blockade using an anti-IL-18R antibody attenuated MAS manifestations in IL-18BP-/- mice and abrogated IFN-γ production and downstream signalling. IFN-γ blockade using an anti-IFN-γ antibody also attenuated the MAS phenotype. Conclusions By using IL-18BP-/- mice, we showed that unopposed IL-18 signalling was detrimental in the TLR9-induced MAS model. Importantly, blocking IL-18, as well as IFN-γ, improved disease in IL-18BP-/- mice. Altogether, our results suggest that IL-18 exerts a pathogenic role in this model of MAS, acting upstream of IFN-γ. Disclosure of interest None declared

Background Disease activity and severity of rheumatoid arthritis (RA) appear to be worse in women than in men [1]. The role of parity on disease activity is controversial, since pregnancy is characterized by a lower disease activity, but the postpartum period by an increase in activity [2]. Radiographic joint damage progression represents the cumulative effect of disease activity and allows us to study the long term effect of parity. Objectives To study the impact of parity on radiographic progression in women with RA. Methods This is an observational cohort study of RA patients included in the Swiss Clinical Quality Management in Rheumatoid Arthritis (SCQM-RA). Patients enrolled are followed-up yearly and have x-rays assessments at regular intervals. Information about female hormonal factors, such as pregnancies, breastfeeding, menstrual cycles and hormonal treatment were retrospectively retrieved using a questionnaire. For this analysis we included women with at least two x-rays and full information on reproductive factors. The primary outcome was the rate of radiographic progression (Ratingen erosion score) and the secondary outcome was functional disability progression (Health Assessment Questionnaire-Disability Index (HAQ-DI)). We compared the rate of progression between parous and nulliparous women using a multilevel regression model for longitudinal data, adjusting for potential confounders, such as age, disease duration, DAS 28 and treatment. In a subanalysis we explored if the x-ray progression was more severe during the active parous period, operationally defined as the 10 years following the first pregnancy or miscarriage. Results A total of 726 women were analysed, of which 438 (60%) were parous, with a median number of pregnancies of 2 (IQR: 2–3), a mean of 4.8 x-rays per patient and 10.9 years of follow-up. Baseline patients and disease characteristics were balanced, but parous women were older than nulliparous (median of 49 vs 45 years, p=0.001) (Table 1). During follow-up, erosion progression did not differ significantly between parous and nulliparous women (p=0.94). In a subanalysis, the radiographic progression during the active parous period was not different [0.6% (95% CI: 0.5 to 0.8) vs 0.5% (95% CI: 0.4 to 0.7) by year, respectively, p=0.28]. The decrease of the HAQ-DI score overtime was not different between parous and nulliparous women (p=0.21), and it was not different during the active parous period [-0.02 (95% CI: -0.03 to - 0.01) vs -0.02 (95% CI:- 0.03 to - 0.01) by year, respectively, p=0.67]. We did not find differences in radiographic progression or HAQ-DI score between women with a single pregnancy and multiparous women. Conclusions In women with RA, the progression of structural damage and of functional disability did not differ between parous and nulliparous women. Among parous women, the active parous period was not associated with more radiographic damage progression. Although postpartum period is associated with increase in disease activity, our results suggest that parity does not have a negative long term impact on structural damage. References Camacho EM, et al. Ann Rheum Dis. 2010; 69:1834–37. Pikwer M, et al. Arthritis Res Ther. 2015 Dec 12;17:358. Disclosure of Interest None declared

Background Psoriasis is a common chronic skin disorder caused by a dysregulated crosstalk between immune and resident cells (eg, keratinocytes). The identification of the pathogenic role of several cytokines in psoriasis led to the development of successful therapies. Recently, IL-36 cytokines, which belong to the IL-1 family, were shown to be involved in the pathogenesis of psoriasis. Mice deficient in IL-36 receptor (IL-36R) were protected from imiquimod (IMQ)-induced skin inflammation, whereas IL-36R antagonist (IL-36Ra) KO mice exhibited a more severe phenotype. The objective of our study was to examine the expression and function of IL-38, a newly discovered IL-1 family member with supposed IL-36 inhibitory properties, in the IMQ model of psoriasis. Materials and methods IL-38 mRNA expression was determined in skin samples, at steady state or after IMQ application. IL-38 KO or IL-36Ra KO mice and their respective WT littermates were challenged with the topical application of IMQ on the left ear during 7 days. The severity of skin inflammation was assessed by daily measurement of ear thickness using a calliper, by semi-quantitative histologic scoring, and by measuring mRNA levels of inflammatory markers. Results At the peak of IMQ-induced skin inflammation, IL-38 mRNA levels were lower than in normal skin, whereas IL-36Ra mRNA levels were increased in IMQ treated skin. The severity of skin inflammation, as assessed by ear thickness, histological changes (leukocyte infiltration and epidermis hyperplasia) and pro-inflammatory mediator transcript levels, was not significantly different in IL-38 KO and WT mice. After cessation of topical IMQ application, the resolution of skin inflammation was also not altered by IL-38 deficiency. As opposed to these findings, IL-36Ra deficient mice displayed more severe pathological changes as compared to WT mice. Conclusions We showed that endogenous IL-38 is not involved in the development and the resolution of IMQ-induced skin inflammation. Our findings suggest that IL-38 does not exert IL-36 inhibitory activities in the skin.

Objectives Interleukin (IL)-38 is a newly characterised cytokine that belongs to the IL-1 family. This cytokine is expressed in the rheumatoid arthritis (RA) synovial tissue and IL-38 deficient mice have exacerbated arthritis. Here, we analysed the effect of IL-38 overexpression in the joints of arthritic mice, in human macrophages and synovial fibroblasts in vitro. Methods Articular injections of an adeno-associated virus (AAV) 2/8 encoding IL-38 were performed in collagen-induced arthritis (CIA), K/BxN serum transfer-induced arthritis (STIA) and antigen-induced arthritis (AIA) in mice. The effect of IL-38 overexpression was evaluated through clinical scores, immunohistochemistry, microCT, Luminex and RT-qPCR analysis. THP-1 macrophages were transduced with a lentiviral vector to overexpress IL-38. Results Clinical inflammatory scores were significantly decreased after AAV IL-38 injection in joints of mice with CIA and STIA, but not AIA. This decrease was accompanied by reduced macrophage infiltration and a decreased expression of Th17 cytokines (IL-17, IL-23, IL-22) and TNFα. However, IL-38 overexpression had no effect on cartilage or bone destruction. In vitro, the THP-1 monocytic cell line expressed less IL-6, TNFα and IL-23 after IL-38 overexpression. Conditioned media from these cells, containing released IL-38, also exert an anti-inflammatory effect on human primary macrophages and synovial fibroblasts from patients with RA. Conclusions This study shows for the first time that IL-38 overexpression attenuates the severity of experimental arthritis. IL-38 may exert its anti-inflammatory effects by decreasing the production of proinflammatory cytokines by macrophages and synovial fibroblasts. This effect can lead to the development of novel treatment strategies in arthritis.

Background A significant proportion of patients with Rheumatoid Arthritis (RA) are negative for rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPA). There is a unmet need for new biomarkers to better define the severity and prognosis of RA. Objectives To study the sensitivity and specificity of numerous auto-antibodies and calprotectin in RA patients compared with patients with spondyloarthritis (SpA) and to investigate the relationship between these biomarkes and measures of disease activity and severity. Methods Data was obtained from the “Swiss Clinical Quality Management” (SCQM) registry which recruits adults with RA, psoriatic arthritis (PsA) and axial spondylarthritis (axSpA). All patients with a Biobank sample were tested for RF (IgM and IgA, QUANTA Flash (QF), Inova), ACPA (anti-CCP2 (Eurodiagnostica) and anti-CCP3 (IgG QF and QUANTA Lite (QL) and IgA QF, Inova), anti-carbamylated proteins (anti-CarP, carbamylated fetal calf serum, prototype ELISA, Inova), anti-peptidyl arginine deiminase type-3 antibodies (anti-PAD3-E.Coli QF and anti-PAD3-insect QF, Inova) and serum calprotectin (QL prototype ELISA, Inova). The control group consisted of patients with PsA (CASPAR criteria positive) and axSpA (ASAS criteria positive). In univariable analyses we tested for associations between the biomarkers and the 28-joint disease activity score (DAS28) and the presence of erosive disease at inclusion. Multivariable analyses were corrected for age, sex, disease duration and anti-CCP2 positivity. Results A total of 1471 patients were included, 969 with RA and 502 in the SpA control group (317 with axSpA and 185 with PsA). Statistical measures of the biomarkers for the diagnosis of RA are shown in Table 1. In univariable analyses, RA patients positive for anti-PAD3-E.Coli, anti-PAD3-insect or calprotectin were significantly more likely to have a moderate to high DAS28 (DAS28>3.2) than patients with negative values (68.2% vs. 54.8%, p=0.001, 67.7% vs. 56.0%, p=0.027 and 70.4% vs. 56.2%, p=0.020 respectively). RA patients positive for anti-CarP, anti-PAD3-E.Coli or anti-PAD3-insect were significantly more likely to demonstrate joint erosions compared with patients with negative values (66.3% vs. 57.7%, p=0.037, 68.5% vs. 57.7%, p=0.015 and 71.6% vs. 58.3%, p=0.020 respectively). In multivariable analyses, the odd ratios (and 95% confidence intervals) for the association of anti-PAD3-E.Coli, anti-PAD3-insect or calprotectin with a moderate to high DAS28 were 1.65 (1.15, 2.39), 1.64 (1.02, 2.64) and 1.90 (1.08, 3.35) respectively, and for the association of anti-CarP, anti-PAD3-E.Coli or anti-PAD3-insect with joint erosions were 1.32 (0.90, 1.93), 1.43 (0.95, 2.15) and 1.82 (1.05, 3.15) respectively. Conclusions Although anti-PAD3 and calprotectin are only present in a minority of RA patients, they are significantly associated with more active disease and anti-PAD3-insect is significantly associated with the presence of joint erosions, independently of ACPA (anti-CCP2). Disclosure of Interest None declared

Background Rheumatoid arthritis (RA) has been associated with an increased risk of cardiovascular (CV) morbidity and mortality. Recent data indicated that this could also be the case in psoriatic arthritis (PsA) and axial spondyloarthritis (AxSpA) patients. Objectives To compare the risk of major adverse cardiovascular events (MACE) in RA, PsA and AxSpa as of disease onset. Methods A mixed retrospective and prospective cohort study was conducted using data from patients included in the Swiss Clinical Quality Management (SCQM) registry. Patients diagnosed with RA, PsA, or AxSpA at an age of ≥18 years without a history of MACE (i.e., non-lethal myocardial infarction, non-lethal stroke and death from cardiovascular event) at the time of first articular symptoms or, if not available, at the time of diagnosis were included. The primary outcome was the incidence of first MACE, using self-reported or physician-reported data. For each patient we recorded the time in years from disease onset until either occurrence of first MACE or end of follow-up. The exposure of interest was the type of rheumatic disease: RA, PSA or AxSpA. In addition to sex and age at disease onset, we assessed traditional CV risk factors: known familial history of MACE before age 50, and presence of ever-hypertension, ever-diabetes, ever-hyperlipidemia, and ever-smoking. We applied Poisson regression to estimate and compare the incidence rates for MACE. Results 5311 patients (3068 RA, 1462 AxSpA, and 781 PsA patients) were eligible and contributed a total follow-up time of 379459 years for RA, 199818 for AxSpA, and 99159 for PsA. Diseases started between 1950 and 2014 and differed significantly with respect to assessed CV risk factors (Table 1). A total of 98 RA, 28 AxSpA, and 11 PsA patients had experienced at least one MACE since disease onset. The unadjusted incidence rate of MACE per 1000 person-years was 2.62 for RA, 1.41 for AxSpA, and 1.20 for PsA (p=0.0011). After adjusting for traditional CV risk factors, age at disease onset, sex, and year of disease onset (categorized as: <1985, 1985–2000, or >2000) the significant difference between PsA and RA persisted (MACE incidence rate ratio (IRR): 0.45, 95% confidence interval (CI): 0.22- 0.91), while the difference between AxSpA and RA was decreased and no longer significant (IRR: 0.86, 95% CI: 0.5 -1.49). Because ever-hypertension, ever-diabetes, ever-hyperlipidemia, and ever-smoking may not reflect the status at time of disease onset, we analyzed the risk of MACE also with only adjusting for sex, age at disease onset, disease onset and family history of early MACE. The results (n=5311) were very similar, suggesting that especially age and sex are important confounders when comparing incidence rates across diseases. Male sex, older age at disease onset, a known family history of early MACE, ever-hypertension, and ever-hyperlipidemia were also significantly positively associated with the risk of MACE. Conclusions Taking into account traditional CV risk factors, age and sex, the type of rheumatic disease remained associated with risk of MACE. RA seems to convey a larger CV risk than PsA, while the difference between RA and AxSpA remains unclear. Disclosure of Interest None declared

Background Statins are widely used serum cholesterol-lowering drugs. Because of the increased CV risk many rheumatoid arthritis (RA) patients are receiving statin therapy. Statins have also been linked to anti-inflammatory effects, but its effects on inflammatory activity have been inconsistent in the literature. More recently, statins have been associated with effects on bone metabolism, such as increased bone mineral density, decreased fracture risk, and lower risk of periprosthetic osteolysis following total hip arthroplasty. The impact of statins on the progression of structural joint damage in RA has not been studied. Objectives To compare the rate of radiographic damage progression in RA patients using concomitant statins or not in a large prospective RA cohort. Methods This is a prospective observational cohort study nested within the Swiss RA registry (SCQM-RA). The SCQM monitors disease activity, radiographic damage, patient characteristics and treatments at regular intervals. All patients in the SCQM-RA database with sequential X-rays and information on statin use were included. The exposure of interest was concomitant statin use as reported by the treating rheumatologist and/or the patient and categorized as ever or never. To minimize exposure misclassification or false negatives, we excluded patients with hypercholesterolemia, but no information on lipid lowering therapy. The primary end point was radiographic disease progression as measured by the rate of change from baseline in radiographic damage scores. The damage score (ERO) was assessed on 38 joints of hands and feet with a validated scoring method (Ratingen score) by a single experienced reader, blinded to clinical history. We analyzed the rate of ERO progression in pts treated with statins or not using a mixed regression model for longitudinal data, adjusting for potential confounding factors. Results 4213 RA patients with a median of 4 [2-6] sequential X-rays/pt and 3.9 [2.0-6.4] years of follow-up/pt were included. 493 (11%) of pts were taking statins during follow-up. Statin users were significantly more often males (34% versus 23%, p<0.001), older (mean age 58 versus 54 years, p<0.001) and overweight (mean BMI 27.0 versus 24.9, p<0.001). RA treatment and disease characteristics were balanced between statin-users and non-users. After adjusting for differences in baseline prognostic factors, we found no significant difference in ERO progression in statin-users compared to non-users (ERO progression in statin-users 0.98% of the maximum score per year (95% CI: 0.46 – 1.49), compared to 0.95% (95% CI: 0.70 – 1.20); p=0.36). Disease characteristics associated with higher ERO were male sex, longer disease durations, rheumatoid factor positivity, higher disease activity and treatment types. Conclusions The results of this study do not support the hypothesis that statins are protecting against progression of structural joint damage and bone erosions in RA patients. Disclosure of Interest None declared

Background Treatment of rheumatoid arthritis (RA) with a combination of methotrexate (MTX)+adalimumab (ADA) is more effective than ADA monotherapy. We assessed the toxicity of different doses of MTX and treatment efficacy of ADA+MTX in two trials. Methods Data originated from CONCERTO, in patients with early RA initiating ADA+ 2.5, 5, 10 or 20 mg/week MTX for 26 weeks; and MUSICA, in patients with an inadequate response to MTX initiating ADA+ 7.5 or 20 mg/week MTX for 24 weeks. Efficacy was assessed by the American College of Rheumatology 50 (ACR50). Patient-reported MTX-related toxicity information was collected at each visit on 18 prespecified MTX-related adverse events (AE) in the MTX label. Results In CONCERTO, ACR50 rates increased over time, ranging from 54% to 68% at week 26, while AE rates remained steady, ranging from 2.4% to 17.8% at week 26. Of 395 patients, 113 (28.6%) reported 345 MTX-related AEs, including one serious AE (SAE, excessive fatigue and/or malaise); 10 AEs (in two patients) led to study discontinuation. In MUSICA, ACR50 rates increased over time, and were 32.3% and 37.5% at week 24, while MTX-related AE rates remained steady and were 6.5% at week 24. Of 309 patients, 71 (23%) reported 185 MTX-related AEs, including 5 SAEs (four infections and one fever/chills); six AEs (in four patients) led to study discontinuation. Conclusion In patients with RA initiating ADA+MTX combination, treatment efficacy was achieved and increased throughout both trials, while rates of MTX-related AEs remained steady. MTX-related AEs were observed in up to 30% of patients and most were mild. MTX was discontinued by 0.5%-1.3% of patients.

Background Clinical and epidemiological evidence implicates elevated lipoprotein (a) [Lp(a)] level as a risk factor for cardiovascular disease (CVD).1 Lp(a) is also implicated as a causal agent in the atherothrombotic process.1 RA pts are at increased risk for CVD and have higher Lp(a) levels than the general population.2 Treatment guidelines recommend reduction of Lp(a) at levels >50 mg/dL.3 Niacin is the most recognized means to lower Lp(a) but is no longer recommended given recent trial results.4 Although some evidence implies a beneficial role for TNF inhibitors in reducing CV risk and mortality in RA pts, their effect on Lp(a) remains unclear. Some studies report a decrease in Lp(a) following treatment with MTX and TNF inhibitors; others show no such effects.5–7 Interleukin-6 (IL-6), a proinflammatory cytokine, has been shown to increase Lp(a) levels in monkey hepatocyte cultures.8 Conversely, inhibition of IL-6 signaling with TCZ decreases serum Lp(a) levels.9,10 Objectives This post hoc analysis examined the response of Lp(a) to TCZ in comparison to ADA and to MTX in two separate clinical trials. Methods ADACTA was a randomized, double-blind, phase 4 study in RA pts (TCZ, n=163; ADA, n=163).11 Pts received TCZ 8 mg/kg IV Q4W or ADA 40 mg SC Q2W for 24 wks, both as monotherapy. MEASURE was a randomized, double-blind, phase 3 study in RA pts (TCZ + MTX, n=69; pbo + MTX, n=63).10 Pts received TCZ 8 mg/kg IV Q4W or pbo IV Q4W, both in combination with MTX. Serum samples were analyzed at baseline (BL) and wk 8 for Lp(a). Change from BL to wk 8 in Lp(a) was summarized within each study split by EULAR and ACR50 response at wk 24. Shifts from >50 mg/dL to ≤50 mg/dL in Lp(a) from BL to wk 8 were summarized for each study. Results A greater reduction in Lp(a) was seen in pts receiving TCZ compared to those receiving ADA (p<0.0001) and in pts receiving TCZ + MTX compared to those receiving MTX (p<0.0001) (Table 1). Although the reduction was greater in both ACR50 and EULAR responders compared to nonresponders, the numerical differences were mostly small. At wk 8, numerically higher proportions of pts with BL Lp(a) >50 mg/dL improved (defined as achieving Lp(a) ≤50 mg/dL) in the TCZ group (11/21 [52.4%]) compared to the ADA group (6/24 [25%]) and in the TCZ + MTX group (7/12 [58.3%]) compared to the MTX group (2/20 [10%]). Conclusions The current analysis showed greater reduction of Lp(a) levels in RA pts treated with TCZ monotherapy compared to ADA and with TCZ + MTX compared to MTX. This effect appears to be partially related to treatment response. These results suggest that IL-6 receptor inhibition may modulate Lp(a) more than MTX or TNF blockade with ADA. More studies are needed to elucidate the hierarchy of signaling pathways that modulate Lp(a) and how this may impact CV risk. References Curr Atheroscler Rep 2013;15:360. Circulation 2003;108:2957. Eur Heart J 2010;31:2844. N Engl J Med 2011;365:2255. Ann NY Acad Sci 2006;1069:414. Arthritis Rheum 2007;56:831. Clin Exp Rheum 2013;31:415-21. Arterioscler Thromb Vasc Biol 1998;18:984. PLoS One 2010;5:e14328. Ann Rheum Dis 2013 doi:10.1136. Lancet 2013;381:1541. Disclosure of Interest C. Gabay Grant/research support: Roche, AbbVie, Merck, Consultant for: Roche, AbbVie, Merck, I. McInnes Grant/research support: Pfizer, Roche, Consultant for: BMS, Pfizer, NovoNordisk, AstraZeneca, Eli Lilly, UCB, Speakers bureau: BMS, Pfizer, NovoNordisk, AstraZeneca, Eli Lilly, UCB, A. Kavanaugh Grant/research support: Roche/Genentech, K. Tuckwell Employee of: Roche, N. Collinson Employee of: Roche, M. Klearman Shareholder of: Roche/Genentech, Employee of: Roche/Genentech, J. Green Employee of: Roche, N. Sattar Consultant for: Roche, Speakers bureau: Roche DOI 10.1136/annrheumdis-2014-eular.3030