S. Ohshima

OP0191 The Fecal Microbiota of Rheumatoid Arthritis Patients Differs from that of Healthy Volunteers and is Considerably Altered by Treatment with Biologics

Background New insights from DNA sequence-based analyses of gut microbial communities and the renewed interest in mucosal immunology suggest that the microbiome is an important environmental factor influencing autoimmune disease manifestations. Objectives To investigate the fecal microbiota of rheumatoid arthritis (RA) patients and healthy volunteers (HVs), we used the Yakult Intestinal Flora-SCAN (YIF-SCAN®), based on reverse transcription–quantitative polymerase chain reaction, with primers specific for target bacterial rRNA. We also investigated whether the fecal microbiota changed in RA patients due to biologic (Bio) treatment. Methods Fecal samples were collected from 55 RA patients and 77 HVs in NHO Osaka Minami Medical Center during 2011–2012. YIF-SCAN® was performed to quantify the bacterial count with 16S or 23S rRNA-targeted group-specific primers. 25 RA patients were re-examined after they underwent Bio treatment (tocilizumab, 10; infliximab, 9; etanercept, 3; adalimumab, 1; abatacept, 2) for 6 mo. Patients who had received antibiotic therapy were excluded. Results The mean age was 57.8 ± 14.5 and 36.3 ± 10.7 years; mean body weight, 54.5 ± 10.0 and 57.0 ± 9.0 kg; and the proportion of women, 72.7% and 90.9%, for the RA patients and HVs, respectively. From the facultative anaerobic groups, the total population levels of Lactobacillus (especially the Lactobacillus gasseri, L. reuteri, and L. fermentum subgroups) and Enterococcus, Enterobacteriaceae were significantly higher in RA. These results support those of our 2012 study [1]. The background of the 25 RA patients reanalyzed after treatment is shown in Table 1. The DAS28CRP was 4.1 ± 1.1 at baseline and 2.5 ± 1.1 at 6 mo. 12 patients (48%)achieved remission (DAS28CRP < 2.3) at 6 mo. The total bacterial count and total population levels of the Clostridium. coccoides group, Bifidobacterium, and the L. gasseri and L. plantarum subgroups significantly decreased at 6 mo (Table 2). Conclusions We confirmed the difference in the fecal microbiota of RA patients and HVs. Bio treatment considerably reduced commensal bacteria in RA patients. To our knowledge, this is the first study describing changes in fecal microbiota due to Bio treatment. Further studies are required to elucidate the clinical significance of these results. References Y.maeda, et al. Ann Rheum Dis 2012;71(Suppl3):496 Disclosure of Interest None Declared

THU0442 Serum Prepsepsin (Soluble CD14-Subtype) as a Novel Useful Biomaker for Infection in Patients with Rheumatoid Arthritis (RA)

Background Infection is one of the serious complications seen in the management of RA patients. The acute inflammatory marker C-reactive protein (CRP) is elevated both during infection and during high disease activity of RA, and this often poses a problem when distinguishing the two. The soluble CD14 subtype, presepsin has been reported to be a novel effective marker for the diagnosis of sepsis but has not been evaluated in RA patients. Objectives To evaluate the use of presepsin in RA patients during an infectious event. Methods 25 RA patients with infections, 21 RA patients with high disease activity, 23 healthy controls (HC) were enrolled in this study. RA patients in whom the pathogens were identified (22 bacterias, 2 viruses, and 1 M. tuberculosis) were designated as the infection RA group (iRA), high disease activity RA patients without infection were designated as the flare RA group (fRA). Presepsin was measured using a chemiluminescent enzyme immunoassay. CRP and procalcitonin (PCT) were also measured. RA disease activity was evaluated using DAS28-CRP. Levels of respective measurements at both pre- and post-treatment were analyzed using the Wilcoxon signed-rank test, and comparisons of levels within each group were analyzed using the Mann-Whitney’s U-test. Additionally, Spearman’s rank correlation coefficient was used to analyze the correlation of levels of presepsin, CRP, and PCT in iRA and correlation of presepsin, DAS28-CRP, and CRP in fRA. Further, AUC was obtained from the ROC analysis. Treatment for iRA included antibiotics, antivirals, and treatment for fRA included corticosteroids, DMARDs, and biologics. Results In fRA, average level of CRP was 2.4±2.1mg/dl, DAS28-CRP was 4.2±1.31. At pre-treatment, levels of presepsin in iRA (2088.4±4243.7pg/ml) was significantly higher compared to in fRA (319.3±321.8pg/ml, p<0.01). Both levels were significantly higher compared to those in HC (136±57.0pg/ml). In iRA, presepsin level correlated with CRP (r=0.65, p<0.01) and PCT (r=0.48, p<0.05). In fRA, presepsin level did not correlate with CRP or DAS28-CRP. After treatment, levels of prepsin (p<0.001), CRP (p<0.001), and PCT (p<0.001) were significant decreased in iRA. On the other hand, in fRA, CRP (p<0.001) and DAS28-CRP (p<0.001) were significantly decreased after treatment, however presepsin level showed no significant change (p=0.37). Furthermore, presepsin levels in fRA with low disase activity after treatment were significantly higher compared to those in HC (p<0.01). ROC analysis of iRA showed that AUC levels for presepsin was 0.817, indicating the efficacy of presepsin for diagnosis of infection in RA. Conclusions Presepsin is an effective diagnostic marker for infection in RA patients. References Y. Yaegashi, et al., J Infect Chemother 2005;11:234-238 T. Shozushima, et al., J Infect Chemother 2011;17:764-769 Disclosure of Interest None Declared

SAT0002 Quantitative CD64 Molecules on Peripheral Blood Monocytes in Systemic Lupus Erythematosus (SLE)

Background The majority of SLE patients show evidence of excess type I interferon (IFN-I) production, a phenotype associated with renal disease and certain autoantibodies. However, detection of IFN-I proteins in serum is expensive examination and time consuming. Research in the past few years has demonstrated that their relative expression on circulating monocytes/macrophages may be modified by cytokines, including interferon (IFN)-α1. Little data exist on the in vivo expression of specific FcγRI, CD64 in SLE. However, the quantitative expression of CD64 on monocytes has not been reported. For the treatment early intervention, more rapid disease activity assessment is required in SLE. Objectives To determine whether quantitative expression of CD64 on circulating monocytes (mCD64) is associated with systemic exacerbation in SLE patients. Methods In this study, 10 SLE patients with high disease activity (aSLE), 20 SLE patients with quiescent activity (qSLE) and 20 healthy controls (HC) were enrolled. We measured the expression level of CD64 per monocyte quantitatively by flow cytometry. Comparisons of levels within each group were analyzed, and levels of respective measurements at both pre- and post-treatment in aSLE were analyzed. The expression level of mCD64 was compared with SLE disease activity index (SLEDAI) and SLE disease activity biomarkers (anti DNA antibody and compliments titer). Results aSLE had a mean SLEDAI of 12.1±7, and qSLE SLEDAI was 3.0±1.6. In aSLE and qSLE, mean expression level of mCD64 was 39800±16609 and 21759±7423 molecules/cell (p<0.01), respectively. Both levels were significantly higher than HC (15325±2731). Moreover, the level of mCD64 was correlated with SLEDAI (r=0.66, p<0.001), anti DNA antibody (r=0.69, p<0.01) and CH50 (r=-0.69, p<0.001). After treatment, the value of mCD64 was significantly decreased in aSLE (p<0.05). Conclusions Our results suggest that quantitative measurement of CD64 expression on monocytes can be useful to detect disease exacerbation in SLE. Flow cytometry analysis of CD64 expression on circulating monocytes is a convenient and rapid approach for estimating IFN-α levels in SLE. Quantitative mCD64 could become a novel biomarker to replace the SLE disease activity marker from that found a correlation with markers of disease activity, and reduced by treatment advantage. Further quantitative measurement of CD64 on flow cytometry will be necessary to obtain consistency of result between facilities. References Li, Y. et al. Monocyte surface expression of FCγreceptor RI (CD64), a biomarker reflecting type-I interferon levels in systemic lupus erythematosus. Arthritis Research & Therapy 2010, 12:R90. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.2432

AB0449 Baseline Procalcitonin (PCT) Level as A Predictive Marker for Clinical Remission (DAS28-ESR, CDAI) at 52 Weeks in Biologic NaÏVe Rheumatoid Arthritis (RA) Patients Treated by TOCILIZUMAB (TCZ); A Single Center Retrospective Study

Background The goal in treating RA has changed from clinical remission to structural remission with the increased use of biological agents (Bio) which target the pro-inflammatory cytokines interleukin (IL)-6 and tumor necrosis factor (TNF)-α. PCT, a useful marker of infection, has been shown to increase in mRNA expression in peripheral blood mononuclear cells by stimulated by pro-inflammatory cytokine, IL-6 or TNF-α. Objectives This study investigates whether the levels of following parameters (PCT, modified HAQ (mHAQ), MMP-3, RF, ESR, CRP and ACPA) at baseline (BL) of TCZ treatment can be used to predict clinical remission (DAS28-ESR, CDAI) at 52 weeks after the start of TCZ treatment. Methods Bio naïve RA patients who can be observed at Week 52 were analyzed in this study. PCT (n=34), mHAQ (n=48), matrix metalloproteinase (MMP)-3 (n=54), RF (n=54), ESR (n=53), CRP (n=56) and ACPA (n=52) were assessed at BL. The patients were divided into 2 groups, based on DAS28-ESR remission (DAS28-ESR<2.6) at Week 52. For each variable where there was a significant difference between the remission and non-remission groups, receiver operating characteristic (ROC) analysis was performed and cut-off values (COV) were found. For each of those variables, 2 groups were formed by dividing the Bio naïve RA at the COV: the under COV (U) group and over COV (O) group, and the Week 52 clinical remission (DAS28-ESR, CDAI) rate in each group were analyzed. Results The variables with a significant difference between the remission and non-remission groups were PCT (p=0.001), mHAQ (p=0.014), RF (p=0.003), ESR (p=0.001) and CRP (p=0.041). The COVs were 0.027 ng/ml for PCT, 100U/ml for RF, 0.38 for mHAQ, 18 mm/hr for ESR, 1.96 mg/dl for CRP. For each variable, the DAS28-ESR remission rate in U group (PCT: 100%, mHAQ: 70.8%, RF: 77.8%, ESR: 87.5% and CRP: 64.3%) was significantly higher than in O group. Additionally, the CDAI remission rate in U group (PCT: 50.0% (p=0.040) and RF: 32.1% (p=0.026)) was significantly higher than in O group, but not others. For PCT at BL, DAS28-ESR or CDAI were no different between the U group and O group and no correlation with DAS28-ESR or CDAI were observed. BL DAS28-ESR was significantly lower in U groups for all other variables. Furthermore, BL PCT had correlation with DAS28-ESR and CDAI at Week 52. Conclusions BL PCT level is a useful predictive marker for clinical remission (DAS28-ESR, CDAI) at week 52 in Bio naïve RA treated by TCZ. Moreover, unlike other variables, it is not affected by the BL disease activity level. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.2437

SAT0003 Quantitative Monocyte CD64 (MCD64) Expression is Useful Biomarker for Disease Activity in Systemic Lupus Erythematosus (SLE) Patients

Background Interferon (IFN)-α has been largely implicated in the ethiopathogenesis of SLE. The activation of IFN-α might be important in the prognosis and activity assessment of the disease. CD64 (FcγRI) is upregulated on monocytes as a response to IFN-I1. Flow cytometry analysis of mCD64 expression (Mean Fluorescence Intensity) is a convenient and rapid approach for estimating IFN-α levels in SLE patients. Additionally, macrophage colony-stimulating factor (M-CSF), which is involved in the differentiation of monocyte/macrophage, affects CD64 expression on monocyte2 has been reported to increase in levels in SLE patients,3 and is also involved in SLE disease activity4. mCD64 can be the quantified easily and the measurement is consistent among the respective facilities. However, the quantitative mCD64 of SLE patients has not been reported. Objectives We investigated the levels of mCD64 by quantitative flow cytometry to assess the usefulness of it as a SLE disease activity biomarker. Methods 30 SLE patients (10 active SLE, 20 inactive SLE) and 20 healthy controls (HC) were in this study. SLE disease activity was evaluated using by SLE-Disease Activity Index (SLEDAI) score. mCD64, SLE activity biomarkers (anti-DNA antibody and complement titer (CH50)), IFN-α and M-CSF were measured in SLE patients. mCD64 was measured by a quantitative flow cytometry using fluorescene microbeads. INF-α and M-CSF levels were measured using an enzyme-linked immunoabsorbent assay (ELISA). Correlational analysis between levels of mCD64, SLEDAI, SLE activity biomarkers, INF-α and M-CSF in each group were evaluated. Results Disease activity markers are shown Table 1. The variables with significant difference between active SLE and inactive SLE were SLEDAI (p<0.05), CH50 (p<0.05), anti DNA antibody (p<0.05), mCD64 (p<0.05), IFN-α (p<0.05), M-CSF (p<0.05).mCD64 levels were correlated with SLEDAI (r=0.6669, p<0.001), anti-DNA antibody (r=0.6887, p<0.05) and CH50 (r=-0.6928, p<0.001), but were not correlated with C3 or C4 levels (data not shown). Further, mCD64 levels were correlated with IFN-α (r=0.6089, p<0.001) and M-CSF (r=0.64443, p<0.001) Table 1. Disease activity markers in SLE patients (mean ± SD) Active SLE Inactive SLE SLEDAI 1.3±30.39 3.0±4.92 Anti-DNA antibody (IU/ml) 144.1±146.7 12.4±10.83 CH50 (U/ml) 25.3±19.16 41.4±13.08 mCD64 (molecules/cell) 38152±15323 21955±5359 IFN-α (pg/ml) 30.4±57.1 7.7±4.9 M-CSF (pg/ml) 752.9±580.7 356.9±124 Conclusions This study suggests that quantitative CD64 molecules expressed on monocytes can be a useful disease activity biomarker in SLE patients. References Li, Y. et al. Arthritis Research & Therapy 2010, 12:R90 Ji XH et al. Acta Pharmacol Sin. 2004; 25(10): 1361-5 Yang PT et al. Ann Rheum Dis. 2008, 67(3): 429-30 Tsuji S. et al. The Journal of Tokyo Medical University vol.70, No.1: 151-8 Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.2824

FRI0091 Neutrophil CD64 (NCD64) as A Useful Marker for Differentiating Organizing Pneumonia (OP) from Bacterial Pneumonia (BP) in Rheumatoid Arthritis (RA)

Background In rheumatoid arthritis (RA), pulmonary involvement is one of the important complication.Both organizing pneumonia (OP) and bacterial pneumonia (BP) can be evident during the course of RA. However, we often have difficulties with differentiating OP from BP clinically, because both present similar infiltrations in chest computed tomography (CT). Corticosteroid is the standard therapy for OP, on the other hand, antibiotics is necessary for BP.CD64 on neutrophils (nCD64) is FcγRI known to be upregulated as a response to infection. Objectives To evaluate the usefulness of nCD64 for differentiating OP from BP. Methods Eleven RA patients diagnosed as pneumonia with infiltrative shadows in their chest CT were enrolled in this study. Five patients were successfully treated only with predonisolone (PSL). Three of them were pathologically confirmed the tissue of lung lesions by transbronchial lung biopsy (TBLB). They were diagnosed as OP. Nine patients were successfully treated with only broad-spectrum antibiotics (ABx), diagnosed as BP. All patients received blood examinations before treatment and following parameters were examined. The expression level of CD64 per neutrophil (nCD64) was quantitatively measured by using flow cytometery. C-reactive protein (CRP) and procalcitonin (PCT) were also measured. Comparisons of levels within each group were analyzed using the Mann-Whitneys U-test. Results nCD64 levels in OP were significantly lower than those in BP (p=0.0164). On the other hand, in CRP and PCT, there were no differences between OP and BP. Among these biomarkers, only nCD64 can distinguish OP from BP, indicating the efficacy of nCD64 for diagnosis of OP. OP (n=5) BP (n=9) p value nCD64 (molecules/cell) 1787±630 9248±7049 0.0164¶ CRP (mg/dl) 7.13±5.6 10.4±5.8 0.23 PCT (ng/ml) 0.056±0.027 0.187±0.14 0.07 ¶ Statistically signifiacnt. Conclusions nCD64 is an effective diagnostic marker for differentiating OP from BP in RA patients. This novel marker would be contribute to both early diagnosis and suitable treatment. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.2901

AB0943 Monocyte/Neutrophil (M/N) CD64 Ratio is Useful for Differentiating Infection from Disease Activity in Systemic Lupus Erythematosus (SLE) Patients

Background Clinical decision-making in the management of fever is difficult when distinguishing infection and high disease activity in SLE. CD64 (FcγRI) is upregulated on monocytes as a response to interferon (IFN)-α in SLE1. In infection, upregulation of CD64 expression on neutrophils occurs with IFN-γ, granulocyte colony stimulating factor, or lipopolysaccharide. CD64 on neutrophils is known to be a sensitive and specific marker for detection of infection in rheumatoid arthritis patients (using a cutoff value of 2000 molecules per cell)2. Objectives We examined the utility of quantitative CD64 molecules expressed on monocytes and neutrophils, and calculated mCD64/nCD64 ratio as a marker for distinction between infection and disease activity in SLE. Methods In this study, 12 SLE patients with infections (iSLE), 10 SLE patients with high disease activity, (aSLE) (SLEDAI 12.1±7) and 20 healthy controls (HC) were enrolled. The expression level of CD64 per monocyte and neutrophil quantitatively were measured by flow cytometry, and the mCD64/nCD64 ratio were calculated. Comparisons of levels within each group were analyzed, and levels of respective measurements at both pre- and post-treatment were analyzed. Results Results of mCD64, nCD64 and m/nCD64 ratio are presented in Table1. The nCD64 value of iSLE and aSLE was 2000 molecules/cell or more together. There was no difference in mCD64 of iSLE and aSLE. m/nCD64 ratio was significantly lower among iSLE than in those without infection, aSLE, or healthy controls (p<0.001). After each treatment, the value of mCD64 and nCD64 were significantly decreased in iSLE and aSLE. But there was no change in mCD64/nCD64 ratio in aSLE, on the other hand it was increased in iSLE (p<0.01). Table 1. Quantitative mCD64 and nCD64, mCD64/nCD64 ratio in SLE patients (mean (± SD) CD64 analysis iSLE aSLE HC (n=12) (n=10) (n=20) mCD64 (molecules/cell) 49415±20544 39800±16609 15325±2731 nCD64 (molecules/cell) 8397±4450 2835±1370 1167±411 mCD64/nCD64 ratio 7.22±3.09 15.08±4.75 14.0±3.65 Conclusions A cutoff value of 2000 molecules per cell on neutrophils did not clearly distinguish infection from exacerbation in SLE. However, our results suggest that calculating m/n CD64 ratio can be useful to distinguish infection or disease activity in SLE. Moreover, combined mCD64 and nCD64 could help to guide therapeutic decision in SLE. References Li, Y. et al. Arthritis Research & Therapy 2010, 12:R90. Matsui, T. et al. The Journal of Rheumatology 2006; 33: 2416-24. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.2779

SAT0144 The Causes of Discontinuation of Biologics(Bio)-Use in the Treatment of Rheumatoid Arthritis (RA) Under Practical Circumstances in Japan: from the “Ninja” Registry

Background With recent astonishing advances in drug therapy, “remission” has become a realistic therapeutic goal in RA. Among them the most powerful driving force might be the advent of “biologics (Bio)” targeting cytokines and lymphocytes. In fact, they have shown outstanding efficacy and become a master of drug. However, some issues remain to be addressed and resolved, including adverse effects such as serious infections, therapeutic failure and high cost. Objectives To examine the causes of discontinuation of Bio in 2,274 Japanese RA patients treated with Bio under practical circumstances. Methods The causes of discontinuation of Bio-use were analyzed by using the data from the “NinJa”, one of the largest Japanese RA patient registry established since 2002 and now consists of the clinical data of 10,368 patients from more than 30 hospitals. Results 2,274 (22.1%) of the total registered patients were treated with some Bio. Among them, 352 (15.5%), 940 (41.3%), 258 (11.3%), 63 (2.8%), 459 (20.2%), and 206 (9.1%) of the patients were treated with IFX, ETN, ADA, TCZ and ABT respectively. The cases (percentages) of discontinuation of Bio-use were 62 (17.6%), 71 (7.6%), 55 (21.3%), 39 (8.5%), and 20 (9.7%) in IFX, ETN, ADA, TCZ and ABT respectively. The causes of discontinuation were analyzed and shown in the following Table. Regarding “Failure”, the discontinuation rate was significantly higher ADA. In “Adverse effects”, there was no significant difference among all Bio, although a tendency of higher rate was observed in IFX and ADA. In contrast, it was relatively low in ABT. In addition, the discontinuation due to “Remission” was significantly higher in IFX compared to other Bio. Conclusions There were some characteristics in the causes of discontinuation among Bio. It seems to be feasible to achieve Bio-free remission in some Bio. With further investigations, these findings might provide us a clue to know how and which cases can be achieved Bio-free remission. Acknowledgements This study was supported by both the grants of NHO (National Hospital Organization) and the Ministry of Health, Welfare and Labor. Disclosure of Interest None Declared

SAT0415 Diagnostic and predictive value of novel method for analyzing IGG galactosylation in rheumatoid arthritis

Background While the emphasis in rheumatoid arthritis (RA) treatment is on early diagnosis and intervention, we cannot provide the certain strategies for diagnosis. It is well known that serum immunoglobulin G (IgG) from RA patient exhibit decreased galactosylation of conservative N-glycans in CH2 domains of the heavy chains. Kondo et al. developed novel high-performance liquid chromatography (HPLC) method for precise analysis of oligosaccharides of IgG1). Objectives To evaluate both diagnostic and predictive value of agalactosyl IgG in RA, we analyze serum IgG oligosaccharide structures using HPLC both in established RA, and in early undifferentiated arthritis as a diagnostic marker for future development ofRA. Methods Firstly, serum samples were collected from 94 patients with RA fulfilled the 1987-ACR-criteria, 16 with OA, 18 with SLE, and 24 with healthy subjects (HS). The percentage agalactosyl IgG oligosaccharides (%G0) were analyzed in serum samples by high performance HPLC method. Secondly, we analyzed %G0 in 144 patients with recent-onset undifferentiated arthritis (UA) defined as not fulfilling existing classification criteria. They were studied for whether UA progressed to RA after 2 year. Results In the first study, serum levels of %G0 were significantly higher in established RA (41.5±9.1) than in HS (21.1±4.4; p<0.01), OA (26.2±7.1; p<0.05), and SLE (28.7±10.4; p<0.05). The second study comprised 144 patients (35 males; 109 females) with early UA. The following parameters were assessed at baseline: %G0, rheumatoid factor (RF), autoantibody to galactose deficient IgG (CARF) and anti-cyclic citrullinated peptide antibodies (ACPA). Of the 144 patients with early UA, 58 (40.3%) developed RA, 9 (6.2%) developed collagen disease, 33 (22.9%) diagnosed as OA, 21 (14.6%) were UA as ever, and 23 (16.0%) were healthy (transient arthropathy) within two years after registration. %G0 predicted progression of UA to RA with high accuracy (AUC=0.807) at a highest sensitivity of 79.3% among 4 diagnostic markers (Table 1). Moreover, combination of %G0 and ACPA predicted progression of UA to RA with highest accuracy (AUC=0.905; 95% CI, 0.835-0.962). Conclusions In conclusion, this novel method for analyzing agalactosyl IgG revealed excellent diagnostic value inRA. %G0 as well as ACPA would contribute to prediction of RA, allowing early therapeutic decisions. Furthermore, agalactosyl IgG may be involved in pathogenesis of RA. References Kondo A, et al. Separation of pyridylamino oligosaccharides by high-performance liquid chromatography on an amine-bearing silica column. Anal Biochem 1994, 219, (1), 21-5. Disclosure of Interest None Declared

SAT0059 Association of Serum Antibody Responses to Porphyromonas Gingivalis and Periodontal Conditions with Clinical Response to Biologics in Rheumatoid Arthritis Patients

Background There is little evidence on prediction of clinical response to biologics in patients with RA. Recently, it has been suggested that clinical response to tumor necrosis factor (TNF) inhibitors was related to periodontal conditions, which might be affected by infection with Porphyromonas gingivalis. Additionally, serum antibody responses to P. gingivalis were associated with RA and periodontitis. Objectives To evaluate whether serum antibody responses to P. gingivalis antigens and periodontal conditions are associated with clinical response to biologics in patients with RA. Methods Sixteen patients with severe RA enrolled in this study were treated with biologics according to the usual regimen (7 patients with TNF inhibitors, 6 with IL-6 blocker and 3 with CTLA-4 Ig). DAS28-CRP was evaluated at 3 months (reassessment) after the initiation of biologics. Periodontal conditions were assessed by two calibrated periodontists for the following measurements; number of teeth present, probing depth (PD), clinical attachment level (CAL), plaque control record (PCR), and bleeding on plobing (BOP). Serum levels of antibodies against P. gingivalis sonicated extracts (SE) and outer membrane protein (OMP) and anti-CCP antibodies were determined by enzyme-linked immunsorbent assay, respectively, prior to the start of medication with biologics (baseline). Results Their mean (SD) age was 55.9 (15) years. Patients were predominantly women (81%) with mean disease duration of 8.2(6) years. Most patients had active disease with DAS28-CRP 4.2 (0.9) at baseline, and showed an improvement of DAS28-CRP 2.77 (0.85) at reassessment. Six patients (38%) achieved remission (DAS28-CRP<2.3) at reassessment. The mean serum levels of anti-CCP antibodies were 474.6 (1001.8), and the mean serum levels of antibodies against P. gingivalis SE and OMP 2.37 (3.41)and 0.605 (0.274), respectively at baseline. Periodontal conditions of the patients were as follows; mean PD was 2.84 (0.37) (mm), mean CAL was 3.2 (0.85) (mm), mean PCR was 0.61 (0.41) (%), and mean BOP was 0.33 (0.16) (%) at baseline. No associations were observed between changes in DAS28-CRP (delta-DAS28-CRP) and serum and periodontal parameter values (serum levels of antibodies against P. gingivalis SE and OMP, PD, CAL, PCR, and BOP). However, a significant correlation was found between delta-DAS28-CRP and serum levels of anti-CCP antibodies (P = 0.02, r = 0.56). Conclusions These results failed to show an association of the serum antibody responses to P. gingivalis antigens and periodontal conditions with clinical response to biologics in patients with RA. However, it is suggested that serum anti-CCP antibody titer constitute a factor predicting clinical response to biologics. Further studies would be required to fully evaluate the association with a larger population. References C, Savioli et al. Ann Rheum Dis 70(Suppl3):566(2011) Kobayashi, T. et al. J Periodontol.77,364-69(2006) Disclosure of Interest None Declared