A. Peltea

Monitoring Drug and Antidrug Levels: A Rational Approach in Rheumatoid Arthritis Patients Treated with Biologic Agents Who Experience Inadequate Response While Being on a Stable Biologic Treatment

Clinical response in patients with rheumatoid arthritis (RA) treated with biologic agents can be influenced by their pharmacokinetics and immunogenicity. The present study evaluated the concordance between serum drug and antidrug levels as well as the clinical response in RA patients treated with biological agents who experience their first disease exacerbation while being on a stable biologic treatment. 154 RA patients treated with rituximab (RTX), infliximab (IFX), adalimumab (ADL), or etanercept (ETN) were included. DAS28, SDAI, and EULAR response were assessed at baseline and reevaluated at precise time intervals. At the time of their first sign of inadequate response, patients were tested for both serum drug level and antidrug antibodies level. At the next reevaluation, patients retreated with RTX that had detectable drug level had a better EULAR response (P = 0.038) with lower DAS28 and SDAI scores (P = 0.01 and P = 0.03). The same tendency was observed in patients treated with IFX and ETN regarding EULAR response (P = 0.002 and P = 0.023), DAS28 score (P = 0.002 and P = 0.003), and SDAI score (P = 0.001 and P = 0.026). Detectable biologic drug levels correlated with a better clinical response in patients experiencing their first RA inadequate response while being on a stable biologic treatment with RTX, IFX, and ETN.

AB1158 Prevalence of comorbidities in psoriatic arthritis: a cross-sectional study

Background Psoriatic arthritis (PsA) is associated with important comorbidities: cardiovascular, gastro-intestinal, infectious, malignant, and psychiatric [1, 2]. However, they are less studied in PsA compared to other chronic inflammatory arthritis. Objectives The objective of this study was to calculate the prevalence of comorbidities and risk factors in a cohort of PsA patients. Methods This was an observational cross-sectional study, including consecutive, unselected adult patients, with a diagnosis of PsA according to their rheumatologist. Data collected: demographical, clinical (affected joints, current psoriasis, axial involvement, enthesitis, dactylitis), biological (acute phase reactants), and treatment related (nonsteroidal anti-inflammatory drugs, synthetic remissive drugs and biologics). Data on comorbidities and risk factors were collected according to the European League Against Rheumatism (EULAR) recommendations on reporting comorbidities in chronic inflammatory rheumatic diseases in daily practice [3]. Results In all, 129 PsA patients were included: 77 (59.7%) women, mean age ± standard deviation 53.5±11.8 years, disease duration 7±7.4 years; 53 (41.1%) had axial involvement, 33 (25.6%) dactylitis, 18 (14%) enthesitis, and 24 (18.6%) current moderate/severe psoriasis. Most of them had low or moderate disease activity and almost a quarter of them (32; 24.8%) were taking a biologic. The most prevalent comorbidities were: dyslipidaemia 103 patients (79.8%), hypertension 67 (51.9%), obesity 44 (34.1%), diabetes 21 (16.3%) and ischemic heart disease 15 (11.6%). Almost a third of patients (42, 32.6%) suffered a cardiovascular event after their PsA diagnosis, of which heart attack 2 patients, stroke 4, cardiac failure 4 and peripheral arterial disease one patient. Cardiovascular events correlated with smoking (r=0.893, p<0.001) and current moderate/severe psoriasis (r=0.218, p=0.013). Regarding infectious comorbidities: 11 patients (8.5%) had a history of tuberculosis after being diagnosed with PsA, 7 (5.4%) chronic viral hepatitis, of which 4 with B virus and 3 with C virus, and 5 patients (3.9%) developed severe infections. Five patients (3.9%) were diagnosed with neoplasia, but no correlation was identified with any of the clinical, biological or treatment related included variables. Only 11 patients (8.5%) were diagnosed with depression, but the prevalence is probably underestimated, since not all patients were screened to this end. Conclusions PsA is associated with a high prevalence of comorbidities, especially cardiovascular diseases. This should be taken into consideration in the therapeutic and the global management of PsA patients. References Husni ME, Mease PJ. Managing comorbid disease in patients with psoriatic arthritis. Curr Rheumatol Rep 2010;12(4):281–7. Ogdie A, Yu Y, Haynes K, et al. Risk of major cardiovascular events in patients with psoriatic arthritis, psoriasis and rheumatoid arthritis: apopulation-based cohort study. Ann Rheum Dis 2015;74(2):326–32. Baillet A, Gossec L, Carmona L, et al. Points to consider for reporting, screening for and preventing selected comorbidities in chronic inflammatory rheumatic diseases in daily practice: a EULAR initiative. Ann Rheum Dis 2016;75(6):965–73. Disclosure of Interest None declared

A10.05 Optimising existing tools for reaching an adequate disease control in patients with spondylarthritis

Background and objectives Disease activity in spondylarthritis (SpA) is widely evaluated through disease activity scores such as BASDAI, ASDAS and PtGA, with no proven superiority between them. The aim of this study was to subdivide patients with anti-TNF therapy according to their disease status indicated by scores in order to evaluate group characteristics with further assessment on the quality of each activity indicator. Materials and methods We included 100 SpA patients under anti-TNF therapy (32 infliximab, 33 adalimumab, 35 etanercept). We collected demographic, clinical (BASDAI, ASDAS, PtGA) and laboratory (ESR, CRP) data. Statistical analysis was performed with SPSS 20.0. Results We evaluated disease activity for the entire study group based on conventional scores: BASDAI, ASDAS-CRP, ASDAS-ESR, acute phase reactants and PtGA. When used as an external criterion PtGA showed that 12% of patients had active disease while 88% were classified as low disease activity (PtGA < 5). Mean ASDAS-CRP/ESR in the active group were 3.39/3.24. Mean BASDAI score in the high activity group was 5.66. We showed that both ASDAS scores had good discriminating capacities, with similar values when using the SMD (ASDAS-CRP and ASDAS-ESR – SMD 2.00). In our study group, based on PtGA, BASDAI outperformed ASDAS scores with a SMD of 3.33. We used ROC curves of the disease activity scores by using the PtGA ≥ 5 as variable of high disease activity state. For ASDAS-CRP, ASDAS-ESR and BASDAI the AUCs (area under curve) were 0.89 (P = 0.05), 0.88 (P < 0.001), and 0.99 (P = 0.009), respectively. For CRP and ESR the AUCs were 0.81 and 0.79 (P = 0.001, P = 0.003). This shows the high accuracy of the three scores in assessing SpA activity. When dividing patients according to BASDAI score (4 as cut-off), 14% showed a more active disease than the rest of 86% who had low disease activity. Mean ASDAS scores in the first group were 3.31 and 3.16, respectively. BASDAI correlated to both ASDAS scores (r = 0.65 and 0.71, P < 0.001) and PtGA stronger to BASDAI(r = 0.912, P < 0.01) than ASDAS(r = 0.67 and 0.71, P < 0.01). Conclusion We proved that disease activity scores have good discriminatory power and that BASDAI and ASDAS perform similarly in assessing and investigating SpA patients. BASDAI outperformed ASDAS when using PtGA as criterion.

A5.1 Vitamin D and subclinical atherosclerosis in systemic sclerosis

Background Whether scleroderma patients have an increased atherosclerotic risk or not is still a matter of dabate. Existing evidence suggests that impaired vitamin D metabolism, common in autoimmune diseases, might contribute to the development of atherosclerosis. Objective To evaluate the relationship between vitamin D status and subclinical atherosclerosis in scleroderma patients. Methods 20 scleroderma patients were evaluated. Radio frequency-based echo-tracking (Aloka alpha-10 Prosound, 15Mhz transducer) was performed to evaluate subclinical atherosclerosis. Several parameters were measured: intima-media-tchikness (IMT), βeta (stiffness parameter), Ep (pressure-strain elasticity modulus), AC (arterial compliance), AI (augmentation index), PWV (pulse wave velocity) in common carotid artery. 25(OH)D level was measured in the sera by an ELISA kit [DIASource® 25-Hydroxyvitamin D Total ELISA assay (Nivelles, Belgium)]. 25(OH)D levels were considered normal for values ≥30 ng/ml, insufficient for values between 20–30 ng/ml and deficient for values under 20 ng/ml. Results The medium level of vitamin D was 20.71ng/ml, 37.5% of the patients being classified as insufficiency and 37.5% as deficiency. Patients with IMT ≥0.9mm had significantly lower levels of 25OH)D than those with normal IMT (13.78ng/ml vs. 25.67ng/ml, p = 0.034). Strong negative correlations were observed between vitamin D status and arterial rigidity parameters: β index (p < 0.001, r = -0.894), elastic modulus (p = 0.001, r = -0.760) and pulse wave velocity (p = 0.001, r = -0.728). Differences between groups were statistically significant when using ANOVA tests for beta index [F(2,13) = 13.069, p = 0.001] and elastic modulus [F(2,13) = 4.45, p = 0.034]. Post hoc Games Howell tests conformed the differences. βeta index was higher in patients with vitamin D insufficiency (7.10 ± 1.52, p = 0.004) or deficiency (11.3 ± 0.56, p = 0.05) compared to those with normal 25(OH) D levels (4.15 ± 0.07) and significantly higher in patients with deficiency compared with those with insufficiency (p < 0.001). Elastic modulus was higher in patients with insufficient (86.16 ± 27.48, p = 0.036) or deficient vitamin D levels (139.25 ± 54.09, p = 0.011) compared to those with normal status (53.25 ± 1.06) and higher in patients with deficiency compared with those with insufficiency (p = 0.04). Arterial compliance is higher in patients with normal vitamin D status than those with insufficient levels, in patients with insufficient levels compared with those with deficient levels but the difference are not significant (p = 0.77, p = 0.54). Coincidence or not, the patient with the lowest 25 (OH)D in the group had significant carotid artery stenosis and developed stroke. Conclusions We identified a strong negative correlation between 25(OH)D levels and arterial rigidity parameters, suggesting that low vitamin D status can be an additional risk factor factor for atherosclerosis in scleroderma patients.

THU0583 Does Eye Gaze Tracking Have the Ability to Assess How Rheumatologists Evaluate Musculoskeletal Ultrasound Images

Background Eye gaze tracking (EGT) is a method used to measure eye position and eye movement, defined as visual attention. Much existing research has used the connection between visual attention and EGT in a broad spectrum of fields such as psychology or cognitive linguistics. The application of EGT in terms of evaluating the knowledge and expertise of individuals has been widely ignored, although a strong frame of study has indicated that this method may be weighed as unbiased indicator of the focus of visual attention. Objectives To evaluate the ability of EGT in assessing quantitatively and qualitatively how rheumatologists analyze musculoskeletal ultrasound images (MSUS). Methods 11 young rheumatologists in training were evaluated in December 2014. Two experts in this field were also asked to participate. 12 ultrasound images of different rheumatologic pathologies were used (displayed on 21inch HD monitor), and 6 other different images were used in order to calibrate the machine. The images were analyzed using The Eye Tribe (Hardware) and Eyeproof (Software). The data was analyzed using SPSS and Microsoft Excel. For every ultrasound image 1 to 5 regions of interest were set by a team of MSUS experts. Items assessed were: Total time of evaluation (TTE), time spent to reach the first region of interest (TSRF), time spent upon the region of interest (TSUR), ratio of time spent in the region of interest (ROI) and time spent outside the region of interest, number of elements identified in the region of interest versus the rest of the image. Results Total time of evaluation - mean (SD) - for experts was 5,4 (1,3) seconds versus 8,9 (2,7) seconds for unexperienced rheumatologists. Time spent to reach the first region of interest for experts was 0,6 (0,8) msec and 3,2 (2,1) msec for non-experts. The unexperienced sonographers focused on a higher number of elements outside the region of interest than experts: 8,6 (3,8) vs. 2 (0,5). There was no significant difference between the two groups of evaluators regarding the time spent upon the region of interest. Conclusions Experienced sonographers spend less time than unexperienced ones when assessing an ultrasound image (mainly because they focus quicker on the ROIs). Once focused on a particular ROI both groups tend to act similarly. EGT can be a useful tool in assessing the way ultrasound images are visually analyzed and can be used to improve MSUS training and optimize the visual analyze pattern of the sonographer. References Robert H. Tai, John F. Loehr, Friederick J. Bringham An exploroation of the use of eye-gaze tracking to study problem-solving on standarized science assessments. International Journal of Research&Method in Education, volume 29, Issue 2, 2006. Carlos H. Morimoto, Marcio R.M. Mimica Eye gaze tracking techniques for interactive applications. Computer Vision and Image understanding Vol 98, issue 1, April 2005 Olmeda, R. A. (2002) Using eye movements to differentiate students with and without ADHD in a simple reading task. Unpublished manuscript, Charlottesville, VA, The University of Virginia Acknowledgements RCRD team. Disclosure of Interest None declared

A1.85 Can rituximab treatment in rheumatoid arthritis patients decrease cardiovascular risk

Background and Objectives Rheumatoid arthritis (RA) is not only a chronic inflammatory joint disease, but also a condition associated with a high risk for cardiovascular diseases, mostly due to the inflammatory burden. Therefore, EULAR recommends using a modified assessment of cardiovascular risk, mSCORE which is obtained by multiplying SCORE by 1.5. The objective of this study is to evaluate the impact of rituximab on cardiovascular risk. Materials and Methods 42 RA patients initiating rituximab in 2010 were included and were prospectively followed for 18 months. They all had active disease and were non-responders to DMARD therapy (synthetic and/or biologic). The cardiovascular risk was evaluated by mSCORE (at baseline and at 18 months). Response to rituximab therapy was assessed using EULAR response at 6 months (ER6) and at 18 months (ER18). The study is still ongoing, in order to assess long-term impact of rituximab on cardiovascular risk in RA patients. Results ER at 6 months was negatively correlated to mSCORE at 18 months (r = -0.353, p = 0.032). Disease duration seems to have a high impact on cardiovascular risk (correlation with mSCORE at baseline: r = 0.993, p < 0.001, with mSCORE at 18 months: r = 0.668, p < 0.001). Drugs such as statins and methotrexate which are known to be protective for cardiovascular diseases do not influence mSCORE at 18 months (r = 0.191, p = 0.239, respectively r = -0,156, p = 0.337). Cardiovascular risk in patients initiating rituximab is still high after a year and a half of treatment, regardless of using cardiovascular protective drugs. The negative correlation between ER6 and mSCORE at 18 months suggests that anti CD20 therapy might be useful for decreasing cardiovascular burden in RA patients. Though, in patients with a longer duration of disease, mSCORE remains high after 18 months of treatment. Conclusions Although an early good clinical and biological response at 6 months to rituximab therapy seems to decrease the cardiovascular risk after 18 months of treatment, the cardiovascular burden is still high (even with protective drugs), probably due to previous inflammatory status.

A1.27 Serum rituximab level and other predictors for good eular response before re-treatment in rheumatoid arthritis

Background and Objectives Predictive factors for rituximab (RTX) response in rheumatoid arthritis (RA) patients are currently still under debate. Repeated re-treatment is now the rule, but it is worth knowing if there are predictive factors for good or moderate EULAR response after a new course of RTX. Materials and Methods Twenty five long term RTX-treated patients with RA were included in this prospective study. Their average time with RTX treatment was 41.79 months. RTX serum level was measured by sandwich ELISA (Promonitor-RTX Ref. PG-PRTX-12700), after 6 months from last re-treatment. Patients were divided into detectable versus non-detectable drug levels based on assay cut-off. Clinical and pharmacological data were collected at the time of dosing serum RTX level and 6 months later. Results At 6-month follow-up, 8 (32%) of the RA patients have achieved a good EULAR response and 7 patients (28%) - a moderate EULAR response. Regarding the serum RTX level, 9 patients (36%) had no detectable drug level. From this group at 6 month after last infusion, 6 (66.6%) patients had no EULAR response. All patients tested negative for anti-RTX antibodies. A significant correlation was found between detectable drug level and EULAR response after re-treatment (p = 0.034, r = 0.424). Patient status of anti-citrullinated protein antibodies (ACPA) was strongly correlated to RTX drug level (p = 0.021, r = 0.460) but not to the EULAR response (p = 0.216, r = 0.256). Another interesting finding was the significant correlation between RTX serum level and number of previous anti tumor necrosis agents used before RTX was initiated (p = 0.009, r = 0.514). Conclusions This observational study suggests that detectable RTX serum level before re-treatment can be predictive for a good EULAR response at 6 month. ACPA positivity and higher number of previous anti TNF agents used are correlated with RTX level. As non-detectable RTX level increases chances for no response, drug level monitoring may be used to optimise treatment in patients with RA.

A1.4 Predictive factors for the eular response in ra patients with rituximab treatment

Background and Objective The prediction of therapeutic response to Rituximab in Rheumatoid Arthritis is still an unmet need. This study aimed to identify useful and accessible predictive factors of non-responders to Rituximab ahead of time. Materials and Methods 42 RA patients initiating Rituximab were included in this study and were prospectively followed for 18 months. The lymphocyte count (LyC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), rheumatoid factor (RF) and anti-CCP antibodies (ACPA) titres, disease duration and number of previous anti-TNFα drugs were assessed at baseline, at 6 and 18 months. The clinical response was evaluated using disease activity score (DAS28) based on ESR and European League Against Rheumatism (EULAR) response at 6 (ER6) and 18 months (ER18). Results LyC at baseline is predictive for ER6 (F=4.597, p=0.041), but not for ER18, and the ER6 is predictive for ER18 (F=8.834, p=0.005). Also DAS28 at 6 months and the number of previous anti-TNFα drugs are predictive for ER18 (F=7.937, p=0.008, respectively F=4.356, p=0.044). Although we obtained a negative correlation between the ESR at 6 months and ER18 (p=0.002), the therapeutic response cannot be predicted based on ESR values. RF and ACPA titres, CRP values and the disease duration did not seem to correlate with the therapeutic response. Conclusions: The lymphocyte count at baseline represents a predictive factor for the EULAR response at 6 months. A better 6 months ER along with a lower DAS28 predict a better ER at 18 months of Rituximab treatment in RA patients.

FRI0026 Correlation between Serum Rituximab Level and Clinical Response in Rheumatoid Arthritis Patients Treated with B Cell Depletion Therapy

Background Rituximab (RTX) sustained efficacy in rheumatoid arthritis (RA) patients can be achieved by repeated courses of RTX (1). The correlation between serum drug level before a new retreatment and clinical response remains to be established. Objectives To investigate the relationship between serum RTX level just before a new retreatment course and clinical response after 6 months. Methods Twenty five consecutive RA patients treated with RTX for more than 12 months were included in this prospective study initiated in May 2013 and followed up until December 2013. Serum samples were collected for assay of drug level and anti-drug antibodies (ADAb) using sandwich ELISA (Promonitor-RTX Ref.PG-PRTX-12700). Patients were divided according to RTX serum level into detectable versus non-detectable drug level based on assay cut-off. Their disease activity was assessed by Disease Activity Score (DAS28) and Simplified Disease Activity Index (SDAI) scores at baseline and after 6 months of follow-up. At follow-up they were classified using DAS28 score in patients with high, moderate, low disease activity and remission. The unpaired t-test or, where appropriate, Wilcoxon Mann-Whitney test, was used for comparisons. Results At baseline, 8 (36%) patients had non-detectable RTX level with mean DAS28 and SDAI score of 3.45 and 21.79 respectively. Six (66.6%) of them had moderate or high disease activity compared to 75% of patients in the RTX detectable group. At the time of follow-up (6 months after dosing RTX), 5 patients with no-detectable RTX level achieved low disease activity or remission, compared to 11 patients with detectable serum RTX. Patients with detectable RTX level showed larger reduction of DAS28 and SDAI score at follow-up (P=0.0104, respectively P=0.0332), versus patients with undetectable RTX level. Lower disease activity class was achieved in patients with detectable drug serum level (P=0.0036). All patients tested negative for ADAb. Conclusions This pilot prospective study suggests that detectable RTX serum level just before re-treatment is correlated to further clinical response in RA patients. The results of our study are promising for the optimization of treatment responses and easy applicable in clinical practice. References Mok CC. Rituximab for the treatment of rheumatoid arthritis: an update. Drug Des. Devel. Ther. 2014;8(87-100). Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.4831