Faculty Bibliography

Background: Methotrexate (MTX) is the most commonly used conventional synthetic DMARD (csDMARD) for the treatment of rheumatoid arthritis (RA). In the phase 3 MOBILITY study (NCT01061736),1 of sarilumab in inadequate responders to MTX, MTX was continued at the dose prescribed at screening, which was assumed to be the maximally tolerated dose. This sub-analysis evaluates whether the different doses of MTX, in association with sarilumab, had an effect on efficacy or safety outcomes in MOBILITY. Objectives: To explore the relationship of MTX dose on the efficacy or safety of sarilumab in the treatment of active RA in MTX-IR patients. Methods: The MOBILITY study design and methods have been reported.2 Patients were required to be on a stable dose of MTX of 10-25 mg/wk for at least 6 weeks, except for Asian-Pacific region patients (6-25 mg/wk) prior to enrolment and to maintain this dose throughout the trial. A sufficient number of subjects with doses of 7.5-25 mg/wk were included: 11 (0.92%) patients were excluded from the analysis due to small numbers receiving doses of <7.5 and >25mg/wk. The relationship between efficacy and safety outcomes for sarilumab and MTX was assessed by appropriate regression models (logistic or generalized linear) in this post hoc analysis. Results: 1186 patients were included. For each dose group in each treatment arm we computed proportions of ACR20 responders at Wk 24 and 52; no apparent difference in achieving an ACR20 response was observed with increasing MTX dose in any treatment group (Figure). A logistic regression assessment confirms this observation: for sarilumab 150 mg and 200 mg. Based on mean change from baseline in HAQ-DI at Wk 24, no apparent relationship with MTX dose was observed. A linear regression model assessment provides confirmation of the results. There was no statistical difference in mean HAQ-DI change from baseline based on MTX dose in the Pbo, sarilumab 150 mg or sarilumab 200 mg groups. Similar findings were observed for changes from baseline for DAS28-CRP, FACIT-Fatigue, and mTSS (change and progression). There was no association of the incidence of treatment-emergent adverse events (AE), serious AEs and changes in ALT, neutrophil counts and lipids with increasing MTX doses. Conclusions: In this post hoc analysis, within the sarilumab 150 mg or sarilumab 200 mg dose groups, efficacy or safety was similar across concomitant MTX Doses. (Figure Presented)

Background: Due to a small number of clinical studies, treatment of Behcet's syndrome (BS) mainly depends on the type and severity of symptoms and may vary substantially. In 2008, EULAR recommendations were published, aiming for an evidence-based approach for the management of BS [1]. Objectives: To assess guideline adherence in treatment of BS in two different geographic areas. Methods: We extracted guideline statements from the 2008 EULAR recommendations [1]. Adherence to these statements in both New York (USA) and Amsterdam (The Netherlands) was evaluated retrospectively by reviewing records from patients fulfilling the ISG criteria. We analyzed data per statement and event, and divided the data according to the year in which an event occurred. We compared events prior to 2009 to those in 2009 or later (after publication of the EULAR recommendations). Results: 474 patients were evaluated, 24 of whom were from Amsterdam (Table 1). Adherence in posterior uveitis was relatively low in clinical practice. However, secondary analysis of the patients in the second time frame showed that 66% of cases with partial- and 77% with non-adherence were on cyclosporine or a biologic DMARD. Colchicine in treatment of arthritis was hardly ever used as monotherapy (3 cases <2009, 9 cases >2009). Other drugs used included prednisone (n=197), Plaquenil (n=63), methotrexate (n=67), azathioprine (n=133) and anti-TNF agents (n=124). Conclusions: Adherence to the guidelines varies substantially across type of events. Adherence in treatment of posterior uveitis was low and a variety of other drugs were used in its treatment, which are also considered DMARDs for this condition. The extensive use of anti-TNF agents might indicate a shift towards more aggressive treatment and acceptance of TNF inhibitors for the treatment of other rheumatic and inflammatory eye conditions in the countries studied. In patients with neuro-Behcet, adherence to the recommendation is quite good in clinical practice. In the majority of patients with arthritis, colchicine is either used in combination with other (biological) DMARDs or not at all. This might indicate that colchicine alone is not sufficient as treatment of arthritis in BS or that arthritis is often combined with other manifestations implying a need for more combination treatment. Our results suggest that a revision of the current guidelines may be due, given widespread use of other immunosuppressive medications and newly available studies of these medications. (Table Presented)

Background: Diagnosis of Behcet's syndrome (BS) is based on clinical signs and symptoms, without the use laboratory or imaging tests and the most commonly used diagnostic criteria are the International Study Group (ISG) criteria. Patients have also been diagnosed with BS not meeting ISG criteria based on the clinical impression of the treating physicians. Objectives: To determine differences in disease activity among BS patients meeting and not meeting the ISG criteria. Methods: Behcet's patients seen at the NYU Behcet's Center had their demographic, clinical features and outcomes data abstracted. Confirmed BS diagnosis was determined if ISG criteria were met at any time during the course of observation. Multiple linear regression estimated any association between meeting ISG criteria and gender with the patient-reported outcomes Behcet's Syndrome Activity Scale (BSAS), Pain Visual Analog Scale (VAS), Fatigue VAS, Patient Global Assessment VAS, and RAPID3 as well as Physician Global Assessment VAS. Results: First observation data on 832 subjects were abstracted for this analysis. 504 (63%) met ISG criteria for Behcet's, 616 (74%) were female with an average age of 35 years (+/-13.8). Mean scores for BSAS, pain, fatigue, patient global, RAPID3 and physician global were uniformly higher in those meeting criteria vs not meeting criteria. Meeting ISG criteria was significantly associated with increases in all clinical outcomes independently of gender, ranging from 5 points on the Physician Global VAS to 13 points on the Fatigue VAS (Table). Conversely, gender was only associated with significant increases in BSAS, Fatigue and RAPID3 independent of meeting ISG criteria. Conclusions: Behcet's patients who fulfill ISG criteria have more active disease, based on all measures of disease activity, including patient and physician based outcomes. Interestingly, females were also at risk for more active disease as measured by composite patient reported outcome measures, reflecting possibly the different character of Behcet's in non-endemic areas, such as the US. (Table Presented)

Background: Behcet's Syndrome Activity Scale (BSAS), a patient reported outcome measure for Behcet's syndrome, has been validated for routine clincial care and have been shown to differentiate active treatment from placebo in clinical trials. Currently, there are no identified thresholds for disease activity levels for BSAS. Objectives: To determine resmission, low, moderate and high disease activity level threshold scores for BSAS. Methods: Behcet's patients seen at the NYU Behcet's Center had their demographic, clinical features and outcomes data abstracted. Confirmed Behcet's diagnosis was determined if ISG criteria were met at any time during the course of observation. Concordance correlation was estimated between the BSAS and the RAPID3; both outcomes were scaled to [0-100]. Proposed BSAS severity categories are as: Near-Remission = [0-10), Low = [10-30), Moderate=[30-60), and High = [60-100]. Weighted Kappa statistics were estimated between BSAS and RAPID3 severity categories. RAPID3 categories were based upon published1 as well as matching the proposed BSAS severity categories. Results: First observation data on 832 subjects were abstracted for this analysis. 504 (63%) met ISG criteria for Behcet's, 616 (74%) were female with an average (Table Presented ) age of 35 years (+/-13.8). Concordance between BSAS and RAPID3 was moderate (CCC =0.518). BSAS severity categories classified 7% Near-Remission, 24% Low, 48% Moderate, and 21% of the study population as High disease severity (Table). Published RAPID3 categories classified 44% of subjects as High, while matching RAPID3 categories only classified 20%. Agreement between categories was moderate for both RAPID3 classifications. Conclusions: BSAS, a patient reported outcome measure for Behcet's syndrome, correlates well with other composite indices of disease activity. Proposed cut off points for near-remission, low, moderate and high activity for BSAS may be used in clinical care for a "treat-to-target" approach to Behcet's treatment

Background: Routine Assessment of Patient Index Data 3 (RAPID3) comprises the three patient-reported ACR RA Core Data Set measures: function (HAQDI), pain and patient global estimate of status (10-cm visual analogue scale). These can be scored in <10 seconds, making it suitable for use in routine clinical practice where it provides quantitative data to supplement qualitative assessments. Significant correlations between RAPID3 scores and DAS28 (CRP) and CDAI have been previously reported.1,2 Objectives: To examine whether RAPID3-defined remission performs similarly to definitions of remission by DAS28 (CRP), CDAI, SDAI or Boolean criteria in patients with early RA. Methods: We performed post hoc analyses on data from the AVERT study, described previously.3 RAPID3 scores were calculated at baseline and 3-monthly intervals thereafter, up to 12 months. Definitions of remission were RAPID3 <1 (scale of 1-10); DAS28 (CRP) <2.6; CDAI <2.8; SDAI <3.3; and Boolean: TJC28 <1, SJC28 <1, patient global assessment of disease activity (0-10 cm) <1 and high-sensitivity CRP <1 mg/dL. Proportions of patients in RAPID3, DAS28 (CRP), CDAI, SDAI and ACR/EULAR Boolean remission at each time point were calculated. Proportions at 3, 6, 9 and 12 months were compared using cross-tabulation analysis. Agreement between RAPID3 disease activity states and those of other measures were assessed using kappa and weighted kappa statistics. Results: Among the total AVERT population, the respective percentages of patients in remission according to each measure at Months 3, 6 and 9 were: RAPID3, 12.3%, 17.4%, 21.9%; Boolean, 8.5%, 12.5%, 21.1%; SDAI, 10.8%, 21.4%, 27.4%; CDAI, 10.0%, 20.8%, 27.1%; and DAS28 (CRP), 25.1%, 35.0%, 42.7%. Similar trends held at Month 12: 29.3%, 28.8%, 32.2%, 33.6%, 50.1%, respectively. When weighted kappa correlations were calculated for RAPID3 versus other remission criteria, RAPID3 remission showed good correlation with Boolean remission, and kappa correlations with SDAI and CDAI remission were (Table presented) higher than those with DAS28 (CRP) (Table). These trends were similar for each treatment arm. Conclusions: RAPID3-defined remission may be as stringent as ACR/EULAR Boolean-defined remission and agrees well with SDAI and CDAI remission criteria. RAPID3 may be used in clinical trials in addition to routine clinical care in early RA, and its ease of use may be an advantage over other indices

BACKGROUND: Oral ulcers, the hallmark of Behcet's syndrome, can be resistant to conventional treatment; therefore, alternative agents are needed. Apremilast is an oral phosphodiesterase-4 inhibitor that modulates several inflammatory pathways. METHODS: We conducted a phase 2, multicenter, placebo-controlled study in which 111 patients with Behcet's syndrome who had two or more oral ulcers were randomly assigned to receive 30 mg of apremilast twice daily or placebo for 12 weeks. This regimen was followed by a 12-week extension phase in which the placebo group was switched to apremilast and a 28-day post-treatment observational follow-up phase. The patients and clinicians were unaware of the study assignments throughout the trial. The primary end point was the number of oral ulcers at week 12. Secondary outcomes included pain from these ulcers (measured on a 100-mm visual-analogue scale, with higher scores indicating worse pain), the number of genital ulcers, overall disease activity, and quality of life. RESULTS: The mean (+/-SD) number of oral ulcers per patient at week 12 was significantly lower in the apremilast group than in the placebo group (0.5+/-1.0 vs. 2.1+/-2.6) (P<0.001). The mean decline in pain from oral ulcers from baseline to week 12 was greater with apremilast than with placebo (-44.7+/-24.3 mm vs. -16.0+/-32.5 mm) (P<0.001). Nausea, vomiting, and diarrhea were more common in the apremilast group (with 22, 9, and 12 incidents, respectively, among 55 patients) than in the placebo group (with 10, 1, and 2 incidents, respectively, among 56 patients), findings that were similar to those in previous studies of apremilast. There were two serious adverse events in patients receiving apremilast. CONCLUSIONS: Apremilast was effective in treating oral ulcers, which are the cardinal manifestation of Behcet's syndrome. This preliminary study was neither large enough nor long enough to assess long-term efficacy, the effect on other manifestations of Behcet's syndrome, or the risk of uncommon serious adverse events. (Funded by Celgene; ClinicalTrials.gov number, NCT00866359.).