Lower body mass and lower adiposity are associated with differential responses to two treatment strategies for rheumatoid arthritis.

PURPOSE: To determine if body mass index (BMI) and adipokine levels identify rheumatoid arthritis (RA) patients most likely to benefit from initiation of tumour necrosis factor inhibitors (TNFi) after methotrexate inadequate response. METHODS: This is a secondary analysis of the Rheumatoid Arthritis Comparison of Active Treatments (RACAT) trial and the (TEAR) trial. Both studies compared treatment strategies starting with conventional disease-modifying anti-rheumatic drugs (DMARDs) (triple therapy) versus etanercept plus methotrexate. We compared response rates between TNFi and triple therapy among patients with different BMI. Adipokines were measured at enrolment and associations with treatment response were examined using regression, adjusting for age, sex, BMI and baseline disease activity. RESULTS: In RACAT (n=306), participants who were normal/underweight were more likely to benefit from TNFi versus triple therapy, with greater change in Disease Activity Score in 28 and greater ACR20 response (ACR 20: 64% vs 23%, p=0.001). In contrast, overweight/obese participants had similar response to TNFi versus triple therapy (p-for-interaction=0.001). Similarly, but modest patterns were observed in TEAR (n=601; ACR20: 67% vs 52%, p=0.05). In RACAT, adipokine scores consistent with lower adiposity also predicted greater response to TNFi (ACR20: 58% vs 37%, p=0.01) with better model fit compared with BMI alone. CONCLUSIONS: Lower BMI and evidence of lower adiposity based on adipokine profiles were associated with a superior response to TNFi compared with triple therapy. There was no difference between treatments among overweight/obese participants. The results support TNFi being a particularly important therapeutic among normal/underweight patients, with implications for clinical decisions and trial design.

Trimethoprim sulfamethoxazole prophylaxis and serious infections in granulomatosis with polyangiitis treated with rituximab.

OBJECTIVE: To assess the association of trimethoprim sulfamethoxazole (TMP-SMX) prophylaxis with serious infections in rituximab-treated patients with granulomatosis with polyangiitis (GPA). METHODS: This retrospective cohort study included adults with GPA (2011-2020) within the United States Merative™ Marketscan® Research Databases with ≥6 months enrolment prior to first (index) rituximab treatment. We defined TMP-SMX prophylaxis as a ≥28-day prescription dispensed after or overlapping the index date. Serious infection was a hospital primary diagnosis for infection (excluding viral or mycobacterial codes). Secondary outcomes were outpatient infection, PJP, and adverse events potentially attributable to TMP-SMX. Cox proportional hazards regression assessed the association of time-varying TMP-SMX with outcomes of interest, adjusting for potential confounders. Individuals were followed until the outcome of interest, end of database enrolment, or Dec 31, 2020. RESULTS: Among 919 rituximab-treated individuals (53% female), mean age was 52.1 years (SD 16) and 281 (31%) were dispensed TMP-SMX within 30 days of index date. Over a median of 496 (IQR 138, 979) days, 130 serious infections occurred among 104 individuals (incidence 6.1 [95% CI 5.0-7.4] per 100 person-years). Time-varying TMP-SMX was negatively associated with serious infection (adjusted HR 0.5; 95% CI 0.3-0.9). The aHR for outpatient infections was 0.8 (95% CI 0.6-1.1). The estimate for PJP was imprecise (13 events, unadjusted HR 0.2; 95% CI 0.03-1.8). TMP-SMX was potentially associated with adverse events (aHR 1.3; 95% CI 0.9-1.9). CONCLUSIONS: TMP-SMX prophylaxis was associated with reduced serious infections in rituximab-treated GPA, but may increase adverse events, warranting further study of optimal prophylaxis strategies.

Incidence of checkpoint inhibitor-associated inflammatory arthritis, immunomodulation and mortality in cancer patients on immunotherapy: a retrospective cohort study.

OBJECTIVES: Immune checkpoint inhibitor (ICI) associated inflammatory arthritis (ICI-IA) occurs in 4-6% of ICI-treated patients based on one observational study. We identified cases of ICI-IA using administrative claims to study its incidence and characteristics at the population level. METHODS: We used the Medicare 5% sample to identify patients initiating ICIs. Cancer patients were identified by having ≥ 2 ICD-9/10-CM diagnosis codes from an oncologist for lung cancer, melanoma, or renal/urothelial cancer. ICI-IA was defined as having two Medicare claims ≥ 30 days apart with combinations of ICD-9/10-CM diagnosis codes that favored specificity. ICI-IA was identified in patients with a musculoskeletal diagnosis after ICI initiation, who had i.) no inflammatory arthritis or inflammatory rheumatic disease before ICI initiation ever, and ii) no musculoskeletal complaint in the one year prior to ICI. We examined DMARD utilization and visits to rheumatology in patients with ICI-IA. Landmark analysis and a time varying Cox proportional hazards model for overall survival was constructed. RESULTS: The incidence of ICI-IA was 7.2 (6.1-8.4) per 100 patient years. Patients with ICI-IA were mean (SD) age 73.5(7.0) years, 48% women, 91% white. Median(IQR) time from ICI initiation to first ICI-IA diagnosis was 124(56, 252) days. Only 24(16%) received care from a rheumatologist, and 24(16%) were prescribed a DMARD (46% by a rheumatologist). The HR for mortality in patients with ICI-IA was 0.86 (95% CI 0.59-1.26, p= 0.45). CONCLUSIONS: The incidence of ICI-IA identified in claims data is similar to that reported in observational studies, however, few patients are treated with a DMARD or see a rheumatologist. There was no difference in overall survival between ICI-treated patients with and without ICI-IA.

Anti-citrullinated protein antibody profiles predict changes in disease activity in patients with rheumatoid arthritis initiating biologics.

OBJECTIVES: To determine whether an expanded antigen-specific ACPA profile predicts changes in disease activity in patients with RA initiating biologics. METHODS: The study included participants from a prospective, non-randomized, observational RA cohort. For this sub-study, treatment groups of interest included biologic-naïve initiating anti-TNF, biologic-exposed initiating non-TNF, and biologic-naïve initiating abatacept. ACPAs to 25 citrullinated peptides were measured using banked enrolment serum. Principal component analysis (PCA) was performed and associations of resulting principal component (PC) scores (in quartiles) and anti-CCP3 antibody (≤15, 16-250 or >250 U/ml) with EULAR (good/moderate/none) treatment response at 6 months were examined using adjusted ordinal regression models. RESULTS: Participants (n = 1092) had a mean age of 57 (13) years and 79% were women. At 6 months, 68.5% achieved a moderate/good EULAR response. There were three PCs that cumulatively explained 70% of variation in ACPA values. In models including the three components and anti-CCP3 antibody category, only PC1 and PC2 were associated with treatment response. The highest quartile for PC1 (odds ratio [OR] 1.76; 95% CI: 1.22, 2.53) and for PC2 (OR 1.74; 95% CI: 1.23, 2.46) were associated with treatment response after multivariable adjustment. There was no evidence of interaction between PCs and treatment group in EULAR responses (P-value for interaction >0.1). CONCLUSION: An expanded ACPA profile appears to be more strongly associated with biologic treatment response in RA than commercially available anti-CCP3 antibody levels. However, further enhancements to PCA will be needed to effectively prioritize between different biologics available for the treatment of RA.

Impact of high rheumatoid factor levels on treatment outcomes with certolizumab pegol and adalimumab in patients with rheumatoid arthritis.

OBJECTIVES: To assess the impact of baseline RF level on drug concentrations and efficacy of certolizumab pegol [CZP; TNF inhibitor (TNFi) without a crystallizable fragment (Fc)] and adalimumab (ADA; Fc-containing TNFi) in patients with RA. METHODS: The phase 4 EXXELERATE study (NCT01500278) was a 104-week, randomized, single-blind (double-blind until week 12; investigator-blind thereafter), head-to-head study of CZP vs ADA in patients with RA. In this post hoc analysis, we report drug concentration and efficacy outcomes stratified by baseline RF quartile (≤Q3 or >Q3). RESULTS: Baseline data by RF quartiles were available for 453 CZP-randomized and 454 ADA-randomized patients (≤Q3: ≤204 IU/ml; >Q3: >204 IU/ml). From week 12, the area under the curve (AUC) of ADA concentration was lower in patients with RF >204 IU/ml vs patients with RF ≤204 IU/ml; the AUC of CZP concentration was similar in patients with RF ≤204 IU/ml and >204 IU/ml. For patients with RF ≤204 IU/ml, disease activity score (DAS28)-CRP was similar between CZP- and ADA-treated patients through week 104. For patients with RF >204 IU/ml, mean DAS28-CRP was lower in CZP- vs ADA-treated patients at week 104. The proportion of patients with RF >204 IU/ml achieving DAS28-CRP low disease activity at week 104 was greater in CZP- vs ADA-treated patients. CONCLUSION: CZP was associated with maintained drug concentration and efficacy in patients with RA and high RF and may therefore be a more suitable therapeutic option than TNFis with an Fc fragment in these patients. TRIAL REGISTRATION: Clinicaltrials.gov, http://clinicaltrials.gov, NCT01500278.

Determining patient preferences for the medical management of osteoporosis using conjoint analysis.

We used conjoint analysis-a method that assesses complex decision making-to quantify patients' choices when selecting an osteoporosis therapy. While 60% of people prioritized medication efficacy when deciding among treatments, the remaining 40% highly valued factors other than efficacy, suggesting the need for personalized shared decision-making tools. INTRODUCTION: In this study, we aimed to examine patient decision-making surrounding osteoporosis medications using conjoint analysis. METHODS: We enrolled osteoporosis patients at an academic medical center to complete an online conjoint exercise which calculated each patient's relative importance score of 6 osteoporosis medication attributes (higher = greater relative importance in decision-making). We used latent class analysis to identify distinct segments of patients with similar choice patterns and then used logistic regression to determine if demographics and osteoporosis disease features were associated with latent class assignment. RESULTS: Overall, 304 participants completed the survey. The rank order of medication attributes by importance score was the following: efficacy at preventing hip fractures (accounted for 31.0% of decision making), mode of administration (17.5%); risk of serious side effects (16.6%); dose frequency (13.9%); efficacy at preventing spine fractures (12.5%); risk of non-serious side effects (8.4%). We found that 60.9% of the cohort prioritized medication efficacy as their top factor when selecting among the therapies. Being a college graduate, having stronger beliefs on the necessity of using medications for osteoporosis, and never having used osteoporosis medicines were the only factors associated with prioritizing medication efficacy for fracture prevention over the other factors in the decision-making process. CONCLUSIONS: While about 60% of patients prioritized efficacy when selecting an osteoporosis therapy, the remaining 40% valued other factors more highly. Furthermore, individual patient characteristics and clinical factors did not reliably predict patient decision making, suggesting that development and implementation of shared decision-making tools is warranted.

Which Educational Topics and Smartphone App Functions Are Prioritized by US Patients With Rheumatic and Musculoskeletal Diseases? A Mixed-Methods Study.

OBJECTIVE: We sought to identify (1) what types of information US adults with rheumatic and musculoskeletal diseases (RMD) perceive as most important to know about their disease, and (2) what functions they would use in an RMD-specific smartphone app. METHODS: Nominal groups with patients with RMD were conducted using online tools to generate a list of needed educational topics. Based on nominal group results, a survey with final educational items was administered online, along with questions about desired functions of a smartphone app for RMD and wearable use, to patients within a large community rheumatology practice-based research network and the PatientSpot registry. Chi-square tests and multivariate regression models were used to determine differences in priorities between groups of respondents with rheumatic inflammatory conditions (RICs) and osteoarthritis (OA), and possible associations. RESULTS: At least 80% of respondents considered finding a rheumatologist, understanding tests and medications, and quickly recognizing and communicating symptoms to doctors as extremely important educational topics. The highest-ranked topic for both RIC and OA groups was "knowing when the medication is not working." The app functions that most respondents considered useful were viewing laboratory results, recording symptoms to share with their rheumatology provider, and recording symptoms (eg, pain, fatigue) or disease flares for health tracking over time. Approximately one-third of respondents owned and regularly used a wearable activity tracker. CONCLUSION: People with RMD prioritized information about laboratory test results, medications, and disease and symptom monitoring, which can be used to create educational and digital tools that support patients during their disease journey.

Agreement of Medication Information Derived From EHR Data Compared to Medicare Insurance Claims: An Analysis of Biologic Disease-Modifying Antirheumatic Drugs.

PURPOSE: Few studies have reported the agreement between medication information derived from ambulatory EHR data compared to administrative claims for high-cost specialty drugs. We used data from a national EHR-enabled registry, the Rheumatology Informatics System for Effectiveness (RISE), with linked Medicare claims in a population of patients with rheumatoid arthritis (RA) to investigate variations in agreement for different biologic disease-modifying agents (bDMARDs) between two data sources (RISE EHR data vs. Medicare claims), categorized by drug, route of administration, and patient insurance factors (dual eligibility). METHODS: Patients ≥ 65 years old, with ≥ 2 visits in RISE with RA ICD codes ≥ 30 days apart, and continuous enrollment in Medicare Parts B and D in 2017-2018 were included. We classified patients as bDMARD users or nonusers in Medicare claims or EHR data in 2018, and we calculated sensitivity, specificity, positive predicted value (PPV), and negative predicted value (NPV) of EHR data for identifying bDMARD users, using Medicare as the reference standard. We also calculated these metrics after stratifying by clinic-administered (Part B) versus. pharmacy-dispensed (Part D) bDMARDs and by patient dual-eligibility. RESULTS: A total of 26 097 patients were included in the study. Using Medicare claims as the reference standard, EHR data had a sensitivity of 75.0%-90.8% for identifying patients with the same medication and route. PPV for Part B bDMARDs was higher compared with Part D bDMARDs (range 94.3%-97.3% vs. 51.0%-69.6%). We observed higher PPVs for Part D bDMARDs among patients who were dual-eligible (range 82.4%-95.1%). CONCLUSION: The risk of misclassification of drug exposure based on EHR data sources alone is small for Medicare Part B bDMARDs but could be as high as 50% for Part D bDMARDs, in particular for patients who are not dually eligible for Medicare and Medicaid.

Extracting forced vital capacity from the electronic health record through natural language processing in rheumatoid arthritis-associated interstitial lung disease.

PURPOSE: To develop a natural language processing (NLP) tool to extract forced vital capacity (FVC) values from electronic health record (EHR) notes in patients with rheumatoid arthritis-interstitial lung disease (RA-ILD). METHODS: We selected RA-ILD patients (n = 7485) in the Veterans Health Administration (VA) between 2000 and 2020 using validated ICD-9/10 codes. We identified numeric values in proximity to FVC string patterns from clinical notes in the EHR. Subsequently, we performed processing steps to account for variability in note structure, related pulmonary function test (PFT) output, and values copied across notes, then assigned dates from linked administrative procedure records. NLP-derived FVC values were compared to values recorded directly from PFT equipment available on a subset of patients. RESULTS: We identified 5911 FVC values (n = 1844 patients) from PFT equipment and 15 383 values (n = 4982 patients) by NLP. Among 2610 date-matched FVC values from NLP and PFT equipment, 95.8% of values were within 5% predicted. The mean (SD) difference was 0.09% (5.9), and values strongly correlated (r = 0.94, p < 0.001), with a precision of 0.87 (95% CI 0.86, 0.88). NLP captured more patients with longitudinal FVC values (n = 3069 vs. n = 1164). Mean (SD) change in FVC %-predicted per year was similar between sources (-1.5 [30.0] NLP vs. -0.9 [16.6] PFT equipment; standardized response mean = 0.05 for both). CONCLUSIONS: NLP of EHR notes increases the capture of accurate, longitudinal FVC values by three-fold over PFT equipment. Use of this NLP tool can facilitate pharmacoepidemiologic research in RA-ILD and other lung diseases by capturing this critical measure of disease severity.

Safety profile of upadacitinib in patients at risk of cardiovascular disease: integrated post hoc analysis of the SELECT phase III rheumatoid arthritis clinical programme.

OBJECTIVE: Increased risk of serious adverse events (AEs) was reported for tofacitinib relative to tumour necrosis factor inhibitor therapy in patients with rheumatoid arthritis (RA) aged ≥50 years enriched for cardiovascular (CV) risk (ORAL Surveillance). We assessed post hoc the potential risk of upadacitinib in a similar RA population. METHODS: Pooled safety data from six phase III trials were evaluated post hoc for AEs in patients receiving upadacitinib 15 mg once a day (with or without conventional synthetic disease-modifying antirheumatic drugs), adalimumab 40 mg every other week with concomitant methotrexate (MTX), or MTX monotherapy in the overall trial population and in a subset of patients with higher CV risk (aged ≥50 years, ≥1 CV risk factor). Higher-risk patients from a head-to-head study of upadacitinib 15 mg versus adalimumab (SELECT-COMPARE) were assessed in parallel. Exposure-adjusted incidence rates for treatment-emergent AEs were summarised based on exposure to upadacitinib or comparators. RESULTS: A total of 3209 patients received upadacitinib 15 mg, 579 received adalimumab and 314 received MTX monotherapy; ~54% of the patients were included in the overall and SELECT-COMPARE higher-risk populations. Major adverse cardiovascular events (MACE), malignancy (excluding non-melanoma skin cancer (NMSC)) and venous thromboembolism (VTE) were more frequent in the higher-risk cohorts versus the overall population but were generally similar across treatment groups. Rates of serious infections in higher-risk populations and herpes zoster (HZ) and NMSC in all populations were higher with upadacitinib 15 mg than comparators. CONCLUSIONS: An increased risk of MACE, malignancy (excluding NMSC) and VTE was observed in higher-risk populations with RA, yet risk was comparable between upadacitinib-treated and adalimumab-treated patients. Higher rates of NMSC and HZ were observed with upadacitinib versus comparators across all populations, and increased rates of serious infections were detected in upadacitinib-treated patients at higher CV risk. TRIAL REGISTRATION NUMBERS: NCT02706873, NCT02675426, NCT02629159, NCT02706951, NCT02706847 and NCT03086343.

Malignancy risk with tofacitinib versus TNF inhibitors in rheumatoid arthritis: results from the open-label, randomised controlled ORAL Surveillance trial.

OBJECTIVES: To evaluate malignancies and their associations with baseline risk factors and cardiovascular risk scores with tofacitinib versus tumour necrosis factor inhibitors (TNFi) in patients with rheumatoid arthritis (RA). METHODS: In an open-label, randomised controlled trial (ORAL Surveillance; NCT02092467), 4362 patients with RA aged ≥50 years with ≥1 additional cardiovascular risk factor received tofacitinib 5 (N=1455) or 10 mg two times per day (N=1456) or TNFi (N=1451). Incidence rates (IRs; patients with first events/100 patient-years) and HRs were calculated for adjudicated malignancies excluding non-melanoma skin cancer (NMSC), NMSC and subtypes. Post hoc analyses for malignancies excluding NMSC, lung cancer and NMSC included risk factors identified via simple/multivariable Cox models and IRs/HRs categorised by baseline risk factors, history of atherosclerotic cardiovascular disease (HxASCVD) and cardiovascular risk scores. RESULTS: IRs for malignancies excluding NMSC and NMSC were higher with tofacitinib (combined and individual doses) versus TNFi. Risk of lung cancer (most common subtype with tofacitinib) was higher with tofacitinib 10 mg two times per day versus TNFi. In the overall study population, the risk of malignancies excluding NMSC was similar between both tofacitinib doses and TNFi until month 18 and diverged from month 18 onwards (HR (95% CIs) for combined tofacitinib doses: 0.93 (0.53 to 1.62) from baseline to month 18 vs 1.93 (1.22 to 3.06) from month 18 onwards, interaction p=0.0469). Cox analyses identified baseline risk factors across treatment groups for malignancies excluding NMSC, lung cancer and NMSC; interaction analyses generally did not show statistical evidence of interaction between treatment groups and risk factors. HxASCVD or increasing cardiovascular risk scores were associated with higher malignancy IRs across treatments. CONCLUSIONS: Risk of malignancies was increased with tofacitinib versus TNFi, and incidence was highest in patients with HxASCVD or increasing cardiovascular risk. This may be due to shared risk factors for cardiovascular risk and cancer. TRIAL REGISTRATION NUMBERS: NCT02092467, NCT01262118, NCT01484561, NCT00147498, NCT00413660, NCT00550446, NCT00603512, NCT00687193, NCT01164579, NCT00976599, NCT01059864, NCT01359150, NCT02147587, NCT00960440, NCT00847613, NCT00814307, NCT00856544, NCT00853385, NCT01039688, NCT02281552, NCT02187055, NCT02831855, NCT00413699, NCT00661661.

Identification of inadequate responders to advanced therapy among commercially-insured adult patients with Crohn's disease and ulcerative colitis in the United States.

BACKGROUND: The purpose of this analysis was to assess the frequency of inadequate response over 1 year from advanced therapy initiation among patients with Crohn's disease (CD) or ulcerative colitis (UC) in the United States using a claims-based algorithm. Factors associated with inadequate response were also analyzed. METHODS: This study utilized claims data of adult patients from the HealthCore Integrated Research Database (HIRD®) from January 01, 2016 to August 31, 2019. Advanced therapies used in this study were tumor necrosis factor inhibitors (TNFi) and non-TNFi biologics. Inadequate response to an advanced therapy was identified using a claims-based algorithm. The inadequate response criteria included adherence, switching to/added a new treatment, addition of a new conventional synthetic immunomodulator or conventional disease-modifying drugs, increase in dose/frequency of advanced therapy initiation, and use of a new pain medication, or surgery. Factors influencing inadequate responders were assessed using multivariable logistic regression. RESULTS: A total of 2437 patients with CD and 1692 patients with UC were included in this analysis. In patients with CD (mean age: 41 years; female: 53%), 81% had initiated TNFi, and 62% had inadequate response. In patients with UC (mean age: 42 years; female: 48%), 78% had initiated a TNFi, and 63% had an inadequate response. In both patients with CD and UC, inadequate response was associated with low adherence (CD: 41%; UC: 42%). Inadequate responders were more likely to be prescribed a TNFi (for CD: odds ratio [OR] = 1.94; p < 0.001; for UC: OR = 2.76; p < 0.0001). CONCLUSION: More than 60% of patients with CD or UC had an inadequate response to their index advanced therapy within 1 year after initiation, mostly driven by low adherence. This modified claims-based algorithm for CD and UC appears useful to classify inadequate responders in health plan claims data.

Identifying inpatient mortality in MarketScan claims data using machine learning.

PURPOSE: Inpatient mortality is an important variable in epidemiology studies using claims data. In 2016, MarketScan data began obscuring specific hospital discharge status types for patient privacy, including inpatient deaths, by setting the values to missing. We used a machine learning approach to correctly identify hospitalizations that resulted in inpatient death using data prior to 2016. METHODS: All hospitalizations from 2011 to 2015 with discharge status of missing, died, or one of the other subsequently obscured values were identified and divided into a training set and two test sets. Predictor variables included age, sex, elapsed time from hospital discharge until last observed claim and until healthcare plan disenrollment, and absence of any discharge diagnoses. Four machine learning methods were used to train statistical models and assess sensitivity and positive predictive value (PPV) for inpatient mortality. RESULTS: Overall 1 307 917 hospitalizations were included. All four machine learning approaches performed well in all datasets. Random forest performed best with 88% PPV and 93% sensitivity for the training set and both test sets. The two factors with the highest relative importance for identifying inpatient mortality were having no observed claims for the patient on days 2-91 following hospital discharge and patient disenrollment from the healthcare plan within 60 days following hospital discharge. CONCLUSION: We successfully developed machine learning algorithms to identify inpatient mortality. This approach can be applied to obscured data to accurately identify inpatient mortality among hospitalizations with missing discharge status.

Accuracy of administrative claims prescription fill data to estimate glucocorticoid use and dose in patients with rheumatoid arthritis.

PURPOSE: To assess accuracy of administrative claims prescription fill-based estimates of glucocorticoid use and dose, and approximate bias from glucocorticoid exposure misclassification. METHODS: We identified adults with rheumatoid arthritis with linked Medicare and CorEvitas registry data. An algorithm identifying glucocorticoid use and average dose over 90 days from Medicare prescription fills was compared to physician-reported measures from a CorEvitas visit during the same period, using weighted kappa to compare doses (none, ≤5 mg, 5-10 mg, >10 mg/day). A deterministic sensitivity analysis examined the effect of exposure misclassification on estimated glucocorticoid-associated infection risk from a prior study. RESULTS: We identified 621 observations among 494 patients. Prescription fills identified glucocorticoid use in 41.9% of observations versus 31.1% identified by CorEvitas physician-report. For glucocorticoid use (yes/no), prescription fills had sensitivity 88.1% (95% CI 82.7-92.3), specificity 79.0% (74.8-82.7), PPV 65.4% (59.3-71.2), NPV 93.6% (90.6-95.9), and 81.8% agreement with CorEvitas, with kappa 0.61 (moderate to substantial agreement). There was 89.5% agreement between prescription fills and physician-reported doses, with weighted kappa 0.56 (moderate agreement). Applying these results to a prior Medicare study evaluating glucocorticoid-associated infection risk [risk ratio 1.44 (95% CI 1.41-1.48)] led to an externally adjusted risk ratio of 1.74 when accounting for exposure misclassification, representing -17% bias in infection risk estimate. CONCLUSIONS: This study supports the use of claims data to estimate glucocorticoid use and dose, but investigators should account for exposure misclassification, which may lead to underestimates of glucocorticoid risks. Our results could be applied to adjust risk estimates in other studies that use prescription fills to estimate glucocorticoid use.

Using pooled electronic health records data to conduct pharmacoepidemiology safety studies: Challenges and lessons learned.

PURPOSE: We assessed the suitability of pooled electronic health record (EHR) data from clinical research networks (CRNs) of the patient-centered outcomes research network to conduct studies of the association between tumor necrosis factor inhibitors (TNFi) and infections. METHODS: EHR data from patients with one of seven autoimmune diseases were obtained from three CRNs and pooled. Person-level linkage of CRN data and Centers for Medicare and Medicaid Services (CMS) fee-for-service claims data was performed where possible. Using filled prescriptions from CMS claims data as the gold standard, we assessed the misclassification of EHR-based new (incident) user definitions. Among new users of TNFi, we assessed subsequent rates of hospitalized infection in EHR and CMS data. RESULTS: The study included 45 483 new users of TNFi, of whom 1416 were successfully linked to their CMS claims. Overall, 44% of new EHR TNFi prescriptions were not associated with medication claims. Our most specific new user definition had a misclassification rate of 3.5%-16.4% for prevalent use, depending on the medication. Greater than 80% of CRN prescriptions had either zero refills or missing refill data. Compared to using EHR data alone, there was a 2- to 8-fold increase in hospitalized infection rates when CMS claims data were added to the analysis. CONCLUSIONS: EHR data substantially misclassified TNFi exposure and underestimated the incidence of hospitalized infections compared to claims data. EHR-based new user definitions were reasonably accurate. Overall, using CRN data for pharmacoepidemiology studies is challenging, especially for biologics, and would benefit from supplementation by other sources.

Improvement or Worsening of Disease Activity After Switch to Sarilumab in Patients With Rheumatoid Arthritis With a Partial Response to Adalimumab.

OBJECTIVE: The aim of this study was to assess the effect of switching from adalimumab to sarilumab monotherapy in partial responders with rheumatoid arthritis from the MONARCH randomized trial and its open-label extension (OLE). METHODS: Partial response was defined as improvement in Clinical Disease Activity Index (CDAI) of 12 or 6 units (baseline score: >22 or >10 and ≤22, respectively). Proportions of adalimumab partial responders with meaningful worsening or improvement at OLE weeks 12 and 24 were evaluated using 2 CDAI thresholds (≥6 and ≥12 points), 28-joint Disease Activity Score using erythrocyte sedimentation rate (≥0.6 and ≥1.2 points), Health Assessment Questionnaire Disability Index (≥0.22 and ≥0.30 points), Simple Disease Activity Index (≥7 and ≥13 points), physician and patient global assessments (≥10 and ≥20), and 28-joint swollen and tender joint counts (≥1 and ≥2 joints). Outcomes were analyzed using mixed-effect models with repeated measures for observed cases. The p values were produced using Wilcoxon tests. RESULTS: Of 369 enrolled patients, 320 (87%) entered the OLE and 155 switched from adalimumab to sarilumab; 59% (91/155) were partial responders. At week 24, 4%-17% and 2%-12% of partial responders experienced a worsening using the lower and higher thresholds, respectively, whereas 47%-78% and 27%-66% experienced improvement. CONCLUSIONS: Partial responders to adalimumab who switched to sarilumab had a low likelihood of experiencing meaningful worsening, with most patients showing meaningful improvement or no change in disease activity. This may help alleviate patients' fears of worsening when considering switching to a treatment with a different mechanism of action.

Effects of Disease-Worsening Following Withdrawal of Etanercept or Methotrexate on Patient-Reported Outcomes in Patients With Rheumatoid Arthritis: Results From the SEAM-RA Trial.

BACKGROUND/OBJECTIVE: The effect of treatment withdrawal on patient-reported outcomes (PROs) in patients with rheumatoid arthritis (RA) whose disease is in sustained remission has not been well described. This analysis aimed to compare PRO changes in patients with RA following medication withdrawal and disease worsening. METHODS: SEAM-RA (Study of Etanercept and Methotrexate in Combination or as Monotherapy in Subjects With Rheumatoid Arthritis) was a phase 3, multicenter, randomized withdrawal, double-blind controlled study in patients with RA taking methotrexate plus etanercept and in remission (Simple Disease Activity Index ≤3.3). Patient's Global Assessment of Disease Activity, Patient's Assessment of Joint Pain, Health Assessment Questionnaire-Disability Index, and 36-Item Short-Form Health Survey were evaluated for 48 weeks following methotrexate or etanercept withdrawal. Treatment differences for patients with versus without disease worsening were evaluated using a 2-sample t test for continuous end points and log-rank test for time-to-event end points. RESULTS: Of 253 patients, 121 experienced disease worsening and 132 did not. All PRO scores were similar to those of a general population at baseline and deteriorated over time across the study population. The PtGA and Patient's Assessment of Joint Pain values deteriorated less in those on etanercept monotherapy compared with methotrexate monotherapy. More patients with versus without disease worsening experienced deterioration that was greater than the minimal clinically important difference (MCID) for all PROs tested. In patients with disease worsening, PtGA deterioration more than the MCID preceded Simple Disease Activity Index disease worsening. CONCLUSIONS: Etanercept monotherapy showed benefit over methotrexate in maintaining PRO scores. Patients with disease worsening experienced a more rapid worsening of PtGA beyond the MCID versus patients without disease worsening.Categories: autoinflammatory disease, biological therapy, DMARDs, rheumatoid arthritis.

Meaningful Improvement and Worsening in Patients Who Do Not Achieve Low Disease Activity and Switch Therapy to a New Biologic or Targeted Disease-Modifying Antirheumatic Drug: Results From the CorEvitas RA Registry.

OBJECTIVE: The aim of this study was to assess the change in disease activity associated with switching from 1 biologic/targeted synthetic disease-modifying antirheumatic drug (b/tsDMARD) to another in patients with rheumatoid arthritis who did not achieve low disease activity (LDA) after 6 to 12 months of their initial treatment. METHODS: This observational study included patients from the CorEvitas Rheumatoid Arthritis Registry, who initiated a b/tsDMARD at the index visit (prebaseline), had any clinical disease activity index (CDAI) improvement but did not achieve LDA/remission at the subsequent visit (baseline), and switched therapy at baseline or between baseline and follow-up visits. Regardless of the preswitch CDAI value, 2 thresholds of CDAI change were used to define meaningful improvement and worsening for all patients: ≥6 units and ≥12 units; no meaningful change was defined as any change between -6 to +6 units and -12 to +12 units, based on respective thresholds. RESULTS: Of 1226 patients fulfilling the inclusion criteria, 93 (7.6%) switched therapy at baseline or between baseline and follow-up, after an inadequate response at the baseline visit. At follow-up, meaningful worsening occurred in 30.1% and 12.9% of switchers, whereas the remaining switchers achieved meaningful improvement (34.4% and 20.4%) or had no meaningful change (35.5% and 66.7%), based on the thresholds of ≥6 and ≥12 units, respectively. CONCLUSIONS: Rheumatoid arthritis patients, who had not achieved LDA and switched b/tsDMARD, were more likely to have meaningful improvement or no change, rather than meaningful worsening. These data may help some patients overcome their hesitancy to switch therapy, potentially improving clinical outcomes.

A Randomized Trial Comparing the Specific Carbohydrate Diet to a Mediterranean Diet in Adults With Crohn's Disease.

BACKGROUND & AIMS: This study compared the effectiveness of the Specific Carbohydrate Diet (SCD) to the Mediterranean diet (MD) as treatment for Crohn's disease (CD) with mild to moderate symptoms. METHODS: Adult patients with CD and with mild-to-moderate symptoms were randomly assigned 1:1 to consume the MD or SCD for 12 weeks. For the first 6 weeks, participants received prepared meals and snacks according to their assigned diet. After 6 weeks, participants were instructed to follow the diet independently. The primary outcome was symptomatic remission at week 6. Key secondary outcomes at week 6 included fecal calprotectin (FC) response (FC <250 µg/g and reduction by >50% among those with baseline FC >250 µg/g) and C-reactive protein (CRP) response (high-sensitivity CRP <5 mg/L and >50% reduction from baseline among those with high-sensitivity CRP >5 mg/L). RESULTS: The study randomized 194 patients, and 191 were included in the efficacy analyses. The percentage of participants who achieved symptomatic remission at week 6 was not superior with the SCD (SCD, 46.5%; MD, 43.5%; P = .77). FC response was achieved in 8 of 23 participants (34.8%) with the SCD and in 4 of 13 participants (30.8%) with the MD (P = .83). CRP response was achieved in 2 of 37 participants (5.4%) with the SCD and in 1 of 28 participants (3.6%) with the MD (P = .68). CONCLUSIONS: The SCD was not superior to the MD to achieve symptomatic remission, FC response, and CRP response. CRP response was uncommon. Given these results, the greater ease of following the MD and other health benefits associated with the MD, the MD may be preferred to the SCD for most patients with CD with mild to moderate symptoms. ClinicalTrials.gov Identifier: NCT03058679.

Incidence and risk factors for herpes zoster in patients with rheumatoid arthritis receiving upadacitinib: a pooled analysis of six phase III clinical trials.

BACKGROUND: Upadacitinib (UPA) is an oral Janus kinase (JAK) inhibitor approved for the treatment of rheumatoid arthritis (RA). JAK inhibitors have been associated with an increased risk of herpes zoster (HZ) in patients with RA. OBJECTIVES: To evaluate the incidence and risk factors for HZ in UPA-treated patients with RA from the UPA phase III clinical trial programme. METHODS: Exposure-adjusted incidence/event rates for HZ were determined in patients receiving UPA (monotherapy or combination therapy) in six randomised phase III trials (data cut-off on 30 June 2020). HZ incidence and event rates were also determined in patients receiving methotrexate (MTX) monotherapy or adalimumab (ADA) + MTX. Multivariable Cox regression analysis was used to identify HZ risk factors in UPA-treated patients. RESULTS: A total of 5306 patients were included in this analysis. The incidence rate of HZ/100 patient-years (95% CI) was 0.8 (0.3 to 1.9), 1.1 (0.5 to 1.9), 3.0 (2.6 to 3.5) and 5.3 (4.5 to 6.2), in the MTX monotherapy, ADA + MTX, UPA 15 mg and UPA 30 mg groups, respectively. The majority of HZ cases with UPA (71%) involved a single dermatome. Prior history of HZ and Asian region were HZ risk factors in UPA-treated patients. CONCLUSION: In the UPA phase III RA clinical programme, HZ incidence and event rates were higher with UPA versus ADA + MTX or MTX monotherapy, and higher with the 30 mg versus 15 mg dose. Patients from Asia and those with a history of HZ may be at increased risk of HZ while receiving UPA.