Crowson, Cynthia S.

Objective To determine the association of allergic conditions with incident rheumatoid arthritis (RA), especially in relation to smoking history and anti-citrullinated peptide antibody (ACPA) status. Methods This case–control study included 3515 incident RA cases and 5429 matched controls from the Epidemiological Investigation of Rheumatoid Arthritis study 1995 to 2016, including questionnaire-based information on eight allergic conditions composed from a list of 59 unique allergies. We used logistic regression and adjusted ORs (aOR) to assess the association between allergic conditions and risk of RA, adjusting for age, sex, residential area, body mass index, education, and smoking, and stratified by smoking and ACPA. Results A history of any reported allergy was equally common in RA (n=1047, 30%) as among population controls (n=1540, 29%), aOR 1.04, 95% CI 0.95 to 1.15. Metal, respiratory, food, plant/pollen and chemical allergies were not associated with risk of RA. By contrast, statistically significant associations were observed for animal dander allergy (6% vs 5%, aOR 1.37, 95% CI 1.03 to 1.82), especially in ACPA-positive RA (aOR 1.46 95% CI 1.06 to 2.01) and for atopic dermatitis, in particular for older and ACPA-negative RA (aOR 2.33, 95% CI 1.37 to 3.96 at age 80). Never smokers with allergic rhinitis also had increased risk of developing RA (aOR 1.30, 95% CI 1.00 to 1.68). Conclusion Most common allergies do not increase risk of RA, nor do they protect against RA. However, some allergic conditions, notably animal dander allergy, atopic dermatitis and allergic rhinitis, were associated with an increased risk for RA.

Objective To comprehensively assess multimorbidity burden in patients with rheumatoid arthritis (RA) in order to unify the multimorbidity definition for RA research and clinical practice. Methods In this population-based study, residents of eight Minnesota counties with prevalent RA on 1 January 2015 were identified. Age, sex and county-matched non-RA comparators were selected from the same population. Diagnostic codes were retrieved for 5 years before 1 January 2015. Using two codes ≥30 days apart, 44 previously defined morbidities and 78 non-overlapping chronic disease categories based on Clinical Classification Software were defined. Prevalence of each morbidity in the RA versus non-RA cohorts was compared using false discovery rate to adjust for multiple comparisons. Morbidities more common in RA than non-RA and those with prevalence ≥5% were retained. Results 1643 patients with RA and 1643 non-RA subjects (72% women; mean age 63.1 years) were studied. Using the 44 morbidities, multimorbidity (defined as 2+ morbidities) was present in 1411 (86%) of RA and 1164 (71%) of non-RA subjects (p<0.001) with 5+ morbidities present in 907 (55%) of RA and 619 (38%) of non-RA (p<0.001). Patients with RA had significantly higher prevalence of 24 of the 44 morbidities compared with non-RA, especially interstitial lung disease, fibromyalgia, osteoarthritis and osteoporosis. Among the additional 78 categories, 7 were significantly higher in RA than non-RA, including organic sleep disorders, vitamin D deficiency and foot ulcers. Conclusion Patients with RA have a higher prevalence of multimorbidity compared with non-RA subjects. These results confirm the list of 44 morbidities and add several other morbidities of interest in RA.

ObjectiveTo identify the incidence, risk factors, and outcomes of rheumatoid arthritis–associated interstitial lung disease (RA‐ILD) and to assess time trends in the incidence and mortality in RA‐ILD.MethodsWe included adult residents of Olmsted County, Minnesota with incident RA between 1999 and 2014. Subjects were followed until death, emigration, or April 30, 2019. ILD was defined as the presence of a radiologist‐defined pattern consistent with ILD on chest computed tomography (CT). When chest CT was absent, the combination of chest radiograph abnormalities compatible with ILD and restrictive pattern on pulmonary function testing was considered consistent with ILD. Potential risk factors included age, sex, smoking, obesity, seropositivity, extraarticular manifestations (EAMs), and medications. For survival analysis, we matched RA‐ILD patients to RA–non‐ILD comparators. The frequency and mortality from clinician‐diagnosed RA‐ILD from 1999 to 2014 was compared against a cohort from 1955 to 1994.ResultsDuring the 1999–2014 time period, 645 individuals (70% women) had incident RA, were a median age of 55.3 years, and 53% never smoked. Twenty‐two patients had ILD before RA, and 51 (67% women) developed ILD during follow‐up. The 20‐year cumulative incidence of RA‐ILD was 15.3%. Ever‐smoking (hazard ratio [HR] 1.92), age at RA onset (HR 1.89 per 10‐year increase), and severe EAMs (HR 2.29) were associated with incident RA‐ILD. The RA‐ILD cases had higher mortality than their matched RA comparators (HR 2.42). Incidence of RA‐ILD was non‐significantly lower from 1999 to 2014 than from 1955 to1994, but mortality was improved.ConclusionsRA‐ILD occurs in nearly 1 in 6 patients with RA within 20 years and is associated with shorter survival. Lack of significant change in RA‐ILD incidence over 6 decades deserves further investigation.

Objective To test the ability of machine learning (ML) approaches with clinical and genomic biomarkers to predict methotrexate treatment response in patients with early rheumatoid arthritis (RA). Methods Demographic, clinical, and genomic data from 643 patients of European ancestry with early RA (mean age 54 years; 70% female) subdivided into a training (n = 336) and validation cohort (n = 307) were used. The genomic data comprised 160 single‐nucleotide polymorphisms (SNPs) previously associated with RA or methotrexate metabolism. Response to methotrexate monotherapy was defined as good or moderate by the European Alliance of Associations for Rheumatology (EULAR) response criteria at the 3‐month follow‐up. Supervised ML methods were trained with 5 repeats and 10‐fold cross‐validation using the training cohort. Prediction performance was validated in the independent validation cohort. Results Supervised ML methods combining age, sex, smoking, rheumatoid factor, baseline Disease Activity Score in 28 joints (DAS28) scores and 160 SNPs predicted EULAR response at 3 months with the area under the receiver operating curve of 0.84 ( P = 0.05) in the training cohort and achieved a prediction accuracy of 76% ( P = 0.05) in the validation cohort (sensitivity 72%, specificity 77%). Intergenic SNPs rs12446816, rs13385025, rs113798271, and ATIC (rs2372536) had variable importance above 60.0 and along with baseline DAS28 scores were among the top predictors of methotrexate response. Conclusion Pharmacogenomic biomarkers combined with baseline DAS28 scores can be useful in predicting response to methotrexate in patients with early RA. Applying ML to predict treatment response holds promise for guiding effective RA treatment choices, including timely escalation of RA therapies.

AbstractBackgroundThe World Trade Center (WTC) general responder cohort (GRC) was exposed to environmental toxins possibly associated with increased risk of developing autoimmune conditions.ObjectivesTwo study designs were used to assess incidence and risks of autoimmune conditions in the GRC.MethodsThree clinically trained professionals established the status of possible GRC cases of autoimmune disorders adhering to diagnostic criteria, supplemented, as needed, by specialists’ review of consenting responders’ medical records. Nested case‐control analyses using conditional logistic regression estimated the risk associated with high WTC exposure (being in the 9/11/2001 dust cloud or ≥median days' response worked) compared with low WTC exposure (all other GRC members'). Four controls were matched to each case on age at case diagnosis (±2 years), sex, race/ethnicity, and year of program enrollment. Sex‐specific and sensitivity analyses were performed. GRC age‐ and sex‐adjusted standardized incidence ratios (SIRs) were compared with the Rochester Epidemiology Project (REP). Complete REP inpatient and outpatient medical records were reviewed by specialists. Conditions meeting standardized criteria on ≥2 visits were classified as REP confirmed cases.ResultsSix hundred and twenty‐eight responders were diagnosed with autoimmune conditions between 2002 and 2017. In the nested case‐control analyses, high WTC exposure was not associated with autoimmune domains and conditions (rheumatologic domain odds ratio [OR] = 1.03, 95% confidence interval [CI] = 0.77, 1.37; rheumatoid arthritis OR = 1.12, 95% CI = 0.70, 1.77). GRC members had lower SIR than REP. Women's risks were generally greater than men's.ConclusionsThe study found no statistically significant increased risk of autoimmune conditions with WTC exposures.

Objective To characterize the incidence and prevalence of childhood‐onset systemic lupus erythematosus (SLE), and to estimate the proportion of patients who are diagnosed with SLE during childhood. Methods A cohort of patients with incident childhood‐onset SLE from 1976 to 2018 from an 8‐county region in the US were identified based on comprehensive medical record review. All patients met the European Alliance of Associations for Rheumatology (EULAR)/American College of Rheumatology (ACR) classification criteria for SLE or the ACR SLE classification criteria from 1997 at or before age 18 years. Incidence rates were estimated using Poisson methods. We estimated the childhood‐onset SLE point prevalence for January 1, 2015. Results were sex and age adjusted to the US 2000 population. Among all the SLE patients living in the 8‐county region on January 1, 2015, the proportion of patients diagnosed at ≤18 years was estimated. Results A total of 13 children were diagnosed with childhood‐onset SLE during the study period (using the EULAR/ACR definition; mean age at diagnosis 15.1 years, 85% female, 69% White). Childhood‐onset SLE overall adjusted incidence rate was 0.7 (95% confidence interval [95% CI] 0.2–1.1) per 100,000 children. The incidence rate in girls was 1.2 (95% CI 0.5–1.9) per 100,000 children, while in boys it was 0.2 (95% CI 0.0–0.5) per 100,000. The adjusted prevalence of childhood‐onset SLE was 1.1 (95% CI 0.0–3.1) per 100,000 children. The proportion of patients with SLE diagnosed as children was 9% (95% CI 6–13%). Conclusion In this population‐based study, both the incidence and prevalence rates of childhood‐onset SLE were ~1 per 100,000 children. One in 10 adults with SLE was diagnosed in childhood. More studies are needed to further characterize the epidemiology of childhood‐onset SLE in minorities.

Abstract Objectives We aimed to estimate the risk of HCQ retinopathy and its risk factors among incident users in the community. Methods Using the Rochester Epidemiology Project, a record-linkage system, a cohort of incident users of HCQ was identified from 27 counties in the American upper Midwest. HCQ retinopathy was defined based on characteristic paracentral automated 10-2 visual field (10-2 AVF) defects and parafoveal retinal photoreceptor layer changes on spectral domain optical coherence tomography. Cumulative incidence rates were estimated adjusting for competing risk of death. Risk factors for HCQ retinopathy were examined using Cox models. Results The study included 634 incident HCQ users (mean age at initial HCQ use was 53.7 years, 79% females, 91% white). Most common indications for HCQ were RA (57%) and SLE (19%). The average follow-up length was 7.6 years. Eleven patients developed HCQ retinopathy (91% females, 91% white). The majority used HCQ for RA (91%). The cumulative incidence rate at year 5 was 0%, which increased to 3.9% (95% CI 2.0, 7.4) by 10 years. Taking an HCQ dose ≥5 mg/kg was associated with a hazard ratio (HR) of 3.59 (95% CI 1.09, 11.84) compared with lower doses. There was a 48% increase [HR 1.48 (95% CI 1.03, 2.14)] in the risk of HCQ retinopathy for each 100 g of HCQ cumulative dose. Conclusion The risk of HCQ retinopathy at 10 years of use is lower compared with previous prevalence-based estimations. A dose ≥5 mg/kg was associated with higher HCQ retinopathy risk.

AbstractObjectiveTo estimate the incidence and time-to-classification of SLE by the 1997 ACR (ACR97) criteria, the SLICC criteria, and the European Alliance of Associations for Rheumatology/ACR (EULAR/ACR) criteria.MethodsWe identified all incident SLE cases from 2000–2018 in the well-defined Olmsted County population. Clinical data included in the ACR97, SLICC and EULAR/ACR criteria were manually abstracted from medical records. All incident cases met at least one of the three classification criteria. Time-to-classification was estimated from the first documented lupus-attributable disease manifestation to the time of criteria fulfilment by each of the three definitions. Annual incidence rates were age or age/sex adjusted to the 2000 US population.ResultsOf 139 incident cases there were 126 cases by the EULAR/ACR criteria, corresponding to an age/sex-adjusted incidence of 4.5 per 100 000 population (95% CI: 3.7, 5.2). The age/sex-incidence was higher than that of the SLICC criteria (113 cases; 4.0 per 100 000 [95% CI: 3.3, 4.7], P = 0.020) and the ACR97 (92 cases; 3.3 per 100 000 [95% CI: 2.6, 3.9], P &lt; 0.001). The median time from first disease manifestation to criteria fulfilment was shorter for the EULAR/ACR criteria (29.4 months) than the ACR97 criteria (47.0 months, P &lt; 0.001) and similar to the SLICC criteria (30.6 months, P = 0.83).ConclusionThe incidence of SLE was higher by the EULAR/ACR criteria compared with the ACR97 and the SLICC criteria, and the EULAR/ACR criteria classified patients earlier that the ACR97 criteria but similar to the SLICC criteria.