Search

We aim to generate an artificial neural network (ANN) model to predict early TNF inhibitor users in patients with ankylosing spondylitis. The baseline demographic and laboratory data of patients who visited Samsung Medical Center rheumatology clinic from Dec. 2003 to Sep. 2018 were analyzed. Patients were divided into two groups: early-TNF and non-early-TNF users. Machine learning models were formulated to predict the early-TNF users using the baseline data. Feature importance analysis was performed to delineate significant baseline characteristics. The numbers of early-TNF and non-early-TNF users were 90 and 505, respectively. The performance of the ANN model, based on the area under curve (AUC) for a receiver operating characteristic curve (ROC) of 0.783, was superior to logistic regression, support vector machine, random forest, and XGBoost models (for an ROC curve of 0.719, 0.699, 0.761, and 0.713, respectively) in predicting early-TNF users. Feature importance analysis revealed CRP and ESR as the top significant baseline characteristics for predicting early-TNF users. Our model displayed superior performance in predicting early-TNF users compared with logistic regression and other machine learning models. Machine learning can be a vital tool in predicting treatment response in various rheumatologic diseases. [ABSTRACT FROM AUTHOR]

Aim: To evaluate the effect of tumor necrosis factor a inhibitors (TNFi) on spinal radiographic progression in patients with ankylosing spondylitis (AS). Methods: Subjects were selected from patients at a single tertiary hospital between 1995 and 2014. Patients who used TNFi with baseline and paired follow-up radiographic data with a minimum interval of 2 years were included. Time to start TNFi was defined as the time from symptom onset to the start of TNFi use. TNFi index was defined as the ratio of the period of TNFi use to the entire period of disease. Radiographic damage was assessed by the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS). Univariable and multivariable linear regression analyses were used to identify factors associated with radiographic progression. Results: A total of 151 patients were included in the analysis. Seventeen (11.3%) patients were female and mean DmSASSS/year was 1.01 units/year. Mean X-ray follow-up duration was 102.9 ± 54.9 months. Mean time from symptom onset to start of TNFi use was 104.8 ± 83.6 months (median 84 months) and mean TNFi index was 42.9 ± 23.8% (median 40.9%). In multivariable analysis, initial mSASSS, initial C-reactive protein, body mass index, current smoker, and delayed start of TNFi use were associated with radiographic progression. Presence of peripheral arthritis and the TNFi index were negatively associated with radiographic progression. Conclusions: A delay in starting TNFi use and low TNFi index were associated with radiographic progression. Early and long-term use of TNFi appear to reduce spinal radiographic progression in patients with AS. [ABSTRACT FROM AUTHOR]

Aim To evaluate the clinical characteristics and outcomes of nonradiographic axial spondyloarthritis (nr-axSpA) in Korean patients. Methods A retrospective analysis evaluated 155 patients with nr-axSpA at a single tertiary hospital between January 2001 and January 2011. Baseline characteristics and clinical courses were reviewed and compared with those of patients with ankylosing spondylitis ( AS). Results The mean age at disease onset was 29.5 ± 10.8 years and 52 (33.5%) patients were female. The mean age at symptom onset was older (29.5 ± 10.8 vs 25.9 ± 9.2, respectively, P < 0.001) and the male-to-female ratio was lower (2 : 1 vs 5 : 1, respectively, P = 0.001) in patients with nr-axSpA compared with patients with AS. The proportion of females was higher among patients with late onset SpA than early-onset nr-axSpA (55.0% vs 30.1%, respectively, P = 0.029). Among 74 patients with nr-axSpA, whose follow-up duration was more than 1.5 years, 29 (39.2%) patients progressed to AS during the follow-up period. The proportion of females was lower in progressors that that of non-progressors (13.8% vs 44.4%, respectively, P = 0.010). Presence of syndesmophyte and minimal X-ray changes at baseline were frequently observed in progressors compared with non-progressors (26.7% vs 0.0%, P = 0.006 and 69.0 vs 35.6%, P = 0.005, respectively). Conclusions The predominance of male patients is more prominent among Korean patients with SpA compared with Caucasians. Female nr-axSpA patients had late symptom onset and less progression to AS. X-ray changes at baseline were associated with radiographic progression. [ABSTRACT FROM AUTHOR]

Background: Patients undergoing anti-tumor necrosis factor (TNF) treatment are at an increased risk of reactivating a latent tuberculosis infection (LTBI). This study evaluated the effectiveness of the QuantiFERON-TB Gold In-Tube (QFT) assay for diagnosing LTBI in arthritis patients undergoing anti-TNF treatment. Methods: We enrolled 342 consecutive patients from August 2007 to October 2013: 176 (51.5%) patients with ankylosing spondylitis and 166 (48.5%) with rheumatoid arthritis. Screening tests included tuberculin skin test (TST) and QFT assay. Positive QFT results, regardless of TST results, were considered an indicator for LTBI treatment. Results: Bacillus Calmette-Guérin scars were found in 236 (69.0%) patients. Of 342 patients, TST and QFT were positive in 122 (35.7%) and 103 (30.1%) patients, respectively, and discordant in 101 (29.5%) patients. During a median follow-up duration of 41.7 months, five patients (1.5%) developed TB in a median of 20.8 months after initiation of anti-TNF treatment (428/100,000 person-years). TB did not occur in 62 TST+/QFT+ patients who received LTBI treatment. Of 41 TST−/QFT+ patients who received LTBI treatment, one (2.4%) developed TB 20.5 months after starting anti-TNF treatment (705/100,000 person-years). Of 60 TST+/QFT− patients who did not receive LTBI treatment, two (3.3%) developed TB 20.8 and 22.0 months after starting anti-TNF treatment (871/100,000 person-years). Of 179 TST−/QFT− patients, two (1.1%) developed TB 7.2 and 22.7 months, respectively, after initiating anti-TNF treatment (341/100,000 person-years). TB incidence rate during the follow-up period did not differ among TST−/QFT+, TST+/QFT−, and TST−/QFT− patients (P = 0.661). Conclusion: QFT might be used instead of TST for diagnosing LTBI in patients before starting anti-TNF therapy in countries, such as Korea, where the TB prevalence is intermediate and the BCG vaccination is mandatory at birth. In the absence of a true gold standard test for LTBI, however, there is still a risk of TB development during anti-TNF treatment. [ABSTRACT FROM AUTHOR]