Deep Learning Enables Automatic Detection of Joint Damage Progression in Rheumatoid Arthritis-Model Development and External Validation.

OBJECTIVES: Although deep learning has demonstrated substantial potential in automatic quantification of joint damage in rheumatoid arthritis (RA), evidence for detecting longitudinal changes at an individual patient level is lacking. Here, we introduce and externally validate our automated RA scoring algorithm (AuRA), and demonstrate its utility for monitoring radiographic progression in a real-world setting. METHODS: The algorithm, originally developed during the Rheumatoid Arthritis 2-Dialogue for Reverse Engineering Assessment and Methods (RA2-DREAM) challenge, was trained to predict expert-curated Sharp-van der Heijde total scores in hand and foot radiographs from two previous clinical studies (n = 367). We externally validated AuRA against data (n = 205) from Turku University Hospital and compared the performance against two top-performing RA2-DREAM solutions. Finally, for 54 patients, we extracted additional radiograph sets from another control visit to the clinic (average time interval of 4.6 years). RESULTS: In the external validation cohort, with a root-mean-square-error (RMSE) of 23.6, AuRA outperformed both top-performing RA2-DREAM algorithms (RMSEs 35.0 and 35.6). The improved performance was explained mostly by lower errors at higher expert-assessed scores. The longitudinal changes predicted by our algorithm were significantly correlated with changes in expert-assessed scores (Pearson's R = 0.74, p< 0.001). CONCLUSION: AuRA had the best external validation performance and demonstrated potential for detecting longitudinal changes in joint damage. Available in https://hub.docker.com/r/elolab/aura, our algorithm can easily be applied for automatic detection of radiographic progression in the future, reducing the need for laborious manual scoring.

Patient-reported outcomes in axial spondyloarthritis and psoriatic arthritis patients treated with secukinumab for 24 months in daily clinical practice.

OBJECTIVES: In patients with axial spondyloarthritis (axSpA) or psoriatic arthritis (PsA) initiating secukinumab, we aimed to assess and compare the proportion of patients achieving 6-, 12- and 24-month patient-reported outcomes (PRO) remission and the 24-month retention rates. PATIENTS AND METHODS: Patients with axSpA or PsA from 16 European registries, who initiated secukinumab in routine care were included. PRO remission rates were defined as pain, fatigue, Patient Global Assessment (PGA) ≤2 (Numeric Rating Scale (NRS) 0-10) and Health Assessment Questionnaire (HAQ) ≤0.5, for both axSpA and PsA, and were calculated as crude values and adjusted for drug adherence (LUNDEX). Comparisons of axSpA and PsA remission rates were performed using logistic regression analyses (unadjusted and adjusted for multiple confounders). Kaplan-Meier plots with log-rank test and Cox regression analyses were conducted to assess and compare secukinumab retention rates. RESULTS: We included 3087 axSpA and 3246 PsA patients initiating secukinumab. Crude pain, fatigue, PGA and HAQ remission rates were higher in axSpA than in PsA patients, whereas LUNDEX-adjusted remission rates were similar. No differences were found between the patient groups after adjustment for confounders. The 24-month retention rates were similar in axSpA vs. PsA in fully adjusted analyses (HR [95 %CI] = 0.92 [0.84-1.02]). CONCLUSION: In this large European real-world study of axSpA and PsA patients treated with secukinumab, we demonstrate for the first time a comparable effectiveness in PRO remission and treatment retention rates between these two conditions when adjusted for confounders.

Similar levels of disease activity and remission rates in patients with psoriatic arthritis and rheumatoid arthritis-results from the Finnish quality register.

OBJECTIVES: To compare the current disease activity and remission rates, and their regional variation in patients with psoriatic arthritis (PsA) and rheumatoid arthritis (RA) in Finland. METHODS: Data of patients' most recent visit in 1/2020-9/2021 were extracted from the Finnish Rheumatology Quality Register. Measures for disease activity and remission included joint counts, DAS28, cDAPSA, CDAI, the Boolean definition, and physician assessment. Regression analyses were applied, adjusted for age and sex. RESULTS: Data of 3598 patients with PsA (51% female, mean age 54 years) and 13,913 patients with RA (72% female, 74% ACPA-positive, mean age 62 years) were included. The median (IQR) DAS28 was 1.9 (1.4, 2.6) in PsA and 2.0 (1.6, 2.7) in RA (p = 0.94); for cDAPSA, the median (IQR) values were 7.7 (3.1, 14) in PsA and 7.7 (3.3, 14) in RA (p < 0.001). In all regions in both diseases, the median DAS28 was ≤ 2.6 and the median cDAPSA < 13. Remission rates included DAS28 < 2.6 in 73% in PsA and 69% in RA (p = 0.17) and Boolean remission in 17% in PsA and 15% in RA (p < 0.001). By other definitions of remission, the rates ranged between 30% and 46%. Methotrexate was currently used by 49% in PsA and 57% in RA (p < 0.001). Self-administered bDMARDs were currently used by 37% in PsA and 21% in RA (p < 0.001). CONCLUSION: The overall disease activity was low and similar in patients with PsA and RA across the country. Remission rates varied between 15 and 73%, depending on the definition but were similar in PsA and RA. Key Points • The disease activity and clinical picture was similar between patients with PsA and RA, in a cross-sectional setting in 1.2020-9.2021. • A significant majority of patients with PsA had low disease activity or were in remission according to cDAPSA. Majority of patients with RA were in remission according to DAS28. • Patients with PsA and RA used methotrexate similarly. The utilization of bDMARDs was more prevalent in patients with PsA.

Development of prediction model for alanine transaminase elevations during the first 6 months of conventional synthetic DMARD treatment.

Frequent laboratory monitoring is recommended for early identification of toxicity when initiating conventional synthetic disease-modifying antirheumatic drugs (csDMARDs). We aimed at developing a risk prediction model to individualize laboratory testing at csDMARD initiation. We identified inflammatory joint disease patients (N = 1196) initiating a csDMARD in Turku University Hospital 2013-2019. Baseline and follow-up safety monitoring results were drawn from electronic health records. For rheumatoid arthritis patients, diagnoses and csDMARD initiation/cessation dates were manually confirmed. Primary endpoint was alanine transaminase (ALT) elevation of more than twice the upper limit of normal (ULN) within 6 months after treatment initiation. Computational models for predicting incident ALT elevations were developed using Lasso Cox proportional hazards regression with stable iterative variable selection (SIVS) and were internally validated against a randomly selected test cohort (1/3 of the data) that was not used for training the models. Primary endpoint was reached in 82 patients (6.9%). Among baseline variables, Lasso model with SIVS predicted subsequent ALT elevations of > 2 × ULN using higher ALT, csDMARD other than methotrexate or sulfasalazine and psoriatic arthritis diagnosis as important predictors, with a concordance index of 0.71 in the test cohort. Respectively, at first follow-up, in addition to baseline ALT and psoriatic arthritis diagnosis, also ALT change from baseline was identified as an important predictor resulting in a test concordance index of 0.72. Our computational model predicts ALT elevations after the first follow-up test with good accuracy and can help in optimizing individual testing frequency.

Initial presentation of early rheumatoid arthritis.

OBJECTIVES: To study the joint distribution and clinical picture of rheumatoid arthritis (RA) at the initial presentation in seropositive (anti-citrullinated protein antibody (ACPA) and/or rheumatoid factor (RF) positive) and negative patients and the effect of duration of symptoms on the clinical picture. METHODS: Data of patients who received reimbursement for DMARDs for newly diagnosed RA in 1/2019 to 9/2021 were extracted from the national databases. Joint counts, presence of symmetrical swelling, other disease activity measures, and patient reported outcomes (PROs) were compared in seropositive and negative patients. Regression analyses were applied to compare clinical variables in patients with duration of symptoms of <3, 3-6, and >6 months, adjusted for age, sex, and seropositivity. RESULTS: Data of 1816 ACPA and RF-tested patients were included. Symmetrical swelling was present in 75% of patients. Seronegative versus positive patients had higher value for all disease activity measures and PROs including median swollen joint count (SJC46 10 versus 5) and DAS28 (4.7 versus 3.7), (p<0.001). Patients diagnosed in <3 months had higher median pain VAS (62 versus 52 and 50, p<0.001) and HAQ (1.1 versus 0.9 and 0.75, p = 0.002) compared to those with a duration of symptoms of 3-6 and >6 months. Patients diagnosed >6 months were ACPA-positive more frequently (77% versus 70% in other groups, p = 0.045). CONCLUSION: Incident RA presents mainly as symmetric arthritis. Seronegative patients have higher disease burden at the initial presentation. Patients experiencing more severe pain and decreased functional ability are diagnosed earlier, regardless of ACPA- status.

Disease burden measured by patient-reported outcomes: does psoriatic arthritis feel worse than rheumatoid arthritis? A cross-sectional nationwide study.

OBJECTIVES: To study the subjective disease burden of patients with psoriatic arthritis (PsA) and rheumatoid arthritis (RA), using patient-reported outcomes (PROs) cross-sectionally. METHODS: Data of 3598 patients with PsA and 13913 with RA were extracted from the database. Measures included the VAS-values of pain, fatigue and patient global assessment (PGA), HAQ, and disease activity at the most recent visit/remote contact in the period 1.2020 to 9.2021. Values were compared between patients with PsA and RA overall, and by sex and age (<50, 50-59, 60-69 and ≥70 years). Regression analyses were applied. RESULTS: The overall median (IQR)-values for pain were 29 (10, 56) for PsA and 26 (10, 51) for RA, 29 (9, 60) and 28 (8, 54) for fatigue, 28 (10, 52) and 29 (11, 51) for PGA, 0.4 (0, 0.9) and 0.5 (0, 1.0) for HAQ (p<0.001 for all comparisons; adjusted for sex and age). The median (IQR)-values for pain, fatigue, PGA and HAQ were higher for PsA vs. RA in most age groups for males and females. All PROs were higher in older patients with both diagnoses. The median values for DAS28, doctor global assessment, ESR and CRP were 1.9 vs. 2.0, 8 vs. 8, 7 vs. 8 and 2 vs. 3 in PsA and RA, respectively. CONCLUSIONS: Overall, both PsA and RA groups showed moderate disease control by patients' perspective, but the burden of disease was higher especially in women with PsA compared to RA. Disease activity was similar and low in both diseases.

Automated Text Message-Enhanced Monitoring Versus Routine Monitoring in Early Rheumatoid Arthritis: A Randomized Trial.

OBJECTIVE: Frequent monitoring of patients with early rheumatoid arthritis (RA) is required for achieving good outcomes. This study was undertaken to investigate the influence of text message (SMS)-enhanced monitoring on early RA outcomes. METHODS: We randomized 166 patients with early, disease-modifying antirheumatic drug-naive RA to receive SMS-enhanced follow-up or routine care. All patients attended visits at 0, 3, and 6 months, and a follow-up visit at 12 months. Treatment was at the physicians' discretion. The intervention included 13 SMSs during weeks 0-24 with questions concerning medication problems (yes/no) and disease activity (patient global assessment [PtGA], scale 0-10). Patients were contacted if response SMSs indicated medication problems or PtGA exceeded predefined thresholds. Primary outcome was 6-month Boolean remission (no swollen or tender joints and normal C-reactive protein levels). Quality of life (QoL; measured by the Short Form 36 survey) and Disease Activity Score in 28 joints (DAS28) were assessed. RESULTS: Six and 12-month follow-up data were available for 162 and 157 patients, respectively. In the intervention group, 46% of the patients (38 of 82) reported medication problems and 49% (40 of 82) reported text message PtGAs above the alarm limit. Remission rates at 6 months (P = 0.34) were 51% in the intervention group and 42% in the control group. These rates were 57% and 43% at 12 months (P = 0.17) in the intervention and control groups, respectively. The respective mean ± SD DAS28 scores for the intervention and control groups were 1.92 ± 1.12 and 2.22 ± 1.11 at 6 months (P = 0.09); and 1.79 ± 0.91 and 2.08 ± 1.22 at 12 months (P = 0.28). No differences in QoL were observed. CONCLUSION: The study did not meet the primary outcome despite a trend favoring the intervention group. This may be explained by the notably high overall remission rates.

High burden of adverse events is associated with reduced remission rates in early rheumatoid arthritis.

Adverse events (AEs) are common during disease-modifying antirheumatic drug (DMARD) treatment, but their influence on treatment results is unclear. We studied AEs in relation to disease activity in early rheumatoid arthritis (RA). Ninety-nine patients started intensive treatment with three conventional synthetic DMARDs (csDMARDs) and oral prednisolone, and were randomized to a 6-month induction treatment with infliximab or placebo. All AEs during the first 12 months of treatment were recorded. We scored each AE based on severity (scale 1-4) and defined the burden of AEs as the sum of these scores. Patients were divided into tertiles according to the burden of AEs. As outcomes, we assessed 28-joint disease activity score (DAS28) levels and remission rates at 12 and 24 months. Three hundred thirty-one AEs in 99 patients were reported, and 27 (8%) were categorized as severe or serious. Mean burden of AEs per patient was 5.4 ± 4.3. Seventy-nine AEs (24%) led to temporary (n = 52) or permanent (n = 27) csDMARD discontinuation. Of discontinuations, 1, 21, and 57 were detected in the first, second, and third tertiles, respectively. DAS28 remission rates decreased across tertiles at 12 months (94, 94, and 76%; p for linearity 0.029) and at 24 months (90, 86, and 70%; p for linearity 0.021). Mean DAS28 levels increased across tertiles at 12 months (1.5 ± 1.0, 1.7 ± 0.9, and 1.9 ± 1.2; p for linearity 0.021) and at 24 months (1.4 ± 0.8, 1.6 ± 1.0, and 1.9 ± 1.1; p for linearity 0.007). High burden of AEs is associated with higher disease activity and lower likelihood of remission in early RA.

Patient-reported outcomes as predictors of remission in early rheumatoid arthritis patients treated with tight control treat-to-target approach.

Identifying prognostic factors for remission in early rheumatoid arthritis (ERA) patients is of key clinical importance. We studied patient-reported outcomes (PROs) as predictors of remission in a clinical trial. We randomized 99 untreated ERA patients to receive remission-targeted treatment with three disease-modifying antirheumatic drugs and prednisolone for 24 months, and infliximab or placebo for the initial 6 months. At baseline, we measured following PROs: eight Short Form 36 questionnaire (SF-36) dimensions, patient's global assessment [PGA, visual analogue scale (VAS)], Health Assessment Questionnaire (HAQ), and pain VAS. We used multivariable-adjusted regression models to identify PROs that independently predicted modified American College of Rheumatology remission at 2 years. Follow-up data at 2 years were available for 93 patients (92%), and 58 patients (62%) were in remission. At baseline, patients who achieved remission had higher radiological score (p = 0.04), lower tender joint count (p = 0.001), lower PGA (p = 0.005) and physician's global assessment (p = 0.019), lower HAQ (p = 0.016), less morning stiffness (p = 0.009), and significantly higher scores in seven out of eight SF-36 dimensions compared with patients who did not. In multivariable models that included all PROs, remission was associated with SF-36 dimensions higher vitality (odds ratio 2.01; 95% confidence interval 1.19-3.39) and better emotional role functioning (odds ratio 1.64; 95% confidence interval 1.01-2.68). PGA, pain VAS, HAQ, and other SF-36 dimensions were not associated with remission. We conclude that self-reported vitality and better emotional role functioning are among the most important PROs for the prediction of remission in ERA.

Intra-articular glucocorticoid injections should not be neglected in the remission targeted treatment of early rheumatoid arthritis: a post hoc analysis from the NEO-RACo trial.

OBJECTIVES: To study the effects of neglecting intra-articular glucocorticoid injections (IAGCIs) into swollen joints in early rheumatoid arthritis (RA). METHODS: Ninety-nine patients with early, DMARD naive RA were treated, aiming at remission, with methotrexate, sulfasalazine, hydroxychloroquine, low-dose oral prednisolone and, when needed, IAGCIs for 2 years, and randomised to receive infliximab or placebo from weeks 4 to 26. During each of the 15 study visits, patients were scored retrospectively 0.2-0.4 points (depending on the number of non-injected joints) if IAGCIs to all swollen joints were not given. Patients were divided into tertiles by their cumulative scores for neglected injections (CSNI) over 24 months. 28-joint disease activity score (DAS28) area under the curve (AUC) between 0-24 months, remission rates, changes in quality of life, and radiological changes during the follow-up were assessed. Trends across tertiles of CSNI were tested with generalised linear models. RESULTS: Higher CSNI was associated with lower strict remission rates (p=0.005), and lower quality of life (p=0.004) at 24 months, and higher DAS28 AUC (p<0.001) during the follow-up. At 24 months, DAS28 remission rates were 90%, 93% and 76% (p=0.081), and strict remission rates were 74%, 77% and 39% by tertiles of CSNI. No significant differences were observed in radiological progression (p=0.089). IAGCIs were well tolerated. CONCLUSIONS: Neglecting IAGCIs into swollen joints is associated with lower remission rates, higher disease activity, and lower quality of life. Hence, IAGCIs should be used as an integral part of the targeted treatment of early RA.

Psychiatric (axis I) and personality (axis II) disorders in patients with burning mouth syndrome or atypical facial pain.

Background and aims Burning mouth syndrome (BMS) and atypical facial pain (AFP) are often persistent idiopathic pain conditions that mainly affect middle-aged and elderly women. They have both been associated with various psychiatric disorders. This study examined current and lifetime prevalence of psychiatric axis I (symptom-based) and II (personality) disorders in patients with chronic idiopathic orofacial pain, and investigated the temporal relationship of psychiatric disorders and the onset of orofacial pain. Method Forty patients with BMS and 23 patients with AFP were recruited from Turku university hospital clinics. Mean age of the patients was 62.3 years (range 35-84) and 90% were female. BMS and AFP diagnoses were based on thorough clinical evaluation, and all patients had undergone clinical neurophysiological investigations including blink reflex and thermal quantitative tests. Current and lifetime DSM-IV diagnoses of axis I and II disorders were made on clinical basis with the aid of SCID-I and II-interviews. The detected prevalence rates and their 95% confidence intervals based on binomial distribution were compared to three previous large population-based studies. Results Of the 63 patients, 26 (41.3%) had had an axis I disorder that preceded the onset of orofacial pain, and 33 (52.4%) had had a lifetime axis I disorder. Rate of current axis I disorders was 36.5%, indicating that only about 16% of lifetime disorders had remitted, and they tended to run chronic course. The most common lifetime axis I disorders were major depression (30.2%), social phobia (15.9%), specific phobia (11.1%), and panic disorder (7.9%). Twelve patients (19.0%) had at least one cluster C personality disorder already before the emergence of orofacial pain. Patients with cluster C personality disorders are characterized as fearful and neurotic. None of the patients had cluster A (characterized as odd and eccentric) or B (characterized as dramatic, emotional or erratic) personality disorders. The most common personality disorders were obsessive-compulsive personality (14.3%), dependent personality (4.8%), and avoidant personality (3.2%). The majority of the patients (54%) had also one or more chronic pain conditions other than orofacial pain. In almost all patients (94%) they were already present at the onset of orofacial pain. Conclusions Our results suggest that major depression, persistent social phobia, and neurotic, fearful, and obsessive-compulsive personality characteristics are common in patients with chronic idiopathic orofacial pain. Most psychiatric disorders precede the onset of orofacial pain and they tend to run a chronic course. Implications We propose that the high psychiatric morbidity, and comorbidity to other chronic pain conditions, in chronic idiopathic orofacial pain can be best understood in terms of shared vulnerability to both chronic pain and specific psychiatric disorders, most likely mediated by dysfunctional brain dopamine activity.