Prediction of ineffectiveness of biological drugs using machine learning and explainable AI methods: data from the Austrian Biological Registry BioReg.

OBJECTIVES: Machine learning models can support an individualized approach in the choice of bDMARDs. We developed prediction models for 5 different bDMARDs using machine learning methods based on patient data derived from the Austrian Biologics Registry (BioReg). METHODS: Data from 1397 patients and 19 variables with at least 100 treat-to-target (t2t) courses per drug were derived from the BioReg biologics registry. Different machine learning algorithms were trained to predict the risk of ineffectiveness for each bDMARD within the first 26 weeks. Cross-validation and hyperparameter optimization were applied to generate the best models. Model quality was assessed by area under the receiver operating characteristic (AUROC). Using explainable AI (XAI), risk-reducing and risk-increasing factors were extracted. RESULTS: The best models per drug achieved an AUROC score of the following: abatacept, 0.66 (95% CI, 0.54-0.78); adalimumab, 0.70 (95% CI, 0.68-0.74); certolizumab, 0.84 (95% CI, 0.79-0.89); etanercept, 0.68 (95% CI, 0.55-0.87); tocilizumab, 0.72 (95% CI, 0.69-0.77). The most risk-increasing variables were visual analytic scores (VAS) for abatacept and etanercept and co-therapy with glucocorticoids for adalimumab. Dosage was the most important variable for certolizumab and associated with a lower risk of non-response. Some variables, such as gender and rheumatoid factor (RF), showed opposite impacts depending on the bDMARD. CONCLUSION: Ineffectiveness of biological drugs could be predicted with promising accuracy. Interestingly, individual parameters were found to be associated with drug responses in different directions, indicating highly complex interactions. Machine learning can be of help in the decision-process by disentangling these relations.

Evaluation of discontinuation for adverse events of JAK inhibitors and bDMARDs in an international collaboration of rheumatoid arthritis registers (the 'JAK-pot' study).

BACKGROUND: In a clinical trial setting, patients with rheumatoid arthritis (RA) taking the Janus kinase inhibitor (JAKi) tofacitinib demonstrated higher adverse events rates compared with those taking the tumour necrosis factor inhibitors (TNFi) adalimumab or etanercept. OBJECTIVE: Compare treatment discontinuations for adverse events (AEs) among second-line therapies in an international real-world RA population. METHODS: Patients initiating JAKi, TNFi or a biological with another mode of action (OMA) from 17 registers participating in the 'JAK-pot' collaboration were included. The primary outcome was the rate of treatment discontinuation due to AEs. We used unadjusted and adjusted cause-specific Cox proportional hazard models to compare treatment discontinuations for AEs among treatment groups by class, but also evaluating separately the specific type of JAKi. RESULTS: Of the 46 913 treatment courses included, 12 523 were JAKi (43% baricitinib, 40% tofacitinib, 15% upadacitinib, 2% filgotinib), 23 391 TNFi and 10 999 OMA. The adjusted cause-specific hazard rate of treatment discontinuation for AEs was similar for TNFi versus JAKi (1.00, 95% CI 0.92 to 1.10) and higher for OMA versus JAKi (1.11, 95% CI 1.01 to 1.23), lower with TNFi compared with tofacitinib (0.81, 95% CI 0.71 to 0.90), but higher for TNFi versus baricitinib (1.15, 95% CI 1.01 to 1.30) and lower for TNFi versus JAKi in patients 65 or older with at least one cardiovascular risk factor (0.79, 95% CI 0.65 to 0.97). CONCLUSION: While JAKi overall were not associated with more treatment discontinuations for AEs, subgroup analyses suggest varying patterns with specific JAKi, such as tofacitinib, compared with TNFi. However, these observations should be interpreted cautiously, given the observational study design.

After JAK inhibitor failure: to cycle or to switch, that is the question - data from the JAK-pot collaboration of registries.

OBJECTIVES: The expanded therapeutic arsenal in rheumatoid arthritis (RA) raises new clinical questions. The objective of this study is to compare the effectiveness of cycling Janus kinase inhibitors (JAKi) with switching to biologic disease-modifying antirheumatic drug (bDMARD) in patients with RA after failure to the first JAKi. METHODS: This is a nested cohort study within data pooled from an international collaboration of 17 national registries (JAK-pot collaboration). Data from patients with RA with JAKi treatment failure and who were subsequently treated with either a second JAKi or with a bDMARD were prospectively collected. Differences in drug retention rates after second treatment initiation were assessed by log-rank test and Cox regression analysis adjusting for potential confounders. Change in Clinical Disease Activity Index (CDAI) over time was estimated using a linear regression model, adjusting for confounders. RESULTS: 365 cycling and 1635 switching patients were studied. Cyclers were older and received a higher number of previous bDMARDs. Both strategies showed similar observed retention rates after 2 years of follow-up. However, adjusted analysis revealed that cycling was associated with higher retention (p=0.04). Among cyclers, when the first JAKi was discontinued due to an adverse event (AE), it was more likely that the second JAKi would also be stopped due to an AE. Improvement in CDAI over time was similar in both strategies. CONCLUSIONS: After failing the first JAKi, cycling JAKi and switching to a bDMARD appear to have similar effectiveness. Caution is advised if an AE was the reason to stop the first JAKi.

Effectiveness of TNF-inhibitors, abatacept, IL6-inhibitors and JAK-inhibitors in 31 846 patients with rheumatoid arthritis in 19 registers from the 'JAK-pot' collaboration.

BACKGROUND: JAK-inhibitors (JAKi), recently approved in rheumatoid arthritis (RA), have changed the landscape of treatment choices. We aimed to compare the effectiveness of four current second-line therapies of RA with different modes of action, since JAKi approval, in an international collaboration of 19 registers. METHODS: In this observational cohort study, patients initiating tumour necrosis factor inhibitors (TNFi), interleukin-6 inhibitors (IL-6i), abatacept (ABA) or JAKi were included. We compared the effectiveness of these treatments in terms of drug discontinuation and Clinical Disease Activity Index (CDAI) response rates at 1 year. Analyses were adjusted for patient, disease and treatment characteristics, including lines of therapy and accounted for competing risk. RESULTS: We included 31 846 treatment courses: 17 522 TNFi, 2775 ABA, 3863 IL-6i and 7686 JAKi. Adjusted analyses of overall discontinuation were similar across all treatments. The main single reason of stopping treatment was ineffectiveness. Compared with TNFi, JAKi were less often discontinued for ineffectiveness (adjusted HR (aHR) 0.75, 95% CI 0.67 to 0.83), as was IL-6i (aHR 0.76, 95% CI 0.67 to 0.85) and more often for adverse events (aHR 1.16, 95% CI 1.03 to 1.33). Adjusted CDAI response rates at 1 year were similar between TNFi, JAKi and IL-6i and slightly lower for ABA. CONCLUSION: The adjusted overall drug discontinuation and 1 year response rates of JAKi and IL-6i were similar to those observed with TNFi. Compared with TNFi, JAKi were more often discontinued for adverse events and less for ineffectiveness, as were IL-6i.

[Pain and pain management in old age: special features and recommendations]. / Schmerzen und Schmerzerfassung im Alter: Besonderheiten und Empfehlungen.

INTRODUCTION: Pain is highly prevalent in older persons and has a variety of causes. In geriatric patients, especially in patients with dementia, pain is often not sufficiently recognized and therefore frequently remains untreated. For the affected patient group this can have far-reaching consequences for their functional and cognitive abilities and may consequently lead to loss of autonomy. OBJECTIVE: Existing deficits of pain assessment for geriatric patients are described, with a primary focus on those patients suffering from cognitive impairments and pain. In addition, the influence of multimorbidity on pain management in old age is considered in detail. METHODS: The diagnostics and measurement of pain in older individuals are described based on recent literature and corresponding instruments used in the assessment of pain are outloned. The authors pay special attention to the possibilities of pain measurement in patients with higher grade cognitive impairments and non-communicative patients. CONCLUSION: A standardized pain assessment should be an integral component in the care and treatment of geriatric patients and individuals suffering from dementia. Validated instruments for pain measurement exist for both groups and should be integrated into daily clinical practice.

[Pain treatment in old age: special features and recommendations]. / Schmerztherapie im Alter: Besonderheiten und Empfehlungen.

BACKGROUND: For geriatric patients with chronic pain, a comprehensive well-coordinated pain management is pivotal to ensure the best possible pain relief and to minimize as far as possible preventable negative side effects of treatment. OBJECTIVE: Description of the difficulties in pain management of geriatric patients with respect to general basic rules that are worth paying attention to and presentation of pharmacological and non-pharmacological treatment options. METHODS: This article describes the special features of pain management in older patients and gives recommendations on the use of analgesics and potential drug interactions in geriatric patients with organ dysfunction. Furthermore, individual substance groups are described with respect to their use in geriatric patients based on the recent literature. CONCLUSION: The aim of an individualized pain treatment in older and multimorbid patients is the relief of pain to an appropriate level, preservation of mobility, self-reliance and autonomy of each individual. The ability to participate in social activities as well as improvement in the quality of life need to be the focus of pharmacological and non-pharmacological treatment.

Guidelines for the conduct of pharmacological clinical trials in hand osteoarthritis: Consensus of a Working Group of the European Society on Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO).

OBJECTIVES: To gather expert opinion on the conduct of clinical trials that will facilitate regulatory review and approval of appropriate efficacious pharmacological treatments for hand osteoarthritis (OA), an area of high unmet clinical need. METHODS: The European Society on Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal diseases (ESCEO) organized a working group under the auspices of the International Osteoporosis Foundation (IOF) and the World Health Organization (WHO). RESULTS: This consensus guideline is intended to provide a reference tool for practice, and should allow for better standardization of the conduct of clinical trials in hand OA. Hand OA is a heterogeneous disease affecting different, and often multiple, joints of the thumb and fingers. It was recognized that the various phenotypes and limitations of diagnostic criteria may make the results of hand OA trials difficult to interpret. Nonetheless, practical recommendations for the conduct of clinical trials of both symptom and structure modifying drugs are outlined in this consensus statement, including guidance on study design, execution, and analysis. CONCLUSIONS: While the working group acknowledges that the methodology for performing clinical trials in hand OA will evolve as knowledge of the disease increases, it is hoped that this guidance will support the development of new pharmacological treatments targeting hand OA.

The Short Form Score for the Assessment and Quantification of Chronic Rheumatic Affections of the Hands in Daily Clinical Routines-Its Sensitivity to Change and Preliminary Patient Relevant Variation Values: A Pilot Study.

OBJECTIVE: The SF-SACRAH was developed to assess the involvement of the hand in rheumatoid arthritis (RA) and hand osteoarthritis (HOA) patients in daily clinical routines. In this pilot study, its sensitivity to change will be assessed longitudinally, and preliminary thresholds for patient relevant changes are derived. METHODS: Ninety-nine outpatients suffering from HOA (n = 55) or RA (n = 44) completed the SF-SACRAH once initially. After approximately 3 months, patients repeated the SF-SACRAH. At both visits, patients rated their satisfaction (PATSAT) with the state of their disease (1 = very good to 5 = unsatisfactory). For assessing its sensitivity to change, SF-SACRAH changes in patients with stable, improving, or worsening conditions according to PATSAT were calculated in HOA and RA patients. The respective medians and highest values were used to estimate patient relevant variation values. SF-SACRAH changes and positive or negative PATSAT changes in HOA as well as RA patients were analyzed by applying the Kruskal-Wallis test. In RA patients, the DAS28 was also calculated. Spearman's rho was calculated to correlate SF-SACRAH changes with the EULAR response criteria. RESULTS: In HOA and RA patients, a statistically high correlation between PATSAT changes and SF-SACRAH values was revealed (p < 0.0001 in HOA and p < 0.01 in RA patients, respectively). The median changes in SF-SACRAH in patients with improving, stable, or worsening conditions according to PATSAT were HOA patients: PATSAT improving: ΔSF-SACRAH -1.6, PATSAT stable: ΔSF-SACRAH +0.8, PATSAT worsening: ΔSF-SACRAH +1.0; RA patients: PATSAT improving: ΔSF-SACRAH -0.9, PATSAT stable: ΔSF-SACRAH +0.2, PATSAT worsening: ΔSF-SACRAH +0.8. In RA patients, there is a moderate, but significant, correlation between DAS28 EULAR response criteria and SF-SACRAH changes (ΔDAS28 improving >0.6: ΔSF-SACRAH -0.4, ΔDAS28 <0.6: ΔSF-SACRAH +0.0, ΔDAS28 worsening >0.6: ΔSF-SACRAH +0.5; r = 0.433, p < 0.01). CONCLUSION: The SF-SACRAH constitutes a reliable tool for the assessment of hand impairment in patients with chronic rheumatic diseases. It proved to be sensitive to change in this short-term evaluation in both HOA and RA patients. Additionally, preliminary patient variation values for improvement (-1.60) and deterioration (+1.0) could be derived.

Improving inflammatory arthritis management through tighter monitoring of patients and the use of innovative electronic tools.

Treating to target by monitoring disease activity and adjusting therapy to attain remission or low disease activity has been shown to lead to improved outcomes in chronic rheumatic diseases such as rheumatoid arthritis and spondyloarthritis. Patient-reported outcomes, used in conjunction with clinical measures, add an important perspective of disease activity as perceived by the patient. Several validated PROs are available for inflammatory arthritis, and advances in electronic patient monitoring tools are helping patients with chronic diseases to self-monitor and assess their symptoms and health. Frequent patient monitoring could potentially lead to the early identification of disease flares or adverse events, early intervention for patients who may require treatment adaptation, and possibly reduced appointment frequency for those with stable disease. A literature search was conducted to evaluate the potential role of patient self-monitoring and innovative monitoring of tools in optimising disease control in inflammatory arthritis. Experience from the treatment of congestive heart failure, diabetes and hypertension shows improved outcomes with remote electronic self-monitoring by patients. In inflammatory arthritis, electronic self-monitoring has been shown to be feasible in patients despite manual disability and to be acceptable to older patients. Patients' self-assessment of disease activity using such methods correlates well with disease activity assessed by rheumatologists. This review also describes several remote monitoring tools that are being developed and used in inflammatory arthritis, offering the potential to improve disease management and reduce pressure on specialists.