Dalifer Freites-Núñez

Introduction/objectives Treat-to-target (T2T) strategies in systemic lupus erythematosus (SLE) emphasise remission or low disease activity together with glucocorticoid minimisation. Real-world evidence evaluating anifrolumab using predefined target-oriented outcomes remains limited. We evaluated outcomes after anifrolumab initiation within a pragmatic T2T framework. Methods This retrospective single-centre cohort included adults with SLE receiving anifrolumab. The study-defined pragmatic low disease activity composite required SLEDAI-2K ≤ 4, Physician Global Assessment ≤ 1, prednisone ≤ 7.5 mg/day, and absence of clinically significant renal activity. Prednisone ≤ 5 mg/day was explored in sensitivity analyses. Outcomes were assessed at baseline and 1, 3, 6, and 12 months. Longitudinal changes in SLEDAI-2K, PGA, and prednisone dose were analysed using linear mixed-effects models. Results Nineteen patients were included (median baseline SLEDAI-2K 12.0 and median PGA 2.0). Composite attainment using prednisone ≤ 7.5 mg/day was 5.3% at 1 month, 47.4% at 3 months, 50.0% at 6 months, and 8/12 (66.7%) at 12 months. Using prednisone ≤ 5 mg/day, attainment was 36.8% at 3 months, 50.0% at 6 months, and 66.7% at 12 months. Disease activity components were met earlier than glucocorticoid thresholds. Mixed-effects models showed significant reductions in SLEDAI-2K, PGA, and prednisone dose during follow-up. One patient permanently discontinued anifrolumab due to influenza A, and one temporarily interrupted treatment because of pneumonia. Conclusions After anifrolumab initiation, the proportion of evaluable patients meeting the pragmatic composite increased alongside reductions in disease activity and glucocorticoid dose. These exploratory findings suggest a potential role for anifrolumab within T2T strategies and warrant confirmation in larger controlled studies using validated treatment targets. Key Points • The proportion of evaluable patients meeting a pragmatic low disease activity composite increased during the 12 months after anifrolumab initiation. • Disease activity criteria were met earlier than glucocorticoid dose thresholds, while later composite attainment coincided with progressive glucocorticoid tapering. • These exploratory findings suggest a potential role for anifrolumab within T2T strategies in SLE and require confirmation in larger controlled studies.

Objectives: To systematically identify and characterize original rheumatology research using large language models (LLMs), quantify reliance on closed versus open models, and assess reporting practices relevant to reproducibility (model versions, prompts, inference settings, and availability of code/data), in order to propose practical recommendations to strengthen transparency.Methods: We searched PubMed/MEDLINE for English-language peer-reviewed original research published from 1 November 2022 to 23 January 2026. Two reviewers independently screened title/abstract and full texts, with a third resolving disagreements. Data were extracted using an LLM-assisted workflow and then independently verified against source articles by the authors. Extracted items included study characteristics, model families and openness, access mode, versioning/timing, and transparency indicators (prompts, code, data).Results: Of 185 screened records, 63 studies were included. Most were research (n = 33/63; 52.38%) or education-focused (n = 24/63; 38.10%). Studies predominantly used closed-source LLMs (n = 50/63; 79.37%), with limited exclusive use of open-weight models (n = 4/63; 6.35%) and some hybrid use (n = 9/63; 14.29%). OpenAI models were most common (n = 55/63; 87.30%). Reporting was heterogeneous: interaction language was often not reported (n = 42/63; 66.67%); access mode was reported in (n = 19/63; 30.16%); output generation date in (n = 33/63; 52.38%). Prompts were shared in (n = 48/63; 76.19%), but code was publicly available in (n = 4/63; 6.35%), and data in (n = 13/63; 20.63%).Conclusions: LLM-based rheumatology research largely depends on closed models with inconsistent reporting of reproducibility-critical details and minimal code sharing. A concise, rheumatology-oriented checklist is proposed to standardize reporting and improve auditability, replication, and cross-study comparability.

Objectives: The aim of this study was to identify robust predictive markers which may help personalize tapering protocols, minimizing flare risk while optimizing long-term disease management in RA patients. Methods: The OPTIBIO trial (EudraCT 2012–004482-40) was a phase IV, randomized, open-label, non-inferiority study conducted in five hospitals in Spain. RA patients in sustained remission on stable bDMARD therapy were randomized 1:1 to standard care or dose reduction. The primary outcome was to compare the proportion of joint flare between baseline and 12 months by a non-inferiority analysis analysed by the intention-to-treat principle and to identify predictors for flare and sustained remission. Results: A total of 195 patients were randomized: 99 to the control group and 96 to the optimization group. Thirty-nine flares occurred (optimization: 22.7%, control: 17.2%), with a risk difference of −5.5% (95% CI: −16.8% to 5.7%; P = 0.33). Two predictive models were developed: one for flares (AUC: 0.84) including 3v-DAS28-CRP, VAS pain, erosions, systolic blood pressure and haemoglobin, and another for sustained remission (AUC: 0.77) including 3v-DAS28-CRP, age and rheumatoid factor. Adding molecular biomarkers improved AUCs to 0.91 and 0.88, respectively. No significant differences in adverse events were observed. Conclusion: bDMARD dose optimization was not non-inferior to standard therapy on the flare rate but demonstrated similar safety. Predictive models for remission and flares were developed, which may help select patients to ensure safe implementation of this strategy, highlighting the need for personalized treatment. Trial registration: www.clinicaltrialsregister.eu; EudraCT 2012–004482-40

Objective: To evaluate the real-world effectiveness, treatment retention, and safety of secukinumab (SEC) uptitration to 300 mg in patients with axial spondyloarthritis (axSpA), including radiographic (r-axSpA) and nonradiographic (nr-axSpA) subtypes, with active disease despite receiving SEC 150 mg every 4 weeks. Methods: This multicenter, retrospective study included patients with axSpA who had received SEC 150 mg for ≥ 3 months and had active disease (Axial Spondyloarthritis Disease Activity Score [ASDAS] ≥ 2.1) at the time of dose escalation. Patients were followed for up to 24 months after escalation. Effectiveness was assessed through changes in ASDAS over time. Treatment retention was analyzed using Kaplan-Meier curves, with factors associated with discontinuation explored by Cox regression. Safety outcomes were reported as incidence of adverse events (AEs). Results: Among 106 patients (77 r-axSpA, 29 nr-axSpA), ASDAS significantly declined within 6 months and was sustained through 24 months. Median ASDAS decreased from 4.1 to 2.0 in patients with r-axSpA and from 4.1 to 2.0 in those with nr-axSpA. Overall, 85.9% of patients achieved ASDAS ≤ 2.1 at least once after escalation. Retention of SEC 300 mg was 87.8% at 6 months and 59.1% at 24 months, with no significant difference between subtypes (log-rank P = 0.48). HLA-B27 positivity showed a nonsignificant trend toward higher discontinuation risk. AE incidence was 15.79 per 100 patient-years, and AEs were mostly mild infections and cutaneous reactions. No new safety signals were identified. Conclusion: SEC uptitration to 300 mg was associated with sustained improvements in disease activity and favorable retention and safety in axSpA patients with active disease after receiving SEC 150 mg, supporting dose escalation as a clinically effective and safe strategy.

Objectives Optimization of biological DMARDs (bDMARDs) in patients with rheumatoid arthritis (RA) may be feasible in those who have maintained remission for at least six months. However, this approach carries the risk of disease flare, underscoring the need for reliable predictors to guide clinical decisions. Methods The OPTIBIO trial (EudraCT 2012-004482-40) was a phase IV, randomized, open-label, non-inferiority study conducted in five hospitals in Spain. RA patients in sustained remission on stable bDMARD therapy were randomized 1:1 to standard care or dose reduction. The primary outcome was to compare to proportion of joint flare at 12 months by a non-inferiority analysis analyzed by the intention-to-treat principle and identify predictors for flare and sustained remission. Results 195 patients were randomized: 99 to the control group and 96 to the optimization group. Thirty-nine flares occurred (optimization: 22.7%, control: 17.2%), with a risk difference of -5.5% (95% CI: -16.8% to 5.7%; P = 0.33). Two predictive models were developed: one for flares (AUC: 0.84) including 3v-DAS28-CRP, VAS pain, erosions, systolic blood pressure, and hemoglobin, and another for sustained remission (AUC: 0.77) including 3v-DAS28-CRP, age, and rheumatoid factor. Adding molecular biomarkers improved AUCs to 0.91 and 0.88, respectively. No significant differences in adverse events were observed. Conclusion bDMARD dose optimization was non-inferior to standard therapy on the flare rate but demonstrated similar safety. Predictive models for remission and flares were developed, which may help select patients to ensure safe implementation of this strategy, highlighting the need for personalized treatment. Key messages bDMARD dose optimization was non-inferior to standard therapy on the flare Clinical models classified patients in terms of disease severity Incorporation of molecular markers to clinical models improved their prediction power

Objectives: Timely access to current rheumatology guidelines at the point of care is challenging. We aimed to develop and evaluate the first Retrieval-Augmented Generation (RAG) system specifically designed for adult rheumatology, integrating European Alliance of Associations for Rheumatology (EULAR) and American College of Rheumatology (ACR) guidelines to provide rheumatologists with timely, evidence-based recommendations at the point of care. Methods: EULAR and ACR management guidelines were selected by rheumatologists based on their clinical relevance for decision making and processed. A RAG system was implemented using LangChain framework, voyage-3 embedding model, and a Qdrant vector database. To evaluate the system, ten questions per guideline were generated using ChatGPT 4.5 . Answers to these guideline-specific questions were subsequently produced by ChatGPT-o3-mini with context retrieval (RAG) and without (baseline). Performance was assessed by an LLM-as-a-judge ( Gemini 2.0 Flash ) using a 5-point Likert scale across five dimensions: relevance, factual accuracy, safety, completeness, and conciseness. The judge also determined preference between the RAG and baseline responses. Statistical significance was established using Wilcoxon signed-rank and Binomial tests. For further validation, two blinded rheumatologists independently evaluated a random sample of questions (15%). Results: After agreement, 74 guidelines were included, and 740 evaluation questions were generated. Analysis revealed that the RAG system significantly outperformed the baseline system across all criteria (p<0.001) in the LLM-as-a-judge evaluation. Manual evaluation by rheumatologists confirmed these findings (p<0.001 for accuracy, safety, completeness). Furthermore, the RAG system was significantly preferred by the LLM-as-a-judge in 92.8% of comparisons (p<0.001) and by the human evaluators in 71.2%-74.8% of comparisons (p<0.001). Conclusion: This study demonstrates the successful development and evaluation of a RAG system integrating extensive EULAR/ACR guidelines for adult rheumatology. The system significantly improves answer quality compared to a baseline LLM. This provides a robust foundation for reliable, AI-driven clinical decision support tools designed to enhance guideline adherence and evidence-based practice in rheumatology by providing clinicians with rapid, context-aware access to recommendations. Graphical Key messages: Large language models, combined with EULAR and ACR guidelines, may enhance rheumatology clinical decision support. Retrieval augmented generation (RAG) responses showed significantly greater accuracy, safety and completeness than baseline LLMs. RAG is a promising architecture for reducing hallucinations and providing grounded, reliable answers.

Background: Rheumatology has experience notably changes in last decades. New drugs, including biologic agents and janus kinase inhibitors, have bloosom. Concepts such as window of opportunity , arthralgia suspicious for progression , or difficult-to-treat rheumatoid arthritis have appeared; and new management approaches and strategies such as treat-to-target have become popular. Statistical learning methods, gene therapy, telemedicine or precision medicine are other advancements that have gained relevance in the field. To better characterise the research landscape and advances in rheumatology, automatic and efficient approaches based on natural language processing should be used. The objective of this study is to use topic modeling techniques to uncover key topics and trends in the rheumatology research conducted in the last 23 years. Methods: This study analysed 96,004 abstracts published between 2000 and December 31, 2023, drawn from 34 specialised rheumatology journals obtained from PubMed. BERTopic, a novel topic modeling approach that considers semantic relationships among words and their context, was used to uncover topics. Up to 30 different models were trained. Based on the number of topics, outliers and topic coherence score, two of them were finally selected, and the topics manually labeled by two rheumatologists. Word clouds and hierarchical clustering visualizations were computed. Finally, hot and cold trends were identified using linear regression models. Results: Abstracts were classified into 45 and 47 topics. The most frequent topics were rheumatoid arthritis, systemic lupus erythematosus and osteoarthritis. Expected topics such as COVID-19 or JAK inhibitors were identified after conducting the dynamic topic modeling. Topics such as spinal surgery or bone fractures have gained relevance in last years, however, antiphospholipid syndrome, or septic arthritis have lost momentum. Conclusions: Our study utilized advanced natural language processing techniques to analyse the rheumatology research landscape, and identify key themes and emerging trends. The results highlight the dynamic and varied nature of rheumatology research, illustrating how interest in certain topics have shifted over time.