Tel Aviv Sourasky Medical Center Ramat Gan, Israel
Dafne Capelusnik1, Diego Benavent2, Désirée van der Heijde3, Robert Landewé4, Denis Poddubnyy5, Astrid van Tubergen6, Louise Falzon7, Victoria Navarro-Compán8 and Sofia Ramiro9, 1Instituto de Rehabilitación Psicofísica (IREP), Buenos Aires, Argentina, 2La Paz University Hospital, Madrid, Spain, 3Department of Rheumatology, Leiden University Medical Center, Leiden, The Netherlands, Leiden, Netherlands, 4Amsterdam University Medical Centers, Amsterdam, Netherlands, 5Department of Gastroenterology, Infectious Diseases and Rheumatology, Charité – Universitätsmedizin Berlin, Berlin, Germany, 6Maastricht UMC+, Maastricht, Netherlands, 7Health Economics and Decision Science, School of Health and Related Research, University of Sheffield, Sheffield, United Kingdom, 8Department of Rheumatology, La Paz University Hospital, IdiPaz, Madrid, Spain, 9Leiden University Medical Center, Leiden, Netherlands
Background/Purpose: The ASAS-SPEAR (SPondyloarthritis EARly definition) project aims to develop a consensual definition of early SpA. Therefore, it is important to know whether treatment earlier in the disease course compared to treatment of established disease leads to better outcomes in axSpA.
We aim to summarize the evidence on the relationship between symptom duration or the presence of radiographic damage and clinical response in patients with axSpA treated with NSAIDs or b/tsDMARDs.
Methods: A SLR was conducted using Medline, EMBASE and the Cochrane Library, supplemented by hand-searches in the FDA and EMA website. Randomized controlled trials (RCTs) and cohort studies in patients with axSpA addressing the impact of symptom duration or disease duration and presence of radiographic damage on treatment response (to NSAIDs, b/tsDMARDs) were included. Based on a cut-off of symptom/disease duration or the absence/presence of radiographic damage, groups of 'early' and 'established' disease were compared. Treatment outcomes were measures of disease activity, function or quality of life.
Two reviewers independently identified eligible studies and extracted the data, including the risk of bias (RoB) assessment. For categorical outcomes we calculated relative risk (RR), relative risk ratio (RRR) and number needed to treat (NNT), and differences in differences (DID) for continuous outcomes.
Results: From the 8769 articles retrieved, 25 were included and 2 added by hand-search, most of them with low RoB. Twenty studies (11 RCTs) compared groups based on symptom (n=4)/disease duration (n=7) and 7 studies (4 RCTs) based comparisons on absence/presence of radiographic damage in posthoc analyses.
When early axSpA was defined by symptom duration in RCTs (n=4), in patients with nr-axSpA, early treatment was associated with higher RR and RRR and lower NNT for ASAS40 in two studies (table 1); a third study showed that patients achieving ASDAS-ID and ASAS-PR had shorter symptom duration than those not achieving this. Lastly, in one study including patients with axSpA patients, no difference in treatment response was observed based on symptom duration. In most of the cohort studies using a definition based on symptom/disease duration (n=9), no association was found between symptom/disease duration and treatment response (n=7). Only in one cohort study,disease duration was a significant predictor of quality of life, and in another cohort study, it was a predictor of functional improvement.
When early axSpA was defined based on disease duration or the presence of radiographic damage, there was no significant difference between early and established axSpA.
Conclusion: Studies addressing treatment response based on symptom duration or radiographic damage in axSpA are scarce.
When defining early axSpA based on symptom duration, in nr-axSpA treatment with bDMARDs may lead to better outcomes compared to established axSpA whereas in axSpA there is no difference in response to treatment between early and established disease.
When early axSpA is defined based on disease duration or radiographic damage, no differences in response to treatment are found between early and established disease. Table 1: Assessment of treatment response in RCTs based on symptom duration Disclosures: D. Capelusnik, Bristol Myers Squibb, Pfizer; D. Benavent, Jannsen, Novartis; D. van der Heijde, AbbVie, Bayer, BMS, Cyxone, Eisai, Galapagos, Gilead, Glaxo-Smith-Kline, Janssen, Novartis, Pfizer, UCB, Imaging Rheumatology bv, Lilly; R. Landewé, AbbVie/Abbott, Bristol-Myers Squibb(BMS), Galapagos, Gilead, GlaxoSmithKlein(GSK), Janssen, Eli Lilly, Novartis, Pfizer, UCB; D. Poddubnyy, AbbVie, Biocad, Bristol-Myers Squibb, Eli Lilly, Gilead, GlaxoSmithKline, MSD, Moonlake, Novartis, Pfizer, Samsung-Bioepis, UCB; A. van Tubergen, Novartis, Galapagos, Novartis, Pfizer, UCB; L. Falzon, None; V. Navarro-Compán, AbbVie, Eli Lilly, Janssen, Merck/MSD, Novartis, Pfizer, UCB Pharma; S. Ramiro, AbbVie/Abbott, Eli Lilly, Galapagos, Merck/MSD, Novartis, Pfizer, UCB, Sanofi.
.jpg)
Table 1: Assessment of treatment response in RCTs based on symptom duration