Session Information
Session Type: Poster Session (Tuesday)
Session Time: 9:00AM-11:00AM
Background/Purpose: Tumor necrosis factor inhibitors (TNFi) have contributed to improved prognosis in patients with psoriatic arthritis (PsA). However, many patients treated with TNFi fail to achieve a treatment target of remission. Hence, the aims of this study were to identify predictors of 6-month Disease Activity in Psoriatic Arthritis in 28 joints (DAPSA28)(1) remission (DAPSA28≤4) in bio-naive PsA patients starting TNFi in clinical practice and investigate the performance of the derived prediction model.
Methods: Pooled data from PsA patients in 13 European registries participating in the EuroSpA Research Collaboration were analyzed(2). Patients with a 6-month follow-up visit (time window 3-9 months) with data allowing for calculation of DAPSA28, were included. The study cohort was divided into a derivation and validation cohort (50% of patients from each registry in each sub-cohort). Logistic regression analyses were applied to identify conventional clinical variables (marked with bold in Table 1) associated with DAPSA28≤4 at 6 months in the derivation cohort. Missing covariate data were imputed with Multiple Imputation with Chained Equations. Variables with a p-value < 0.25 in univariate analyses were included in the initial multivariable model. A priori it was decided to adjust for age, gender and country. Purposeful selection guided removal of variables from the multivariable model. Model fit was tested in the validation cohort by area under the Receiver Operating Curve (ROC) and misclassification error.
Results: Of the 16,230 PsA patients in the EuroSpA database 7,975 patients initiating 1st TNFi had a registered follow-up visit with registered variables for DAPSA28 calculation and were included in the study. The study cohort had slightly higher baseline disease activity than the patients without DAPSA28 assessment at follow-up (Table 1). At 6 months, 1,956 (24.5%) patients were in DAPSA28 remission. Based on univariate analyses, all tested variables except concomitant csDMARD and time since diagnosis were included in the initial multivariate model. The final multivariate model identified that current or past smoking, treatment start prior to 2009, normal CRP, high 28TJC, high global, fatigue and HAQ scores decreased the probability of DAPSA28 remission at 6 months (Table 2). The regression coefficients of the model were used to derive a prediction index for each patient in the validation cohort, determining their predicted probability of DAPSA28 remission at 6 months. The ability of the model to correctly predict DAPSA28 remission using different cut-offs for predicted probability are shown with the ROC (Figure 1). The fit was deemed reasonable (median area under the curve 0.75, misclassification error 0.22).
Conclusion: A prediction model based on conventional clinical variables correctly predicted DAPSA28 remission status at 6 months in 3 out of 4 patients in a validation cohort. Future studies should investigate the potential of improving the model by addition of imaging and soluble biomarkers.
Acknowledgements:
Novartis Pharma AG and IQVIA for supporting the EuroSpA collaboration.
References:
- Michelsen et al. Ann Rheum Dis 2018;77(12):1736-41
- Brahe et al. Arthritis Rheum 2018; 70 (suppl 10)
20190604_eurospa_abstract4.1_table1
Table 2_EuroSpA_study4.1_table2
20190603_EuroSpA_abstract4.1_figure1
To cite this abstract in AMA style:
Ørnbjerg L, Georgiadis S, Jacobsson L, Loft A, Iannone F, Moeller B, Sexton J, Mann H, Santos M, Pombo-Suarez M, Eklund K, Tomsic M, Gudbjornsson B, Erten �, Codreanu C, van der Horst-Bruinsma I, Wallman J, Sebastiani M, Nissen M, Kristianslund E, Pavelka K, Vieira-Sousa E, Sánchez-Piedra C, Trokovic N, Rotar Z, Love T, Yolbas s, IONESCU R, van de Sande M, Jones G, Michelsen B, Østergaard M, Lund Hetland M. Predictors of DAPSA28 Remission at 6 Months in Bio-Naive Patients with Psoriatic Arthritis Starting a TNF Inhibitor in Clinical Practice– Results from the EuroSpA Collaboration [abstract]. Arthritis Rheumatol. 2019; 71 (suppl 10). https://acrabstracts.org/abstract/predictors-of-dapsa28-remission-at-6-months-in-bio-naive-patients-with-psoriatic-arthritis-starting-a-tnf-inhibitor-in-clinical-practice-results-from-the-eurospa-collaboration/. Accessed .« Back to 2019 ACR/ARP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/predictors-of-dapsa28-remission-at-6-months-in-bio-naive-patients-with-psoriatic-arthritis-starting-a-tnf-inhibitor-in-clinical-practice-results-from-the-eurospa-collaboration/