Background Little is known about how patient factors may impact the efficacy of biologics used for psoriatic arthritis (PsA).
Objectives The objective of this study was to determine which clinical factors modify treatment effects of biologics used for PsA.
Methods A systematic literature review was performed in Medline, EMBASE, Cochrane Central and clinicaltrials.gov. Randomized controlled trials (RCTs) assessing the efficacy of biologics in adults with PsA were eligible. The efficacy outcomes of interest included 20% and 50% improvements in the American College of Rheumatology response criteria (ACR20 and ACR50, respectively) and a 75% reduction in the Psoriasis Area and Severity Score (PASI75). Clinical factors such as age, gender, ethnicity, PsA duration, psoriasis (PsO) duration, swollen and tender joint counts, steroid use, methotrexate use, prior anti-TNF use, Disease Activity Score (DAS) and study year were evaluated for effect on biologic treatment response.
Clinical factors were considered potential effect modifiers (PEM) if they graphically demonstrated associations via forest and l'Abbé plots. PEMs were further investigated through meta-regression analyses where covariate effects were assumed to be similar for all treatments.
Results Twelve placebo-controlled RCTs were identified as eligible for analyses. Treatments assessed in these trials included adalimumab, etanercept, infliximab, ustekinumab, abatacept, certolizumab pegol, golimumab and secukinumab.
For ACR20 at 12 and 24 weeks, age, PsO duration, PsA duration, prior anti-TNF use and study year were identified as PEM. In meta-regression, trials with a higher proportion of patients with shorter PsO duration, shorter PsA duration, higher percent of prior anti-TNF use and more recent study year tended to have more patients achieving ACR 20 at 12 weeks. Trials which had a higher proportion of patients with shorter PsA duration and more recent study year tended to have fewer patients achieving ACR20 at 24 weeks.
Gender, PsA duration, swollen joint count, prior anti-TNF use and study year were identified as PEM for ACR50 at 12 weeks. In a meta-regression, trials with less males, higher percent of prior anti-TNF use and more recent study year tended to have fewer patients achieving ACR50 at 12 weeks. At 24 weeks, gender, PsO duration, PsA duration, swollen joint count, DAS and study year were identified as PEM for ACR50. Meta regression showed trials with a higher proportion of patients with lower PsA duration and more recent study year tended to have fewer patients achieving ACR50 at 24 weeks.
None of the PEM identified for PASI75 at 12 weeks were significant in meta-regression. Among the factors identified as PEM for PASI75 at 24 weeks (i.e. gender, swollen joint count, tender joint count, and prior anti-TNF use), trials with higher proportions of patients with lower swollen joint count, lower tender joint count and higher percent of prior anti-TNF use tended to have fewer patients reaching this outcome in the meta-regression analysis.
Conclusions These analyses identified key clinical factors that modify treatment effects of biologics for PsA. Adjusting for these clinical factors in future comparative effectiveness research or health economic analyses may be important to ensure the best evidence is used for coverage and biologic treatment decision-making.
Disclosure of Interest J. Palmer Employee of: Novartis Pharmaceuticals Corporation, S. Kanter Consultant for: Novartis Pharmaceuticals Corporation, E. Druyts Consultant for: Novartis Pharmaceuticals Corporation, Y. Tsang Employee of: Novartis Pharmaceuticals Corporation, V. Herrera Employee of: Novartis Pharmaceuticals Corporation
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.