Switching of TNFis Linked to Lower Drug Survival and Response Rates in Psoriatic Arthritis

psoriatic arthritisin the hand
psoriatic arthritisin the hand
Researchers assessed the long-term effectiveness of tumor necrosis factor inhibitors (TNFis) in patients with psoriatic arthritis who were exposed to at least 1 TNFi.

Switching to a second or third tumor necrosis factor inhibitor (TNFi) may be associated with significantly lower drug survival and response rates in patients with axial and peripheral psoriatic arthritis (PsA), according to research results published in Journal of Rheumatology.

Investigators abstracted data from Reuma.pt, an electronic records database of patients with PsA treated with disease-modifying antirheumatic drugs at rheumatology departments across mainland Portugal, Madeira, and the Azores Islands. All patients with PsA who had registered with the database between 2001 and 2017 and were exposed to at least 1 TNFi were eligible for inclusion in the study.

Researchers assessed patient demographic and clinical characteristics at baseline, including disease activity and function, according to the 4-variable, 28-joint count Disease Activity Score using erythrocyte sedimentation rate (DAS28-4vESR) and Health Assessment Questionnaire-Disability Index, respectively. They measured response to TNFi treatment at 3 months and 6 months by composite indices, including the European League Against Rheumatism (EULAR) and Disease Activity Index for Psoriatic Arthritis (DAPSA) criteria. Patients’ rheumatologists determined treatment nonresponse and logged it into Reuma.pt. Drug retention was defined as the time from TNFi initiation until discontinuation due to adverse events or ineffectiveness. Investigators used Kaplan-Meier curves to model drug retention over time and constructed Cox proportional hazard models to evaluate predictors of TNFi discontinuation.

Researchers collected data from 750 patients with PsA, of whom 377 (50.3%) were women. Mean age at first TNFi initiation was 47.6±11.6 years. Baseline peripheral disease activity was high, according to average DAS28-4vESR (4.9±1.4) and DAPSA (29.9±15.4) scores. Overall retention of TNFis was 48.9±40.8 months for the first prescribed drug, which decreased to 35.5±33.1 months for the second TNFi (P <.001) and to 22.7±22.9 months for the third TNFi (P <.001).

According to Cox models, baseline characteristics significantly associated with discontinuation of first-line therapy included being women (hazard ratio [HR], 2.1 [95% CI, 1.29-3.41]; P =.003), greater baseline DAS28-4vESR score (HR. 1.18 [95% CI, 1.01-1.39]; P =.039), and the use of infliximab as the first-line TNFi (HR, 2.00 [95% CI, 1.14-3.52]; P =.015). In models specific to the reasons for discontinuation, being women, and baseline DAS28-4vESR score were each associated with an increased risk for discontinuation due to ineffectiveness, not adverse events.

After 6 months of first-line TNFi treatment, 48.7% of patients achieved a good EULAR response criteria and 20.9% achieved DAPSA remission. Response rates to the second and third TNFis were significantly inferior compared with response rates of the first TNFi. Female sex and obesity were each associated with lower odds of achieving a good EULAR response at 6 months (P =.005 and P =.006, respectively).

Related Articles

These data highlight several important predictors of drug retention in a cohort of patients with PsA. Although TNFi treatment displays efficacy in the treatment of PsA, long-term retention remains a challenge.

Study limitations included predictors of discontinuation not being determined for patients who switched TNFis and response ratesnot being available for third-line TNFis.

According to the researchers, further study using data obtained from clinical trials is necessary to explore the effects of sex and obesity on TNFi efficacy.


Vieira-Sousa E, Eusébio M, Ávila-Ribeiro P, et al. Real-world longterm effectiveness of tumor necrosis factor inhibitors in psoriatic arthritis patients from the rheumatic disease Portuguese register [published online August 1, 2019]. J Rheumatol. doi:10.3899/jrheum.181272