AB1082 EARLY CLINICAL RESPONSE AS A PREDICTOR OF LONG-TERM HEALTH-RELATED QUALITY OF LIFE IMPROVEMENTS IN PATIENTS WITH PSORIATIC ARTHRITIS AND TNFI-IR RECEIVING GUSELKUMAB (COSMOS)

Abstract

BackgroundImprovement in health-related quality of life (HRQoL) is a key goal of psoriatic arthritis (PsA) therapy.[1,2]Here, we assess whether early clinical response predicts long-term improvement in HRQoL for patients (pts) with PsA and inadequate response to 1–2 tumour necrosis factor inhibitors (TNFi-IR).ObjectivesAssess the association between early clinical response across various PsA domains and HRQoL at Week (W)48, as well as identify baseline (BL) characteristics that could predict early response in guselkumab (GUS)-treated TNFi-IR PsA pts in the Phase 3b COSMOS trial.MethodsIn the randomized controlled COSMOS trial (NCT03796858),[3]adults with active PsA (swollen/tender joint counts [SJC/TJC] each ≥3) and TNFi-IR were randomized 2:1 to receive subcutaneous injections of either GUS 100 mg (at W0, W4, then every 8 weeks through W44) or placebo (at W0, W4, W12, W20, followed by GUS at W16 [early escape] or at W24 [planned], W28, W36, W44). Only pts randomized to GUS were included in this analysis. Long-term improvements in HRQoL were defined as changes from BL to W48 in 36-item short-form survey (SF-36) physical and mental component summary (PCS and MCS) and Dermatology Life Quality Index (DLQI) scores. Early clinical response was defined as achievement of the following criteria at W4 or W8: American College of Rheumatology (ACR)20, pt pain on a visual analogue scale (VAS) ≤15,SJC ≤1, skin VAS ≤20, and health assessment questionnaire – disability index.(HAQ-DI) ≤0.5. In addition, Psoriasis Area and Severity Index (PASI) ≤1 was considered at W16 – the earliest PASI assessment. Analyses were restricted to pts not meeting the respective early response criteria at BL. Long-term HRQoL improvements were compared between pts achieving vs not achieving early response criteria by means of Student’s.t-test and by multivariate linear regression models adjusting for demographic and BL pt disease characteristics. Results from the multivariate linear regression analyses are presented here. Demographic and BL pt disease characteristics predicting early clinical response were investigated using multivariate logistic regression.ResultsOverall, 189 pts were randomized to GUS, with a mean age of 49.1 years, and 45.5% were male. Among pts not meeting the respective early response criteria at BL, GUS led to.2.7–19.0% and 4.3–38.4% of pts achieving one of the clinical responses of interest as early as W4 and W8, respectively. SF-36 PCS improvement from BL to W48 was significantly associated with ACR20 response, SJC ≤1 and HAQ-DI ≤0.5 achievement at W4 as well as at W8. There were no significant findings for SF-36 MCS. DLQI improvement from BL to W48 was significantly associated with ACR20 and skin VAS ≤20 at W8 (Figure 1). Improvements in SF-36 PCS, SF-36 MCS and DLQI at W48 were all significantly associated with achievement of PASI ≤1 at W16. Multivariate logistic regression identified significant (P<0.05) associations between males and early clinical response at W8 (ACR20: odds ratio [OR]=1.98; HAQ-DI ≤0.5: OR=3.71), BL SJC and.SJC ≤1 at W8 (OR=0.84), BL HAQ-DI and HAQ-DI ≤0.5 at W8 (OR=0.23), and BL skin VAS and skin VAS ≤20 at W8 (OR=0.98).ConclusionFor pts receiving GUS, ACR20 response at W4 and W8 was positively associated with SF-36 PCS improvement from BL to W48, and also at W8 with DLQI improvement from BL to W48. Therefore, early clinical response is relevant for HRQoL improvements over time. These results may help in shared decision-making processes.