Secukinumab shows Higher Symptomatic improvement Versus Etanercept in Psoriatic Arthritis: Comparative Effectiveness Up To 24 Weeks Assessed by Matching-Adjusted Indirect Comparison

• Published in: Value in health 2017-10, Vol.20 (9), p.A936
• Main Authors: Mease, P, Choy, E, Thom, H, Kalyvas, C, Lopes, N, Pricop, L, Gandhi, K, Jugl, SM, Nash, P
• Format: Article
• Language: English
• Subjects: Arms; Arthritis; Body weight; Clinical trials; Comparative studies; Dominance; Etanercept; Matching; Medical treatment; Methotrexate; Monoclonal antibodies; Necrosis; Patients; Psoriasis; Psoriatic arthritis; Race; Rheumatology; Sensitivity analysis; Skin diseases
• ISSN: 1098-3015; 1524-4733
• DOI: 10.1016/j.jval.2017.08.2961

OBJECTIVES: When populations across different trials are heterogeneous, Matching-Adjusted Indirect Comparison (MAIC) can be used to assess comparative effectiveness; it is supported by NICE DSU methodological guidance. The objective of this MAIC was to assess the comparative effectiveness of secukinumab 150mg (SEC; fully human anti-interleukin-17A) and etanercept 25mg twice weekly (ETN; tumor necrosis factor inhibitor [TNFi]) up to 24 weeks in biologic-naive patients with psoriatic arthritis (PsA). METHODS: In this MAIC, individual patient data from the pooled SEC arms of FUTURE 1 (Fl) and FUTURE 2 (F2; n=302) were weighted to match baseline characteristics of the ETN arm of NCT00317499 (n=101); placebo arms were also matched. Before matching, one notable difference was the proportion of biologic-naive patients (67.2% [F1/F2] versus 100% [NCT00317499]). Logistic regression was used to determine weights for age, body weight, sex, race, PsA disease duration, presence of psoriasis (>3% body surface area), mean HAQ-DI score, swollen joint count (SJC), CRP, methotrexate use and previous TNFi therapy failure. Recalculated outcomes from F1/F2 (SEC, effective sample size [ESS]=79; placebo, ESS=94) were compared with NCT00317499. Pairwise comparisons using odds ratios (ORs) were performed for American College of Rheumatology (ACR) 20, 50 and 70 responses at week 12 (placebo-adjusted) and week 24 (non-placebo-adjusted). Placebo-adjustment at week 24 was not possible as patients could receive active treatment from week 16. RESULTS: At week 12, no differences in placebo-adjusted ACR 20,50 and 70 responses between SEC and ETN were observed. At week 24 (non-placebo-adjusted), ACR 70 responses were higher with SEC than ETN (OR [95% CI]: 3.06 [1.30-7.19], p= 0.010). A sensitivity analysis excluding PsA disease duration, SJC and CRP from the matching confirmed these results. CONCLUSIONS: In this MAIC, secukinumab showed evidence of superiority for symptomatic improvement (non-placebo-adjusted ACR 70) over etanercept for active PsA at 24 weeks.