OBJECTIVES: The ELATIVE and RESPONSE trials assessed elafibranor and seladelpar, respectively, compared to placebo for primary biliary cholangitis (PBC) in patients with inadequate response or intolerance to ursodeoxycholic acid. This analysis investigated the relative efficacy of these treatments.

METHODS: Following a feasibility assessment, a Bayesian network meta-analysis (NMA) was conducted to estimate elafibranor’s treatment effect versus seladelpar. Patients not meeting the alkaline phosphatase (ALP) eligibility criteria of RESPONSE were excluded from the ELATIVE patient level data; outcome definitions were aligned to RESPONSE. Random effects NMAs were performed for the base case analysis with informative priors on the between-study standard deviation. Vague priors were applied for treatment and baseline study effects. Posterior probabilities (PPs) of elafibranor having more favourable outcomes than seladelpar were generated.

RESULTS: Elafibranor had significantly greater odds (median odds ratio [mOR] [95% credible interval (CrI)], PP) of achieving the primary cholestasis response composite endpoint (ALP<1.67x upper limit of normal [ULN], ALP reduction ≥15% from baseline, and TB≤ULN) at Week 52 (13.02 [1.45-420.20], 0.991) versus seladelpar. The remaining analyses did not identify significant differences between the treatments; odds of ALP normalisation were numerically lower (0.42 [0.00-282.30], 0.380) and reductions were lower (median difference [95% CrI], PP) in pruritus according to the worst itch numerical rating scale (0.46 [-0.83-1.75], 0.236) and 5-D Itch (0.93 [-1.62, 3.49], 0.233) with elafibranor than seladelpar. Conversely, there was a greater reduction in pruritus with elafibranor than seladelpar according to PBC-40 Itch (-0.33 [-2.29, 1.64], 0.642). Pruritus as a treatment-emergent adverse event was less likely (mOR [95% CrI], PP) with elafibranor than seladelpar (0.73 [0.14, 3.81], 0.649), though all-cause discontinuation was more likely with elafibranor than seladelpar (1.30 [0.25, 6.90], 0.368).

CONCLUSIONS: Elafibranor had significantly greater odds of cholestasis response compared to seladelpar, without significant differences for other endpoints. This NMA may inform healthcare resource allocation and treatment decisions.