Efficacy and Safety of Eliapixant in Refractory Chronic Cough: The Randomized, Placebo-Controlled Phase 2b PAGANINI Study

**Background:** Refractory chronic cough (RCC) affects approximately 10% of the global adult population and has no approved treatments in most countries. P2X3 receptors play a key role in sensory neural dysregulation underlying RCC. While the P2X3 antagonist gefapixant showed efficacy, it caused substantial taste-related side effects (up to 81% in phase 2b), attributed to P2X2/3 receptor blockade. Eliapixant is a highly selective P2X3 antagonist with potentially fewer taste effects. A prior phase 2a study showed efficacy with lower taste-related AE rates.

**Methods:** PAGANINI was a randomized, double-blind, placebo-controlled, parallel-group, dose-finding phase 2b study conducted at 99 centers across 19 countries. Adults with RCC lasting ≥12 months and cough severity ≥40 mm on a 100 mm VAS were randomized 1:1:1:1 to eliapixant 25 mg, 75 mg, 150 mg, or placebo twice daily for 12 weeks. The primary endpoint was change from baseline in 24-hour cough count at Week 12, analyzed using a multiple comparison procedure modeling (MCP-Mod) approach with a one-sided α=0.10. Secondary endpoints included cough severity VAS, LCQ, and responder analyses. The study was not powered for individual pairwise comparisons.

**Key Results:** 310 participants were randomized (25 mg n=75, 75 mg n=78, 150 mg n=80, placebo n=77); 276 (89%) completed treatment. Baseline characteristics were balanced (mean age ~59 years, ~78% female, mean cough duration ≥10 years in ~43%). A statistically significant dose-response signal was detected for the primary endpoint (all four candidate models, adjusted p<0.1). At Week 12, 24-hour cough count reductions relative to placebo were: 25 mg: -12% (95% CI -30 to 11); 75 mg: -27% (95% CI -41 to -9); 150 mg: -18% (95% CI -33 to <1). The Emax model estimated the minimum effective dose (MED) for a 20% relative improvement over placebo at ~58 mg twice daily. A ≥30% reduction in cough count was achieved by 46% of placebo vs. 52-64% of eliapixant groups; the 75 mg group showed a significant difference vs. placebo (18% treatment difference, 95% CI 2 to 34, p=0.03). Cough severity VAS showed numeric but non-significant reductions vs. placebo. LCQ total score improvements were similar across groups. Adverse events occurred in 51% (placebo) vs. 57-65% (eliapixant). Dysgeusia was dose-related: 1% placebo, 1% (25 mg), 12% (75 mg), 16% (150 mg). Any taste/smell disorder AE occurred in 6% placebo vs. 4% (25 mg), 15% (75 mg), 24% (150 mg). No participants rated taste effects as "extremely" bothersome. A moderate drug-induced liver injury (DILI) of hepatocellular origin occurred in one participant receiving 150 mg eliapixant, leading to study drug discontinuation; liver enzymes normalized after cessation. Dose-dependent increases in alkaline phosphatase, fibrinogen, and antithrombin III were observed.

**Clinical Implications:** Eliapixant demonstrated efficacy in reducing objective cough frequency in RCC with a substantially lower rate of taste-related AEs than reported for gefapixant (up to 24% vs. up to 81% in phase 2b). The 75 mg twice-daily dose appeared optimal, with no additional benefit at 150 mg. However, the occurrence of DILI at the 150 mg dose led Bayer to discontinue the entire eliapixant development program across all indications. This highlights the challenge of balancing P2X3 receptor selectivity for efficacy with off-target hepatic safety. The findings underscore the continued unmet need for safe, well-tolerated treatments for RCC.