ACADIA Pharmaceuticals (ACAD) investor relations material

Study design and patient population

• RADIANT is a global, randomized, double-blind, placebo-controlled phase II/III program evaluating remlifanserin in Alzheimer's disease psychosis (ADP), with 363 patients randomized 1:1:1 to 60 mg, 30 mg, or placebo for 6 weeks.

• The modified full analysis set included patients with baseline SAPS-H+D ≥10 and at least one post-baseline assessment.

• Demographics were balanced across groups, with mean age around 75–76, majority White, and two-thirds female.

Efficacy results

• The 60 mg dose narrowly missed statistical significance on the primary endpoint (SAPS-H+D, p=0.0603) but achieved nominal significance on the key secondary endpoint (CGI-S-ADP, p=0.0077), with effect sizes of 0.26 and 0.37, respectively.

• Reductions from baseline were 12.4 vs 10.4 (SAPS-H+D) and 1.3 vs 0.9 (CGI-S) for 60 mg vs placebo.

• Consistent efficacy trends were observed across endpoints, with increasing separation from placebo through week 6.

• The 30 mg dose showed minimal improvement relative to placebo.

• Subgroup analysis of patients with higher baseline psychosis (80% of trial) showed greater effect sizes for both endpoints.

Dose selection and population refinement

• Efficacy and safety data support advancing the 60 mg dose to phase III, with the 30 mg arm being removed.

• Refining inclusion criteria to modestly higher baseline psychosis could increase effect sizes while retaining 80% of the trial population.

• Further enrichment strategies are under consideration to optimize phase III outcomes.