BioVie (BIVI) Study result summary

Study design and objectives

• ADDRESS-LC was a phase II, multicenter, randomized, double-blind, placebo-controlled, signal-finding, proof-of-concept trial for bezisterim in Long COVID, enrolling 203 adults to evaluate 22 clinical outcomes and over 380 proteomic biomarkers, focusing on fatigue, post-exertional malaise, and cognitive impairment.

• The study was exploratory, with no single pre-specified primary endpoint, aiming to identify effective endpoints and patient subgroups for future phase III design.

• Subgroup analyses were pre-specified in the statistical plan submitted to the FDA before unblinding.

• The trial was fully funded by a $13.13 million U.S. Department of War grant.

• Bezisterim is a first-in-class oral small molecule targeting ERK/NFκB-mediated inflammatory signaling and crosses the blood-brain barrier.

Patient population and baseline characteristics

• 203 patients were randomized; treatment and placebo arms were balanced in age, sex, and baseline symptom severity.

• 78% of patients had at least one severe symptom; subgroups included high fatigue (55%), high post-exertional malaise (36%), and high cognitive impairment (48%).

• Most patients were on multiple concomitant medications, with 96% taking at least one and 873 unique medications reported.

Key efficacy findings/results

• Statistically significant improvements were observed in pre-specified subgroups with high baseline fatigue, post-exertional malaise, or cognitive impairment, with benefits seen in multiple endpoints per symptom group.

• In high fatigue subgroup, significant benefit was seen on five endpoints and positive trends on three others; high malaise subgroup had significant improvements on seven endpoints and trends on four more; high cognitive impairment subgroup saw significant gains on four endpoints and a trend on one.

• Bezisterim showed a consistent advantage over placebo on nearly all endpoints in the intent-to-treat population, though most did not reach statistical significance due to symptom heterogeneity.

• Effect sizes (Cohen's d) were moderate to high in responders, with effect sizes in subgroups more than doubling compared to the overall population.

• Improvements included fatigue, sleep, malaise, cognition, and overall patient and clinician impressions.