Diverse Screening vs. Diverse Enrollment — The Amyloid Positivity Gap

Overview

Acumen Pharmaceuticals screened more than 50% diverse participants by race/ethnicity in Phase 1 but final enrollment was only ~25% diverse. The drop-off was largely driven by differential amyloid positivity rates across populations — not a recruitment failure but a protocol biology issue. A critical lesson: Diversity in the funnel does not guarantee diversity in enrollment, and the solution may require changes to protocol design, not just outreach.

Methods

Leveraged a combination of site-led and sponsor-led recruitment tactics across 17 Alzheimer’s disease (AD) trial sites, including site databases, physician referrals, community outreach, and digital advertising campaigns
Prioritized recruitment through existing site databases and established patient relationships, which proved particularly effective for reaching historically underrepresented racial and ethnic populations.
Monitored diversity at multiple stages of the recruitment funnel, including pre-screening, screening, eligibility determination, and enrollment.
Used amyloid positron emission tomography (PET) imaging as a key enrollment criterion to confirm AD pathology in potential participants.
Analyzed where participant diversity was gained or lost throughout the recruitment and screening process to identify barriers to equitable enrollment.

Metrics

1,025 potential participants were pre-screened across 17 U.S. study sites.
260 participants completed study screening; overall eligibility rate was 26.9% (70 of 260 screened participants).
Screened participants were 52.7% from historically underrepresented racial and ethnic groups.
Only 21.4% of eligible participants were from historically underrepresented racial and ethnic groups.
Participants from historically underrepresented racial and ethnic groups had significantly higher screen-failure rates (87.8%) than non-Hispanic white participants (59.9%).
Amyloid-negative PET scans were the leading cause of ineligibility, accounting for 67.6% of screen failures among historically underrepresented participants versus 53.7% among non-Hispanic white participants.

Key Lessons

Recruitment success does not automatically translate into enrollment diversity. A diverse screening population may still yield a less diverse enrolled population when biological eligibility criteria disproportionately affect certain groups.
Site databases and established patient relationships were particularly effective for reaching historically underrepresented populations, highlighting the importance of trusted local connections.
Diversity initiatives should evaluate the entire participant journey, including protocol inclusion criteria and screening requirements, rather than focusing exclusively on recruitment tactics.
Biomarker-driven AD trials may unintentionally reduce enrollment diversity when amyloid positivity requirements exclude clinically symptomatic individuals from some racial and ethnic groups.
Sponsors and investigators should consider both scientific rigor and equity when developing eligibility criteria and evaluating enrollment outcomes.
Measuring diversity at multiple points in the recruitment funnel (pre-screened, screened, eligible, and enrolled) provides more meaningful insight than enrollment statistics alone.