RWE is finding a role earlier in drug development than ever before, but the data quality, integration, and organizational alignment required to make it regulatory-grade are still catching up to the ambition.

In rare disease drug development where no registry or natural history dataset exists, real-world evidence quality depends on treating evidence engineering as a design-stage decision, defining intended regulatory use and standardizing endpoints, harmonization, and data provenance upfront rather than reconciling gaps after collection.

The pilot pairs drug sponsors with qualified research institutions to compress the path from drug identification to first-in-human study through rolling submission review and earlier coordination of institutional review board and site activation activities.

Hierarchical composite endpoints analyzed through pairwise comparisons more accurately reflect multifaceted treatment benefit than time-to-first-event composites, but transparent outcome prioritization, patient involvement in ranking, and reporting of Net Treatment Benefit remain underutilized despite their importance to interpretation.

The new effort combines adaptive platform trial design, AI-enabled site activation, nationwide data infrastructure, and patient data contribution tools to reduce timelines, costs, and patient burden across clinical development.

AI agents in clinical operations acquire broader autonomous capability through expanded permissions, tools, memory, and delegated authority, requiring governance focused on whether effective capability has shifted outside approved boundaries rather than whether software has changed.

In this video interview, Claire Riches, VP of clinical solutions at Citeline, makes the case for AI as a sophisticated strategic advisor in trial design while arguing that humans must remain in the lead—especially when factors the model can't fully account for are at stake.

In this video interview, Claire Riches, VP of clinical solutions at Citeline, walks through the categories of hidden protocol risk that AI-driven simulations can identify—from overly restrictive eligibility criteria to patient dropout patterns and structural trial assumptions.