Empowering clinical trial design with agentic intelligence and real-world data

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Not reported — no quantitative effect size is provided for any outcome measure

• Primary Outcome: Qualitative demonstration of EmulatRx's ability to autonomously refine clinical trial protocols and generate RWE reports across six disease areas (septic shock, acute heart failure, acute pulmonary edema, acute kidney injury, Alzheimer's disease, Parkinson's disease)

• Nnt Or Sensitivity: Not applicable — no diagnostic accuracy, therapeutic effect, or prognostic metrics are reported; framework performance metrics (sensitivity, specificity, concordance with expert consensus) are absent

• Confidence Interval: Not reported

Clinical Application

Implementation feasibility is currently limited by the requirement for access to large, structured EHR databases, computational infrastructure, and domain expertise to validate outputs. The framework has not been tested outside the development team's institutional environment. Regulatory acceptance of EmulatRx-generated protocols by bodies such as the FDA or EMA has not been established. EmulatRx has not been validated on Australian EHR data. Australian clinical trial infrastructure operates under TGA oversight, and trial protocols must comply with ICH E6(R2) GCP guidelines as adopted by the TGA. The INSIGHT Network's NYC-based data does not map to Australian ICD-10-AM coding, PBS medication lists, or AIHW population health datasets. RACGP and ACCRM guidelines for chronic disease management (relevant to Alzheimer's and Parkinson's disease arms) reflect different care pathways than US practice. Before any Australian application, validation against datasets such as the APDC, MBS/PBS linked data, or state-based EHR systems (e.g., NSW HealtheNet) would be essential. The Australian Clinical Trials Alliance (ACTA) and NHMRC would need to evaluate regulatory and ethical acceptability of autonomous protocol generation tools. Researchers and clinical trialists working with large-scale EHR datasets for trial design in acute care (ICU/emergency medicine) and chronic neurological disease settings; not directly applicable to individual patient care decisions

Abstract

Clinical trial design (CTD) is a time-consuming process that requires substantial domain expertise. Large-scale real-world data (RWD), such as electronic health records (EHR), encodes practice-based evidence that is of tremendous value to CTD. In recent years, many machine learning methods have been developed to extract such real-world evidence (RWE) from the RWD to inform CTD, but they still need to be communicated with the domain experts extensively in an iterative manner to be further refined and ultimately useful. In this paper, we introduce EmulatRx, an agentic framework that derives RWE for helping with CTD. Through the iterative conversation and analysis across agents with different roles, EmulatRx can autonomously refine trial protocols and finally generate a robust report containing insights that inform better CTD. We applied EmulatRx on the CTD process for both acute diseases (e.g., septic shock, acute heart failure, acute pulmonary edema, and acute kidney injury) using the MIMIC-IV data and chronic diseases (e.g., Alzheimer's disease and Parkinson's disease) using the INSIGHT Network across five New York City health systems. The results demonstrate EmulatRx's capabilities in facilitating and accelerating the CTD process.

References

1. 1.Li, H., Pan, W., Rajendran, S., Zang, C., & Wang, F. (2026). Empowering clinical trial design with agentic intelligence and real-world data. Nature Communications. https://doi.org/10.1093/aje/kwq439