Clinical Report: AI-Selected Trial Candidates are More Diverse

Overview

An AI system effectively identified eligible patients for a clinical trial in transthyretin amyloid cardiomyopathy, demonstrating high accuracy and increasing diversity among candidates. The platform's ability to streamline chart reviews could enhance recruitment efficiency and broaden participation from underrepresented populations.

Background

Recruitment for clinical trials often faces challenges, particularly in reaching diverse populations. Traditional methods of patient identification can be labor-intensive and may overlook eligible candidates, leading to underrepresentation in clinical studies. The integration of AI in this process presents an opportunity to improve both the efficiency of patient selection and the inclusivity of trial participants.

Data Highlights

Key Findings

• The AI system processed 1,476 patient records in six days, significantly faster than manual reviews.
• It achieved 96.2% accuracy in answering trial-relevant questions compared to physician review.
• Of 30 AI-identified patients, 37% were Black, compared to only 7% identified manually.
• The AI system correctly excluded ineligible patients 99% of the time.
• Most eligible patients identified by AI had not been detected through standard screening processes.

Clinical Implications

The use of AI in clinical trial recruitment can enhance the identification of eligible patients, particularly from underrepresented groups. This approach not only streamlines the recruitment process but also addresses biases inherent in traditional methods, potentially leading to more equitable trial participation.

Conclusion

The findings suggest that AI-assisted screening could revolutionize patient recruitment for clinical trials, making the process more efficient and inclusive. Further evaluation is necessary to compare these methods directly with conventional recruitment strategies.

References

1. Automating Chart Review Using an Artificial Intelligence−Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility, ScienceDirect, 2023 -- AI-Selected Trial Candidates are More Diverse
2. AI and innovation in clinical trials, npj Digital Medicine, 2023 -- AI and innovation in clinical trials
3. ASCO-ACCC Piloting Project to Test Tools to Diversify Cancer Clinical Trials, The ASCO Post, 2021 -- ASCO-ACCC Piloting Project to Test Tools to Diversify Cancer Clinical Trials
4. Novel AI Platform May Help Identify Patients Likely to Benefit Most From Clinical Trials, The ASCO Post, 2025 -- Novel AI Platform May Help Identify Patients Likely to Benefit Most From Clinical Trials
5. Non-Biopsy Diagnosis of Cardiac Transthyretin Amyloidosis - American College of Cardiology, 2016 -- Non-Biopsy Diagnosis of Cardiac Transthyretin Amyloidosis
6. The ASCO Post — Efforts to Broaden Eligibility Criteria for Clinical Trials Seek to Include More Racial and Ethnic Minority Patients
7. AI Can Unlock EHR Data to Determine Trial Eligibility
8. Diversity Action Plans to Improve Enrollment of Participants from Underrepresented Populations in Clinical Studies | FDA
9. Non-Biopsy Diagnosis of Cardiac Transthyretin Amyloidosis - American College of Cardiology
10. Tafamidis in Transthyretin Cardiomyopathy Clinical Trial - American College of Cardiology
11. Automating Chart Review Using an Artificial Intelligence−Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility - ScienceDirect