Leveraging EA8191 to Assess AI-Augmented EHR Abstraction

Leveraging a Penn-based Cancer Trial (EA8191) to Assess the Prospective Performance of Artificial Intelligence Augmented Electronic Health Record (EHR) Data Abstraction for Clinical Trial Patient Screening and Selection

The goal of this prospective study is to assess the performance of AI (artificial intelligence) augmentation (compared against historical controls) to identify oncology patients who meet inclusion criteria for a clinical trial. The study staff will leverage a natural language processing (NLP)-based AI algorithm that rank-orders patients most likely to meet inclusion criteria for a trial. We hypothesize that this collaborative Human+AI workflow can improve the efficiency, accuracy, and diversity of trial prescreening.

Study Overview

• Status: Active, not recruiting
• Conditions: Prostate Cancer
• Intervention / Treatment: Other: Chart review

Detailed Description

The objective of this prospective study is to assess the accuracy and efficiency of a clinical research coordinator utilizing AI augmentation to identify oncology patients who meet the inclusion criteria for participation in clinical trials. The clinical research coordinator utilizing AI augmentation ("Human+AI") will leverage an autonomous natural language processing (NLP)-based AI algorithm (Mendel AI) developed by artificial intelligence startup company Mendel.ai. The algorithm Mendel AI serves as a supportive tool in the decision-making process by providing the clinical research coordinator a rank-ordered list of patients most likely to meet inclusion criteria for a trial, as well as a list of elements abstracted by the AI algorithm for each patient. The performance of "Human+AI" would be compared against historical control from the selected clinical trial: EA8191/INDICATE (Penn IRB #848795), a national, Phase III, randomized, prostate cancer clinical trial on the use of PET scan findings to direct local and systemic treatment intensification in patients with post-prostatectomy biochemical recurrence.

The electronic health records (EHRs) reviewed as part of this clinical trial screen would draw from randomly-selected Penn patients with upcoming genitourinary radiation oncology and medical oncology appointments at the Perelman Center for Advanced Medicine. Given a randomly selected batch of EHRs from these patients viable for prescreening, the research team aims to determine how well (and also if better, how much better) clinical research staff leveraging Mendel's AI algorithm can identify those patients who met the eligibility criteria for the EA8191 trial (as compared to historical averages for the EA8191).

The study primarily aims to compare (1) the efficiency of the Human+AI collaboration relative to historical efficiency from a Human-alone workflow, (2) the accuracy of the Human+AI collaboration relative to historical accuracy from a Human-alone workflow and (3) the diversity of eligible patients identified by the Human+AI prescreening workflow, compared to historical diversity from a Human-alone workflow.

Our central hypothesis is that workflows that merge traditional CRC-driven prescreening with automated AI - "Human+AI" workflows - can improve the efficiency, accuracy and diversity of trial prescreening.

The identification of eligible patients for clinical trials is a critical component of clinical research, as it directly impacts patient recruitment, study enrollment, and the generalizability of research findings. Currently, the process of identifying eligible patients often relies on manual chart review by clinical research staff, which can be time-consuming, labor-intensive, and prone to human error. Consequently, eligible patients may be overlooked, and opportunities for trial participation may be missed. The integration of AI technology into the patient identification process has the potential to enhance the accuracy and efficiency of this critical task, leading to improved clinical trial recruitment and outcomes.

This study holds important implications for the field of clinical research by evaluating the effectiveness of AI-augmented patient identification compared to traditional manual methods and autonomous AI algorithms. By examining the strengths and limitations of each approach, the study will provide valuable insights into the optimal integration of AI technology in clinical research processes. Furthermore, the results of this study have the potential to benefit patients by improving their access to clinical trials and increasing awareness of available treatment options. For clinical research institutions, enhancing the efficiency of patient identification can lead to more effective use of research resources and the potential for accelerated clinical trial timelines. Ultimately, the findings of this study may contribute to advancements in clinical research practices, promoting more equitable access to trials and facilitating the development of innovative treatments for patients with cancer.

• Study Type: Observational
• Enrollment (Estimated): 300

Contacts and Locations

Study Locations

• United States
  • Pennsylvania
    • Philadelphia, Pennsylvania, United States, 19104
      • University of Pennsylvania

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Probability Sample

Study Population

Patients with a diagnosis code of prostate cancer and with an upcoming genitourinary radiation oncology or medical oncology appointments at the Perelman Center for Advanced Medicine

Description

Inclusion Criteria:

• Patient must be male and >= 18 years of age.
• Patient must have had a radical prostatectomy (RP) as definitive therapy for histopathologically-proven prostatic adenocarcinoma

• Patient must have biochemical recurrence (BCR) after RP, defined as follows:

  • If time to BCR, defined as time to first detectable PSA ( > lower limit of normal for assay used) after RP, is < 12 months, a minimum PSA level of >= 0.2 ng/mL and a confirmatory reading of >= 0.2 ng/mL is required, per the American Urological Association (AUA) definition (Note: patients with a persistent PSA reading of at least 0.2 ng/mL are eligible)
  • If time to BCR, defined as time to first detectable PSA (> lower limit of normal for assay used) after RP, is >= 12 months, a minimum absolute PSA of 0.5 ng/mL is required
  • If the patient has a detectable PSA (> lower limit of normal for assay used) at any time after RP AND has an eligible baseline SOC PET (PET1) with at least one positive lesion in any location, then there is no minimum PSA requirement

• Patients must have no definite evidence for extrapelvic metastatic disease by conventional imaging modalities (CIM) (CT abdomen/pelvis or MRI abdomen/pelvis AND bone scintigraphy, or equivalent), within 26 weeks prior to Step 0 registration. If a patient only has a study-eligible PET/CT or PET/MR (i.e., PET done without prior CIM): if the PET is negative for extrapelvic lesions, then baseline CIM is NOT required. If the PET positive for extrapelvic lesions, then patient should have a baseline CT/MRI for soft tissue lesions and/or a bone scan for osseous lesions

  • Study eligible = PET using FDA-approved radiotracer and performed within 16 weeks prior to study registration
• Extra-pelvic metastases is defined as any osseous metastases and/or any extrapelvic soft tissue, lymph nodes and organ metastases; extra-pelvic is defined as superior to common iliac bifurcation, outside of standard prostate bed + whole pelvis nodal RT fields. Baseline PET/CT or PET/MR scan (PET1) is eligible for this study if the SOC PET scan is completed with an FDA approved radiotracer for prostate cancer after Step 0 registration and prior to Step 1 randomization OR up to 16 weeks prior to Step 0 registration
• Patient must be a candidate for SOC post-prostatectomy radiation therapy (RT) to the prostate bed and pelvic nodes with androgen deprivation therapy (ADT)
• Patient must have the ability to understand and the willingness to sign a written informed consent document. Patients with impaired decision-making capacity (IDMC) who have a legally authorized representative (LAR) or caregiver and/or family member available will also be considered eligible
• Patient must have an Eastern Cooperative Oncology Group (ECOG) performance status 0-2
• Patient must not have started ADT for biochemical recurrence prior to baseline PET (PET1) imaging. A short course of low-dose anti-androgen such as bicalutamide, given after baseline study PET/CT but prior to study registration, is permitted as a brief temporizing measure in advance of starting protocol-approved SOC ADT.
• Patient must not be enrolled in another therapeutic clinical trial
• Patient must be able to lie flat and still for approximately 20-30 minutes or otherwise tolerate a PET scan and radiation treatment planning and delivery
• Patients undergoing a PET/MR must meet local institutional safety guidelines for MRI
• Patient must not have history of seizures or known condition that may cause predisposal to seizures (e.g., stroke or head trauma resulting in loss of consciousness) within 1 year prior to registration
• Patient must not have history of inflammatory bowel disease or any gastrointestinal disorder affecting absorption that is expected to increase risk of complication from radiotherapy
• Hemoglobin (Hgb) >= 9.0 g/dL (independent of transfusion and/or growth factors within 3 months prior to Step 0 registration) (obtained within 8 weeks prior to Step 0 registration)
• Leukocytes >= 3,000/mcL (obtained within 8 weeks prior to Step 0 registration)
• Absolute neutrophil count >= 1,500/mcL (obtained within 8 weeks prior to Step 0 registration)
• Platelets >= 100,000/mcL (obtained within 8 weeks prior to Step 0 registration)
• Total bilirubin < 1.5 x institutional upper limit of normal (ULN) (patients with Gilbert's syndrome, if total bilirubin is > 1.5 x ULN, must have a direct bilirubin of < 1.5 x ULN to be eligible) (obtained within 8 weeks prior to Step 0 registration)
• Aspartate aminotransferase (AST) (serum glutamic oxaloacetic transaminase [SGOT])/alanine aminotransferase (ALT) (serum glutamate pyruvate transaminase [SGPT]) =< 2.5 x institutional ULN (obtained within 8 weeks prior to Step 0 registration)
• Creatine < 1.5 x institutional ULN (or measured creatinine clearance > 30 mL/min)
• Human immunodeficiency virus (HIV)-infected patients on effective anti-retroviral therapy with undetectable viral load within 6 months are eligible for this trial
• Patients with known history or current symptoms of cardiac disease, or history of treatment with cardiotoxic agents, should have a clinical risk assessment of cardiac function using the New York Heart Association Functional Classification. To be eligible for this trial, patients should be class I or II (by patient symptoms) or A or B (by objective assessment)
• Patient must not have completed a course of prior pelvic radiation therapy for any reason
• Patient must agree not to father children while on study
• Patient must be English or Spanish speaking to be eligible for the QOL component of the study

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Human + AI

Other: Chart review; Patients eligible for prescreening will be identified using structured criteria (e.g., anyone with an appointment in the upcoming 2 months with a relevant provider). All individual charts will be de-identified. De-identified EHRs will then be transferred to Mendel's secure data abstraction platform, where trial eligibility criteria will be abstracted. Patients will be rank-ordered by number of eligibility criteria met, with patients meeting the most eligibility criteria at the top of the queue. The study team will then review eligibility criteria and flag eligible patients.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Efficiency	The number of eligible patients per 10 charts prescreened	1 year

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Accuracy	The proportion of total charts identified correctly as eligible	6 months
Diversity	The proportion of eligible individuals who identified from groups historically underrepresented in clinical trials	6 months

Collaborators and Investigators

• Sponsor: University of Pennsylvania
• Collaborators: Mendel AI

Study record dates

Study Major Dates

• Study Start (Actual): April 4, 2025
• Primary Completion (Estimated): May 1, 2026
• Study Completion (Estimated): August 1, 2026

Study Registration Dates

• First Submitted: August 16, 2024
• First Submitted That Met QC Criteria: August 16, 2024
• First Posted (Actual): August 19, 2024

Study Record Updates

• Last Update Posted (Actual): March 12, 2026
• Last Update Submitted That Met QC Criteria: March 10, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Urogenital Diseases; Genital Diseases; Genital Neoplasms, Male; Urogenital Neoplasms; Neoplasms by Site; Neoplasms; Genital Diseases, Male; Prostatic Diseases; Male Urogenital Diseases; Prostatic Neoplasms
• Other Study ID Numbers: 854974

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: No individual participant data will be shared.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No