Leveraging EA8191 to Assess AI-Augmented EHR Abstraction

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.

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.