PURPOSE Less than 5% of patients with cancer enroll in clinical trials, and 1 in 5 trials are stopped for poor accrual. We evaluated an automated clinical trial matching system that uses natural language processing to extract patient and trial characteristics from unstructured sources and machine learning to match patients to clinical trials. PATIENTS AND METHODS Medical records from 997 patients with breast cancer were assessed for trial eligibility at Highlands Oncology Group between May and August 2016. System and manual attribute extraction and eligibility determinations were compared using the percentage of agreement for 239 patients and 4 trials. Sensitivity and specificity of system-generated eligibility determinations were measured, and the time required for manual review and system-assisted eligibility determinations were compared. RESULTS Agreement between system and manual attribute extraction ranged from 64.3% to 94.0%. Agreement between system and manual eligibility determinations was 81%-96%. System eligibility determinations demonstrated specificities between 76% and 99%, with sensitivities between 91% and 95% for 3 trials and 46.7% for the 4th. Manual eligibility screening of 90 patients for 3 trials took 110 minutes; system-assisted eligibility determinations of the same patients for the same trials required 24 minutes. CONCLUSION In this study, the clinical trial matching system displayed a promising performance in screening patients with breast cancer for trial eligibility. System-assisted trial eligibility determinations were substantially faster than manual review, and the system reliably excluded ineligible patients for all trials and identified eligible patients for most trials.

No abstract

6501 Background: IBM Watson for Clinical Trial Matching (CTM) is a cognitive computing solution that uses natural language processing (NLP) to help increase the efficiency and accuracy of the clinical trial matching process. This solution helps providers locate suitable protocols for their patients by reading the trial criteria and matching it to the structured and unstructured patient characteristics when integrated with the Electronic Medical Record (EMR). It is also designed to determine which sites have the most viable patient population and identify inclusion and exclusion criteria that limit enrollment. Methods: This project was a collaboration among Highlands Oncology Group (HOG), Novartis and IBM Watson Health to explore the use of CTM in a community oncology practice. HOG is in Northeast Arkansas and has 15 physicians and 310 staff members working across 3 sites. During the 16-week pilot period, data from 2,620 visits by lung and breast cancer patients were processed by the CTM system. Using NLP capabilities, CTM read the clinical trial protocols provided by Novartis, and evaluated the patient data against the protocols’ inclusion and exclusion criteria. Watson excluded ineligible patients, determined those that needed further screening, and assisted in that process. Feedback on the user experience was also obtained. Results: In an initial pre-screening test, the HOG clinical trial coordinator (CTC) took 1 hour and 50 minutes to process 90 patients against 3 breast cancer protocols. Conversely, when the CTM screening solution was used, it took 24 minutes. This represents a significant reduction in time of 86 minutes or 78%. Watson excluded 94% of the patients automatically, providing criteria level evidence regarding the reason for exclusion, thus reducing the screening workload dramatically. Conclusions: IBM Watson CTM can help expedite the screening of patient charts for clinical trial eligibility and therefore may also help determine the feasibility of protocols to optimize site selection and enable higher and more efficient trial accruals.

134 Background: Understanding and gaining familiarity with the adverse event (AE) and drug-drug interaction (DDI) potential of novel agents can be challenging, cumbersome, and time consuming for clinicians. A comprehensive and secure AE management/DDI mobile application (CT Scholar [Novartis Pharmaceuticals Corp]) was developed to ease the management of clinical trial participants with hormone receptor–positive, human epidermal growth factor receptor 2–negative advanced breast cancer receiving ribociclib. Methods: CT Scholar was designed specifically for the ribociclib clinical program to provide easy access to information on AE and DDI management. The AE management module characterizes and quickly informs drug-specific AE management based on protocol guidelines and supportive care guidance. The DDI module includes a database of > 100 drugs searchable by both trade and generic name. Number of application visits, unique users, and screen views were monitored from Jan to Sept 2016. Results: There were 1055 visits to CT Scholar (117/month) by 145 unique users (28/month), with 5764 screen views (640/month) and 5.5 screens viewed/visit. Among the most-viewed modules were hematologic AE management (1132 views), gastrointestinal AE management (771), and DDI (734). Diarrhea, neutropenia, and nausea were the most viewed submodules (308, 230, and 205 views, respectively). The information on CT Scholar was consistently shown to be accessed more quickly than information obtained from printed clinical protocols. No patients discontinued from the ribociclib studies due to AEs following the introduction of CT Scholar. Conclusions: CT Scholar was widely used by clinicians, and it responded to their needs by providing essential information, especially on prevalent AEs, which could be easily updated and rapidly accessed. Additionally, CT Scholar provides a platform showing how technology can assist clinicians and patients in managing care during a trial, which can inform the development of future supportive materials.