Introduction/background
Clinical trials are the cornerstone of therapeutic advances in healthcare. In oncology—one of the most active areas of clinical research—more than 19 000 clinical trials are actively recruiting in the United States. 1–4 As therapeutic options become increasingly targeted, thousands of new trials emerge annually, offering patients expanded access to investigational treatments. 5 However, connecting patients to appropriate trials remains a persistent challenge, particularly within community oncology. Barriers such as complex eligibility criteria, frequent updates to trial availability and protocols, and limited institutional infrastructure contribute to low enrollment rates. 6 , 7 Community oncology programs report a 4.1% participation rate in cancer treatment trials, compared with 21.6% at National Cancer Institute (NCI) designated Comprehensive Cancer Centers and 5.4% at academic (non-NCI) centers. 8
Healthcare providers are uniquely positioned to support trial enrollment, yet studies indicate that many clinicians lack the time and resources to remain current with the evolving trial landscape. 9–11 In practice, trial information is often scattered across portals, spreadsheets, and email communications, making it difficult for providers to surface relevant trial options during time-constrained visits. Because patients most often learn about clinical trials directly from their providers, this fragmentation can result in promising trials being overlooked or not discussed in time. 12 For providers, clear trial information, consistent engagement, and institutional-specific points of contact are critical to supporting trial discussions and sustaining enrollment. 13–18
To address these barriers, we developed a clinical trial knowledge management platform tailored to the workflows of day-to-day oncology providers. 19–21 Unlike automated trial-matching tools that focus on identifying candidate studies based on patient characteristics, a clinical trial knowledge management platform is designed to support the “last-mile” execution of trials once a study has been identified. Specifically, it equips clinicians with institution-specific, point-of-care knowledge needed to contextualize eligibility criteria, understand trial logistics, identify appropriate contacts, and engage in informed discussions with patients. The platform is built around 3 core functions: (1) providing real-time, institutional-specific trial information; (2) enabling bidirectional knowledge exchange between clinicians and trial teams; and (3) streamlining trial coordination and referrals toward enrollment.
Following successful deployment of this platform at a National Comprehensive Cancer Center, where it supported 138 oncology trials and maintained user engagement for over 20 months, 20 this study evaluates the feasibility of developing and validating institution-specific clinical trial content within a community oncology network.
Methods
Study setting
This project was conducted at the Hartford HealthCare Cancer Institute (HHCI), a regional network of 7 hospitals and 16 outpatient oncology clinics across Connecticut. The network includes 40 medical oncologists and 14 advanced practice providers who care for more than 9,000 newly diagnosed cancer patients annually, in addition to a large cohort of patients receiving ongoing cancer care.
Platform development and data curation
Three independent research coordinators collaborated with disease-site oncology teams to compile an inventory of actively recruiting trials. Foundational data fields—trial title, disease area, biomarkers, stage and line of therapy, participating sites, and recruiting status—were structured and entered into the platform. Protocol summaries and eligibility elements were extracted using an embedded retrieval-augmented large language model, 21 and subsequently reviewed and validated by research coordinators to ensure accuracy, completeness, and institutional relevance. Cancer center leadership, oncologists, and trial coordinators further refined content to optimize clarity for point-of-care decision making.
Workflow integration and administrative structure
The initial implementation phase required approximately 2 to 4 weeks, after which disease-site study teams assumed administrative ownership of their trial content. Ongoing responsibilities included managing trial availability, updating recruitment status, validating protocol information, and responding to provider inquiries. Study teams set the status of their trials to open (recruiting), open (not recruiting), or closed and were responsible for adding new trials. Research coordinators were designated to support real-time updates and serve as workflow liaisons as needed, with the intent of maintaining engagement and facilitating timely referrals in future phases. The platform did not require EHR integration. Trial content creation and maintenance were incorporated into existing research workflows and personnel responsibilities to minimize operational burden.
Feasibility assessment
Feasibility was assessed by evaluating whether clinical trial content could be systematically compiled and validated within the platform across the HHCI community oncology network. Descriptive statistics characterized the scope and distribution of trials uploaded, including cancer type, department, biomarker specificity, and stage of disease. Implementation metrics such as the number of recruiting trials represented and coverage across oncology service lines served as indicators of feasibility and breadth of deployment.
Results
A total of 53 clinical trials across 10 oncology disease groups were systematically compiled and validated into the platform. At the time of reporting, 48 trials (91%) were actively recruiting. Trials represented 28 cancer conditions, with the highest representation in breast (25%), gastrointestinal (13%), and non–small cell lung cancer (11%). One-third of trials (30%) included biomarker-specific eligibility criteria, and most targeted advanced (38%), metastatic (28%), or recurrent (6%) disease.
Trial content was structured using standardized data elements aligned with existing research workflows, resulting in consistent representation across oncology service lines. Three primary administrative roles were defined during content development: research coordinators (initial configuration and support), oncologists (content review and clinical validation), and clinical trial staff (content maintenance). Following initial configuration, disease-site teams assumed responsibility for maintaining trial content, including recruitment-status updates and responses to provider inquiries.
Initial platform configuration required limited incremental resources beyond existing clinical research operations. Configuration was completed over a 2- to 4-week period per disease group, averaging approximately 6-8 hours of effort per week by a single team member, and relied on existing personnel roles without the need for additional staffing, dedicated informatics support, or new electronic health record builds. The platform was designed such that ongoing trial content updates could be incorporated into existing workflows. See Figure 1 for current user interface. Summary trial characteristics are presented in Table 1 .
Table 1.
Clinical trial knowledge management application: clinical trial characteristics.
* Trial may include more than one condition. † Trial may include more than one clinical setting.
| Clinical trial characteristics | Overall, N (%) |
| All trials | 53 |
| Open (Recruiting) | 48 (91%) |
| Open (Not Recruiting) | 5 (9%) |
| Closed | 0 (0%) |
| Oncology services | |
| Breast | 8 (15%) |
| Gastrointestinal | 7 (13%) |
| General | 9 (17%) |
| Genitourinary | 6 (11%) |
| Gynecologic | 7 (13%) |
| Head and neck | 1 (2%) |
| Hematologic | 2 (4%) |
| Melanoma | 3 (6%) |
| Neurologic | 2 (4%) |
| Thoracic | 8 (15%) |
| Conditions * | |
| Bladder cancer | 4 (8%) |
| Breast cancer | 13 (25%) |
| Cervical/vaginal cancer | 1 (2%) |
| Chronic myelomonocytic leukemia | 1 (2%) |
| Colon cancer | 1 (2%) |
| Endometrial cancer | 1 (2%) |
| Fallopian tube cancer | 3 (6%) |
| Gastroesophageal cancer | 1 (2%) |
| Gastrointestinal cancer | 7 (13%) |
| Genitourinary cancer | 1 (2%) |
| Glioma | 2 (4%) |
| Gynecologic (general) cancer | 2 (4%) |
| Head and neck cancer | 1 (2%) |
| Hepatocellular cancer | 1 (2%) |
| Lymphoma | 0 (0%) |
| Melanoma | 3 (6%) |
| Multiple myeloma | 1 (2%) |
| Myelodysplastic syndrome | 1 (2%) |
| Non-small cell lung cancer | 6 (11%) |
| Neuro-oncology cancer | 1 (2%) |
| Ovarian cancer | 3 (6%) |
| Pancreatic cancer | 2 (4%) |
| Peritoneal cancer | 3 (6%) |
| Prostate cancer | 4 (8%) |
| Rectal cancer | 1 (2%) |
| Small cell lung cancer | 1 (2%) |
| Thoracic (general) cancer | 4 (8%) |
| Urothelial cancer | 1 (2%) |
| Biomarkers | |
| No | 37 (70%) |
| Yes | 16 (30%) |
| Clinical settings † | |
| Advanced | 20 (38%) |
| Early stage | 4 (8%) |
| Locally advanced | 12 (23%) |
| Metastatic | 15 (28%) |
| Muscle invasive | 2 (4%) |
| Newly diagnosed | 1 (2%) |
| Non-muscle invasive | 1 (2%) |
| Preventative | 1 (2%) |
| Recurrent | 3 (6%) |
| Supportive care | 5 (9%) |
| Unresectable | 1 (2%) |
Discussion
In this study, we demonstrated the feasibility of developing and validating institution-specific content within a clinical trial knowledge management application across a regional community oncology network. Using standardized data elements, 53 trials spanning 10 oncology disease groups were systematically compiled, structured, and validated. Content development was supported by existing clinical and research infrastructure, without the need for additional personnel or informatics resources.
A key finding is that trial content development and validation can occur outside the EHR while remaining aligned with established clinical and research workflows. Use of an external knowledge management application enabled structured content configuration without competing for EHR development capacity. AI-assisted abstraction accelerated initial extraction of protocol summaries and eligibility elements; however, human review was necessary to ensure accuracy and institutional relevance, reinforcing the importance of a human-in-the-loop approach for trial content development.
This work adds to the growing literature on informatics tools supporting clinical trial operations. Prior efforts have focused largely on trial registries and automated patient-trial matching tools, which emphasize trial discovery and eligibility determination. 22–27 In contrast, a clinical trial knowledge management platform focuses on surfacing the local trial information required once a study has been identified–such as practical eligibility nuances, recruitment status, and operational contacts–that are critical to support informed discussion of trials and not routinely captured in registries. Establishing this type of institution-specific trial content is particularly relevant in community oncology settings, where limited infrastructure and fragmented information have historically contributed to disparities in access to clinical research. 28 , 29
Several study limitations warrant consideration. This analysis was limited to feasibility of content development and validation and did not assess application usability, workflow adoption, provider behavior, referral activity, or trial accrual. The findings reflect experience within a single community oncology network with established research infrastructure and may not generalize to settings with fewer resources.
Despite these limitations, this work demonstrates the feasibility of developing and validating institution-specific clinical trial content within a knowledge management application for the community oncology setting. Establishing this content layer is a prerequisite for subsequent evaluations of the application usability, implementation barriers and facilitators, and its downstream effects on clinical trial referral and enrollment. Ongoing studies are underway to address these dimensions as part of a broader research program. Future research will also explore how such platforms may be optimized and extended to additional stakeholders, including patients and trial sponsors, as part of a more coordinated clinical research ecosystem.
Acknowledgements
We would like to thank all of the patients, providers, and research and administrative staff members who supported this study.
Author contributions
Elizabeth O’Brien (Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Validation, Visualization, Writing—original draft, Writing—review & editing), Aanshi Thumar (Data curation, Validation, Visualization), Melanie Manning (Methodology, Project administration, Resources, Supervision, Writing—review & editing), Krishna Gunturu (Methodology, Project administration, Resources, Supervision, Validation, Writing—review & editing), Peter Yu (Project administration, Resources, Supervision, Validation, Visualization, Writing—review & editing), and Tony Kin Wai Hung (Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing)
Funding
This work was supported by the 2024 Conquer Cancer—Janssen Research & Development LLC Career Development Award in Health Equity (AWD00003905).
Conflicts of interest
T.K.W.H. is the developer of the AI-powered application evaluated in this study and founder of TeamX Health. A formal research agreement and a conflict-of-interest management plan were executed with Hartford HealthCare to address this relationship. The other authors have no competing interests to declare.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
1 ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). Accessed July 30, 2025. https://clinicaltrials.gov/
2 Tran G , Harker M , Chiswell K , et al Feasibility of cancer clinical trial enrollment goals based on cancer incidence . JCO Clin Cancer Inform . 2020 ; 4 : 35 - 49 . 10.1200/CCI.19.00088 31977253
3 Gresham G , Meinert JL , Gresham AG , Piantadosi S , Meinert CL. Update on the clinical trial landscape: analysis of ClinicalTrials.gov registration data, 2000-2020 . Trials . 2022 ; 23 : 858 . 36203212 10.1186/s13063-022-06569-2 PMC9540299
4 Hirsch BR , Califf RM , Cheng SK , et al Characteristics of oncology clinical trials: insights from a systematic analysis of ClinicalTrials. gov . JAMA Intern Med . 2013 ; 173 : 972 - 979 . 23699837 10.1001/jamainternmed.2013.627
5 Pant J , Aparna J , Roshan S , Gayathri R. Post-pandemic outlook of the oncology clinical trial landscape . J Clin Oncol . 2023 ; 41 : e13549 . 10.1200/JCO.2023.41.16_suppl.e13549
6 Wong AR , Sun V , George K , et al Barriers to participation in therapeutic clinical trials as perceived by community oncologists . JCO Oncol Pract . 2020 ; 16 : e849 - e858 . 32240068 10.1200/JOP.19.00662 PMC7489483
7 Ebrahimi H , Megally S , Plotkin E , et al Barriers to clinical trial implementation among community care centers . JAMA Netw Open . 2024 ; 7 : e248739 . 10.1001/jamanetworkopen.2024.8739 38683608 PMC11059033
8 Unger JM , Shulman LN , Facktor MA , Nelson H , Fleury ME. National estimates of the participation of patients with cancer in clinical research studies based on commission on cancer accreditation data . J Clin Oncol . 2024 ; 42 : 2139 - 2148 . 10.1200/JCO.23.01030 38564681 PMC11191051
9 Hillyer GC , Beauchemin M , Hershman DL , et al Discordant attitudes and beliefs about cancer clinical trial participation between physicians, research staff, and cancer patients . Clin Trials . 2020 ; 17 : 184 - 194 . 32009456 10.1177/1740774520901514 PMC7211123
10 Knelson LP , Cukras AR , Savoie J , et al Barriers to clinical trial accrual: perspectives of community-based providers . Clin Breast Cancer . 2020 ; 20 : 395 - 401.e3 . 32605813 10.1016/j.clbc.2020.05.001
11 Comis RL , Miller JD , Colaizzi DD , Kimmel LG. Physician-related factors involved in patient decisions to enroll onto cancer clinical trials . J Oncol Pract . 2009 ; 5 : 50 - 56 . 20856718 10.1200/JOP.0922001 PMC2790653
12 Albrecht TL , Eggly SS , Gleason ME , et al Influence of clinical communication on patients’ decision making on participation in clinical trials . J Clin Oncol . 2008 ; 26 : 2666 - 2673 . 10.1200/JCO.2007.14.8114 PMID: 18509178; PMCID: PMC3807688. 18509178 PMC3807688
13 Farrar N , Elliott D , Jepson M , et al The role of healthcare professionals’ communication in trial participation decisions: a qualitative investigation of recruitment consultations and patient interviews across three RCTs . Trials . 2024 ; 25 : 829 . 39695876 10.1186/s13063-024-08656-y PMC11653767
14 Coffey T , Duncan E , Morgan H , Gillies K. What influences communication about retention in randomised trials: a multi-trial, theory-based analysis exploring trial staff perspectives . BMC Med Res Methodol . 2022 ; 22 : 231 . 36002801 10.1186/s12874-022-01708-4 PMC9404662
15 Leggett H , Vinall-Collier K , Csikar J , et al Legacy lessons from the COVID-19 era to improve trial participation and retention: views from trial participants, PPIE contributors and trial staff across the NIHR portfolio . PLoS One . 2024 ; 19 : e0296343 . 38381743 10.1371/journal.pone.0296343 PMC10880997
16 Abshire M , Dinglas VD , Cajita MI , Eakin MN , Needham DM , Himmelfarb CD. Participant retention practices in longitudinal clinical research studies with high retention rates . BMC Med Res Methodol . 2017 ; 17 : 30 . 28219336 10.1186/s12874-017-0310-z PMC5319074
17 Chhatre S , Jefferson A , Cook R , et al Patient-centered recruitment and retention for a randomized controlled study . Trials . 2018 ; 19 : 205 . 29587805 10.1186/s13063-018-2578-7 PMC5870194
18 Viergever RF , Karam G , Reis A , Ghersi D. The quality of registration of clinical trials: still a problem . PLoS One . 2014 ; 9 : e84727 . 24427293 10.1371/journal.pone.0084727 PMC3888400
19 Issa M , Yang O , Doeve J , Arcot S , Chang A , Amara S , Bhakta SH , Baber I , Shali K , Dweik Q , Hung TKW. Development and evaluation of a human-in-the-loop data curation training program to support a digital clinical trial platform: descriptive feasibility study . JMIR Form Res. 2026 ; 10 : e81257 . 10.2196/81257 PMID: 41880602; PMCID: PMC13016434. 41880602 PMC13016434
20 Hung T , Mao JJ , Ho AL , et al Real-world evaluation of user engagement with an artificial intelligence-powered clinical trial application in oncology . J Am Med Inform Assoc . 2025 ; 32 : 1562 - 1569 . 10.1093/jamia/ocaf129 40720634 PMC12451935
21 Hung T , Kuperman GJ , Sherman EJ , Ho AL , Weng C , Pfister DG , Mao JJ. Performance of retrieval-augmented large language models to recommend head and neck cancer clinical trials . J Med Internet Res. 2024 ; 26 : e60695 . 39405514 10.2196/60695 PMC11522650
22 Pfiffner PB , Oh J , Miller TA , Mandl KD. ClinicalTrials.gov as a data source for semi-automated point-of-care trial eligibility screening . PLoS One . 2014 ; 9 : e111055 . 25334031 10.1371/journal.pone.0111055 PMC4205089
23 von Itzstein MS , Hullings M , Mayo H , Beg MS , Williams EL , Gerber DE. Application of information technology to clinical trial evaluation and enrollment: a review . JAMA Oncol . 2021 ; 7 : 1559 - 1566 . 34236403 10.1001/jamaoncol.2021.1165
24 Chow R , Midroni J , Kaur J , et al Use of artificial intelligence for cancer clinical trial enrollment: a systematic review and meta-analysis . J Natl Cancer Inst . 2023 ; 115 : 365 - 374 . 36688707 10.1093/jnci/djad013 PMC10086633
25 Chen H , Li X , He X , et al Enhancing patient-trial matching with large language models: a scoping review of emerging applications and approaches . JCO Clin Cancer Inform. 2025 ; 9 : e2500071 . 10.1200/CCI-25-00071 40489722 PMC12169815
26 Loaiza-Bonilla A , Kurnaz S , Tuysuz E , et al Transforming oncology clinical trial matching through multi-agent AI and an oncology-specific knowledge graph: a prospective evaluation in 3,800 patients . J Clin Oncol . 2025 ; 43 : 1554 . 10.1200/JCO.2025.43.16_suppl.1554
27 Cohen NA , Esfahani MS , Stehr H. Accuracy of large language models for matching cancer patients to biomarker-driven clinical trials based on molecular profiles . J Clin Oncol . 2025 ; 43 : e13627 . 10.1200/JCO.2025.43.16_suppl.e13627
28 Corneli A , Hanlen-Rosado E , McKenna K , et al Enhancing diversity and inclusion in clinical trials . Clin Pharmacol Ther . 2023 ; 113 : 489 - 499 . 10.1002/cpt.2819 Epub 2023 Jan 11. PMID: 36628990. 36628990
29 DePuccio MJ , Torrez D , Hsu E , et al ; A2CPS Consortium . A multilateral qualitative study of perspectives on enhancing clinical trial diversity among historically underrepresented groups . Clin Ther . 2025 ; 47 : 971 - 980 . 10.1016/j.clinthera.2025.06.006 40628618 PMC13151987