Clinical Trial Screening in Gynecologic Oncology: Defining the Need and Identifying Best Practices

Clinical trial recruitment and enrollment are multifaceted and require careful representation of diverse populations; clinical trials focusing on gynecologic cancer tend to be underfunded compared with other conditions, thus underscoring the need for established best practices when it comes to clinical trial recruitment and enrollment. Screening for clinical trials varies based on site, but many rely on manual processes that are not streamlined or standardized, and which can introduce screening bias. The best screening processes will focus on reducing bias and be based on real-world implementation strategies, efficiency, and cost-effectiveness to make clinical trials more equitable and generalizable. This article is a review focusing on current practices in gynecologic oncology for clinical trial screening, including the need for standardization, the future of screening methods, and lessons learned from the survey of current practices. Not having standardized methods for clinical trial screening represents significantly increased costs to patients, scientists, and institutions in terms of both time and money. These increased costs can include less effective use of resources that are allocated to clinical trials, downstream diminishing returns and early termination of clinical trials, less than optimal participation or sample sizes, and the effort and cost of recruitment methods that are not standardized. Each of these costs, in turn, costs investors and funding organizations money due to reduced efficiency and productivity. Evidence for best practice in clinical trial screening for gynecologic cancer is sparse and incomplete, but site-specific data are sometimes available. One study showed that implementing clinical trial prescreening for all patients led to a 2.5-fold increase in screening in general and a 5-fold increase in screening for minority patients. Additionally, accelerated prescreening increased patient understanding and screening success. Other studies have shown similar results with simple interventions such as prescreening or patient education. A survey among gynecologic oncologists surrounding clinical trial screening practices showed that most clinics used manual screening processes, with a few incorporating electronic health records (EHRs) or artificial intelligence (AI). Patients infrequently request screening themselves, and the responsibility for screening was imposed on the treating clinician in roughly 30% of cases; the majority of screenings by far were undertaken by clinical research coordinators or other research personnel (63%). Some clinical trials have begun using AI to assist in screening processes, which shows great potential in the ability to identify patients using EHR information. An AI approach can identify patients with more accuracy and drastically reduce screening costs. There are some concerns, however, about patient privacy and security as well as bias that remains in AI as it is continuing to develop. Other tools that have potential to enhance clinical trial screening include next-generation sequencing and trial matching with biomarkers, as well as open-access clinical trial matching programs. Lessons learned from this review include the possibility of screening that is what the authors call "site agnostic,"meaning it is standardized across sites to optimize time and financial resources. Additionally, metrics that pertain to organizations or specialties may provide opportunities to streamline screening processes. Institutions can also independently assess screening processes and implement targeted interventions to improve clinical trial screening and increase the diversity of patients enrolled in clinical trials. The authors conclude that best practices should include awareness at the site level of screening practices and failures, engaging with partners and sponsors with respect to screening and overall budget, using technology and other solutions to reduce cost while protecting patient privacy and define screening processes that are objective and not subject to bias.