June 15, 2014 (Vol. 34, No. 12)
Last August, the FDA issued its “Guidance for Industry Oversight of Clinical Investigations: A Risk-Based Approach to Monitoring.” Agency officials said they believed that risk-based monitoring (RBM) could improve sponsor oversight of clinical trials.
To get a better understanding of best practices for RBM and to discover the types of technologies that are involved, GEN interviewed experts at six contract research organizations (Aptiv Solutions, Covance,
INC Research, Parexel Informatics, PPD, and Quintiles) and a company that provides information technology business consulting services (Cognizant). These experts are Rob Davie, Ph.D., Vice President and General Manager, Clinical Development Europe, Covance; Lori Eberhardt, Vice President Clinical Development, PPD; Jill Collins, Senior Director Integrated Clinical Processes, Clinical Innovation, INC Research; Barinder Marhok, Head of LIfe Sciences, R&D Practice, Cognizant; Martin Giblin, Vice President Global Data and Safety Systems, Quintiles; Drew Garty, Senior Director, Product Management, Platform Solutions, Parexel Informatics; and Andy Grieve, Ph.D., Senior Vice President Clinical Trial Methodology, Aptiv Soltions.
GEN: How do you view RBM and why is it critical to clinical trial management?
Existing clinical trial methodology is not always successful in the critical areas of patient safety and data integrity. To meet this essential pharmaceutical company need, Covance is one of a number of CROs developing RBM methodology, a holistic approach to clinical trial management that proactively identifies risk of failure.
Most important, the process embraces the entire trial process, not only clinical monitoring. Using a “quality by design” approach to remove initial risk, RBM uses adaptive methodologies to manage or mitigate emergent risk during the lifetime of the trial. Focus of resources to the most risky aspects of a clinical trial should not only allow a safer process, but may lead to potential efficiency gains, reducing the overall cost to the trial sponsor.
The industry is shifting its approach from traditional on-site monitoring practices to risk–based methods that offer an ability to increase data quality and protect patient safety, while helping to control the costs of clinical trials. Instead of all sites being managed equally, in a risk-based approach, monitoring efforts are more intensely directed to sites and subject visits with the greatest risk. There is also less emphasis on source data verification, thereby increasing clinical research associate time and priority to overall site performance, such as quality of documentation and data, adherence to protocol, and site processes and procedures.
RBM offers a leaner, more strategic way to oversee clinical investigations by allocating resources across a study based on data criticality, patient safety, data integrity, protocol compliance, and impact to operational delivery. Strategic Data Monitoring is INC’s holistic approach to risk-based monitoring.
It begins with the evaluation of the protocol risks and the planning of a fit-for-purpose monitoring strategy that includes a targeted approach for source data verification, and data-driven monitoring activities that are triggered by indicators of potential risk, as well as on-site, off-site, remote, and central monitoring. This strategic approach is critical to successful clinical trial management because it drives improvements in data integrity, while streamlining activities that focus on improving patient safety oversight and data integrity.
RBM is a strategy to focus appropriate attention toward areas of a clinical study that exhibit a defined characteristic that indicates a potential concern. The goal is to allow clinical trial managers to focus the necessary resources toward the risk and to eliminate or mitigate concerns before they become significant issues.
The FDA “encourages sponsors to develop monitoring plans that manage important risks to human subjects and data quality and address the challenges of oversight in part by taking advantage of the innovations in modern clinical trials.” A successful RBM program improves the quality and efficiencies of sponsor oversight and facilitates continual improvement within the project.
RBM is a method of conducting clinical trials and leveraging a combination of on-site and remote processes supported by data-driven insights to deliver optimized trial execution—all while maintaining, or enhancing, the quality of data and patient safety. In fact, the name risk-based monitoring can be misleading because we’re actually de-risking our monitoring by using technology and data analysis to better understand our trials and manage potential risks.
RBM is important because it’s imperative for sponsors to demonstrate adequate oversight. By tracking changes in study data over time, and comparing the changes to the level of monitoring activity, CROs can show both sponsors and regulatory authorities that monitoring is being carried out in a responsible and effective manner.
A risk-based approach to monitoring is not intended to imply either that monitoring of clinical trials data is unimportant, or that sponsors can be lax in overseeing clinical studies. Instead, it focuses sponsor oversight activities on the prevention or reduction of important risks to data quality and in particular to those processes which are critical to the protection of subjects and to the integrity of the study. Moreover, RBM is dynamic and can therefore be part of a process of continuous improvement in study conduct.
GEN: What kinds of technology and methodologies do CROs need to employ to conduct efficient RBM?
RBM relies on technology with the capability to collect, integrate, store, report, and analyze big data from clinical trials. The RBM processes should incorporate standardized methodology to proactively identify potential risks in areas such as protocol design, site management, and emergent risk assessment. This should allow interpretation of data by cross-functional teams of physicians, statisticians, and clinical scientists.
Typically, these experts would review trends, out-of-tolerance outliers, and other concerns within the data. To deliver the trial in a state of control, the project team should be able to make decisions, based on this interpretation, to intervene in a timely manner with ongoing centralized data and risk review, closed-loop issue management, and real-time adaptive monitoring focusing on critical data.
RBM requires technologies that compile data from multiple systems and sources. Many CROs have already developed such innovations. Solutions must be capable of receiving all the data in standardized format so that regardless of the source (e.g., third party vendors, multiple CROs), the collective information can be imaged back to various teams in the form of dashboards and reports for early identification of risks and for more immediate decisions and actions to improve study quality and efficiency.
RBM methodology begins with an assessment of the protocol when each monitoring plan is customized based on a detailed evaluation of the protocol risks. Utilizing analytic dashboards, such as PPD’s Preclarus™, clinical teams identify data anomalies, view site health assessments, and analyze key risk indicators. Remote and on-site visit schedules are subsequently risk-adjusted through the life of the trial using both qualitative clinical assessments and quantitative reports.
When it comes to developing a monitoring plan, there is no one-size-fits-all. However, CROs that apply a holistic approach to their operational strategy—one that begins with well-defined plans and processes, clearly articulated roles and responsibilities, and advanced technology that allows the right role to take the right action at the right time—can ensure more efficient RBM.
INC’s strategic data monitoring approach includes an operational plan containing the rationale for the monitoring strategy based on the risk evaluation, resources, and process workflows using data to drive actions. A key component in facilitating the workflow. We also apply our ability to integrate data from disparate sources and the application of statistical data analytics, enabling the collection and analysis of data for early identification and mitigation of potential risks and trends. Sophisticated electronic data capture technology enables this approach by providing efficient targeted source data verification functionality.
Traditional RBM techniques require the manual collection and analysis of data from multiple clinical systems and manual analysis of key risk indicators. CROs define particular characteristics, such as the number of adverse events associated with a site and focus attention if a defined alert level is exceeded. Additionally, several risk categories can be combined to provide an overall view of the study.
Once the potential risk is identified, the CRO can respond to it. Unfortunately, risk mitigation strategies have little value unless they are executed, monitored and analyzed continuously throughout the clinical trial lifecycle. The obvious solution is to use an automated tool to review clinical data and flag areas of concern based on defined metrics in real time.
Cognizant has developed SmartTrials which accelerates the gathering and refining of clinical trial data. SmartTrials provides real-time, continuously analyzed data and configured workflows that greatly reduce or eliminate the potential for individual bias in issue management and decision making.
The deployment and configuration of any technology supporting RBM must be based on the comprehensive risk assessment performed at the outset of the trial. Then, an effective RBM solution should incorporate three technology components: data integration from multiple trial data sources to provide a comprehensive insight into activity at the site and patient level; a carefully designed workflow that enables the CRO to manage appropriate actions based on that data; and advanced data analytics available in real-time that create visual analyses to identify trends, safety signals, and patterns.
Quintiles’ approach to RBM, Data-driven Trial Execution (DTE), harnesses the power of our Quintiles Infosario® technology platform to capture each of these elements and more. We see RBM as a transformational shift for our industry, which is why DTE reflects a holistic approach to RBM, leveraging risk-based strategies across a study’s lifecycle to extract the maximum knowledge about our sites and patients.
To implement a data-driven approach to monitoring, CROs must use data from many sources and technology solutions. At a minimum, these should include CTMS, EDC, and Safety, although data points from other systems may also be useful in calculating site risk. Parexel’s Data-Driven Monitoring solution captures and calculates raw data, and presents it in a way that is easy for the end user to understand and respond to. Being able to rely on consistent, up-to-date, and quality data is crucial for implementing a comprehensive data-driven approach.
One-hundred percent source data verification (SDV) is costly and, for clinical sites that produce high-quality data, unnecessary—it would require the verification of vast amounts of good data. The use of a statistical sampling approach can substantially reduce SDV workload, increasing productivity and allowing monitors to spend time on activities that will improve the performance of clinical sites.
These estimates could be used for future site-selection. In addition, such on-site data verification should accompany effective measures to remotely oversee data and site quality using modern EDC systems capabilities and close, consistent site contact and management, delivering a site performance matrix that focusses attention and resources to critical sites and issues.
This story has been corrected from an earlier version, which mistakenly identified PPD's Vice President of Clinical Development Lori Eberhardt as “Lee Eberhardt”. GEN regrets the error.