International Council for Harmonisation Integrated Addendum to ICH E6 R1 - otherwise known as ICH E6 R21 - suggests centralized monitoring be incorporated as an important monitoring methodology in clinical trials. Different companies are taking different approaches to deploy centralized monitoring and these differences will require more detailed discussions when outsourcing clinical trial services.
In this article, we will cover some of the complexities observed in contracting for centralized monitoring services, and we will address some of the types of information that should be considered in outsourcing decisions, including additional details that may need to be specified in requests for proposals (RFP).
An Evolving Process
It is important to understand the evolution of centralized monitoring to realize that it has been a dynamic concept particularly within the last three years. In 2013, the U.S. Food and Drug Administration (FDA) introduced centralized monitoring by stating that “Centralized monitoring is a remote evaluation carried out by sponsor personnel or representatives (e.g., clinical monitors, data management personnel, or statisticians) at a location other than the sites at which the clinical investigation is being conducted. Centralized monitoring processes can provide many of the capabilities of on-site monitoring as well as additional capabilities. FDA encourages greater use of centralized monitoring practices, where appropriate, than has been the case historically, with correspondingly less emphasis on on-site monitoring.”2
TransCelerate, a non-profit organization of biopharmaceutical companies working to identify and recommend solutions for challenges in the implementation of research and development, published a paper in 20143 on defining a centralized monitoring capability, which looked at different organizational structures. In 2015, TransCelerate followed with a second paper,4 this time looking at the people, the processes and the types of data used to conduct centralized monitoring. In November 2016, the International Conference on Harmonisation (ICH) released ICH E6 R2, its final guidance for good clinical practice, which was amended to include quality management and centralized monitoring.
In ICH E6 R2 Section 5.18.3, centralized monitoring is described as a monitoring method that may incorporate the following elements:
- Identify missing data, inconsistent data, data outliers, unexpected lack of variability and protocol deviations.
- Examine data trends such as the range consistency, and variability of data within and across sites.
- Evaluate for systematic or significant errors in data collection and reporting at a site or across sites; or potential data manipulation or data integrity problems.
- Analyze site characteristics and performance metrics.
- Select sites and/or processes for targeted on-site monitoring.
Many of these elements have been conducted as part of standard data management, clinical monitoring, biostatistics or medical reviews for many years. However, these elements often are conducted manually, in isolation or at the end of the trial just prior to (or sometimes after) database lock.
Assessing Future Practices
Today the industry is evolving as to when and how these types of analyses are conducted. The prevailing trend has been to move toward tools and dashboards that use statistical methods to alert users to anomalies early, often and through different lenses. This addendum to ICH E6 R1 helps to advance these best practices as industry standards, while the specific wording offers flexibility for companies to use one, all or a combination of these elements (or even additional elements) to form a centralized monitoring approach.
Most sponsors and CROs will be assessing these elements and comparing them to their existing cross-functional data cleaning and monitoring practices and refining or establishing their centralized monitoring approaches, as well as evaluating the capabilities of their technology vendors to enhance delivery using improved tools. In addition, many of these same entities also will be assessing their organizational structures to support centralized monitoring, while CROs will be determining if they should include centralized monitoring as a stand-alone service offering, which will increase the complexities associated with outsourcing services.
ICH E6 R2 guidance implies that centralized monitoring can be performed to supplement on-site monitoring or, in some cases, to replace it. As a result, we can expect to see centralized monitoring as a service offering with deliverables such as a centralized monitoring plan and centralized monitoring reports at certain frequencies. Table 1 illustrates some of the different models emerging to support a centralized monitoring capability.
Table 1
CROs are developing organizational structures and processes that can be adapted to the use of homegrown or external tools and dashboards. Sponsors are creating internal centralized monitoring functions and developing proprietary tools and dashboards or selecting and using a single external tool or dashboard. Technology vendors are developing tools and dashboards, but lack alignment with specific roles, processes and workflows.
A Few Typical Scenarios
Sponsors will need to determine not only which of these elements their CRO and technology partners use to facilitate a centralized monitoring approach when outsourcing centralized monitoring, clinical monitoring or even data management services, but they will also need to determine how their CRO partners are organized structurally and in which functional service each of these elements are delivered. Let’s explore a few scenarios as they relate to Table 1.
CRO A chooses to execute pieces of centralized monitoring with the tools and processes that exist within a number of different functions, such as data management, clinical management, biostatistics, data analytics and pharmacovigilance. This approach works well when clinical management, data management and biostatistics services are sourced either through different organizations or all within the same organization. The benefit of this segregated approach allows the CRO to coordinate centralized monitoring with the services it has been contracted to perform. This method reduces the need for frequent transfers of data and the cost associated with data transfers, as well as the potential duplication of effort across functions. For example, components A-C in Table 1 might best be performed by the group or organization that has access to the electronic data capture (EDC) datasets, while components D and E might be best conducted by the group or organization that has access to the clinical trial management system (CTMS) data or operational metrics. Careful delineation of who does what and when, how findings are shared with clinical management for action and follow-up, and how reporting will be handled are critical elements when this model is used. This approach offers the most flexibility, but requires significant discussions and planning to prevent duplication or gaps in coverage. In this model, CRAs may retain accountability for finding issues and fixing them through process improvement at the site. Other models might impact the role of the CRA. Questions to assist in navigating this structure include: Is there a difference in centralized monitoring and remote monitoring conducted by the clinical management team? Do all functions write a report outlining their activities and findings or does one function consolidate activities and findings for a centralized monitoring report? When does each function complete its activities and how can we limit escalations of findings to sites to specific points in time?
CRO B decides to form a new and separate function to conduct centralized monitoring. The benefits of this model are maximum efficiency in execution by reducing any potential for duplication across functions. In this model, centralized monitoring services can be contracted as part of a full-service outsourcing to a single provider or could be contracted separately from clinical monitoring or data management services. When this model is used, it is possible that clinical management and data management services could be supported by different CROs that use different models for centralized monitoring. As a result, detailed discussions around roles and responsibilities are key. For example, if centralized monitoring is contracted as a separate service, do the clinical monitoring and data management teams need to perform listing reviews? Are there other standard data cleaning and monitoring tasks within clinical management and data management that should no longer occur to reduce duplication of effort? Will the centralized monitoring team contact the site to review findings or will those contacts remain the responsibility of the clinical monitoring team via remote monitoring visits to work with sites on root cause and resolutions? Will the clinical monitoring team take direction from the centralized monitoring team on when to perform on-site or remote monitoring visits (MV) or what data to review when on-site? At what frequency will findings be escalated to the clinical team for action, and how can the clinical team minimize the number and frequency of site contacts to investigate and resolve issues? This model might remove the responsibility of finding issues from the CRA and better focus CRAs on fixing issues uncovered during centralized monitoring through process improvement with the sites. This model most likely will require data transfers to power the analytical tools used and may necessitate new contracted units for deliverables such as data transfers, technology licensing, customization of key risk indicators (KRI) and centralized monitoring reports.
CRO C decides to execute centralized monitoring responsibilities within data management. In this scenario, it only may be possible for the centralized monitoring to occur when data management services also have been contracted. This model might remove the CRA’s responsibility of finding issues and instead focus them on fixing issues uncovered during centralized monitoring through process improvement with the sites. This reduces the need for data transfers, but might require contracted units for technology licensing and centralized monitoring reports. In this model, components A-C in the guidance may be very well managed, but components D and E may be best conducted by the clinical monitoring function as these activities impact on-site and remote monitoring visit decisions. Questions similar to those above will need to be addressed. However, additional questions might arise, such as: Will the data management team investigate findings or pass those on to clinical monitoring for investigation? What systems will be used to track findings and outcome of investigations and how will those be shared with clinical management for follow-up? If clinical monitoring is contracted to a separate CRO, how many investigations or escalations might they expect to see and at what frequency? If centralized monitoring also includes many remote data review activities as well, are remote monitoring visits by the clinical team still required to communicate findings with sites or do they become obsolete? The answers to these questions will certainly have an impact on the effort and resources needed by the CRO performing the clinical monitoring function.
With many small and large pharma creating centralized monitoring functions of their own, and with each of them incorporating different elements of ICH E6 R2, Section 5.18.3 A-E, even typical outsourcing of data management and clinical monitoring services can become more complicated. For example, will the sponsor’s centralized monitoring teams issue queries and track deviations? Will they share centralized monitoring reports with data management and clinical monitoring staff? Will the sponsor’s centralized monitoring teams contact sites directly? Will they investigate findings or forward them to data management and clinical monitoring teams? At what frequency might data management and clinical monitoring teams expect to be escalated findings for investigation, how will they report back and how many escalations might they expect to receive? Will these escalations direct the focus of on-site and remote MVs down to the specific data level or perhaps just direct the timing of the visits and the method of the visit (on-site or remote)? Will sponsor tools and dashboards for centralized monitoring need to be used/accessed by CRO staff and which roles in particular? Will those roles needing access require extensive training and will they even align with how the CROs use those roles to conduct centralized monitoring?
What Can We Expect?
While many questions remain about how monitoring will be conducted in these new models, there are some things that we can expect to see. Outsourcing strategies likely will need to change to account for the complexities introduced by ICH E6 R2. What has typically been relegated to clinical monitoring and data management, respectively, may no longer be the case. Data management staff, clinical monitoring staff and centralized monitoring staff now may have overlapping responsibilities. Detailed communications between sponsors and all CROs involved in providing services for a clinical trial will need to have detailed discussions around structure, roles and accountabilities as models evolve to align with ICH guidance.
Here are a few key recommendations to aid in aligning outsourcing decisions with the emerging centralized monitoring service offerings:
- Discuss in detail with CRO partners/vendors and agree on:
- Terminology (remote vs. centralized monitoring)
- Centralized monitoring plan (roles, workflow, frequency, etc.)
- Reporting (communication and tracking of findings)
- Clear delineation of tasks to avoid overlap (data management, clinical management, data review, pharmacovigilance, etc.)
- Discuss access to data and required system access for various roles
- Consider new RFP specifications concerning:
- Centralized monitoring plan development
- Centralized monitoring units and reports
- Number of alerts/triggers needing investigation
- Make RBM expert-to-expert connections, which may require including several people from different functions
The most important concept to note is that many of the tried-and-true standard services, roles, procedures, tools and communication pathways that have been used over the last 30 years are changing. They are changing in different ways and at a different pace across organizations. Very detailed discussions with contributing vendors will be necessary to ensure the services are packaged and grouped appropriately to address sponsor expectations around centralized monitoring as the practices and processes continue to evolve.
References
- http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E6/ E6_R2__Step_4.pdf
- http://www.fda.gov/downloads/Drugs/Guidances/UCM269919.pdf
- http://journals.sagepub.com/doi/pdf/10.1177/2168479014546335
- http://journals.sagepub.com/doi/pdf/10.1177/2168479015618696