Guidance is Applicable to Clinical Investigations of Human Drugs, Biologics, Medical Devices, and Combination Products
On August 7, 2013, the Food and Drug Administration (FDA) announced in the Federal Register1 the availability of a final guidance entitled “Oversight of Clinical Investigations–A Risk-Based Approach to Monitoring.” The intent of the guidance is to assist sponsors of FDA-regulated clinical investigations to develop and implement monitoring strategies that focus on the most critical components of study conduct and reporting, in order to improve human subject protection and the quality of data. The guidance represents FDA’s most current thinking regarding the application of risk-based principles to the design and implementation of study monitoring plans, as well as regarding the incorporation of centralized monitoring techniques.
Sponsors are required to monitor the performance of clinical investigators and the progress of investigations in Investigational New Drug (IND) studies pursuant to 21 C.F.R. §§ 312.50, 312.53(d), and 312.56(a), and Investigational Device Exemption (IDE) studies pursuant to 21 C.F.R. §§ 812.40, 812.43(d), and 812.46. However, the regulations do not specify the frequency of monitoring or methods to be used by sponsors to ensure the protection of human subjects as well as the integrity of the data. In this guidance, FDA makes clear that sponsors can use a variety of approaches to meet their legal responsibilities for monitoring. However, FDA observes that the evolution of modern clinical trials—including complexity, disparate geographies, variability in investigators’ experience, as well as the availability of electronic systems and records—may favor a “quality risk management” approach that addresses the critical parameters of individual studies. FDA also specifically encourages the increased use of centralized electronic monitoring systems, where appropriate, as an adjunct to, or for partial replacement of, traditional on-site monitoring. The guidance clarifies that centralized monitoring, to the extent appropriate and feasible, may be particularly useful in monitoring study quality by identifying “outlier” trends in data (e.g., missing data or minimal variance in data) and site performance (e.g., high rate of subject withdrawals or eligibility violations), as well as verifying completion of administrative and regulatory activities (e.g., initial and continuing review by IRBs).
FDA’s guidance was informed in part by a survey2 of monitoring practices carried out by the Clinical Trials Transformation Initiative (CTTI), a public-private partnership of FDA and other entities including manufacturers. The guidance acknowledges that, although surveyed participants employed a variety of techniques, the typical monitoring strategy used by manufacturers was on-site monitoring, including review of data for each enrolled subject at several week intervals, because of the perception that this approach has “historically has been FDA’s preferred way for sponsors to meet their monitoring obligations.” FDA discusses that the guidance was also informed by a PhRMA analysis of monitoring practices3; however, no input from medical device professional organizations is cited. King & Spalding observes, therefore, that although FDA deems the general principles and strategies of risk-based monitoring in the guidance to be applicable to both drug and medical device trials, the guidance may not have been fully informed by a contemporary, in-depth analysis of best practices in monitoring, or the use of centralized monitoring, in medical device trials.
FDA observes that the agency’s bioresearch monitoring (BIMO) compliance guidance manuals are compatible with the approaches discussed in the guidance; however, the manual addressing sponsors, contract research organizations (CROs), and monitors (CPGM 7348.810) refers to site visits but does not discuss the inclusion of centralized monitoring.
Risk-Based Monitoring: Critical data and Processes to be Monitored
FDA advises that a risk-based monitoring plan, which is study-specific, should be designed to mitigate important and likely sources of error in obtaining and reporting critical data as well as error in processes that are needed for human subject protection and trial integrity. A risk-based plan should be based on:
- prospective identification of critical data and processes,
- performance of a risk assessment to identify and understand the risks that could impact critical data collection or processes, and
- development of a monitoring plan that focuses on important and likely risks to critical data and processes.
In the development of the plan, the identified risks should be assessed and prioritized by considering the likelihood of errors occurring, the impact on human subject protection and trial integrity, and the extent to which such errors would be detectable. The sponsor should also consider the types and intensity of monitoring that is needed to address the identified risks. The sponsor is advised to consider whether an identified risk could be removed by modifying the study protocol rather than being managed by monitoring activities.
FDA provides the following examples of data, processes, and procedures that should “ordinarily be identified as critical” during the risk assessment.
- Verification that informed consent was obtained appropriately.
- Adherence to protocol eligibility criteria.
- Appropriate accountability and administration of the investigational product (e.g., integrity of randomization at the site level, if applicable).
- Conduct of procedures and assessments related to:
- Study endpoints;
- Protocol-required safety assessments;
- Evaluating, documenting, and reporting adverse events, including serious adverse events and unanticipated adverse device effects, deaths, and withdrawals from the study, especially when a withdrawal may be related to an adverse event; and
- Conduct and documentation of procedures essential to trial integrity (e.g., ensuring that the blind is maintained and appropriate referral of specific events for adjudication).
During the risk assessment, FDA recommends that the sponsor consider additional factors in considering the type (e.g., on-site, centralized, or mixed), frequency, and extent (e.g., 100% verification, targeted, or random review of data) of monitoring. The agency cites several factors that may require more intensive on-site monitoring, including (1) complex study design, (2) study endpoints that are subjective, rather than objective and more suitable for remote verification (e.g., death, hospitalization, standardized laboratory data), (3) study endpoints where subject withdrawal or lack of follow-up may impede study evaluation, (4) complex study population involving subjects that are seriously ill or vulnerable, (5) diverse geographic settings with differences in standards of medical practice or “less established clinical trial infrastructure,” (6) less experienced clinical investigators, including investigators in a medical device study using a novel or innovative medical device or surgical procedure associated with the device, (7) study of a drug or device with significant safety concerns or for which there is no prior clinical trial experience, and (8) early stages of the study (e.g., where a tapered approach of more intensive monitoring at study initiation might be followed with less intensive monitoring once procedures are established).
The agency also observes that the availability of electronic data capture and the capability for centralized real-time analysis to assess quality metrics (e.g., rates of missing data, data errors, protocol violations) may help sponsors identify higher risk sites to be targeted for more intensive monitoring as the trial progresses. As the quantity of data collected increases (e.g., due to large size or duration of the study), centralized monitoring may also be useful.
Components of the Monitoring Plan
FDA advises that the monitoring plan should include a brief description of the study, its objectives, and the critical data and procedures, including those that are outside of routine clinical practice and require study staff training. The plan should identify the specific risks to be addressed by monitoring and describe (1) the monitoring methods, (2) responsibilities, and (3) requirements for the trial. The plan, which is specific for the individual study, may also reference other existing standard operating procedures (e.g., those that describe general monitoring processes or the analysis and resolution of issues identified by monitoring). All sponsor and CRO personnel involved with monitoring or determining actions to be taken on issues should review the plan and associated documents.
FDA observes that sponsors of medical devices are required to submit written monitoring procedures as a component of an IDE application. Sponsors of medical device clinical investigations may solicit feedback on proposed monitoring processes prior to the submission of an IDE by submitting a Pre-Submission4 or contacting CDRH’s Division of Bioresearch Monitoring. Sponsors of drug clinical studies may raise specific questions about the monitoring plan in a request to CDER or CBER for a formal End-of-Phase 2 meeting.
The guidance provides specific recommendations regarding the components of a risk-based monitoring plan, including:
- Description of Monitoring Approaches, including:
- Description of monitoring methods and how they will be used;
- Criteria for timing, frequency, and extent of planned monitoring activities;
- Specific activities required for each monitoring method (e.g., reference to tools, logs, or templates);
- Identification of possible deviations critical to study integrity and how they are to be recorded and reported;
- Documentation of monitoring activities, whether conducted on-site or centrally, which should generally include sufficient detail to allow verification that the monitoring plan was followed:
- Date of the activity and the individuals participating in the monitoring;
- Summary of data or activities reviewed;
- Description of any observed or potential noncompliance, data irregularities, or other deficiencies;
- Description of any actions taken, to be taken or recommended, including the responsible party.
- Communication of Monitoring Results, including:
- Format, content, timing, and archiving of monitoring documentation and reports;
- Processes for communication, including routine reporting to management and other clinical trial functions, as well as immediate reporting of significant monitoring issues to FDA and others.
- Management of Noncompliance, including:
- Processes for addressing unresolved or significant issues, including those at single sites or systemically across sites;
- Ensuring that root cause analyses are conducted for important deviations and that corrective and preventative actions are implemented;
- Incorporation of other quality management practices applicable to clinical investigations (e.g., reference to other company documents regarding appropriate actions for non-compliance).
- Ensuring Quality Monitoring, including:
- Specification of training required for personnel involved in monitoring activities;
- Planned audits to ensure that the sponsor and monitors (including CRO staff) conduct monitoring in accordance with the plan and applicable regulations, and that the monitoring is effective in ensuring human subject protection and data integrity.
- Amending the Monitoring Plan, including criteria for review and potential amendment (e.g., a protocol amendment).
FDA observes that actions to ensure study quality should include other activities in addition to monitoring, including the design of protocols that are not ambiguous or unnecessarily complex, and ensuring that the Case Report Forms (CRFs) accurately capture the data required by the protocol. FDA cautions that on-site monitoring visits have traditionally played a major role in clinical investigator training at study start-up and during the study. Thus, sponsors who plan less intense or limited on-site monitoring should consider additional methods for providing and documenting appropriate training and feedback of monitoring findings for investigators and their study personnel.
Delegating Monitoring Responsibilities to a CRO
The guidance observes that, if a sponsor of an IND delegates monitoring responsibilities to a CRO pursuant to 21 C.F.R. § 312.52, the regulations require a written transfer of obligations and require the CRO to comply with the regulations. FDA cautions that sponsors retain responsibility for overseeing the work conducted by a CRO that has assumed monitoring obligations, and advises that a sponsor should evaluate the CRO’s compliance with regulatory requirements and contractual obligations in an ongoing manner during the study. Sponsors and CROs should also have procedures in place for timely exchange of relevant information (e.g., significant monitoring observations or changes in risk for the study). FDA reminds medical device manufacturers that the IDE regulations under 21 C.F.R. Part 812 do not contain any provision for delegation of monitoring obligations to a CRO.
Considerations for Manufacturers
Manufacturers who intend to sponsor an FDA-regulated clinical investigations now have more clarity regarding the agency’s current expectations regarding the development of a risk-based monitoring plan, as well as the potential for greater use of centralized monitoring. As part of the monitoring plan, sponsors should ensure rigorous preparation of the risk assessment document that defines the data and the processes critical for the specific clinical investigation; this risk assessment is expected to provide the basis for the design and implementation of the monitoring plan. In addition, sponsors should carefully consider seeking FDA review of a novel risk-based monitoring plan prior to IND or IDE submission, especially if the planned clinical investigation is characterized by any of the specific risk factors identified in the guidance, or if the monitoring plan will deviate from the company’s customary practices in the degree of reliance on centralized rather than on-site monitoring.
Sponsors should also be aware that the guidance includes some elements that are not yet a component of the FDA’s Good Clinical Practice regulations. As an example, FDA recommends the use of a quality system process, to conduct a formal root cause analysis and implement corrective and preventive actions when monitoring identifies important deviations in clinical trial conduct. The guidance also discusses the use of planned audits of monitoring functions. We recommend that sponsors consider formal training and processes for both activities pertinent to clinical investigations if such procedures are not yet in place.