The Conduct of Clinical Trials: A Food and Drug Agency Perspective : ASAIO Journal

Secondary Logo

Journal Logo


The Conduct of Clinical Trials: A Food and Drug Agency Perspective

Sapirstein, Wolf

Author Information
  • Free

This review considers the conduct of clinical trials as they relate to the regulation of medical devices by the Center for Devices and Radiological Health. The Food and Drug Administration (FDA) is mandated by Congress to ensure devices marketed to the health care environment are safe and effective as determined from valid scientific evidence. The clinical trial provides the seminal evidence for this determination. Trials using devices, particularly if implanted, will of necessity differ significantly from those used for drug evaluations. The Clinical Trial provides pivotal evidence for the safety and effectiveness of an intervention, especially when in vivo and in vitro evaluations have been exhausted with unresolved clinical questions remaining.

Regulatory Requirements for Conduct of Clinical Trials

The regulation of medical devices by the FDA, and the requirement for clinical trial-based data in specific situations, has undergone incremental changes in the 30 or so years since the enactment of the Medical Device Amendment of 1967 to the Food, Drug, and Cosmetics Act of 1938. The Safe Medical Device Act, amended in 1992, and the Food and Drug Administration Modernization Act of 1997 (FDAMA), have codified these changes, which are primarily aimed at streamlining the review process and reducing the delays in availability of technological advances to health care providers. However, where clinical data is necessary, the clinical study of design appropriate to provide clear evidence of safety and effectiveness remains the keystone in the bridge to device approval.

The information derived from a clinical study must be compared with control data derived from the best available alternative therapy. When no treatment alternative exists, the natural history of the condition on which the device is intended to impact should serve as the control.

Clinical data to support approval for marketing a device, via the Pre Marketing Approval (PMA) process, or clearance, by demonstrating substantial equivalence to a marketed device, will require an Investigational Device Exemption (IDE) to use an unapproved device in a human study. The FDA will approve such IDE only after satisfactory review of the detailed protocol submitted for the proposed study. The intensity of this review is second only to that undertaken for evaluating the outcome of the study itself in determining device approval for marketing. This is reflected in the 30–40% disapproval of first time submissions of applications for IDEs to the Division of Cardiovascular and Respiratory Devices. To improve this procedural outcome, the Division has introduced a pre–IDE review process, encouraging sponsors to provide preliminary plans for a study protocol. The Agency will engage in an interactive critique of the pre–IDE proposal with the intention that a carefully crafted study design will be generated. This should provide the pathway most likely to provide assurance to the FDA for safety in marketing, and at the same time provide data necessary for device approval that satisfies the sponsor’s desire for the most expeditious route to marketing approval with the desired labeling indications. In an IDE application, the protocol for conduct of a clinical trial that the FDA will review, should be constructed of the following components.

Background Information

This section should provide an outline of the utility envisioned for the device with relevant literature that supports this thesis. The result of relevant in vivo and in vitro testing, including biocompatibility and bench testing, must be sufficiently detailed to instill confidence that risk associated with human application is acceptable and the low probability of an adverse event occurring is outweighed by the perceived clinical benefits for the intervention.

All available clinical experience with the device for other indications or for use outside the United States for indications similar to that intended in the study, should be submitted. This includes all applicable literature reports.

Clinical Trial Protocol

The submission must provide a clear statement of the objective sought from the study and the derived hypotheses that will be tested to validate achievement of that objective.

The actual trial design must be such that the data derived will be free of bias and sufficiently robust to support claims that the device, for its intended use, will demonstrate either equivalence or superiority to the control treatment. When the alternative therapy is non–interventional, e.g., medical, the need for surgery and/or device implantation will generally require conduct of a superiority trial to justify the possible morbidity or even mortality with use of the device. An equivalence study design on the other hand, could be acceptable for comparison with an alternative device intervention, particularly if other benefit advantage is anticipated, e.g., a lower adverse event rate.

Depending upon several factors, such as the knowledge base that exists concerning the study objective, the natural life history of the clinical condition being addressed, the efficacy of alternative therapies, and finally, the potential impact on public health of the device, the design for conduct of the study may be experimental or observational. An appropriate control will however always be required, the type being dependent upon the same factors determining study design. Thus, the control cohort may be randomized or concurrent, historic, or based on Objective Performance Criteria (OPC) derived from a literature meta-analysis.

Where important clinical questions remain concerning the premises underlying the study objective, a limited initial Feasibility study may be necessary to justify safety. A Pilot study may be required to demonstrate likelihood of effectiveness and to validate final device design before commencing the pivotal trial. In addition, a run-in phase with an extremely limited number of cases can be permitted for each center to assure preparedness for the definitive study. Patients treated in this latter phase are included for analysis of procedural safety but may be excluded from effectiveness determination.

Entrance and exclusion criteria for recruitment of study cohorts must take into account the target population, as the labeling for intended use will rest on the study experience. The importance of addressing gender balance in the sample has been of increasing concern when evaluating study outcomes.

An appropriate sample size estimation should be statistically computed that is likely to assure not only adequate significant outcome differences, but will also have sufficient power to demonstrate a real benefit, if such truly exists, for the patients enrolled in the study.

Endpoints should be carefully defined for safety and effectiveness and Case Report Forms designed that will capture these in an easily accessible format. Where assessment of these endpoints are highly dependent upon subjective interpretation, an independent Core Laboratory surveillance should be obtained.

While the sponsor of the study must monitor study progress with suitably trained personnel, the appointment of a Data Safety Monitoring Committee composed of individuals independent of the sponsor or investigators, should undertake regular data analyses with rules for early stopping of the study, generally acceptable only for safety concerns.

The Institutional Review Boards of the study centers, acting as surrogates for the FDA, must approve the study for assurance of both patient safety and conformance to the institutions’ treatment policies. They must continue monitoring the study progress, reporting problems of significance to the Agency, and with the prerogative of stopping the study at their institution.

Investigators have a responsibility for ensuring the scientific integrity of the study and for conforming to the protocol requirements. FDAMA requires that investigators are certified with regard to any financial relationship with the sponsorship of a device that they are studying.

Informed Consent is an important element in the conduct of the clinical trial. The study population must be provided with a clear understanding of the nature of the trial, and the risks and benefits associated with participation. The alternative therapy available should be clearly explained. This should reflect the societal and clinical equipoise that justifies study approval by the FDA.

Alternative Regulatory Pathways

The clinical trial requirements have been liberalized for certain situations as required by FADAMA. Less stringent adherence to protocol inclusion criteria is allowed for Emergency Use of experimental devices with approved IDEs. After the completion of an IDE approved study, continued use of the device under a Treatment protocol can be permitted while awaiting PMA approval.

The Humanitarium Device Exemption permits use of an unapproved device in patients for whom no alternative therapy is available and where the disease incidence is less than 4000 cases per year and a study capable of generating statistically meaningful data is precluded. An Individual Treatment exemption is allowed for treatment of patients on a case by case assessment for compassionate indications.


The conduct of a clinical trial required by FDA regulations should be executed in conformity with a protocol developed as a cooperative endeavor involving the Agency, the sponsor requiring device approval for marketing, and the investigators. This is the process most likely to assure a scientifically satisfactory study design that will address the primary concerns of all the parties. More importantly, timely access to safe and effective technologic advances will be made available to the health provider community based on the best scientific evidence.

Copyright © 2000 by the American Society for Artificial Internal Organs