On March 26, 2015, uveitis subspecialists, pharmaceutical industry representatives, members of the National Eye Institute (NEI), and representatives of the Food and Drug Administration (FDA) participated in the “NEI-FDA Workshop on Clinical Trial End-points for Inflammatory Eye Diseases.” The workshop was organized to consider the following issues: strategies for improved clinical trial design and choice of endpoints, based on current evidence; classification and stratification of study patient populations; choice of clinically meaningful surrogate or interim endpoints; and identification of important gaps in research funding. Information specific to uveitis, scleritis, and other inflammatory eye diseases will be summarized in a white paper describing proceedings of the workshop; however, an apparent lack of understanding among attendees about FDA procedures in general suggested a need to summarize that information for a general audience as well. Such information is relevant to clinical trials in all subspecialties, and should be of interest to the Journal ‘s readership.
FDA premarket review of drugs, as outlined in the Federal Food, Drug, and Cosmetic Act (FDCA), requires “substantial evidence” of safety and efficacy as the basis of drug approval. Substantial evidence is defined in part as consisting of adequate and well-controlled investigations. In other words, the FDA seeks to confirm that a drug does specifically what the manufacturer claims that it does, without undue risk to the patient. It is not expected that the evidence replicates all aspects of treatment that may involve the drug.
To be regarded as “adequate and well-controlled,” studies must demonstrate a number of qualities: (1) they should have a clear statement regarding objectives of the investigation; (2) the study design should enable valid comparisons with a control, providing quantitative assessments of drug effect; (3) the method of selecting study participants should provide assurance that they have the condition being studied; (4) the method of assigning participants to treatment and control groups should minimize bias and assure comparability of groups; (5) measures should be taken to minimize bias on the part of participants, observers, and analysts; (6) methods for assessing treatment response should be well-defined and reliable; and (7) analysis of study results should be adequate to assess drug effects. Once a sponsor believes it has gathered sufficient evidence of efficacy and safety, data are submitted in support of a New Drug Application to the FDA. If approved, the product can then be marketed for the specific indications for which it was evaluated.
A major topic of discussion at the workshop was use of ophthalmic devices to collect objective data as evidence in clinical trials, in lieu of subjective measures, such as semiquantitative scoring of anterior chamber cells and flare by slit-lamp biomicroscopy or vitreous haze by indirect ophthalmoscopy. The article by Zarranz-Ventura and associates in this issue of the Journal touches upon this topic; it describes use of optical coherence tomography (OCT) for objective quantification of vitreous inflammatory reactions. While the study has limitations, it found statistical associations between OCT measurements and semiquantitative scores assigned to “vitreous haze” using a well-accepted, standardized clinical grading technique. There was substantial overlap between OCT values and different clinical grades, and thus, their value in guiding clinical care is yet to be determined. At present, there are no devices approved by the FDA specifically for quantitating posterior uveitis. The comparative study by Zarranz-Ventura and associates provides an important step toward supporting an additional indication for OCT and its expanded use in uveitis trials. Further refinements may indeed yield a more useful and objective alternative scale for inflammation.
The FDA defines a medical device as something that is “intended to diagnose, cure, mitigate, treat, or prevent a disease/condition, or intended to affect the structure or function of the body, and does not achieve its intended use through chemical action or metabolism.” Medical devices are regulated in the Center for Devices and Radiological Health (CDRH), which serves to protect and promote public health through ensuring that “patients and providers have timely and continued access to safe, effective, and high-quality medical devices and safe radiation-emitting products.”
The CDRH employs a risk-based paradigm whereby classification of a device is determined by its risk to patients from factors such as design and intended use. Devices are categorized into Classes 1, 2, or 3. Class 1 devices represent the lowest risk (eg, perimeter). Class 2 devices represent moderate risk (eg, tonometers). Class 3 devices represent the highest risk (eg, intraocular lenses). All devices are subject to “general controls,” regardless of classification, including registration and listing with the FDA and labeling requirements. For Class 2 devices, such as OCT devices, in addition to general controls, “special controls” are employed, which may include compliance with specific performance standards or FDA guidance documents. Most Class 2 devices require marketing clearance through the premarket notification process, described in section 510(k) of the FDCA ; 510(k) marketing clearance for a new device is based on a determination of substantial equivalence to a legally marketed device (predicate device) of the same intended use and similar technology. Substantial equivalence requires that the new device has the same intended use and either that it has the same technological characteristics or that different characteristics do not raise new types of safety and effectiveness questions. Considerations within this assessment include the intended end-user populations, conditions of device use, device reliability, and anticipated benefits to health vs risk of injury from its use.
Relevant to these procedures is the fact that diagnostic devices are assessed on factors such as intended measurements (what OCT is designed to detect anatomically), instructions for use, conditions of use, and indicated populations. Currently, OCT devices are legally marketed as both quantitative and qualitative diagnostic devices, to aid in detection and management of some ocular diseases, but cannot be promoted or labeled for indications that were not included in the cleared indications statement as part of the marketing application; thus, OCT cannot currently be promoted for quantification of inflammation. To expand indications for use, performance data are needed to confirm that the device truly measures what it is claimed to measure, repeatedly and reproducibly. Requests for clearance of additional indications are typically submitted by the device manufacturer, as proprietary device information is needed for the review process; however, ophthalmologists can help by collecting data necessary for alternative indications, using agreed-upon standard study protocols.
The FDA is interested in working with investigators and sponsors throughout the device development process. The CDRH has a presubmission program that is intended to offer feedback regarding important issues such as study design, nonclinical testing, and risk determination. Presubmissions are intended to serve as an efficient pathway from device concept to market, facilitating the Agency’s goal of fostering the development of new medical devices. In addition, there is an Early Feasibility Study Program for devices that may be early in development or for marketed devices for a novel clinical application. The goal of the program is to allow for early clinical evaluation of such devices, providing proof of principle and initial clinical safety data in a small number of subjects.
It is hoped that this Editorial will help not only investigators who plan to evaluate new drugs, but clinicians who must interpret the relevance of results from clinical trials leading to FDA clearance and approval of new medical products.