April 1, 2011 (Vol. 31, No. 7)
Jeffrey N. N. Gibbs
FDA May Be Considering More Restrictive Approach with Research Use Only Assays
Research use only (RUO) products are ubiquitous in the life science industry. They are widely sold, prominently displayed at major conferences, extensively advertised (including in GEN), and routinely used by laboratories.
Yet, as common as RUOs are, their regulatory status is not well understood. That is not surprising because they have long occupied an anomalous regulatory position.
The FDA adopted a regulation covering RUOs more than 30 years ago. The regulation, 21 C.F.R. § 801.109, states in full: “For a product in the laboratory research phase of development, and not represented as an effective in vitro diagnostic product, all labeling bears the statement, prominently labeled, ‘For Research Use Only. Not for use in diagnostic procedures.’”
That is the extent to which FDA has established regulatory requirements for RUOs. If an RUO product is properly labeled, then it has fully satisfied the applicable FDA requirements.
This very light regulatory touch is appropriate. RUOs are not medical devices and therefore are not subject to FDA jurisdiction. The statutory definition of a “device” includes products intended to be used in the “diagnosis…of disease or other conditions.” By definition, RUOs are not intended to be used diagnostically. Hence, they are not devices and therefore not within FDA’s regulatory domain. This means that the basic FDA regulatory requirements, such as premarket review, the Quality System Regulation, device listing and the submission of reports to FDA, are not applicable.
What Are RUO’s?
However, saying RUOs are not devices leaves open a major question: What, then, are RUOs? A product that is intended to be used diagnostically is not an RUO. For what purposes are RUOs intended to be used?
Decades after FDA issued the RUO regulation, the answer to that fundamental question is still not entirely clear. A product that is intended by the manufacturer to be used by researchers in understanding basic biology would qualify as an RUO. Similarly, an assay whose clinical relevance is unknown should also be eligible for RUO status.
For example, a monoclonal antibody that is being sold because it has biologically interesting characteristics but no known diagnostic use could fall into the RUO category. Instruments or assays that are intended to enable biological discovery or basic research could also be positioned as an RUO product.
In practice, FDA’s approach to regulatory RUOs has been based on exclusion, rather than on affirmatively defining the characteristics of an RUO. That is, FDA essentially has focused on what products are not RUOs, rather than identifying what attributes make a product an RUO.
Given this, it is not surprising that FDA’s enforcement activities involving RUOs have been directed at the claims made for the products. For example, a product labeled RUO but advertised “as an aid in diagnosing lung cancer” would forfeit its RUO status and be regulated as an in vitro diagnostic. Conversely, a product labeled as RUO could be used by researchers to determine whether it might have utility in the detection of lung cancer.
Making matters more complicated, the product might be considered an “investigational device” if the manufacturer intended for the antibody’s clinical performance for a specific diagnosis to be evaluated. Once the intended use shifts from evaluating whether the analyte might potentially have diagnostic utility to evaluating its diagnostic performance for a specific disease, the product’s regulatory status may move from RUO to investigational.
FDA has never adapted a list of terms that companies should not use when promoting RUO products. The agency has, however, intermittently sent warning letters or taken other enforcement actions against companies selling RUOs. Based on these actions, certain principles can be derived.
For example, companies selling RUOs should not claim diagnostic utility. An assertion of diagnostic usefulness is not compatible with the RUO designation. Similarly, companies should not make claims relating to clinical sensitivity, clinical specificity, negative predictive value, or positive predictive value. Nor should the company compare the performance of its RUO product to that of an assay or instrument marketed for diagnostic use, since that would constitute an implicit diagnostic claim.
RUO products are, in fact, used diagnostically. Laboratories buy products labeled as RUO and then use the results in generating diagnoses. Historically, from an FDA prospective, that use did not alter the regulatory status of the RUO product. One well-established regulatory principle is that the regulatory classification of a product hinges on the manufacturer’s objective intent for the product rather than how it is ultimately used. This is a fundamental principle when it comes to devices, biologicals, and drugs that have been cleared or approved by FDA. It is the use intended by the manufacturer that controls the regulatory classification, not the use to which the product is put.
There are some indications that FDA personnel would like to adopt a more restrictive approach for RUOs. In a few instances, FDA officials have stated that a product labeled as RUO loses that status if a laboratory uses it diagnostically. That approach would be at odds with one of the basic tenets of FDA law. The RUO status of a properly promoted product should not be altered simply because a laboratory chooses to use an RUO product diagnostically.
In earlier draft guidance documents, FDA proposed that companies selling RUO products obtain a signed certificate from the laboratory stating that it would not use the product for diagnostic purposes. There is currently no requirement that manufacturers obtain a certificate from a laboratory customer. However, to reinforce the product’s intended use as RUO, some manufacturers do ask laboratories to sign a form acknowledging that the product is not intended to be used diagnostically. At a minimum, the materials sent to customers and potential customers, including invoices and purchase orders, should reiterate the RUO designation of the product.
RUOs and FDA Approval
Although laboratory tests using RUO products are not generally subject to FDA regulation, there are situations where RUO assays do intersect with the premarket review process. One area of overlap involves the use of RUO assays or instruments in support of a marketing application by an IVD manufacturer. FDA has not yet publicly and formally articulated its policy on marketing applications that incorporate an RUO product. Applicants do need to be aware that, at a minimum, they will need to bring the RUO product into their own quality system. Because RUO products are not subject to the Quality System Regulation, the applicant will need to develop appropriate quality oversight for the RUO component.
In some circumstances, it may even be necessary to obtain FDA clearance for the RUO product in order for the application to succeed. For other lower risk products sold as RUO, FDA’s concerns sometimes can be satisfied by having the third-party supplier relabel its product as an “in vitro diagnostic” and list it with FDA as a device. IVD companies incorporating RUO products into their product should consult with FDA at an early stage to address potential issues involving the use of an RUO product. This is particularly important if the RUO product plays a fundamental role in the potential diagnostic product.
Pharmaceutical companies also need to be alert to the issues stemming from the use of RUO products in clinical trials. There is no bar against the use of RUOs during a clinical investigation. An RUO product could, for example, be used to help evaluate changes in a novel biomarker in a Phase II study. One major problem, though, is whether the resulting data would be acceptable to FDA reviewers to support approval. Using an RUO test to help explore and understand biomarker relationships is fine, but relying on this data in a regulatory submission can be problematic. If a companion diagnostic is needed for approval of the drug, an RUO product will not be acceptable.
FDA has recognized that more detailed guidance involving RUOs would be helpful. In the early 1990s, FDA proposed several draft guidance documents. That effort was eventually abandoned. More recently, FDA has indicated that it is planning to issue a draft guidance soon. When that draft document will be released or what it will say remains uncertain.
RUOs are an important class of products. Over the years, they have raised many questions and provoked considerable regulatory controversy. FDA’s pending draft guidance document is unlikely to answer all the questions or end the controversies.
Jeffrey N. Gibbs (firstname.lastname@example.org) is director at Hyman, Phelps & McNamara.