Historically, pharmaceutical and diagnostic companies in the U.S. have inhabited parallel regulatory universes. Although both sets of firms are overseen by the FDA and are subject to some identical provisions, they have been regulated independently. For the pharma industry, regulatory requirements that apply to in vitro diagnostic (IVD) device manufacturers have been little more than an idle curiosity. That is no longer true.
For many reasons, the manner in which FDA regulates IVDs is becoming increasingly important to pharmaceutical companies. Pharmacogenomics, proteomics, and related disciplines mean that diagnostic tests will assume greater significance for regulators, researchers, and pharmaceutical companies alike. Another well-known reason is the Critical Path Initiative (www.fda.gov/oc/initiatives/criticalpath/opportunities06.pdf), for which diagnostic testing is a key aspect.
And then there is the push for personalized medicine. The Secretary of the Department of Health and Human Services has identified this as a top priority (Leavitt M. Personalized Health Care: Delivering Value to Patients. Presented at Personalized Medicine Coalition 3rd Annual Luncheon, Washington, D.C., March 23, 2007; www.hhs.gov/myhealthcare). Personalized medicine will be difficult to accomplish without determining the key characteristics of a patient’s condition, which, in turn, will often require an IVD.
Companies that do not have at least a basic understanding of FDA statutes on IVDs will be at a regulatory and commercial disadvantage. Firms familiar with IVD regulation outside the U.S. should not extrapolate that knowledge to the U.S., since FDA’s regime differs in many material respects from similar organizations in other countries.
IVD firms are developing tests to help doctors make therapeutic decisions, e.g., tests for CYP450 genotypes that can affect drug metabolism. In many instances, IVD manufacturers will want to reference a marketed drug. FDA has already approved several drugs that reference diagnostic tests in their labeling, like Herceptin and HER2/neu as well as Camptosar and UGT1A1. FDA-mandated labeling relating to diagnostic tests is likely to become more common.
What Governs IVD Regulation?
Under the Federal Food, Drug, and Cosmetic Act (FDC Act), FDA regulates devices. A device includes a product “intended for use in the diagnosis of disease or other conditions.” (21 USC 321(h)(2)). An in vitro diagnostic test kit will be regulated as a device even if it is not intended to diagnose the presence of a disease but only to help a doctor decide which drug to administer to a patient with an already diagnosed disease.
The technology of the test is also irrelevant as to whether a product falls within the definition of a device. FDA’s jurisdiction is unaffected by the product’s methodology such as PCR, mAbs, or molecular diagnostics. Also, what is tested doesn’t matter, be it genes, antibodies, or morphology. These factors only affect the data required and the nature of the regulation.
Nor does the diagnostic product need to be in vitro to be regulated by FDA as a device. Although pharmaceutical companies will primarily interact with IVD manufacturers, the device definition covers all diagnostic products such as radiological and other imaging modalities.
This dichotomy—IVD versus non-IVD devices—impacts which division of the agency will regulate the product. The Center for Devices and Radiological Health (CDRH) has created the Office of In Vitro Diagnostic Device Evaluation and Safety (OIVD) to regulate IVDs. Non-IVDs will be regulated in another division of CDRH. Because a majority of diagnostic tests that affect pharmaceutical companies are IVDs, this article focuses exclusively on these products.
Further complicating jurisdictional matters is that not all IVDs are regulated by CDRH. The Center for Biologics Evaluation and Research (CBER) has jurisdiction over some IVDs. For example, diagnostic tests intended to be used in blood screening or banking fall within CBER’s purview. While CBER does try to maintain consistency with CDRH, it does not always interpret the statutes precisely the same way OIVD does.
Many Paths You Can Go Down
The opportunity for companies to select from different regulatory pathways or at least to influence the path is one major difference between IVD and pharmaceutical regulation. Thus, a pharmaceutical company dealing with an IVD manufacturer should not be surprised if the IVD company is initially unclear as to which regulatory pathway its product will traverse (Gibbs JN. Exploring Other Options, Part 1: The trend toward alternative market pathways, IVD Technology, May 18, 2005; Gibbs JN. Exploring Other Options, Part 2: Facilitating the FDA Review Process, IVD Technology, June 22, 2005).
A company with a new drug knows that it will need to submit an NDA or a BLA. There may be flexibility as to whether the application can be a 505(b)(2) NDA or a full NDA, but the NDA is the only available route. Even before this, to conduct a clinical study in the U.S., the sponsor must obtain FDA approval of an IND application. Again, there is no choice.
Diagnostic assay manufacturers, though, have more options. Some well-established, low-risk assays are actually exempt from the need for FDA premarket authorization, like Mycoplasma-detection media and components. New pharmacogenomic tests, though, will need to obtain FDA clearance because of their higher risk levels. IVDs for which the manufacturer can identify a predicate device may be the subject of a 510(k) premarket notification. If there is no predicate device, then the company needs to submit a premarket approval application (PMA), unless FDA allows the company to use the de novo reclassification process instead.
The manufacturer may also decide that instead of selling a complete kit it will sell the key building block(s) to laboratories for use in laboratory-developed tests (LDTs). Alternatively, a company may develop an LDT for in-house use and sell only the testing service, not a product. The IVD company may even decide not to pursue FDA clearance for the kit but instead sell it for research use only (RUO).
Supporting Marketing Applications
IVD applications for new types of assays will almost always need supporting clinical data. The regulatory framework for these studies, however, is different than for pharmaceuticals. Occasionally, the sponsor will need to submit an investigational device exemption (IDE), the device counterpart of an IND. This primarily occurs when the investigational IVD’s results will be the basis for a therapeutic decision. FDA receives only a handful of IVD IDEs each year.
Many IVD studies are not even subject to IDE regulations. If a study meets a four-part test, then the IDE regulations are entirely inapplicable (21 CFR 812.2(c)(3); FDA. Draft Guidance for Industry, Clinical Laboratories, and FDA Staff—In Vitro Diagnostic Multivariate Index Assays, July 2007). A majority of IVD studies fall into this category.
Also very often, an IVD study will be deemed a nonsignificant risk study and will be covered by the abbreviated IDE process (21 CFR 812.2(b)). These studies must comply with only a few elements of the IDE regulation, including having an investigator agreement as well as reporting and maintaining certain records. More important, the study can commence without prior FDA approval. The company can itself deem the study to be a nonsignificant risk study, but each institutional review board (IRB) must agree. These studies must comply with FDA’s informed consent regulations (21 CFR Part 50) and IRB regulations (21 CFR Part 56). IVD studies often use banked specimens, in which case informed consent may not be required (21 CFR Part 56).
Thus, because IVD studies rarely need FDA approval, many investigations are initiated and completed without any interaction between the manufacturer and FDA. It is usually better, though, to obtain the agency’s feedback before beginning the study.
The company can obtain this feedback via the pre-IDE process. This mechanism allows the IVD firm to submit its protocol and meet with FDA to discuss study design, intended use, preclinical testing, regulatory pathway, and other issues. FDA’s comments on a pre-IDE are not binding, and its views can change. Pre-IDE reviews are not subject to statutory time frames.
Just as is the case for pharmaceuticals, IVD companies need to define the clinical endpoints carefully. The terms safe and effective in the IVD context have different meanings than in the drug industry. Since IVDs are not administered to patients, there is no direct impact on patient health. FDA may assess among other parameters, positive predictive value, negative predictive value, clinical specificity, clinical sensitivity, linearity, reproducibility, and limit of detection.
In sum, the route to the market for the diagnostic industry differs in innumerable ways from the pharmaceutical industry. The next article, which will appear in the August issue of GEN, will discuss these pathways in detail.
Jeffrey Gibbs, Esq., is director, Hyman, Phelps & McNamara. Web: www.hpm.com. Email: email@example.com.
Critical Path Initiative
21 USC 321
Federal Food, Drug, and Cosmetic Act
21 CFR 812.2
FDA Medical Device Exemptions
21 CFR Part 50
Protection of Human Subjects
21 CFR Part 56
Institutional Review Boards
Exploring other options Part 1: The trend toward alternative market pathways
(another article by this author on this topic)