Introduction
Human genome research will revolutionize our understanding of human genetics and health. It has already resulted in powerful new diagnostic tools, and more diagnostics are continually being developed. In time, this research will almost surely result in successful new therapeutic products as well. New understanding of the human genome offers the prospect of large rewards for companies that successfully commercialize technology based on it. Marketing new diagnostics and therapeutics, however, will also be accompanied by several risks.
Major areas of risk relate to Food and Drug Administration (FDA) that pervasively regulates therapeutics and diagnostics. Regulatory risks take a variety of forms, including judicial sanctions. Also, companies that commercialize these products face liability risks. Indeed, such risks arise even before products enter commerce, and liability risks for therapeutics are increased by FDA regulation.
This paper will summarize some such risks and discuss the intersection between regulatory violations and liability litigation.
Regulatory RisksClinical Investigations
Before a new therapeutic can be marketed in the U.S., it must undergo clinical testing. Its sponsor must show, through adequate and well-controlled clinical trials, that it is safe and effective for its intended use.1 Ordinarily, these proceed in three phases.2
In Phase I, healthy, normal volunteers are tested to determine the safety of the therapy. In Phase II, subjects with the disease are administered the therapy to obtain safety and effectiveness data. In Phase III, a larger group of subjects are given the therapy to try to determine more about the product's therapeutic profile, including contraindications, warnings and limitations on use.
This traditional model may not always be well-suited for products developed through the genome project, such as gene therapy. Nevertheless, companies will not be able to obtain FDA marketing clearance without conducting significant clinical research. Given the novelty of the therapy and the enormous number of unknowns, companies are likely to face especially stringent FDA scrutiny as they embark upon clinical trials. While the regulations do not treat products differently based upon the technology, FDA tends to be more conservative when presented with new, untested treatment modalities.
FDA can regulate virtually all aspects of clinical research. Before the clinical study can begin, the sponsor must submit an investigational new drug (IND) application. The IND regulations impose many regulatory obligations upon sponsors, such as the need to monitor the study, select qualified investigators, ensure that institutional review board (IRB) approval is obtained for each investigational site, update investigators and IRBs when there is significant new information and submit reports of adverse events to FDA under certain circumstances.3
Violation of these regulations can result in a variety of sanctions. FDA can place a clinical hold on the study. It can send a warning letter to the sponsor or to third parties who participated and have committed violations, e.g., investigators. It can withdraw approval of the IND. Data gathered in violation of regulations may be rejected, which can delay -- and even jeopardize -- approval. Several biotechnology companies have already learned that the failure to comply with FDA's regulations governing clinical trials can doom a marketing application, at great cost to themselves, their employees and their shareholders. In extreme cases, FDA could criminally prosecute the company and individuals responsible for the violations.
Risk of FDA sanctions is not theoretical. In the last several years, it has devoted more resources to monitoring clinical trials. FDA investigators conduct several hundred inspections of clinical study sites annually. Historically, FDA has reported finding the majority of clinical investigators not in full compliance with regulations, and approximately 10% of clinical investigators are believed to be in substantial non-compliance. Moreover, dozens of IRBs have received warning letters for alleged violations of FDA regulations.
Even when violations are not the direct result of their employees' conduct, sponsors can experience adverse regulatory consequences. A violation that occurs during a study can raise questions regarding its validity, even if errors were made by independent clinical investigators.
Sometimes, FDA will find that a sponsor has failed to discharge its own obligations, such as reporting serious adverse events in a timely manner. A regulation proposed as a result of injuries that occurred during recent clinical trials would significantly increase reporting obligations. Sponsors of clinical trials need to be very sensitive to monitoring and reporting obligations for therapeutic products derived from gene therapy and cannot take their responsibilities lightly: Failure to meet them can have heavy regulatory consequences.
For diagnostics that qualify as devices, clinical regulation is less rigorous. Device sponsors generally need an investigational device exemption (IDE) to conduct clinical trials.4 While IDE regulations impose many of the same responsibilities as IND regulations, diagnostics are exempt from these requirements if they meet certain criteria.5 A key element is that investigators not base diagnostic decisions on the new test. A study designed to meet this element is likely to be largely exempt from FDA regulation. If the four-part regulatory test cannot be satisfied, then the sponsor will be subject to the IDE regulation. The amount of regulation will depend on the nature of the study.6
Post-Approval Regulation
The goal of clinical trials is to generate data needed to support marketing approval. Some early stage companies seem to believe that by obtaining FDA approval, they have left their major regulatory hurdles behind. That perception is wrong.
FDA extensively regulates marketed products under the Federal Food, Drug, and Cosmetic Act (FDC Act)7 and the agency's implementing regulations. If a company fails to comply with these requirements, it faces a range of sanctions. The FDC Act imposes numerous restrictions on companies. There are literally hundreds of ways in which a manufacturer can violate the law.
Four of the most important regulatory requirements are:
1. A company must comply with the terms of its product approval. For example, most gene therapy products will be regulated as biologicals. This means that the approval will take the form of a product license application (PLA) and establishment license application (ELA). Among other items, these two documents will describe, in excruciating detail, the manner in which the product will be made, the procedures that will be used by the facility making the product, the data supporting the application, and the labeling. Once the PLA/ELA are approved, a sponsor is restricted in its ability to make changes. Instituting a major change without prior FDA approval is unlawful.
Similarly, some diagnostics will be approved through the premarket approval (PMA) process. A PMA spells out the manufacturing process, labeling and other details. Some types of product changes can be made only by obtaining FDA approval of a PMA supplement. Failure to obtain a supplement can result in FDA action.8
2. Therapeutic and diagnostic products must be manufactured in accordance with the good manufacturing practice (GMP) regulations that establish detailed standards for manufacturers.9 Regulations differ for therapeutic products and diagnostic devices, but they share many core elements. The GMPs cover, among other things, the creation of a quality assurance group; adequate training of personnel; adequate facilities; calibration of equipment; detailed recordkeeping requirements; validation of key processes and procedures; control of packaging and labeling; and complaint handling and investigation. The failure to comply with the GMP regulations causes the product to be adulterated.10 The sale of adulterated therapeutics and diagnostics violates the FDC Act.
FDA investigators regularly inspect health care manufacturers for GMP compliance. The investigators are entitled to access to a wide variety of records. (Refusing to allow an investigator access to records that FDA is entitled to see is itself a violation of the FDC Act.11) Based on this record review, they will determine whether the facility is in GMP compliance. A high percentage of facilities are found to be in non-compliance with one or more elements of the GMP regulations.
3. FDA has broad powers to regulate promotional materials used by manufacturers. If these materials fail to conform with the FDC Act or FDA's implementing regulations, the product may be misbranded. The sale of misbranded therapeutics and diagnostics violates the Act.
FDA has the authority to regulate "labels." Labels are the written, graphic, or printed materials on the immediate product container.12 Labels are relatively short but they must contain all of the elements established by the FDC Act and FDA's regulations.
More important, FDA can also regulate "labeling." This term includes "all labels and other written, printed, or graphic matter (1) upon any article or any of its containers or wrappers, or (2) accompanying such article."13 This definition has been expanded by FDA and the courts to include materials distributed separate and apart from the product.14 Also, despite the reference to "written, printed, or graphic matter," oral remarks have been deemed to be "labeling."
Product labeling is subject to extensive regulation. The Act forbids labeling from being, e.g., "false or misleading in any particular."15 Thus, labeling is subject to challenge if it contains even a single misleading statement. It can also be found misleading based on omissions.16 Also, statements can be literally true but misleading.
In addition, labeling must bear "adequate directions for use."17 If a manufacturer makes a claim that FDA believes is not substantiated by adequate data, the agency may allege that this provision has been violated. FDA has also adopted regulations which set forth an enormous number of different ways in which the labeling of a therapeutic product can be misbranded; these regulations mean that marketing materials must be reviewed extremely carefully.18
4. Therapeutic products are approved with relatively small patient samples. Sometimes, adverse effects are not discovered until the product has been more widely used. FDA has therefore imposed post-approval reporting requirements on biological companies. The failure to comply with these requirements is a violation of the FDC Act.19
These four areas might seem to be extensive enough. Yet FDA's regulations also contain numerous other booby traps for the unwary. For example, there are regulations governing retention samples, designating a "responsible head," and maintaining non-GMP records.20 Violation of these provisions also subjects the company and responsible employees to FDA sanctions.
Neither the FDC Act nor FDA's implementing regulations include any restrictions specific to products developed through research into the human genome. Rather, the provisions that apply to therapeutic and diagnostic products generally also apply to the products arising out of human genome research.
These novel products will raise novel regulatory issues. How will the GMP regulations be applied? What kind of labeling should be written? What kinds of marketing claims should be allowed when there may be unanswered questions? Should there be special reporting requirements? Based on its past handling of new technologies, FDA is likely to take a conservative approach in applying its regulations to these new products.
Regulatory Sanctions
Thus, therapeutic and diagnostic products can violate the FDC Act in many different ways. If FDA concludes that a violation has occurred, it can impose a variety of punishments.
During an inspection, an investigator will look for deficiencies, particularly in the area of GMP compliance. If the investigator concludes that there have been deviations, he or she may issue a Form FDA-483 Notice of Observations. This is not formal enforcement action; it represents only the investigator's own observations. Yet, as a practical matter, an FDA-483 with many observations is often viewed as an adverse regulatory event. FDA-483s can be obtained by third parties through the Freedom of Information Act (FOIA).
If FDA believes that the violations are more serious, the agency might issue a warning letter. This is a letter that states that the company has been found in violation of the law, and threatens court action if corrections are not made. Warning letters are ugly in tone, conveying a stern message. They are made publicly available each week. Warning letters can be perceived as serious enforcement actions. For a start-up company attempting to commercialize new products developed through genomic research, receiving a warning letter would be a very negative experience. Even if the matter is resolved with FDA, it may cause investors to become skittish; the warning letter may also serve as the company's "one bite at the apple" with FDA.
Generally, companies are able to resolve FDA's concerns after receiving a warning letter. Sometimes, however, FDA concludes that even more serious enforcement action is warranted. One FDA option is to seize the product. The agency can go to court and file an ex parte motion asking that the product be held by the U.S. Marshal.21 Once seized, the product cannot be used or sold until the case has been resolved. Sometimes, the parties can settle the matter; sometimes, the seizure is litigated in federal district court.
A seizure affects only units that are seized. It does not prevent the company from continuing to manufacture and sell other units. Thus, FDA sometimes will ask the court to enjoin the continued manufacture and distribution of the product.22
FDA's most severe sanction is criminal prosecution. In theory, any violation of the FDC Act is a criminal violation.23 In practice, FDA reserves criminal prosecution for those instances where it believes the violations have been most serious, e.g., there has been patient injury, the company has a history of repeated violations, or the company has falsified records or lied to the FDA.
FDA can prosecute and enjoin not only companies, but also individual employees. The company can invoke these sanctions against employees who participated in the wrongful acts. Furthermore, under a unique feature of the FDC Act, FDA can punish an individual who did not personally commit a violation, but who was in a "responsible position." Thus, FDA can bring actions against company presidents based primarily on their title, rather than direct culpability.24
Another sanction is to withdraw approval. FDA can withdraw the product license or revoke the establishment license. The effect is to preclude the continued sale and distribution of the product.
At least initially, the manufacture and sale of breakthrough products derived from human genome research is likely to be politically sensitive and carefully watched. If a product causes an injury to a patient (or clinical research subject), and it is determined that this injury was caused by a regulatory violation, the company -- and its employees -- may face a strong regulatory response by FDA.
Product Liability and the FDA
FDA regulation is intended to be separate from the civil liability system. Indeed, in many official pronouncements, FDA has stated that its regulatory decisions are not intended to have any bearing on product liability suits.
Yet, FDA regulation does play a role in product liability litigation. And the reality is that FDA's role tends to help plaintiffs far more than it helps manufacturers of therapeutic products. There is no reason to expect that manufacturers utilizing the products of human genome research will fare any better in court than companies using more traditional technologies.
Health care manufacturers are regularly product liability targets. Numerous courts have held that there is no private right of action under the FDC Act.25 Thus, plaintiffs cannot sue directly based on FDC Act violations. Nevertheless, FDA enforcement actions and the FDC Act itself can play a pivotal role in a product liability trial.
For a while, device companies found that FDC Act could play an very beneficial role. In the Medical Device Amendments of 1976, Congress prohibited states from establishing requirements that are "different from, or in addition to any requirements under this Act" if the requirement relates to safety, effectiveness or to any other requirement under the FDC Act.26 For many years, this provision lay dormant. Then preemption became a potent defense. Many courts held that provision to broadly insulate device companies from civil suits. In August 1995, a court held that preemption applied even when the manufacturer of the device had plead guilty to FDA felony charges.27
The Supreme Court substantially narrowed the effect of the explicit device preemption provision last April.28 Moreover, pharmaceuticals have no explicit preemption provision. Therapeutic manufacturers have had little to no success arguing for protection through implied preemption. Thus, at least under current law, manufacturer of both diagnostic and therapeutic products derived from genomic research are vulnerable to state tort liability.
In such suits, FDA regulation occasionally can be modestly helpful. If a company shows that it complied with FDA requirements; this sometimes may be considered by juries in deciding the case. This evidence of compliance, though, is not decisive. Many courts have ruled that FDA regulations establish a minimum, not a ceiling, and that states may impose higher standards. Thus, a company meeting FDA standards can still be found negligent or strictly liability.
Conversely, if a plaintiff can introduce evidence showing that FDA has found a violation of a relevant provision of the Act, this evidence can go a long way toward, or even be decisive in, establishing liability. Defendants may exclude evidence of FDA enforcement activity on grounds such as relevance. However, plaintiffs can often get evidence of FDA noncompliance before the jury.
Sometimes, proof of regulatory violations is evidence of negligence per se. If that is found, chances of a favorable defense verdict drop dramatically, since negligence is essentially no longer an issue.
Even if a violation is not negligence per se, it is likely to be given substantial weight by a jury. Suppose, for example, that a plaintiff argues that product labeling contains inadequate warnings. Both sides present medical experts, leaving the jury thoroughly confused. In an otherwise close case, an FDA warning letter could easily tip the balance to plaintiff. The letter might accuse the manufacturer of "serious violations of the Act" because its "promotional material is false and misleading due to inaccurate statements in the labeling." If this warning letter is allowed into evidence by the judge, it will certainly be prominently featured in the plaintiff's case. A stronger enforcement action, such as a seizure or injunction, may be even more persuasive that the company failed to meet the applicable standard of care.
Even if examples of FDA enforcement actions are not allowed as direct evidence for the plaintiff, at the very least, they can restrict defendants' maneuvering room, and give plaintiffs ammunition for cross-examination. Suppose, for instance, that a defense expert testifies about the high quality of the manufacturing process used in developing and controlling a form of gene therapy. Plaintiff may then introduce an FDA-483 issued by the FDA investigator which alleges that "the manufacturing process has not been adequately validated or documented." This type of evidence, even if admitted by the judge only to impeach a witness' credibility, may infiuence jurors. Jurors may be more likely to discount experts hired by defendants when their testimony is contradicted by presumably unbiased, independent government employees.
Defendant's difficulties may be compounded by its written response. In answering an FDA-483, manufacturers sometimes say, e.g., "we acknowledge that we have violated the GMPs and will take prompt corrective action." That language, while it may appease the FDA, can backfire in civil litigation.
FDA enforcement actions can play an important role in yet another way. The plaintiff's attorney wants to know in advance as much as possible about defendant's problems and how it conducts its regulatory business. FDA can be a marvelous repository of seemingly damaging information. FDA-483s, containing nothing but allegations of violations, are always available through FOIA. Another internal FDA document can contain even more detailed negative information.
At the conclusion of inspections, investigators write up narrative Establishment Inspection Reports (EIRs) that contain details of investigators' findings. An EIR will, among other things, elaborate upon the negative findings in the FDA-483, summarize the corporate structure, report the comments of management and make general observations regarding the company.
Some of these details may seem innocuous in the FDA context. Yet, in a liability suit, they may open new lines of attack for a plaintiff's attorney. For instance, it may help the lawyer frame questions during depositions or zero in on critical documents in a request for production of documents. A careful reading of the EIR may also help the plaintiff's lawyer learn a considerable amount about the company and its employees. The EIR will list a series of documents collected by the investigator; this, too, can be helpful in developing discovery strategies.
Moreover, the EIR may create opportunities for cross-examination and impeachment. FDA investigators are trained to try to elicit damaging admissions from company employees and record that information. EIRs will sometimes contain quotes, selected by the investigator because they can be used against the company.29
For example, many product liability cases raise the adequacy of product labeling. The company may argue that its labeling was proper, and that FDA was aware of all product problems but had not asked for any changes. Supporting this defense, company employees may testify that all adverse events were properly reported. Relying on the EIR, plaintiff's attorney may then ask, "Ms. Smith, you told the jury that you always submitted adverse events to the FDA, but didn't you admit to FDA Investigator Jones during an inspection that the company had, in fact, failed to submit some required reports." Complicating matters is that while FDA-483 allegations are often challenged by the company, the statements (or misstatements) in EIRs generally are not refuted. A statement in an EIR, even if incorrect, could handicap the defense.
The proof of the value of EIRs and other FDA documents comes from FDA's FOIA log. All FOIA requests are entered and may be reviewed by the public. Hundreds of such requests are submitted by plaintiffs' lawyers seeking helpful documents.
Thus, FDA enforcement -- and its entire complex regulatory structure -- can be very helpful to plaintiffs in product liability suits. Companies involved with products spawned by human genome research, whether therapeutic or diagnostic, need to be aware of the broader implications of FDA regulation. The way in which a company handles its regulatory status will not only affect its relationship with FDA, but may also have a potent impact on civil liability.
Currently, Congress is debating whether to adopt product liability reform, such as allowing compliance with FDA standards to be a defense in civil litigation. Some members of Congress are also contemplating a radical restructuring of the FDA.
Perhaps legislation will be adopted that will reduce manufacturers' exposure to civil suits or restrict FDA enforcement activities. Perhaps Congress will decide that the current regulatory structure unduly inhibits the development of invaluable new health care products. Until new legislation is adopted, however, companies seeking to commercialize therapeutic or diagnostic products derived from human genome research will need to understand and appreciate the civil liability and regulatory risks that they face.
Notes
* Mr. Gibbs is a partner in Hyman, Phelps & McNamara, P.C., Washington, D.C. and was formerly Associate Chief Counsel for Enforcement at FDA. He graduated from Princeton University and New York University School of Law.
1 21 U.S.C. Sec. 355(b)(1).
2 See 21 C.F.R. Sec. 312.21.
3 See 21 C.F.R. Part 312.
4 21 C.F.R. Part 812.
5 21 C.F.R. Sec. 812.2(c)(3).
6 Some studies qualify for abbreviated IDE status. 21 C.F.R. Sec. 812.2(b). These studies are subject to less regulation than studies that are subject to a full IDE.
7 21 U.S.C. Secs. 321, et seq.
8 Some diagnostic devices may be able to obtain marketing clearance through a "510(k) premarket notification." This clearance mechanism requires less data, and allows companies greater flexibility in making product changes. See 21 C.F.R. Sec. 807.81(a)(3).
9 See 21 C.F.R. Parts 210, 211, and 820.
10 21 U.S.C. Sec. 351(a)(2)(B) & (h).
11 21 U.S.C. Sec. 331(f).
12 21 U.S.C. Sec. 321(k).
13 21 U.S.C. Sec. 321(m).
14 See, e.g., United States v. Kordel, 335 U.S. 345 (1948).
15 21 U.S.C. Sec. 352(a).
16 21 U.S.C. Sec. 321(n).
17 21 U.S.C. Sec. 352(f)(1).
18 21 C.F.R. Part 202. FDA has not adopted parallel regulations for diagnostic devices.
19 21 C.F.R. Sec. 314.80. Medical devices are also subject to post-market reporting requirements. 21 C.F.R. Part 803. While diagnostic devices are not exempt from these requirements, there tends to be fewer reportable events with these products.
20 See 21 C.F.R. Part 600.
21 21 U.S.C. Sec. 334.
22 21 U.S.C. Sec. 332.
23 See 21 U.S.C. Sec. 333.
24 United States v. Dotterweich, 320 U.S. 277 (1943).
25 See, e.g., National Womens Health Network v. A.H. Robins Co., 545 F. Supp. 1177 (D. Mass. 1982).
26 21 U.S.C. Sec. 360k.
27 Talbot v. C. R. Bard, 63 F.3d 25 (1st Cir. 1995).
28 Medtronic, Inc. v. Lohr, 64 USLW 4625; to be reported at 116 S.Ct. 2239 (1996) (plurality decision; Justice O'Connor, joined by three others, dissented in part, and Justice Breyer wrote a third opinion.
29 Unlike FDA-483s, EIRs are not always available through FOIA requests. FDA will not release an EIR if there is a pending enforcement action against the company.