Development of the generic drug industry in the US after the Hatch-Waxman Act of 1984

Garth Boehm (a), Lixin Yao (a), Liang Han (a,b), Qiang Zheng (a,c,*)

a. Center for Pharmaceutical Information and Engineering Research, College of Engineering, Peking University, Beijing 100871, China
b. Institute of Molecular Medicine, College of Engineering, Peking University, Beijing 100871, China
c. Department of Industrial Engineering and Management, College of Engineering, Peking University, Beijing 100871, China

Received 17 April 2013; revised 24 June 2013; accepted 18 July 2013

Abstract The key events in the development of the US generic drug industry after the Hatch-Waxman Act of 1984 are systematically reviewed, including the process of approval for generic drugs, bioequivalence issues including “switchability”, bioequivalence for complicated dosage forms, patent extension, generic drug safety, generic substitution and low-cost generics. The backlog in generic review, generic drug user fees, and “quality by design” for generic drugs is also discussed. The evolution of the US generic drug industry after the Hatch-Waxman Act in 1984 has afforded several lessons of great benefit to other countries wishing to establish or re-establish a domestic generic drug industry.

KEY WORDS: Generic drugs; Drug PriceCompetition and Patent Term Restoration Act; Abbreviated new drug application; Bioequivalence; Drug quality; Generic drugsubstitution

1. Introduction
The Drug Price Competition and Patent Term Restoration Act of 1984 (US Public Law 98-417), commonly known as the Hatch-Waxman Act, was signed into law on September 24th 1984 following a vote of 362-0 in favor in the House of Representatives of the 98th Congress and passage through the Senate on by voice vote (1,2). The Hatch-Waxman Act amended the Federal Food, Drug and Cosmetic Act (FDCA) and the Patent Act, established an abbreviated new drug application (ANDA) process, provided for filing of generic drug applications 60 days later, and so created the modern US generic drug industry (3). Although the Hatch-Waxman Act was passed with overwhelming support in the US Congress, it was, and remains, an uneasy compromise and a delicate balance between the interests of the brand-name drug industry and the generic drug industry (Table 1 (4)). The legislation is complex and has given rise to many unforeseen situations as the industry has developed over the subsequent years. The history of the US generic drug industry after the enactment of the Hatch-Waxman Act has presented several lessons which are of benefit to other countries wishing to establish or re-establish a domestic generic drug industry, and especially so for countries like China, where generic drugs constitute the largest share of the pharmaceutical industry and drug consumption.

Prior to passage of the Hatch-Waxman Act, there were relatively few generic drug products in the US. The 1962 amendments to the Food, Drug and Cosmetic Act (FD&C Act) had some unintended consequences (3). The requirements imposed by the amendments to gain approval to market a new drug had made the approval process costly and lengthy. With the exception of antibiotics, generic drugs were approved via a “paper NDA” process which required filing scientific literature to support the safety and efficacy of a generic drug, since the FDA regarded the safety and efficacy data filed by the innovator as proprietary. However, for the majority of branded drug products, excluding the antibiotics that were not subjected to the requirement, the innovator companies did not publish sufficient scientific literature to enable justification of safety and efficacy via the “paper NDA” route (3). Hence in 1983 only 35% of top-selling branded drugs with expired patents had generic competition, and the generic market share was only 13% (5,6). These generic drug products required that a prescription be written for the generic.

The Hatch-Waxman Act addressed the shortcomings of the post-1962 amendments to the FD&C Act situation by providing a less arduous approval route for generic products but restoring a new drug patent term lost by the post-1962 NDA process (1). Thus, and as suggested by the name, the Hatch-Waxman Act is a compromise between the interests of the brand and generic industries (7).

Title I of the Hatch-Waxman Act amended Section 505 of the FD&C Act to create an Abbreviated New Drug Application (ANDA) which allowed approval of generics as equivalent products to an existing brand product (1) (called a reference listed drug, RLD) on the basis of bioequivalence. It allowed for some variance in the RLD provided this was approved via a petition before filing.

Title II of the Hatch-Waxman made two changes to Title 35 of the United States Code regarding patent law: it amended the statute to provide for restoration of that part of the patent term lost to the time taken for FDA required pre-market testing and review, up to a maximum of 5 years for new drug applications. It amended the statute to make using an invention solely for the purposes of generating information to file an application not an act of infringement and that filing an ANDA or paper NDA that challenges a patent could be deemed an act of infringement, albeit an artificial infringement.

The Hatch-Waxman Act grants generic manufacturers the ability to mount a validity challenge without incurring the cost of entry or risking enormous damages flowing from any possible infringement. In addition, the Hatch-Waxman Act requires that the FDA, among other things, makes publicly available a list of approved drug products with therapeutic equivalence evaluations with monthly supplements, commonly known as the Orange Book. This list also included patent and exclusivity listings for drug products where those were in force, which were provided by the drug application owner, and the FDA is obliged to list them (8). Because the FDA-published list included drug products designated as therapeutically equivalent to an original drug product, it became possible for health care providers to substitute a generic equivalent for a brand product3. This allowed the creation of a substitution system where state legislation would allow or mandate the substitution of generic equivalents, where they exist, for prescriptions written for brand products. The only exceptions to this substitution are if the prescription is marked “Do Not Substitute” or the patient refuses a generic substitution. This substitution system created the generic industry marketing system where it is only necessary to get a pharmacy to stock generic products to ensure their selling to patients, and physicians need not know that a generic exists or that it will be taken by their patients. Because the US drug distribution and retail pharmacy industries are concentrated, a generic company requires relatively few people to market its product. In addition, the high prices for branded products means that pharmacy profit margins for generic products are higher as low priced generics can tolerate a higher markup by the pharmacy (9).

This substitution procedure created an extremely efficient marketing and distribution system and ensured the rapid “pull through” of new generic products into the distribution chain due to their higher profitability. Studies have shown that patients and doctors prefer brand name drugs, although pharmacy computer systems default to substitute generic for brand-name drugs. Studies also found problems with health insurance companies and poor communication with the doctors' offices, leading to patient confusion and poorer drug treatment (10).

In 2012, generics reached 84% of dispensed prescriptions, and spending in this segment grew by $8 billion (11). The fourth annual Generic Drug Savings Study revealed remarkable reductions in health care costs over the previous 10 years (from 2002 to 2011) (12). Clearly, despite all the attempts by the brand industry to counter generic product development and use after the enactment of the Hatch-Waxman Act, generic drugs have risen to become a significant majority of the US prescription pharmaceutical market by volume. This has been driven entirely by cost. Because the brand pharmaceutical industry has chosen to maintain very high costs for products dispensed through retail pharmacies, it has created a huge incentive for payers to switch to generics and for retail pharmacies to dispense generics (13–16).

There is no doubt that the US generic industry has been successful beyond the wildest dreams of those who formulated the Hatch-Waxman Act. Even though successful, the development of the generic drug industry has been anything but smooth and the rest of this paper will discuss some key events since its enactment.

2. The generic drug scandal
The beginning of the modern generic drug industry was marked by fraud and other criminality on the part of some companies that almost destroyed the industry before it got started. The fraud was pervasive from 1984 to 1989 and became collectively known as the Generic Drug Scandal (17,18). The generic drug scandal reduced consumers' perception of the quality of generic drugs (19,20). With the passage of the Hatch-Waxman Act, the field for new generic drug products was wide open with many top selling brand-name drugs available for generic competition (21). The Hatch-Waxman Act granted 180-day marketing exclusivity to the first filed ANDA containing a paragraph IV certification (a patent challenge). If the first filed paragraph IV applicant was sued and won in court, they would get 180 days of exclusive marketing of the generic. Generic companies knew that the first approved generic product would attract relatively high prices and take the majority of the generic market share (22), and those who entered later would reap lower margins. Therefore the race was on to develop products and file first to gain first approval for a given product. Also driving the race to file was the fact that FDA had a “first in, first reviewed” policy, although it would later become known that some FDA reviewers were bribed to manipulate this policy (23).The stakes were high for the new industry and tens or hundreds of millions of dollars in potential profits were at stake.

Fraud began on day one of the new industry. One company, Bolar Pharmaceuticals, was reported to have driven to the FDA and filed 40 ANDAs on November 23rd 1984. It would later be found that all of these ANDAs were fraudulent, fabricated for the purpose of filing first to ensure a timely approval (24). One generic company, Mylan Laboratories, had complained to the FDA-CDER Division of Generic Drugs (DGD) that ANDAs were not being reviewed according to the “first in, first reviewed” policy and that some applicants were receiving favored treatment (17). By 1988, Mylan became frustrated with the lack of response to their complaints of favoritism and hired a private detective to investi-gate. Evidence of bribery of DGD reviewers was found and turned over to the US House of Representatives Energy and Commerce Committee's Subcommittee on Oversight and Investigations (the Subcommittee). The Subcommittee began an investigation that revealed bribery and fraud, and resulted in charges against FDA officials and generic drug companies and some of their executives, managers, and employees. The investigation continued for several years and investigators from the Department of Justice and Department of Health and Human Services discovered that not only had there been bribery, but that some companies had submitted fraudulent data, substituting brand product for generic product as samples in bioequivalence testing (25,26).

In all, thirty individuals and nine companies were either found guilty or admitted their role in FDA corruption. At one point, in the subcommittee investigation during a press briefing, it was reported that subcommittee staff stated that “of 39 generic drug companies…(investigated) … only about a half dozen appear to be free of criminal or regulatory taint”(27). Representative John Dingell, Chairman of the Subcommittee, declared that the generic drug industry was “the most pervasively corrupt this subcommittee has ever uncovered”(28).

Clearly by late 1990 the public's faith in generic drugs and in FDA's ability to regulate the drug industry was severely shaken (29). Among the 1009 consumers of a broad range of ages surveyed by Gallup in October 1989 to ascertain their attitudes toward generic drugs after the scandal, 51% feared that generic drugs were not manufactured to the same standards as brand medications and more than 70% indicated that the scandal had affected their confidence in generic drugs to some degree (19). Realizing the risks of lack of trust in FDA and to the fledgling generic drug industry, FDA acted very aggressively to root out fraud. Indeed so aggressive was FDA's approach that one industry analyst reported “everybody is scared to death about the FDA because they know the FDA means business”(30). This aggressive approach was successful in restoring public con-fidence in both FDA's ability to regulate the drug industry and in generic drug products, although it took many years before public confidence in generic drugs returned to pre-scandal levels.

Two major steps were taken to rectify the problems revealed by the Generic Drug Scandal and to restore public faith in generic drugs. First was the passage of the Generic Drug Enforcement Act (GDEA) which gave FDA the ability to take actions against persons or corporations abusing FDA regulations (31). These actions included debarment, withdrawal of product approval, suspension of product distribution, and the ability to levy civil penalties (32). Second was a large product analysis effort aimed at determining whether generic drug products obtained from the market met product specifications. By November of 1989 the FDA had analyzed over 2500 product samples representing the 30 most prescribed generic products. Less than 1% failed to meet product specifications and none was deemed a health threat (33).

The broad-scale unreliability of data submitted to support marketing approval applications, particularly fraudulent data in records submitted in premarket approval applications during the generic drug scandal, resulted in the establishment of the applica-tion integrity policy by FDA in the early 1990s (34). In addition to the very public actions, FDA took numerous other actions in the wake of the Scandal in an effort to prevent a recurrence in the future (6,35), including:

* Institution of a new system to control new drug sponsors' access to application reviewers. The new system required formal requests for meetings and those meetings, which are now usually held by telephone, had to be held with a Project Manager and the Reviewer's Supervisor. Uncontrolled and unsupervised access to reviewers had facilitated the bribery that had occurred as part of the scandal.

* Establishment of astrict application queue system to assign applications to “the next available reviewer” on a first-come basis. Although such a system had existed previously, this was a much more robust system that had better visibility within the Office of Generic Drugs (OGD).

* All ANDAs had to be complete when filed. A “received for filing” procedure was instituted to ensure that each ANDA received for filing by OGD was complete. Previously applications could be filed incomplete and amended with additional data, such as additional strengths test batches, stability test results, bioequivalence studies, etc. This created situations where fraudulent information could be filed to get a position in the review queue. If subsequent to initial filing failures occurred, there was a temptation for the sponsor to overlook or manipulate failing data to keep the review moving forward. Inaddition, post filing failures wasted review resources.

* Establishment of the Office of Generic Drugs for ANDA review. Previously this function resided in the Division of Generic Drugs which was not a separate office within the Office of Pharmaceutical Sciences. This action raised the status and management visibility of generic drugs within the Center for Drug Evaluation and Research (CDER). Pre-viously generic drug review had been seen as a sub-branch of the review divisions and this lack of visibility may have been a contributing factor to the failure to recognize the problems for so long.

* Establishment of the pre-approval inspection (PAI) system. This system was put into place to verify the accuracy of data filed in an application. A number of firms filed in their ANDAs fraudulent data which were made to appear genuine on the surface. The PAI system sent FDA compliance investigators into firms to verify the accuracy of filed data.

* Institution of measures to address bioequivalence studies including the establishment of the retention sample require-ments, financial interest disclosure requirements, and increased inspection of contract research organizations con-ducting generic drug bioequivalence studies. During investi-gations that followed the initial disclosure of fraud and bribery, it was discovered that some firms had substituted brand-name product samples for generic product samples in bioequivalence studies (26). These measures were meant to prevent bioequivalence fraud.

* Establishment of the FDA Office of Ombudsman reporting directly to the Commissioner. The Ombudsman receives and investigates complaints from both within and outside the FDA and provides remedies where necessary. Mylan Labora-tories had complained repeatedly to DGD that they believed there were irregularities in the review queue order and the order of approvals, and claimed these complaints were ignored by those responsible for DGD management (17,22). Had the early complaints been properly investigated, the outcome in the public view would have been much more favorable for FDA.

The root cause of the generic drug scandal was the large number of generic product opportunities suddenly presented by the Hatch-Waxman Act and the profits that could be made by companies securing competitive FDA application approvals for these new generic drug products. Some companies engaged in criminal activities in order to gain an advantage in being first to market and to reap large profits (32). The fact that this was not isolated to one or two companies but was apparently widespread in the fledgling generic drug industry brought the whole industry into disrepute and severely shook confidence in generic drug pro-ducts (26). It is apparent that FDA was too trusting of the new generic drug industry and of its own staff. In retrospect, the filing of such a large number of ANDAs in the months following passage of the Hatch-Waxman Act should have been a warning sign that all was not as it seemed with these applications (36). The lesson from the scandal is that when you do not have a history of an industry, such as the largely new generic drug industry, and there are potentially large “windfall” profits to be made, you need to put in place robust systems to prevent bribery and fraud.

3. “First-in, first reviewed” and the “bundling” of approvals

According to FDA procedures for the review of generic drug applications, applications should be reviewed in the order in which they are received (23). There are several common situations with regard to possible time-of-approval scenarios for multiple ANDAs for the same RLD (37,38).

3.1. Approval limited by a patent expiry where patent life post-RLD-approval is 45 years

In this situation, applications can be filed over a fairly long period of time and reviewed in the normal course of OGD ANDA review. Several applications can be reviewed and ready for approval on the day of patent expiry, so multiple generic approvals can be expected regardless of the initial quality of the ANDA filings because there are several years in which to address and correct deficiencies. This situation provides a low competitive scenario and little incentive to file high quality applications. Provided the filing is made several years ahead of patent expiry even poor quality applications should be approved at market formation.

3.2. Approval is limited by new chemical entity (NCE) exclusivity, no listed patents

In this situation, the time between filing and NCE expiry is limited and so only better quality applications, if reviewed in the normal course of OGD ANDA review, would be expected to receive approval at the expiry of NCE exclusivity. In this situation it is possible to “bundle” approvals to ensure multiple approvals at NCE expiry. Although one may expect that multiple approvals will ensure low prices and remove the temptation of a windfall profit, it also removes the incentive to produce a high quality filing since filings of lower quality are “pushed through” to ensure multiple day 1 approvals.

3.3. Approval is sought by patent challenge, the so-called “Paragraph IV” filings

In this situation, one or more Orange Book listed patents is challenged as not infringed, invalid, or unenforceable. The law provides that the first person to file an application containing a Paragraph IV certification is entitled to 6 months of marketing exclusivity if the applicant is sued for patent infringement and successful in winning that suit (21). Obviously the prospect of 6 months of exclusive marketing is a potentially large windfall profitandan enormous incentive to achieve the first filed position. This situation strongly encourages filing quickly rather than after good product development; the reward is for being first, not for being good.

Clearly what is occurring here in the first two situations is that FDA is attempting to remove the windfall profit incentive through ensuring competition in as many situations as possible as a way of eliminating the commercial incentive that was the cause of the fraudulent applications during the first 5 years after the enactment of the Hatch-Waxman Act. Unfortunately, this also largely removes the incentive for doing good development and filing complete, high quality applications, since that approach costs more and provides no advantage. Policy makers should consider the use of commercial advantage through more timely approvals to encourage the behavior they want to see, high quality filings and good development of robust products. High quality development is more expensive and takes longer so the best way to ensure it occurs is to make it more profitable (39).

In the third situation, the first-to-file Paragraph IV, the legisla-tion potentially rewards the first filer, and in doing so effectively punishes other filers. This is perhaps the ultimate encouragement to do bad product development as speed to file is the sole determinant of success. It is virtually impossible in this situation to do good product development and succeed in filing first. The current state is that many ANDA filings contain Paragraph IV patent certifications rather than the few in the first decade of the Hatch-Waxman Act (21). The incentive to challenge patents should be the possibility of obtaining a competitive approval if both your product development and your legal basis are of high quality.

If a regulatory authority wants to encourage good product development and high quality filings, they should align these aims with the commercial success gained from high quality product development. If approvals are bundled together regardless of filing quality, or if filing first rather than filing best is rewarded, the senior corporate managers will not spend the time and money required for a high quality development.

4. Bioequivalence and switchability

The system of marketing generic drugs in the United States is one of “switching” from a brand product to a generic equivalent (usually an AB-rated generic in the Orange Book). This substitu-tion is performed by the pharmacist, the prescription is most often written as a brand name product. One consequence of this system of substitution is that patients have their medication “switched” either from brand to generic or from one generic to another generic without input from or knowledge of the physician or the patient (40,41).

The system of determining equivalents depends on the product type (42,43). The most frequent types are:

* Bioequivalence determined by single dose blood level studies, with the test and reference products determined to be bioequivalent if Cmax and AUC meet confidence interval requirements of 80%–125% at the 90% level. AUC measures the extent of drug absorption (or exposure), and Cmax is a surrogate measure of rate of absorption (that is over what time period the drug is absorbed).

* Bioequivalence determined by clinical equivalence studies either with or without blood level studies. This system typically is used when the action of the drug in the drug product is not a result of systemic absorption of the drug. For example topically applied drug products or inhaled drug products generally fall into this category.

* Bioequivalence determined by the rate of in vitro drug dissolution from the drug product. This is generally applied to situations where the drug is highly soluble and no in vivo nonequivalence issues are expected.

* Bioequivalence is determined because the drug products are true solutions of the drug substance. Generally, if the true solution is not a solution for oral administration, then the formulation of the product must also be the same as the RLD (with certain limited exceptions). Some drug products in this class also require equivalent systems of administration (e.g., nasal spray, inhaled nebulizer, ophthalmic dropper).