December 2004
Follow-on Drug Development: Wasteful Imitation or Productive Competition?
Several recent books have criticized the pharmaceutical industry for developing too many "follow-on" or "me-too" drugs--a name given to drugs that work the same way as pioneering drugs that create a new class of treatments. Do follow-on drugs raise costs while diverting R&D funds from true innovation, and should the FDA raise the bar for approving follow-on drugs? Or are these drugs just competition at work, generating lower prices and better products? These and other questions were addressed at a December 15 AEI conference that included a presentation by Joseph A. DiMasi of the Tufts Center for the Study of Drug Development, who has recently completed a new study with colleague Cherie Paquette on "The Economics of Follow-on Drug Research and Development: Trends in Entry Rates and the Timing of Development." The study was published in PharmacoEconomics in December 2004.
Joseph A. DiMasi
Tufts University
"Me-too" drugs are the basis of one of the most contentious charges leveled at the pharmaceutical industry. Critics cite these drugs as evidence that much pharmaceutical research and development (R&D) is wasteful.
A common caricature of the "me-too" (or less pejoratively "follow-on") drug phenomenon is that once a pharmaceutical firm produces a breakthrough drug and establishes a new subclass, other manufacturers scramble to tweak the drug's formulation and market their own copycats. These criticisms of the pharmaceutical industry are not new. More than forty years ago, the Kefauver hearings contended that this type of research could be eliminated without stifling what they deemed real innovation.
Follow-on drugs deserve closer inspection. Are they duplicates created after a pioneer drug hits the market, or are they the products of concurrent, competing R&D efforts? Are they perfect substitutes, or do their diverse safety and side-effect profiles suit them better to some patients over others? Is the breakthrough drug necessarily the best? Can competition among follow-ons bring prices down?
In a study published four years ago, I examined the prices of twenty drugs entering existing classes between 1995 and 1999. The data suggested that follow-on drugs did induce some competition. On average, "follow-ons" were priced 26 percent below the price leader (maximum price) and 14 percent below the class average. Some even undercut the average price by more than 40 percent. Having several drugs in a therapeutic class gives managed care plans greater leverage to negotiate discounts from manufacturers, but those additional rebates were not captured in these data.
Market exclusivity for brand-name drugs has declined considerably over the last decades. One dramatic depiction of this trend shows thirteen years of exclusivity for one drug launched in 1965, compared to just three months for one that reached the market in 1999. While exclusivity periods have indeed shrunk over time, the true magnitude of that shift has not been carefully documented.
Our recent study on follow-on drugs examined drug approvals in the United States since 1960 to identify therapeutic subclasses. We constructed groups of drugs prescribed for the same indication that were chemically similar or had the same mechanism of action. We identified seventy-two subclasses where the pioneer drug was approved between 1960 and 1998, and tracked 235 follow-on drugs in those subclasses that were approved through 2003. Each subclass contained between two and sixteen drugs, with a mean of 4.3 and a median of 3, including the originator.
Pioneer drugs in a therapeutic subclass have enjoyed increasingly shorter periods of market exclusivity before a follow-on entered the class. In the 1960s and 1970s, breakthrough drugs faced no competition from therapeutic equivalents for seven or eight years, on average. By the late 1990s, first-in-class drugs generally had less than two years of market exclusivity.
Several developments have facilitated the entry of follow-on drugs into existing subclasses. The Hatch-Waxman Act of 1984 made it easier to get generics approved, dramatically reducing the costs of imitation. With generic competition approaching sooner, follow-on drug manufacturers face additional pressure to get their drugs to market. The growth in prescription drug coverage has also expanded markets and may have increased the incentives to speed the development of follow-ons. Biomedical advances have also fostered more targeted drug research.
For internal management, the Food and Drug Administration (FDA) assigns a therapeutic rating to all the compounds it reviews. "Priority" molecules are considered more innovative than "standard" ones and are given an expedited review. Of 228 follow-on drug submissions, the FDA rated one-third "priority," and most subclasses (57 percent) have at least one follow-on drug with a priority rating. This challenges the characterization that "me-too" drugs are not innovative or that they add no value to existing medicines. It also contests the notion that the first drug in a particular class is necessarily the best.
The production of follow-on drugs can be better characterized as a development race where only one can be the winner than as the imitation of a pioneer drug. Since the 1980s, by the time the breakthrough drug was approved for each class, at least one follow-on drug had already been synthesized and was undergoing initial pharmacological testing. By the 1990s, most follow-ons were already in clinical testing by the time the first-in-class drug was approved. Other evidence of parallel development comes from data on when pioneer drugs and their competitors reached certain clinical milestones. About one-third of follow-on drugs had filed an investigational new drug application (IND) before the breakthrough drug in their class had done so.
Critics of follow-on drugs have proposed a number of policies to redirect firms' R&D efforts to breakthrough drug development. Two demand-side options considered have been to slash reimbursements for follow-ons or to reward breakthroughs with price premia. Such policies aim to imitate the market, but they would do so imperfectly. A better solution would be to improve information in the marketplace, so that consumers, physicians, and payers could make better decisions when they buy drugs. The most troubling policy proposals involve creating registration hurdles for drugs entering an existing class.
These policy fixes--especially those targeting the supply of follow-ons--are problematic. Chemically similar drugs can function differently--better or worse in different patients--so picking winners among them would inevitably hurt some patients. Moreover, drugs can be proven harmful long after their approval and then pulled from the market. If we relied on only one drug in each class and that drug was recalled for safety reasons, patients would be left with no alternative. When multiple firms pursue therapies for a particular indication, it is more likely that a drug with a favorable benefit-risk ratio will emerge.
Some recent proposals have suggested that the registration process somehow distinguish between breakthrough and follow-on drugs. Our analysis shows that most follow-on development is already underway when breakthroughs are approved, so how would the FDA recognize the pioneers and follow-ons beforehand? If the registration process were amended to require demonstration of a drug's superiority over its therapeutic equivalents, development costs for those drugs would soar. An even greater deterrent, however, would be the sharp increase in uncertainty, as firms would have to revamp their development plans midstream as new drugs were approved. The FDA would be asking firms to hit a moving target, and many of them would simply not take the risk.
Alan Lyles
University of Baltimore
The pharmaceutical industry was not always concerned with the threat of follow-on competitors. In 1952, George Merck was featured on the cover of Time magazine after donating the patent for the streptomycin antibiotic to Rutgers University so that it could be more widely distributed and available at lower cost.
In recent decades, several developments have influenced pharmaceutical innovation, the convergence of scientific interests, and collaboration among players. The Bayh-Dole Act of 1980 (and amended in 1984) authorized the transfer of technology between universities and industry. Cooperative Research and Development Agreements (CRADAs) between industry and government emerged in 1986. The Prescription Drug User Fee Act (PDUFA) of 1992 (and amended in 1997 and 2002) responded to a bottleneck in drug approvals at the FDA. Between 1993 and 1995, the FDA approved an average of twenty-five new molecular entities (NMEs) each year. By 1996, the expedited reviews and additional reviewers funded under fees generated by PDUFA began to take effect; fifty-three NMEs were approved that year. From 1997 to 1999, an average of 34.7 NMEs were approved each year. Evidence suggests that industry and government cooperation to speed medicines to the market had worked.
These developments also impacted the number of first-in-class approvals. In the 1960s, eight breakthrough drugs were launched, but between 1995 and 1998, there were eighteen first-in-class approvals. Over this period, however, breakthrough drugs enjoyed shorter periods of market exclusivity. This decline is due to various factors, and we can reasonably expect this trend to abate somewhat in the future.
Not all follow-on drugs are alike. Professor DiMasi's study excludes a class of incrementally modified drugs (IMDs) that are the classic "me-toos," as opposed to the new molecular entities (NMEs) that DiMasi examined. A National Institute of Health Care Management study that included IMDs showed that between 1989 and 2000, only 15 percent of more than one thousand approvals were for priority NMEs.
Industry consolidation has made financial stakeholders expectant of one blockbuster drug each year. The "Shark Fin Project" describes the expected trajectory of an innovator drug's sales when it faces generic entry down the road. Amid efforts to counter the pressure on sales, firms have invested huge sums in direct-to-consumer advertising of third- and fourth-in-class drugs. The Wall Street Journal estimated in 2002 that follow-on drug marketing of the product in the Shark Fin Project cost between $300 and $500 million.
The growth of prescription drug coverage and perhaps rising wealth have expanded pharmaceutical markets and boosted profitability. If a rising tide raises all boats, a follow-on strategy of genuine me-too drugs may be financially attractive.
Mergers and acquisitions have retarded the drug development pipeline, as R&D offices have not integrated smoothly. In an industry that has been reliant on blockbuster products, we may begin to see more planned follow-on development and greater investment in IMDs in order to secure market share and predictable revenue increases.
Scott Gottlieb
AEI
The culture inside the FDA with regard to follow-ons may be instructive in this debate. The FDA recognizes that similar molecules can be markedly different. If one drug shows some extra benefit, that claim is not extended to all drugs in its class. The same is true of one drug's detriment. Vioxx's recall has not brought down all the Cox-IIs, and the heightened risks associated with one antidepressant were not ascribed to all SSRIs.
Firms do not set out to develop follow-ons. They chase a target simultaneously, but only one drug can be the first in a new class. EGFR-inhibitors to treat cancer and protease inhibitors for Hepatitis C, among many others, were developed this way.
From a clinical perspective, there is great value in variety. Even seemingly identical molecules can have sharply different therapeutic effects. Among the interferons prescribed for multiple sclerosis, there is a trade-off between efficacy and tolerability. Even if the efficacy of various atypical antipsychotics is similar, their side-effect profiles can be widely different. Some induce weight gain, others are stimulating, and still others are sedating. For this class of medicines in particular, patients often choose their therapy based on its side effects, though pricing can vary substantially among them as well. IIB and IIA inhibitors given to patients suffering a heart attack can vary widely in potency and duration of action. If a patient is headed to surgery, a shorter-acting drug is preferable, but that advantage would be lost without class competitors. Even medications for erectile dysfunction can differ significantly in how fast they take effect, how long they act, and how much they cost.
Subtle differentiation is extremely important to patient choice. Payers decry follow-on R&D and lay blame on the pharmaceutical industry. In fact, clinicians and patients drive these developments with their own judgments of competing drugs' values. The Medicare Modernization Act of 2003 is likely to further encourage follow-on development because pharmaceutical companies are now poised to negotiate entire panels of their drugs with insurers.
AEI research assistant Ximena Pinell prepared this summary.
