Chapter 1: Drug Invention and the Pharmaceutical Industry

The first edition of this textbook, published in 1941, is often credited with organizing the field of pharmacology, giving it intellectual validity and an academic identity ^*. That first edition began: "The subject of pharmacology is a broad one and embraces the knowledge of the source, physical and chemical properties, compounding, physiological actions, absorption, fate, and excretion, and therapeutic uses of drugs. A drug may be broadly defined as any chemical agent that affects living protoplasm, and few substances would escape inclusion by this definition." These two sentences still serve us well. This first section of the 12th edition of this textbook provides the underpinnings for these definitions by exploring the processes of drug invention and development into a therapeutic entity, followed by the basic properties of the interactions between the drug and biological systems: pharmacodynamics, pharmacokinetics (including drug transport and metabolism), and pharmacogenomics. Subsequent sections deal with the use of drugs as therapeutic agents in human subjects.

We intentionally use the term invention to describe the process by which a new drug is identified and brought to medical practice, rather than the more conventional term discovery. This significant semantic change was suggested to us by our colleague Michael S. Brown, MD, and it is appropriate. In the past, drugs were discovered as natural products and used as such. Today, useful drugs are rarely discovered hiding somewhere waiting to be found; rather, they are sculpted and brought into being based on experimentation and optimization of many independent properties. The term invention emphasizes this process; there is little serendipity.

Man's fascination—and sometimes infatuation—with chemicals (i.e., drugs) that alter biological function is ancient and arose as a result of experience with and dependence on plants. Most plants are root-bound, and many have become capable of elaborate chemical syntheses, producing harmful compounds for defense that animals learned to avoid and man learned to exploit. Many examples are described in earlier editions of this text: the appreciation of coffee (caffeine) by the prior of an Arabian convent who noted the behavior of goats that gamboled and frisked through the night after eating the berries of the coffee plant, the use of mushrooms or the deadly nightshade plant (containing the belladonna alkaloids atropine and scopolamine) by professional poisoners, and a rather different use of belladonna ("beautiful lady") to dilate pupils. Other examples include the uses of the Chinese herb ma huang (containing ephedrine) for over 5000 years as a circulatory stimulant, curare-containing arrow poisons used for centuries by South American Indians to paralyze and kill animals hunted for food, and poppy juice (opium) containing morphine (from the Greek Morpheus, the god of dreams) for pain relief and control of dysenteries. Morphine, of course, has well-known addicting properties, mimicked in some ways by other problematic ("recreational") natural products—nicotine, cocaine, and ethanol.

While many terrestrial and marine organisms remain valuable sources of naturally occurring compounds with various pharmacological activities, especially including lethal effects on both microorganisms and eukaryotic cells, drug invention became more allied with synthetic organic chemistry as that discipline flourished over the past 150 years. This revolution began in the dye industry. Dyes, by definition, are colored compounds with selective affinity for biological tissues. Study of these interactions stimulated Paul Ehrlich to postulate the existence of chemical receptors in tissues that interacted with and "fixed" the dyes. Similarly, Ehrlich thought that unique receptors on microorganisms or parasites might react specifically with certain dyes and that such selectivity could spare normal tissue. Ehrlich's work culminated in the invention of arsphenamine in 1907, which was patented as "salvarsan," suggestive of the hope that the chemical would be the salvation of humankind. This arsenic-containing compound and other organic arsenicals were invaluable for the chemotherapy of syphilis until the discovery of penicillin. During that period and thanks to the work of Gerhard Domagk, another dye, prontosil (the first clinically useful sulfonamide) was shown to be dramatically effective in treating streptococcal infections. The era of antimicrobial chemotherapy was born, and the fascination with dyes soon spread to the entire and nearly infinite spectrum of organic chemicals. The resulting collaboration of pharmacology with chemistry on the one hand, and with clinical medicine on the other, has been a major contributor to the effective treatment of disease, especially since the middle of the 20th century.

Small Molecules Are the Tradition

With the exception of a few naturally occurring hormones such as insulin, most drugs were small organic molecules (typically <500 Da) until recombinant DNA technology permitted synthesis of proteins by various organisms (bacteria, yeast) and mammalian cells, starting in the 1980s. The usual approach to invention of a small-molecule drug is to screen a collection of chemicals ("library") for compounds with the desired features. An alternative is to synthesize and focus on close chemical relatives of a substance known to participate in a biological reaction of interest (e.g., congeners of a specific enzyme substrate chosen to be possible inhibitors of the enzymatic reaction), a particularly important strategy in the discovery of anticancer drugs.

While drug discovery in the past often resulted from serendipitous observations of the effects of plant extracts or individual chemicals administered to animals or ingested by man, the approach today relies on high-throughput screening of libraries containing hundreds of thousands or even millions of compounds for their ability to interact with a specific molecular target or elicit a specific biological response (see "Targets of Drug Action" later in the chapter). Chemical libraries are synthesized using modern organic chemical synthetic approaches such as combinatorial chemistry to create large collections of related chemicals, which can then be screened for activity in high-throughput systems. Diversity-oriented synthetic approaches also are of obvious value, while natural products (plant or marine animal collections) are sources of novel and sometimes exceedingly complex chemical structures.

Several variables affect the frequency of hits obtained in a screen. Among the most important are the "drugability" of the target and the stringency of the screen in terms of the concentrations of compounds that are tested. The slang term "drugability" refers to the ease with which the function of a target can be altered in the desired fashion by a small organic molecule. If the protein target has a well-defined binding site for a small molecule (e.g., a catalytic or allosteric site), chances are excellent that hits will be obtained. If the goal is to employ a small molecule to mimic or disrupt the interaction between two proteins, the challenge is much greater.

From Hits to Leads

Only rarely do any of the initial hits in a screen turn out to be marketable drugs. Initial hits often have modest affinity for the target, and lack the desired specificity and pharmacological properties of a successful pharmaceutical. Skilled medicinal chemists synthesize derivatives of the hits, making substitutions at accessible positions, and begin in this way to define the relationship between chemical structure and biological activity. Many parameters may require optimization, including affinity for the target, agonist/antagonist activity, permeability across cell membranes, absorption and distribution in the body, metabolism of the drug, and unwanted effects. While this approach was driven largely by instinct and trial and error in the past, modern drug development frequently takes advantage of determination of a high-resolution structure of the putative drug bound to its target. X-ray crystallography offers the most detailed structural information if the target protein can be crystallized with the lead drug bound to it. Using techniques of molecular modeling and computational chemistry, the structure provides the chemist with information about substitutions likely to improve the "fit" of the drug with the target and thus enhance the affinity of the drug for its target (and, hopefully, optimize selectivity of the drug simultaneously). Nuclear magnetic resonance (NMR) spectroscopy is another valuable technique for learning the structure of a drug-receptor complex. NMR studies are done in solution, with the advantage that the complex need not be crystallized. However, the structures obtained by NMR spectroscopy usually are not as precise as those from X-ray crystallography, and the protein target must not be larger than roughly 35–40 kDa.

The holy grail of this approach to drug invention will be to achieve success entirely through computation. Imagine a database containing detailed chemical information about millions of chemicals and a second database containing detailed structural information about all human proteins. The computational approach is to "roll" all the chemicals over the protein of interest to find those with high-affinity interactions. The dream gets bolder if we acquire the ability to roll the chemicals that bind to the target of interest over all other human proteins to discard compounds that have unwanted interactions. Finally, we also will want to predict the structural and functional consequences of a drug binding to its target (a huge challenge), as well as all relevant pharmacokinetic properties of the molecules of interest. We are a long way from realization of this fabulous dream; however, we are sufficiently advanced to imagine it and realize that it could someday be a reality. Indeed, computational approaches have suggested new uses for old drugs and offered explanations for recent failures of drugs in the later stages of clinical development (e.g., torcetrapib; see below) (Kim et al., 2010; Kinnings et al., 2009; Xie et al., 2007, 2009).

Large Molecules are Increasingly Important

Protein therapeutics were uncommon before the advent of recombinant DNA technology. Insulin was introduced into clinical medicine for the treatment of diabetes following the experiments of Banting and Best in 1921. Insulin could be produced in great quantities by purification from porcine or bovine pancreas obtained from slaughter houses. These insulins are active in man, although antibodies to the foreign proteins are occasionally problematic.

Growth hormone, used to treat pituitary dwarfism, is a case of more stringent species specificity: only the human hormone could be used after purification from pituitary glands harvested during autopsy. The danger of this approach was highlighted when patients who had received the human hormone developed Creutzfeldt-Jakob disease (the human equivalent of mad cow disease), a fatal degenerative neurological disease caused by prion proteins that contaminated the drug preparation. Thanks to gene cloning and the ability to produce large quantities of proteins by expressing the cloned gene in bacteria or eukaryotic cells grown in enormous (30,000-liter) bioreactors, protein therapeutics now utilize highly purified preparations of human (or humanized) proteins. Rare proteins can now be produced in quantity, and immunological reactions are minimized. Proteins can be designed, customized, and optimized using genetic engineering techniques. Other types of macromolecules may also be used therapeutically. For example, antisense oligonucleotides are used to block gene transcription or translation, as are small interfering RNAs (siRNAs).

Proteins utilized therapeutically include various hormones, growth factors (e.g., erythropoietin, granulocyte-colony stimulating factor), and cytokines, as well as a rapidly increasing number of monoclonal antibodies now widely used in the treatment of cancer and autoimmune diseases. Murine monoclonal antibodies can be "humanized" (by substituting human for mouse amino acid sequences). Alternatively, mice have now been "engineered" by replacement of critical mouse genes with their human equivalents, such that they make completely human antibodies. Protein therapeutics are administered parenterally, and their receptors or targets must be accessible extracellularly.

The earliest drugs came from observation of the effects of plants after their ingestion by animals. One could observe at least some of the effects of the chemical(s) in the plant and, as a side benefit, know that the plant extract was active when taken orally. Valuable drugs were discovered with no knowledge of their mechanism or site of action. While this approach is still useful (e.g., in screening for the ability of natural products to kill microorganisms or malignant cells), modern drug invention usually takes the opposite approach—starting with a statement (or hypothesis) that a certain protein or pathway plays a critical role in the pathogenesis of a certain disease, and that altering the protein's activity would therefore be effective against that disease. Crucial questions arise:

• can one find a drug that will have the desired effect against its target?

• does modulation of the target protein affect the course of disease?

• does this project make sense economically?

The effort that may be expended to find the desired drug will be determined by the degree of confidence in the answers to the latter two questions.

Is the Target "Drugable"?

The drugability of a target with a low-molecular-weight organic molecule relies on the presence of a binding site for the drug that can be approached with considerable affinity and selectivity. If the target is an enzyme or a receptor for a small ligand, one is encouraged. If the target is related to another protein that is known to have, for example, a binding site for a regulatory ligand, one is hopeful. However, if the known ligands are large peptides or proteins with an extensive set of contacts with their receptor, the challenge is much greater. If the goal is to disrupt interactions between two proteins, it may be necessary to find a "hot spot" that is crucial for the protein-protein interaction, and such a region may not be detected. Accessibility of the drug to its target also is critical. Extracellular targets are intrinsically easier to approach and, in general, only extracellular targets are accessible to macromolecular drugs.

Has the Target Been Validated?

This question is obviously a critical one. A negative answer, frequently obtained only retrospectively, is a common cause of failure in drug invention. Based on extensive study of a given biological process, one may believe that protein X plays a critical role in pathological alterations of that process. However, biological systems frequently contain redundant elements, and they are adaptable. When the activity of protein X is, e.g., inhibited by a drug, redundancy in the system may permit compensation. The system also may adapt to the presence of the drug, perhaps by regulating the expression of the target or of functionally related gene products. In general, the more important the function, the greater the complexity of the system. For example, many mechanisms control feeding and appetite, and drugs to control obesity have been notoriously difficult to find. The discovery of the hormone leptin, which suppresses appetite, was based on mutations in mice that cause loss of either leptin or its receptor; either kind of mutation results in enormous obesity in both mice and people. Leptin thus appeared to be a marvelous opportunity to treat obesity. However, obese individuals have high circulating concentrations of leptin and appear quite insensitive to its action.

Modern techniques of molecular biology offer new and powerful tools for validation of potential drug targets, to the extent that the biology of model systems resembles human biology. Genes can be inserted, disrupted, and altered in mice. One can thereby create models of disease in animals or mimic the effects of long-term disruption or activation of a given biological process. If, for example, disruption of the gene encoding a specific enzyme or receptor has a beneficial effect in a valid murine model of a human disease, one may believe that the potential drug target has been validated. Mutations in humans also can provide extraordinarily valuable information. For example, loss-of-function mutations in the PCSK9 gene (encoding proprotein convertase subtilisin/kexin type 9) greatly lower concentrations of LDL cholesterol in plasma and reduce the risk of myocardial infarction (Horton et al., 2009). This single powerful observation speaks to a well-validated drug target. Based on these findings, many drug companies are actively seeking inhibitors of PCSK9 function.

Is This Drug Invention Effort Economically Viable?

Drug invention and development is extraordinarily expensive, as discussed later in the chapter. Economic realities influence the direction of science. For example, investor-owned companies generally cannot afford to develop products for rare diseases or for diseases that are common only in economically underdeveloped parts of the world. Funds to invent drugs targeting rare diseases or diseases primarily affecting developing countries (especially parasitic diseases) can come from taxpayers or very wealthy philanthropists; such funds generally will not come from private investors involved in running for-profit companies.

Following the path just described can yield a potential drug molecule that interacts with a validated target and alters its function in the desired fashion (either enhancing or inhibiting the functions of the target). Now one must consider all aspects of the molecule in question—its affinity and selectivity for interaction with the target, its pharmacokinetic properties (absorption, distribution, excretion, metabolism), issues with regard to its large-scale synthesis or purification from a natural source, its pharmaceutical properties (stability, solubility, questions of formulation), and its safety. One hopes to correct, to the extent possible, any obvious deficiencies by modification of the molecule itself or by changes in the way the molecule is presented for use.

Before being administered to people, potential drugs are tested for general toxicity by monitoring the activity of various systems in two species of animals for extended periods of time. Compounds also are evaluated for carcinogenicity, genotoxicity, and reproductive toxicity. Animals are used for much of this testing, although the predictive value of results obtained in nonhuman species is certainly not perfect. Usually one rodent (usually mouse) and one non-rodent (often rabbit) species are used. In vitro and ex vivo assays are utilized when possible, both to spare animals and to minimize cost. If an unwanted effect is observed, an obvious question is whether it is mechanism-based (i.e., caused by interaction of the drug with its intended target) or due to an off-target effect of the drug. If the latter, there is hope of minimizing the effect by further optimization of the molecule.

Before clinical trials of a potential new drug may proceed in the U.S. (that is, before the drug candidate can be administered to a human subject), the sponsor must file an IND (Investigational New Drug) application, which is a request to the U.S. Food and Drug Administration (FDA; see the next section) for permission to administer the drug to human test subjects. The IND describes the rationale and preliminary evidence for efficacy in experimental systems, as well as pharmacology, toxicology, chemistry, manufacturing, and so forth. It also describes the plan for investigating the drug in human subjects. The FDA has 30 days to review the application, by which time the agency may disapprove the application, ask for more data, or allow initial clinical testing to proceed. In the absence of an objection or request for more information within 30 days by the FDA, a clinical trial may begin.

The FDA is a regulatory agency within the U.S. Department of Health and Human Services. As its mission statement indicates, the FDA:

• is responsible for protecting the public health by assuring the safety, efficacy, and security of human and veterinary drugs, biological products, medical devices, our nation's food supply, cosmetics, and products that emit radiation. The FDA is also responsible for advancing the public health by helping to speed innovations that make medicines and foods more effective, safer, and more affordable; and helping the public get the accurate, science-based information they need to use medicines and foods to improve their health (FDA, 2009).

One of the agency's primary responsibilities is to protect the public from harmful medications. However, the FDA clearly faces an enormous challenge, especially in view of the widely held belief that its mission cannot possibly be accomplished with the available resources. Moreover, harm from drugs that cause unanticipated adverse effects is not the only risk of an imperfect system; harm also occurs when the approval process delays the marketing of a new drug with important beneficial effects. Determining safety and efficacy prior to mass marketing requires careful consideration.

The Conduct of Clinical Trials

Clinical trials (as applied to drugs) are investigations in human subjects intended to acquire information about the pharmacokinetic and pharmacodynamic properties of a potential drug. Depending on the nature and phase of the trial, it may be designed to evaluate a drug's safety, its efficacy for treatment or prevention of specific conditions in patients, and its tolerability and side effects. Efficacy must be proven and an adequate margin of safety established for a drug to be approved for sale in the U.S. The U.S. National Institutes of Health notes seven ethical requirements that must be met before a clinical trial can begin. These include social value, scientific validity, fair and objective selection of subjects, informed consent, favorable ratio of risks to benefits, approval and oversight by an independent review board (IRB), and respect for human subjects.

FDA-regulated clinical trials typically are conducted in four phases. The first three are designed to establish safety and efficacy, while Phase IV post-marketing trials delineate additional information regarding new indications, risks, and optimal doses and schedules. Table 1–1 and Figure 1–1 summarize the important features of each phase of clinical trials, especially the attrition at each successive stage over a relatively long and costly process.

When initial Phase III trials are complete, the sponsor (usually a pharmaceutical company) applies to the FDA for approval to market the drug; this application is called either a New Drug Application (NDA) or a Biologics License Application (BLA). These applications contain comprehensive information, including individual case-report forms from the hundreds or thousands of individuals who have received the drug during its Phase III testing. Applications are reviewed by teams of specialists, and the FDA may call on the help of panels of external experts in complex cases. The use of such external advisory committees greatly expands the talent pool available to assist in making important and difficult decisions.

Under the provisions of the Prescription Drug User Fee Act (PDUFA, enacted initially in 1992 and renewed in 2007), pharmaceutical companies now provide a significant portion of the FDA budget via user fees, a legislative effort to expedite the drug approval review process. PDUFA also broadened the FDA's drug safety program and increased resources for review of television drug advertising. The larger FDA staffing now in place has shortened the time required for review; nevertheless, the process is a lengthy one. A1-year review time is considered standard, and 6 months is the target if the drug candidate is granted priority status because of its importance in filling an unmet need. Unfortunately, these targets are not always met.

Before a drug is approved for marketing, the company and the FDA must agree on the content of the "label" (package insert)—the official prescribing information. This label describes the approved indications for use of the drug and clinical pharmacological information including dosage, adverse reactions, and special warnings and precautions (sometimes posted in a "black box"). Promotional materials used by pharmaceutical companies cannot deviate from information contained in the package insert. Importantly, the physician is not bound by the package insert; a physician in the U.S. may legally prescribe a drug for any purpose that she or he deems reasonable. However, third-party payers (insurance companies, Medicare, and so on) generally will not reimburse a patient for the cost of a drug used for an "off-label" indication unless the new use is supported by one of several compendia such as the U.S. Pharmacopeia. Furthermore, a physician may be vulnerable to litigation if untoward effects result from an unapproved use of a drug.

Determining "Safe" and "Effective"

To demonstrate efficacy to the FDA requires performing "adequate and well-controlled investigations," generally interpreted to mean two replicate clinical trials that are usually, but not always, randomized, double-blind, and placebo-controlled. Is a placebo the proper control? The World Medical Association's Declaration of Helsinki (2000) discourages use of placebo controls when an alternative treatment is available for comparison. What must be measured in the trials? In a straightforward trial, a readily quantifiable parameter (a secondary or surrogate end point), thought to be predictive of relevant clinical outcomes, is measured in matched drug- and placebo-treated groups. Examples of surrogate end points include LDL cholesterol as a predictor of myocardial infarction, bone mineral density as a predictor of fractures, or hemoglobin A_1c as a predictor of the complications of diabetes mellitus. More stringent trials would require demonstration of reduction of the incidence of myocardial infarction in patients taking a candidate drug in comparison with those taking an HMG-CoA reductase inhibitor (statin) or other LDL cholesterol-lowering agent, or reduction in the incidence of fractures in comparison with those taking a bisphosphonate. Use of surrogate end points significantly reduces cost and time required to complete trials but there are many mitigating factors, including the significance of the surrogate end point to the disease or condition that the candidate drug is intended to treat.

The concept of drug safety is perhaps even more complex (Institute of Medicine, 2007). No drug is totally safe; all drugs produce unwanted effects in at least some people at some dose. Many unwanted and serious effects of drugs occur so infrequently, perhaps only once in several thousand patients, that they go undetected in the relatively small populations (a few thousand) in the standard Phase III clinical trial (Table 1–1). To detect and verify that such events are in fact drug-related would require administration of the drug to tens or hundreds of thousands of people during clinical trials, adding enormous expense and time to drug development and delaying access to potentially beneficial therapies. In general, the true spectrum and incidence of untoward effects becomes known only after a drug is released to the broader market and used by a large number of people (Phase IV, post-marketing surveillance). Drug development costs, and thus drug prices, could be reduced substantially if the public were willing to accept more risk. This would require changing the way we think about a pharmaceutical company's liability for damages from an unwanted effect of a drug that was not detected in clinical trials deemed adequate by the FDA.

While the concept is obvious, many lose sight of the fact that extremely severe unwanted effects of a drug, including death, may be deemed acceptable if its therapeutic effect is sufficiently unique and valuable. Such dilemmas can become issues for great debate. The sufficiency of a therapeutic effect in the presence of an unwanted effect of a drug may be quite subjective. One person's meat may indeed be another person's poison. Great effort may be made to quantify the ratio of risks to benefits, but the answers are frequently not simple.

There is no doubt that drugs can save lives, prolong lives, and improve the quality of people's lives. Like adequate nutrition, vaccinations and medications are important for public health. However, in a free-market economy, access to safe and effective drugs (or any kind of healthcare, for that matter) is not equitable. Not surprisingly, there is a substantial tension between those who would treat drugs as entitlements and those who view drugs as high-tech products of a capitalistic society. Supporters of the entitlement position argue that the constitutional right to life should guarantee access to drugs and other healthcare, and they are critical of pharmaceutical companies and others who profit from the business of making and selling drugs. Free-marketers point out that, without a profit motive, it would be difficult to generate the resources and innovation required for new drug development.

The media tend to focus on public policy with regard to the ethics of drug testing, the effectiveness of government regulations, and conflicts of interest on the part of researchers, physicians, and others who may have a personal stake in the success of a drug. In addition, high-profile legal battles have been waged recently over access to experimental (non-FDA-approved) drugs and over injuries and deaths resulting from both experimental and approved drugs. Clearly the public has an interest in both the pharmaceutical industry and its oversight. Consequently, drug development is not only a scientific process but also a political one in which attitudes can change quickly. Little more than a decade ago Merck was named as America's most admired company by Fortune magazine seven years in a row—a record that still stands. Today, Johnson and Johnson is the only pharmaceutical company in the top 50 of the most-admired list, and this likely reflect their sales of consumer products, such as band-aids and baby oil, rather than pharmaceuticals. The next sections explore some of the more controversial issues surrounding drug invention and development and consider some of the more strident criticisms that have been leveled at the pharmaceutical industry (Angell, 2004).

Mistrust of Scientists and Industry

Those critical of the pharmaceutical industry frequently begin from the position that people (and animals) need to be protected from greedy and unscrupulous companies and scientists (Kassirer, 2005). They can point to the very unfortunate (and highly publicized) occurrences of graft, fraud, and misconduct by scientists and industry executives, and unethical behavior in university laboratories and community physicians' offices. These problems notwithstanding, development of new and better drugs is good for people and animals. In the absence of a government-controlled drug development enterprise, our current system relies predominantly on investor-owned pharmaceutical companies that, like other companies, have a profit motive and an obligation to shareholders.

Pricing and Profitability

The price of prescription drugs causes great consternation among consumers, especially as many health insurers seek to control costs by choosing not to cover certain "brand name" products. Further, a few drugs (especially for treatment of cancer) have been introduced to the market in recent years at prices that greatly exceeded the costs of development, manufacture, and marketing of the product. Many of these products were discovered in government laboratories or in university laboratories supported by federal grants. The U.S. is the only large country in the world that places no controls on drug prices and where price plays no role in the drug approval process. Many U.S. drugs cost much more in the United States than overseas. The result is that U.S. consumers subsidize drug costs for the rest of the world, including the economically developed world, and they are irritated by that fact.

Who Pays?

Healthcare in the U.S. is funded by a mix of private payers and government programs. Correspondingly, the cost of prescription drugs is borne by consumers ("out-of-pocket"), private insurers, and public insurance programs like Medicare, Medicaid, and the State Children's Health Insurance Program (SCHIP). Recent initiatives by major retailers and mail-order pharmacies run by private insurers to offer consumer incentives for purchase of generic drugs have helped to contain the portion of household expenses spent on pharmaceuticals; however, more than one-third of total retail drug costs in the U.S. are paid with public funds—tax dollars.

Healthcare in the U.S. is more expensive than everywhere else, but it is not, on average, demonstrably better than everywhere else. However, the U.S. is considerably more socio-economically diverse than many of the countries with which comparisons are made. Forty-five million Americans are uninsured and seek routine medical care in emergency rooms. Remedies are the current subjects of complex medical, public health, economic, and political debates. Solutions to these real problems must recognize both the need for effective ways to incentivize innovation and to permit, recognize, and reward compassionate medical care.

Intellectual Property and Patents

Drug invention, like any other, produces intellectual property eligible for patent protection. Without patent protection, no company could think of making the investments necessary for drug invention and development. With the passage of the Bayh-Dole Act (35 USC 200) in 1980, the federal government created strong incentives for scientists at academic medical centers to approach drug invention with an entrepreneurial spirit. The Act transferred intellectual property rights to the researchers themselves and in some instances to their respective institutions in order to encourage the kinds of partnerships with industry that would bring new products to market, where they could benefit the public. This resulted in the development of "technology transfer" offices at virtually every major university, which help scientists to apply for patents and to negotiate licensing arrangements with industry (Geiger and Sá, 2008). While the need to protect intellectual property is generally accepted, the encouragement of public-private research collaborations has given rise to concerns about conflicts of interest by scientists and universities (Kaiser, 2009).

Despite the complications that come with university-industry relations, patent protection is enormously important for innovation. As noted in 1859 by Abraham Lincoln (the only U.S. president to ever hold a patent [# 6469, for a device to lift boats over shoals]), by giving the inventor exclusive use of his or her invention for limited time, the patent system "added the fuel of interest to the fire of genius, in the discovery and production of new and useful things." The U.S. patent protection system mandates that when a new drug is invented, the patent covering the property lasts only 20 years from the time the patent is filed. During this period, the patent owner may bring suit to prevent others from marketing the product, giving the manufacturer of the brand-name version exclusive rights to market and sell the drug. When the patent expires, equivalent products can come on the market, where they are sold much more cheaply than the original drug, and without the huge development costs borne by the original patent holder. The marketer of the so-called generic product must demonstrate "therapeutic equivalence" of the new product: it must contain equal amounts of the same active chemical ingredient and achieve equal concentrations in blood when administered by the same routes.

Note, however, that the long time course of drug development, usually more than 10 years (Figure 1–1), dramatically reduces the time during which patent protection functions as intended. Although The Drug Price Competition and Patent Term Restoration Act of 1984 (the "Hatch-Waxman Act") permits a patent holder to apply for extension of a patent term to compensate for delays in marketing due to FDA approval processes, patents can be extended only for half the time period consumed by the regulatory approval process, for a maximum of 14 years. The average new drug brought to market now enjoys only ~10-12 years of patent protection. Some argue that patent protection for drugs should be shortened, based on the hope that earlier generic competition will lower healthcare costs. The counter-argument is that new drugs would have to bear higher prices to provide adequate compensation to companies during a shorter period of protected time. If that is true, lengthening patent protection would actually permit lower prices. Recall that patent protection is worth little if a superior competitive product is invented and brought to market at any time in the patent cycle.

Drug Promotion

In an ideal world, physicians would learn all they need to know about drugs from the medical literature, and good drugs would thereby sell themselves; we are a long way from the ideal. Instead we have print advertising and visits from salespeople directed at physicians, and extensive so-called "direct-to-consumer" advertising aimed at the public (in print, on the radio, and especially on television). There are roughly 100,000 pharmaceutical sales representatives in the U.S. who target ~10 times that number of physicians. It has been noted that college cheerleading squads are attractive sources for recruitment of this sales force. The amount spent on promotion of drugs approximates or perhaps even exceeds that spent on research and development. Pharmaceutical companies have been especially vulnerable to criticism for some of their marketing practices.

Exploitation or "Medical Imperialism"

There is concern about the degree to which U.S. and European patent protection laws have restricted access to potentially life-saving drugs in developing countries. Because development of new drugs is so expensive, private-sector investment in pharmaceutical innovation naturally has focused on products that will have lucrative markets in wealthy countries such as the U.S., which combines patent protection with a free-market economy. However, to lower costs, companies increasingly test their experimental drugs outside the U.S. and the E.U., in countries such as China, India, Russia, and Mexico, where there is less regulation and easier access to large numbers of patients. If the drug is successful in obtaining marketing approval, consumers in these countries often cannot afford the drugs they helped to develop. Some ethicists have argued that this practice violates the justice principle articulated in The Belmont Report (1979), which states that "research should not unduly involve persons from groups unlikely to be among the beneficiaries of subsequent applications of the research." On the other hand, the conduct of trials in developing nations also frequently brings needed medical attention to underserved populations. Some concerns about the inequitable access to new pharmaceuticals in the very countries where they have been tested have been alleviated by exemptions made to the World Trade Organization's Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS) agreement. The TRIPS agreement originally made pharmaceutical product patent protection mandatory for all developing countries beginning in 2005. However, recent amendments have exempted the least developed countries from pharmaceutical patent obligations at least through 2016. Consequently, those developing countries that do not currently provide patent protection for pharmaceutical products can legally import less expensive versions of the same drugs from countries such as India where they are manufactured.

Product Liability

Product liability laws are intended to protect consumers from defective products. Pharmaceutical companies can be sued for faulty design or manufacturing, deceptive promotional practices, violation of regulatory requirements, or failure to warn consumers of known risks. So-called "failure to warn" claims can be made against drug makers even when the product is approved by the FDA. Although the traditional defense offered by manufacturers in such cases is that a "learned intermediary" (the patient's physician) wrote the prescription for the drug in question, the rise of direct-to-consumer advertising by drug companies has undermined this argument. With greater frequency, courts are finding companies that market prescription drugs directly to consumers responsible when these advertisements fail to provide an adequate warning of potential adverse effects.

"Me Too" Versus True Innovation: The Pace of New Drug Development

"Me-too drug" is a term used to describe a pharmaceutical that is usually structurally similar to one or more drugs that already are on the market. The other names for this phenomenon are "derivative medications, "molecular modifications," and "follow-up drugs." In some cases, a me-too drug is a different molecule developed deliberately by a competitor company to take market share from the company with existing drugs on the market. When the market for a class of drugs is especially large, several companies can share the market and make a profit. Other me-too drugs result coincidentally from numerous companies developing products simultaneously without knowing which drugs will be approved for sale.

Critics of the pharmaceutical companies argue that they are not innovative and do not take risks and, further, that medical progress is actually slowed by their excessive concentration on me-too products. Figure 1–2 summarizes a few of the facts behind this and some of the other arguments just discussed. Clearly, smaller numbers of new molecular entities have been approved by the FDA over the past decade, despite the industry's enormous investment in research and development. This disconnect has occurred at a time when combinatorial chemistry was blooming, the human genome was being sequenced, highly automated techniques of screening were being developed, and new techniques of molecular biology and genetics were offering novel insights into the pathophysiology of human disease. Some blame mismanagement of the companies. Some say that industry science is not of high quality, an argument readily refuted. Some believe that the low-hanging fruit has been plucked, that drugs for complex diseases, such as neural degeneration or psychiatric and behavioral disorders, will be harder to develop. The biotechnology industry has had its successes, especially in exploiting relatively obvious opportunities that the new recombinant DNA technologies made available (e.g., insulin, growth hormone, erythropoietin, and more recently, monoclonal antibodies to approachable extracellular targets). Despite their innovations, the biotechnology companies have not, on balance, been more efficient at drug invention or discovery than the traditional major pharmaceutical companies.

Whatever the answers, the trends evident in Figure 1–2 must be reversed (Garnier, 2008). The current path will not sustain today's companies as they face a major wave of patent expirations over the next several years. Acquisition of other companies as a business strategy for survival can be successful for only so long. There are arguments, some almost counter-intuitive, that development of much more targeted, individualized drugs, based on a new generation of molecular diagnostic techniques and improved understanding of disease in individual patients, could improve both medical care and the survival of pharmaceutical companies. Finally, many of the amazing advances in genetics and molecular biology are still very new, particularly when measured in the time frame required for drug development. One can hope that modern molecular medicine will sustain the development of more efficacious and more specific pharmacological treatments for an ever wider spectrum of human diseases.