Making a new medicine

Discovering and developing a new medicine requires combining the skills of biology, chemistry, clinical medicine and, of course, pharmacology. There are three key decision points along the way:

Medicinal therapeutics rests on the two great supporting pillars of pharmacology:

Once a target has been selected the process of finding the right chemical to act as an antagonist, agonist, inhibitor, activator or modulator of the protein function depends on a process of screening. For decades, the rational discovery of new medicines depended on modifications of the structures of natural chemical mediators. This may still be the route to find the drug, but more often now large libraries of compounds are screened against the target in robotic high throughput screens. However, it is worth remembering that the exact molecular basis of drug action may remain unknown, and this book contains frequent examples of old drugs whose mechanism of action remains mysterious. The evolution of molecular medicine (including recombinant DNA technology) in the past 30 years has led to identification of many thousands of potential drug targets, but the function of many of these genes remains unknown. The hope was that the identification of targets identified by genetics coupled with high throughout screening would lead to a great increase in the productivity of drug discovery. This has turned out not to be the case since it is clearly important to understand both the function of the target and the nature of the interaction of the chemical drug with the target in order to make a selective and safe medicine. The skills of quantitative pharmacology are key.

The chances of discovering a truly novel medicine, i.e. one that does something valuable for patients that had previously not been possible (or that does safely what could previously have been achieved only with substantial risk), are increased when the discovery and development programme is founded on precise knowledge of the biological processes it is desired to change. The commercial rewards of a successful product are potentially enormous and provide a great incentive for developers to invest and risk huge sums of money. Most projects in drug discovery and development fail. Indeed the chances of making it through from target selection to having a medicine on the market are under 1 in 100.

The huge increase in understanding of molecular signalling – both between cells and within cells – has opened many new opportunities to develop medicines that can target discrete steps in the body’s elaborate pathways of chemical reactions.³ The challenge, of course, is to do so in a way that produces benefit without harm. The more fundamental the pathway targeted the more likely there is to be a big effect, whether beneficial, harmful or both. No benefit comes without some risk.

The molecular, industrialised and automated approach to drug discovery that followed sequencing of the genome and application of high throughput chemical approaches led to two consequences:

Pharmacogenetics has gained momentum from recent advances in molecular genetics and genome sequencing, due to:

Expectations of pharmacogenetics and its progeny, pharmacoproteomics (understanding of and drug effects on protein variants) remain high, but the applicability will not be universal. They include the following:

Consequences of these expectations include: smaller clinical trial programmes with well-defined patient groups (based on phenotypic and genotypic characterisation), better understanding of the pharmacokinetics and dynamics according to genetic variation, and improved monitoring of adverse events after marketing.

New drug development proceeds thus:

The (critical) phase of progress from the laboratory to humans is often termed translational science or experimental medicine. It was defined as ‘the application of biomedical research (pre-clinical and clinical), conducted to support drug development, which aids in the identification of the appropriate patient for treatment (patient selection), the correct dose and schedule to be tested in the clinic (dosing regimen) and the best disease in which to test a potential agent’.⁴

It will be obvious from the account that follows that drug development is an extremely arduous, highly technical and enormously expensive operation. Successful developments (1% of compounds that proceed to full test eventually become licensed medicines) must carry the cost of the failures (99%).⁵ It is also obvious that such programmes are likely to be carried to completion only when the organisations and the individuals within them are motivated overall by the challenge to succeed and to serve society, as well as to make money. A previous edition of this chapter included a quote from a paper I wrote from my time in academia and I leave it here:

Techniques of discovery

(See Fig. 3.1)

Fig. 3.1 Drug discovery sources in context. Different types of chemical compounds (top left) are tested against bioassays that are relevant to therapeutic targets, which are derived from several possible sources of information (right). The initial lead compounds discovered by the screening process are optimised by analogue synthesis and tested for appropriate pharmacokinetic properties. The candidate compounds then enter the development process involving regulatory toxicology studies and clinical trials.

The newer technologies, the impact of which has yet to be fully felt, include the following:

Drug quality

It is easy for an investigator or prescriber, interested in pharmacology, toxicology and therapeutics, to forget the fundamental importance of chemical and pharmaceutical aspects. An impure, unstable drug or formulation is useless. Pure drugs that remain pure drugs after 5 years of storage in hot, damp climates are vital to therapeutics. The record of manufacturers in providing this is impressive. Much of the early work in drug discovery is spent trying to identify the right molecule with the appropriate physical and chemical properties to make it reliable as a medicine.

Studies in animals⁹

Generally, the following are undertaken:

Ethics and legislation

Controversy surrounding the use of animals in scientific research is not new. The renowned Islamic physician Avicenna (980–1037) was aware of the issues for he held that ‘the experimentation must be done with the human body, for testing a drug on a lion or a horse might not prove anything about its effect on man’.¹¹ Leonardo da Vinci (1452–1519) predicted that one day experimentation on animals would be judged a crime, but Descartes¹² asserted that ‘Animals do not speak, therefore they do not think, therefore they do not feel’. Later, Jeremy Bentham (1748–1832), the founding father of utilitarian philosophy, asked of animals: ‘The question is not, Can they reason? nor Can they talk? but Can they suffer?’.

In our present world, billions of animals are raised to provide food and many to be used for scientific experiments. The arguments that evolve from this activity centre on the extent to which non-human animals can be respected as sentient beings of moral worth, albeit with differences between species. In recent years, a boisterous animal rights movement, asserting the moral status of animals, has challenged their use as experimental subjects.¹³ Mainstream medical and scientific opinion around the world accepts that animal research continues to be justified, subject to important protections. This position is based on the insight that research involving animals has contributed hugely to advances in biological knowledge that have in turn allowed modern therapeutics to improve human morbidity and mortality. However, each experiment must be justified and its results must be expected to deliver insight into safety or efficacy. Animal models do contribute to the understanding of human physiology and disease because we share so many biological characteristics and a medicine when introduced into the organism is exposed to a vast array of conditions that we do not fully understand and are unable to reproduce outside the living body. The study of a drug in the whole organism remains an essential step in the process of discovery and development of medicines.

Safety testing in animals is at present the only reliable way to evaluate risks before undertaking clinical trials of potentially useful medicines in humans. The investigation of reproductive effects and potential carcinogenicity would not be undertaken in humans for both ethical and practical reasons. Animal testing eliminates many unsafe test materials before clinical testing on humans, and minimises the risk of possible adverse effects when people are exposed to potential new medicines. In other words, experiments in animal models provide a critical safety check on candidate drugs; potentially hazardous or ineffective drugs can be eliminated and for those drugs that do progress to clinical trials, target organs identified in animal studies can be monitored.

Animal research has contributed to virtually every area of medical research, and almost all best known drug and surgical treatments of the past and present owe their origins in some way to evidence from animals. The antibacterial effectiveness of penicillin was proved in tests on mice. Insulin came about because of research on rabbits and dogs in the 1920s. Poliomyelitis epidemics, which until the 1950s killed and paralysed millions of children, were consigned to history by vaccines resulting from studies on a range of laboratory animals, including monkeys. Major heart surgery, such as coronary artery bypass grafts and heart transplants, was developed through research on dogs and pigs. The BCG vaccine for tuberculosis was developed through research on rats and mice. Meningitis due to Haemophilus influenzae type b, formerly common especially in children, is now almost unknown in the UK because of a vaccine developed through work on mice and rabbits. Almost all of the highly effective drug treatments we currently use were developed using animals: β-adrenoceptor blockers, angiotensin-converting enzyme inhibitors, cytotoxics, analgesics, psychotropics, and so on.

Given this evidence, there is broad public support for the position that experiments on animals are a regrettable necessity that should be limited to what is deemed essential while alternatives are developed. In the UK, for example, this reservation is expressed in progressively more stringent legislation. The Animals (Scientific Procedures) Act 1986 makes it an offence to carry out any scientific procedure on animals except under licence, the requirements of which include that:

Despite the continued necessity of animal studies in drug discovery and development there is now a very clear aim to move medicines into the clinic as soon as safely possible. Detailed clinical experimentation is often a better way to sort out questions related to effects on human biology and pharmacokinetics.

Safety prediction

Knowledge of the mode of action of a potential new drug obviously greatly enhances prediction from animal studies of what will happen in humans. Whenever practicable, such knowledge should be obtained; sometimes this is quite easy, but sometimes it is impossible. Many drugs have been introduced safely without such knowledge, the later acquisition of which has not always made an important difference to their use (e.g. antimicrobials). Pharmacological studies are integrated with those of the toxicologist to build up a picture of the undesired as well as the desired drug effects.

In pharmacological testing, the investigators know what they are looking for and choose the experiments to gain their objectives.

In toxicological testing, the investigators have a less clear idea of what they are looking for; they are screening for risk, unexpected as well as predicted, and certain major routines must be done. Toxicity testing is therefore liable to become a routine to meet regulatory requirements to a greater extent than the pharmacological studies. The predictive value of special toxicology (above) is particularly controversial. All drugs are poisons if enough is given, and the task of the toxicologist is to find out whether, where and how a compound acts as a poison to animals, and to give an opinion on the significance of the data in relation to risks likely to be run by human beings. This will remain a nearly impossible task until molecular explanations of all effects can be provided.

Toxicologists are in an unenviable position. When a useful drug is safely introduced, they are considered to have done no more than their duty. When an accident occurs, they are invited to explain how this failure of prediction came about. When they predict that a chemical is unsafe in a major way for humans, this prediction is never tested. The easiest decision to make in drug discovery is to stop a project, and it is the only decision that can never be shown to be wrong. However, it is also a decision that may deny the world a new and effective medicine.

Orphan drugs and diseases

A free-market economy is liable to leave untreated both rare diseases, e.g. some cancers (in all countries), and some common diseases, e.g. parasitic infections (in poor countries).

In order to stimulate drug discovery and development in rare diseases, legislation in some countries has provided specific incentives to industry to develop medicines for ‘orphan diseases’. Interestingly, many rare diseases are monogenic and the cause is clear. This, unlike most common diseases, means that the drug target is absolutely clear. In view of the greater certainty around cause, the smaller clinical trials and the more receptive regulatory environment, several pharmaceutical companies are beginning to work more assiduously in this area. It is also the case that some rare diseases give significant insights into common diseases and how to approach them.