Pharmacogenetics opens up new horizons in clinical trials

The tailor-made medicine of tomorrow

Obstacles to proof of efficacy

The effectiveness of a drug has to be determined — and demonstrated to drug regulatory authorities — by means of clinical trials. The concept appears straight-forward at first: a test substance is administered to various groups of voluntarily participating patients and observations are made to determine whether the substance works. In practice, however, it is rare for clinical trials to provide an unequivocal answer to the question of whether a substance ‘works’ or ‘doesn’t work’. One of the most famous experiments of Louis Pasteur was an exception in this regard. In the experiment concerned, all the patients treated with a new anti-rabies vaccine survived. As the death rate of untreated rabies is 100 percent, the evidence for the effectiveness of the new vaccine was compelling.

Only a tiny minority of investigational substances turn out to be miracle drugs. In most cases the main effect of a drug is to relieve symptoms such as pain, anxiety, or angina or, for example, to reduce high blood pressure or cholesterol levels. Such effects are complex and difficult to demonstrate.

For a variety of reasons, the results of clinical trials must always be interpreted with a great deal of caution and with reference to the complex circumstances under which they were obtained.

In the first place, diseases are often unpredictable. The course of many chronic ailments such as arthritis, multiple sclerosis, depression, and asthma is difficult to predict. The patient’s condition may improve or deteriorate for no apparent reason. Moreover, a great many diseases are due to a combination of factors and, as in the case of heart attacks and strokes, are influenced by external factors such as lifestyle and environmental conditions.

In the past, attempts were made to overcome these problems by recruiting clinical trial participants on the basis of certain external criteria (age, sex, height, weight, etc.) or characteristics related to lifestyle (e.g. smoker or nonsmoker). In addition, great use was, and still is, made of statistics. Provided that the number of participants and the duration of the study have been correctly chosen, statistically significant results can be obtained despite inter-individual differences and variations over time.

Genetic individuality

However, it is only thanks to advances in molecular genetics made in the past few decades that we know that whether and how a drug works in an individual patient, and thus the results of a clinical trial, can be crucially influenced by differences at the molecular level. Genetic individuality accounts for the fact that even patients who scarcely differ in terms of their external characteristics and symptoms may react quite differently to the same drug.

The existence of ‘chemical individuality’ was first recognised by the British physician Archibald Garrod, who coined this term shortly after 1900. He observed, for example, that susceptibility to infectious diseases varies between members of a family, and he even recognised that this ‘chemical individuality’ is an inheritable characteristic of human beings. Today, a hundred years later, science has identified the basis of this genetic uniqueness. Thanks to findings obtained via genome research, the nature of the chemical individuality of a human being can now be precisely defined.

It has been found that human genes can exist in many forms, in other words that segments of genetic material that have the same function can be made up of different components (polymorphism). A variation that occurs with a frequency of more than one percent in a given group of people (population) is known as a single nucleotide polymorphism, or SNP (pronounced ‘snip’). Unlike random point mutations, SNPs are stable and are passed on from one generation to the next.

Advances in pharmacogenetics change the face of clinical trials

The discipline of pharmacogenetics investigates the relationship between variations in DNA and the way in which human beings react to medicines. Pharmacogenetics is a subdiscipline of pharmacogenomics, which investigates the genome and gene products insofar as these are related to the discovery and development of drugs.

Pharmacogenetics looks into the question of how SNPs influence the proteins that are responsible for breaking down drugs in body cells. These detoxification systems contribute to the tolerability of drugs. The human body contains more than a hundred different proteins that break down drugs into tiny molecular fragments. Each detoxification protein is encoded by a gene, and each such gene may contain a SNP. For example, cytochrome P450, a protein found in the liver, plays a role in the metabolic breakdown of about a quarter of all drugs. Some fifty genetic variants of the gene that codes for this protein have now been found, including almost twenty variants of cytochrome P450-2 D6 alone. The metabolism of over fifty different drugs, from cough remedies to anti-hypertensives, can thus vary between individuals.

Individual variations in base sequence (SNPs) can explain why certain drugs are effective in some patients but ineffective or even harmful in others. For example, individual asthmatics react quite differently to the drug albuterol. The critical gene in this case carries the information for the (beta-2-adrenergic) receptor at which the drug acts. If the base present at position 16 of this gene is adenine, albuterol is able to exert its effect. If, on the other hand, guanine is present at this position, the receptor fails to perform its function.

Pharmacogenetics is thus able to provide information on the effectiveness of drugs. In addition, knowledge of polymorphisms can provide important information on the correct dosage of drugs. Thus, SNPs are responsible for a number of variants of the protein N-acetyl-transferase-2, which plays a role in the breakdown of certain drugs. In particular, a rapidly metabolising and a slowly metabolising variant can be distinguished, the corresponding individuals being known as rapid and slow acetylators, respectively. It has been found that the former category of patient can be given high doses of the drugs concerned, whereas smaller doses are indicated in the latter patients. This is because the drug persists for longer and in higher concentrations in the cells of patients with the slowly metabolising variant of the protein.

The fact that genetically determined diversity causes drugs to be ineffective in certain people can be exploited in the testing and development of new drugs. Performance of SNP analyses at an early stage of clinical trials can permit selection of patients who are biochemically equipped to respond to the investigational substance. Suitable trial participants can then be identified by means of genetic tests that provide information on metabolic status, drug target polymorphism, and the presence of specific diseases. Even small groups of subjects selected in this way can yield statistically significant results. This could permit a reduction in the number of large-scale clinical trials involving several thousand participants and thus a reduction in the number of patients required for clinical trials. In addition, determination of the SNP profile of potential subjects could make it possible to exclude individuals whose participation would only expose them to unnecessary side effects.

Ethical questions

The value of pharmacogenomics and pharmacogenetics for the discovery and development of new drugs goes far beyond the application of these disciplines to clinical trials.

As attractive as these new technologies may be to pharmaceutical manufacturers, it must not be forgotten that the acquisition of even relatively simple genetic data can have major implications for the individual trial participant. Though genetic data do not differ from other medical data in this regard, the question of the social effects of genetics is far from resolved.

Roche has responded to the debate on the ethics of genetics by drawing up the Roche Charter on Genetics, which sets out the company’s principles regarding the use of genetics and genetic information. The Charter emphasises the right of every individual to self-determination, privacy, and confidentiality regarding the procurement and use of genetic information. Roche undertakes to prevent any misuse of genetic information acquired in the course of its research activities.

In accordance with the Charter, Roche will exploit technical developments in pharmacogenetics to the maximum possible extent in order to provide effective and safe pharmaceutical products while at the same time ensuring optimal protection of the privacy of trial participants.

The best example of this is the Roche Sample Repository, in which anonymised DNA samples obtained in clinical trials sponsored by Roche are stored together with the corresponding medical data. It goes without saying that only samples obtained from patients who give appropriate consent are stored. All samples entered into the Roche Sample Repository are assigned a new, independent code within the database system. The key that links clinical identifiers to the identifiers in the Repository database is then deleted. This anonymisation process is effective by itself, however in order to totally exclude the possibility that the presence of ‘exceptional’ genetic information might permit identification of an individual subject, the data in the Roche Sample Repository are analysed only in aggregate. With the aid of samples from the Roche Sample Repository, the effectiveness of drugs or the severity of diseases can be determined. DNA is investigated to see whether it contains specific genetic variations that can influence response to an investigational drug or that may have contributed to a disease.