The history of clinical testing and its regulation

A jungle of clauses

The law stipulates that anyone who wishes to market a new drug has first to demonstrate the safety and efficacy of the drug in clinical trials. In the process, the applicant has to negotiate a maze of statutory regulations, international specifications, and ethical recommendations. There is no detail that is not regulated, nothing that does not have to be carefully documented and monitored. And though there is – and has to be – a certain margin of freedom with regard to the design of trials, the formal requirements are nevertheless clearly set out.

For example, a protocol must be drawn up which meticulously describes the design of the planned trial. Importance is also attached to the choice of a suitable control method (should the new drug be tested in comparison to placebo or an existing medication?), appropriate randomisation and blinding procedures, and the choice of clinical endpoints to elucidate the efficacy and safety of the test drug. During the trial, no deviations from the protocol are allowed. In addition, sound statistical analysis of the data is required. All these requirements serve to ensure that unbiased and scientifically reliable information is obtained on the efficacy and safety of the new drug.

The pharmaceutical industry is well disposed towards these regulations and in fact is closely involved in their elaboration. It has a vested interest in the whole procedure. After all, products must be effective, safe, and of high quality in order to be successful. Throughout the history of clinical testing, scientifically sound methodology for clinical trials has developed hand in hand with an unambiguous and legally binding body of regulations.

A historical excursion

Clinical trials came a long way before they took on their modern form – prospective, controlled, blind, randomised, statistically verified – and their roots stretch far back in history. The literature even mentions the Bible as one of the first sources. For example, the Book of Daniel in the Old Testament contains a description of a clinical trial carried out by Nebuchadnezzar II. Children of royal blood, including Daniel, were ordered to follow a strict diet of meat and wine for three years. But Daniel persuaded the king to give him and three others bread and water for ten days instead. It is reported that after this period the four bread eaters and water drinkers looked more radiant and well nourished than the meat eaters and wine drinkers.

In the early days of medicine, the discovery of new treatments that proved more beneficial than established ones was usually serendipitous. A famous example is the unintentional clinical trial carried out by the Renaissance surgeon Ambroise Paré. In the middle of a siege battle in 1537, Paré ran out of boiling hot oil, the usual remedy for treating open wounds at the time. As a substitute, he mixed a tincture of egg yolk, oil of rose, and turpentine. The following day he noted that the wounds treated by the conventional method were greatly swollen and painful, whereas the wounds to which the new tincture had been applied were neither swollen nor inflamed. Paré concluded that the new remedy was preferable to treatment with boiling oil.

Another chance discovery was the observation by seafarers travelling to the East Indies in 1600 that lemon juice is effective at preventing scurvy. However, the finding remained disputed for a long time, probably due to the high cost of provisioning ships’ crews with lemon juice. Only about 150 years later was the effect proven by James Lind. His experiment of 1747 is regarded as one of the first controlled clinical trials, i.e. a trial with a parallel control group that is given an alternative treatment.

Where no alternative treatment exists, or where an alternative treatment can be stopped without risk for a short time, modern trials generally employ a placebo as a means of dispelling all doubts as to the efficacy of a remedy. Trials with placebos emerged in the 19th century. For example, in 1865 Sir William Gull and Henry Sutton gave mint water – an ineffective remedy – to patients with rheumatic fever. They found that patients with the disease are likely to recover spontaneously and concluded that the most suitable treatment for rheumatic fever had yet to be discovered.

Randomised trials were not carried out until the 20th century. One of the first was Amberson’s 1931 study on the treatment of tuberculosis with sanocrysin. Twenty-four patients were divided into two groups of equal size. A coin was tossed to determine which group would receive sanocrysin and which placebo. It was also a blind trial, since none of the patients knew which group they had been assigned to.

The concept of multicentre studies, which are carried out at various sites but according to a common protocol, is mentioned for the first time in a 1944 publication on the evaluation of patulin in the treatment of the common cold (Patulin Clinical Trials Committee).

First regulations and ethical issues

In the 1930s the authorities in Great Britain realised the necessity of conducting clinical trials to evaluate new drugs and therefore set up the Therapeutic Trials Committee. “The Therapeutic Trials Committee will be prepared to consider applications by commercial firms for the examination of new products, submitted with the available experimental evidence of their value, and appropriate clinical trials will be arranged in suitable cases.”

For centuries, the history of clinical testing was shaped by methodological and medical considerations, whereas the concerns of the individuals involved in the studies was of subsidiary importance. The Nuremberg trials drew attention to the unscrupulous, barbaric experiments inflicted on humans during the Nazi period in Germany and kindled a worldwide ethical discussion about the performance of clinical trials. Finally, in 1947, the Nuremberg Codex laid down ten basic tenets for the protection of subjects and patients. Among other things, these provide for a voluntary declaration of consent by trial participants; the right of trial participants to comprehensive information on the nature, purpose, and potential risks of the experiment; and the right of trial participants to withdraw from the trial at any time. In addition, performance of a trial must be warranted by the expected results, and the risk involved must not be disproportionate to the social and humanitarian significance of the problem being addressed.

The Nuremberg Codex was followed in 1964 by the World Medical Association’s Declaration of Helsinki, which was further developed in 2000 and serves today as the ethical basis for clinical trials throughout the world. [See also the article “In the focus of discussion. The role of ethics in clinical trials.”]

Responding to a tragedy

What really triggered the deluge of statutory regulations on drug development was the thalidomide tragedy in Europe in 1962, when this sedative and hypnotic caused deformities in around 8000 children. This disaster showed that the new generation of synthetic drugs that was revolutionising medicine at the time could be both a blessing and a curse.

In the USA, proof of efficacy of pharmaceuticals was required for the first time with the passage of the Keefauer-Harris Drug Amendment. Until then, the authorities had required only proof of safety for a drug to be marketed. The earlier law (the Federal Food, Drug, and Cosmetic Act of 1938) was passed in response to a tragic error made in formulating a cough syrup for children.

Thalidomide had not yet been approved for use in the United States. The new drug law was intended to prevent such cases from occurring in the first place. The law also regulated the performance and documentation of clinical trials, introducing important elements of today’s requirements for good clinical practice (GCP).

Developments in Europe

Europe too witnessed a rapid expansion of laws, regulations, and recommendations in the 1960s and 1970s. Starting in 1986, individual countries – Great Britain, France, Germany, and the Scandinavian countries – issued GCP recommendations. These were eventually standardised in 1989 in the GCP recommendations of the European Community. These took on force of law with the passage of an EU directive in 1991. The compulsory criteria for the approval of a medicinal product for sale on the European market are its quality, safety, and efficacy, of which the latter two in particular must be demonstrated in humans in clinical trials.

Good clinical practice

Good clinical practice describes the requirements for the protection of participants in clinical trials. It also stipulates detailed requirements for investigators and sponsors with regard to the planning, execution, documentation, analysis, and reporting of studies in order to ensure the quality and logical consistency of the data, findings, and conclusions.

In the interest of protecting trial participants, every trial requires foregoing and ongoing approval by an independent ethics review board, or ethics committee. The ethics review board must determine whether the trial is ethically justified, whether the study site is qualified to conduct it, and whether the written material used for obtaining the subjects’ consent is appropriate. GCP sets out guidelines on the composition and procedures of ethical review boards and determines what documents have to be submitted to the board.

Investigators and sponsors must meet detailed GCP requirements concerning the content of the protocol and the qualifications of the investigators. Guidelines exist that govern the supply, checking, and tracking of test drugs, the collection and documentation of data, the reporting of adverse events, and the monitoring of trials.

In addition, GCP requires that the sponsor maintain or commission an independent quality-assurance unit whose task is to ensure that appropriate procedures are used for the planning, execution, documentation, analysis, and reporting of trials and to confirm the quality of trials by means of independent audits.

The stringent and detailed regulatory requirements have changed the face of clinical trials over the past ten to 15 years. Health authorities now set great store by statistical analysis as a means of evaluating the efficacy of a new drug as reliably as possible. The statistical requirements are very demanding, and biostatisticians have therefore come to play a key role in the planning and evaluation of clinical drug trials.

Towards harmonisation

Globalisation of the pharmaceutical industry has made harmonisation and standardisation of the disparate guidelines and approval requirements an urgent concern – and not just for pharmaceutical manufacturers. In 1990 the International Conference on Harmonization (ICH), a joint project initiated by industrial associations and governmental agencies, was called into being with the aim of eliminating differences in the technical requirements for drug development in the three major pharmaceutical markets, namely the EU, Japan, and the USA. Greater efficiency in the develop-ment and regulation of drugs is also in the interest of the general public and of patients, who can thereby gain earlier access to possibly life-saving drugs. It also helps to reduce costs – since it obviates the need for clinical trials on humans to be conducted twice – and reduces the number of animal experiments to a minimum.

In 1997, at the end of the first phase of the ICH’s activities, far-reaching joint recommendations were issued and implemented in the participating countries. These were divided into four categories, namely quality (questions of chemical and pharmaceutical quality assurance), safety (especially with regard to preclinical studies), efficacy (in relation to clinical trials performed on humans), and multidisciplinary matters (e.g. medical terminology, standards for electronic data transmission). Many other countries, notably Australia, Canada, and the EFTA states, have subsequently adopted these recommendations.

In November 2000 another milestone was reached with the agreement on a Common Technical Document (CTD). This is a uniform dossier to be used in Europe, Japan, and the USA for submitting data to the respective regulatory authorities. Use of the CTD is to be introduced in the member countries within the next few years.

Approval procedures The ICH recommendations are implemented in Europe by the Committee for Proprietary Medicinal Products (CPMP) and in the United States by the Food and Drug Administration (FDA).

The FDA is the central regulatory authority in the United States. In the EU there are currently two approval procedures. In the centralised marketing authorisation procedure, approval is granted not by a national regulatory authority but by the EU Commission in Brussels. This procedure is coordinated by the London-based European Agency for the Evaluation of Medicinal Products (EMEA). The centralised procedure is mandatory for biotechnology products and optional for innovative drugs. It leads to Europe-wide marketing authorisation. In the decentralised procedure, the pharmaceutical company first submits an application for authorisation in only one EU member state. It can then apply for authorisation in other EU countries, which must recognise the first national authorisation within 90 days unless there are serious reasons for opposing it (mutual recognition procedure).