Clinical trials and tribulations

Mike Clark explores the science and the ethics of the drug tests that left previously fit volunteers seriously ill

The fact that six previously healthy volunteers are now lying seriously ill in a London hospital after trials of a new type of antibody-based drug raises many questions. It is just one of several high-profile cases that demonstrate the ethical pressures scientists are under. While the media have deliberated the safety of drug trials and the use of human volunteers, protests have raged in Oxford over animal testing.

At the other end of the spectrum, patients clamour for new antibody-based treatments - drugs such as Herceptin, the antibody that targets breast cancer.

The six volunteers were taking part in a first-phase clinical trial of a monoclonal antibody-based therapeutic agent called TGN 1412, manufactured by TeGenero AG, a German biopharmaceutical company. TGN 1412 is an antibody that is targeted against a molecule on human T-cells, which play an important role in the immune system. It was selected many years ago based on observations in laboratory research that, unlike many other anti-T-cell antibodies, TGN 1412 activated some of the cells, inducing them to secrete molecules that could regulate inflammatory responses. It was suggested that this property gave it the potential to treat autoimmune diseases such as rheumatoid arthritis and that it might also be effective in enhancing responses against some cancers, such as leukaemia.

Antibodies and T-cells are essential parts of our natural immune system, and they generally act to keep us healthy. They circulate in our bodies, ready to respond rapidly to infection. But, very occasionally, the rapid activation of this response is harmful: for example, when a patient has a severe allergic reaction or suffers septic shock during a bacterial infection. One possible explanation for the severe reactions reported for the volunteers in this trial is that TGN 1412 triggered an overreaction of their immune systems.

Does this mean that we should be cautious of all other antibody-based therapies? It would be wrong to think that problems encountered with one antibody-based drug make it highly likely that similar problems might occur with all others. This is because of the intrinsic properties of antibodies as a class of therapeutic agent. Each antibody has a different binding site (the antibody variable regions) to enable it to bind to and recognise different targets. Antibodies also come in different subtypes bearing what are called Fc regions. These antibody Fc regions can bind to so-called Fc receptors on cells of the immune system and activate a response to the target. Some Fc regions can also bind to and activate another part of the immune system called complement.

Every antibody potentially behaves in a different way, depending on the target it recognises and on whether it is also able to bind to other parts of the immune system. Even two antibodies directed at the same target may behave differently and provoke different side-effects.

It is difficult to reliably predict from in vitro experiments how antibodies will behave in vivo . The situation is exacerbated by the fact that different animals show variations in their antibodies, their Fc receptors and other parts of their immune system. When injected into a laboratory animal, a human antibody may still bind to the target if it is present in the same form, but it may not bind to the animal's Fc receptors in the same way as it would a human's, hence the outcome may be different.

It is interesting to look at this factor in the context of opposition to the use of animals in research. Anti-animal research campaigners argue that advances in cell culture techniques, and the use of in vitro experiments, combined with more sophisticated in silico approaches to the understanding of biological processes, make many animal studies redundant. But the outcome of the clinical trial of TGN 1412 demonstrates that the actions of a new drug in vivo may not be so easily predicted from studies on isolated human cell lines carried out in culture. Those opposed to animal experimentation might counter that the antibodies may behave differently in different animals. That is true, but if enough animal species are studied, there is a greater chance of observing phenomena that give a better prediction and understanding of human biology.

Regardless of the extensiveness and nature of pre-clinical testing, ultimately the drugs have to be trialled on humans. And so we come to the first-phase clinical testing in human volunteers. These volunteers are often healthy, but sometimes patients who can't benefit from alternative conventional licensed therapy take part. Volunteers are required to give informed consent for their participation. The conduct of clinical trials is highly regulated. In the UK, the Medicines and Healthcare products Regulatory Agency is responsible for overseeing clinical trials and for monitoring the continued safety profile of approved products. Protocols for a trial are considered against the background of all the available in vitro and animal testing, as well as by reference to related clinical products.

Progression through the stages of the trials (phases one, two and three) is monitored carefully. If unexpected and serious adverse reactions are reported, the trials are halted. Consideration is given as to whether or not to abandon development of the drug or to continue with a modified protocol.

In Europe and the US, there are more than 20 antibodies approved for therapy or in the late stages of clinical trials. Very many more are entering early stage clinical trials, and hundreds are in pre-clinical development. Nearly all the monoclonal antibodies approved for therapy or in the late stages of clinical trials are intended to work in one of two ways. Many of them are intended to kill cells, usually cancer cells, most often by recruiting immune-killing functions using Fc receptors or other parts of the immune system. Occasionally, diseased cells are killed by monoclonal antibodies that have chemical agents or radioactive isotopes attached. The remaining antibodies are mainly intended to work as blocking agents. They bind to their target antigen and stop it from functioning.

Antibodies of this type include those that bind to cell surface receptors and others that bind to secreted molecules. Considerable life-saving benefit has been seen with several of the approved antibodies.

News stories have highlighted the pressure being applied by some groups of patients for wider approval of some antibodies for use in earlier stages of cancer or autoimmunity. There has been a lot of attention paid to the early approval for Herceptin, which has shown positive results. Patients with diseases such as breast cancer believe that antibody drugs offer better prospects than conventional drugs. It would be a great shame if the publicity surrounding the highly unusual results in the clinical trials of TGN 1412 had a major adverse impact on the development of life-saving antibody-based therapeutics.

Mike Clark is a lecturer in therapeutic and molecular immunology at Cambridge University.