We need more drugs testing – on animals and humans

We mustn't let the disastrous trials at Northwick Park hospital blind us to the need for further medical innovation.

Stuart Derbyshire


We can all sympathise with Nav Modi, one of six previously healthy young men who suffered serious side effects after undergoing trials for a new drug at Northwick Park hospital in north London. Modi, who has been referred to as ‘the Elephant Man’, such was the effect of the drug on his head and body, is currently telling his story in the Sun. But we should also remember that human drugs trials are essential, and they must continue.

The disastrous trial of the antibody drug TGN1412 has provoked an inevitable debate about human drugs trials. Various conclusions can be drawn from the tragic events, but the following strike me as the most pertinent:

  • There is a need for innovation in human trials and for an increase in animal testing to minimise the potential risks;
  • Phase 1 human trials, especially those involving highly novel classes of drug such as TGN1412, have definite associated risks;
  • The process of consent for human subjects involved in phase 1 trials should be placed back into the hands of medical professionals;
  • Payment for human subjects involved in phase 1 trials should reflect the risk involved with effective compensation mechanisms for when things go wrong.

TGN1412 is a drug that manipulates the immune system by activating regulator and/or killer T-cells in a manner that is clearly inadequately understood. In animal studies the drug appeared to enhance the immune system in those animals that had reduced function, such as occurs during leukaemia and other cancers, but appeared to inhibit the immune system in those animals that had abnormally increased function, such as occurs during rheumatoid arthritis.

Furthermore, these very happy outcomes occurred in the absence of serious side effects. In rats, mice, rabbits and rhesus monkeys the drug provided no cause for alarm, and this was true of animals that were healthy as well as animals that were in some way immuno-compromised. The only reported side effect, evident in the rhesus monkeys, was an increase in the size of the lymph nodes, which was accepted as a normal response to the increased production of T-cells.

The extensively positive results of these preclinical tests in animals, as well as tests in laboratory models, were key components in the success of TeGenero, the German pharmaceutical company that made TGN1412, in obtaining a license for a phase 1 clinical trial. Phase 1 trials typically involve a small sample of volunteers who agree to receive usually small doses of a drug that has never before been given to humans. In the UK, approval for phase 1 trials is provided by the Medicines and Healthcare Regulatory Authority (MHRA). Predictably, approval of the TGN1412 trial by the MHRA has come under fire, with one unnamed immunologist suggesting that nobody should be surprised by what has happened (1).

This criticism, delivered with 20-20 hindsight, is unfair: everyone is tremendously surprised and shocked by what has happened. Given the preclinical results, there was no reason to suspect such a catastrophic adverse reaction in a human population. Indeed, the Paul Ehrlich Institute in Langen, Germany, had also independently approved TGN1412 for a human trial (2).

The MHRA issues around 300 licences every year for phase 1 trials. The aim of these first human trials is to monitor for negative effects and to understand how the new drug is absorbed and metabolised. Although about 20 to 30 per cent of new drugs fail to proceed past phase 1 because of adverse effects, dramatic effects such as that observed with TGN1412 are largely unprecedented. The only comparable recent incidents are the death in 1999 of Jesse Gelsinger who suffered multiple organ failure after an experimental genetic therapy, and the death in 2001 of Ellen Roche who suffered acute failure of her lungs and kidneys after an experimental asthma therapy.

The current situation, and these other examples, illustrate that early human drugs trials can be extremely dangerous to the point of being lethal. This is likely to be especially true of drugs that are highly innovative and reach beyond current classes of medication. As has been reported on spiked and elsewhere (3), most ‘new’ drugs that come to market are variations of drugs already available (see After Northwick Park: we need more research, not less, by Dr Michael Fitzpatrick). Slight tweaks in the formula, changes in the dosage and even alterations in the colour of the tablets can result in a drug entering or re-entering phase 1 trials as a new drug. Phase 1 trials of this type are very unlikely to result in unforeseen adverse outcomes, because they have, in effect, already been through numerous ‘safety trials’ and have a known low-risk profile.

In contrast, TGN1412 is a totally novel agent that directly stimulates the immune system; currently available immune system drugs do not stimulate the system but merely damp it down. The expectation was that the stimulation of the immune system by TGN1412 would include regulatory processes to prevent runaway destruction, as had been observed in the animal models tested (4). Tragically, this expectation was not borne out in humans and the volunteers suffered massive damage as their immune system essentially attacked their internal organs and body tissue.

Because the animal models and laboratory data provided for no expectation of this highly adverse outcome, many animal rights activists and others have called for a rethink of animal testing. In fact, we should demand increased animal testing when moving novel drug agents towards human trials. All newly proposed drugs must pass through animal testing before progressing to humans and the very low level of serious adverse effects in human trials indicates that this procedure is highly effective. Where a drug stretches scientific understanding, however, then the possibility for an adverse event is more likely and there is a need for additional confidence. The reticence to use animals and the focus of researchers on the three R’s – reducing the number of test animals, refining the procedures to induce the minimum of suffering, and replacing higher primates with lower species of tissue testing – is a burden on animal research generally (5) but is highly inappropriate in the pursuit of innovative therapies. In the case of TGN1412, the immune system was assumed to be largely preserved across species but it is now clear that subtle differences between humans and the animals tested must exist.

Increasing the number of animals used, and using higher species, particularly higher primates, increases the possibility of detecting an adverse effect that may have implications for human trials. Had preclinical trials involved 50 rhesus monkeys instead of 20, and also involved chimpanzees as well as rats, mice and rabbits, then the potential for a catastrophic event in humans might have been detected. At the same time, newly developed computer simulations, live human tissue models and synthetic animal tissue chips can also be employed to predict a negative outcome.

Ultimately, however, if we want new drugs then they will have to be tested on humans. Reasonably simple precautions and new innovations can minimise the potential for adverse reaction when delivering drugs to human beings for the first time. An obvious precaution, which appears to have not been followed in the TGN1412 trial, is the slow delivery of the drug to volunteers. Although the drug was delivered at a dose 500 times less than that given to the rhesus monkeys, and was provided via a series of injections, all the injections were delivered inside 20 minutes and to all six volunteers at the same time. Slower dosing is likely to have minimised the serious reactions that occurred and serially testing the volunteers would have almost certainly prevented five of the six volunteers from receiving the drug. A procedure currently being developed is ‘microdosing’, which involves delivery of extremely tiny doses of drug followed by accelerator mass spectrometry that can detect the minute traces of the drug and monitor the metabolic process (6).

Even with these precautions and advances in procedure, however, eventually a therapeutic dose of drug must be delivered – and the possibility for a serious adverse effect will remain. Volunteers need to be encouraged to step forward in the full knowledge of the risks they face. I have argued elsewhere that consent for many medical procedures involves highly complicated forms that are distant from both patients and medical practitioners (see Unethical committees, by Stuart Derbyshire). Patients do not understand the forms and medical practitioners only relate to them through a committee designed to minimise liability rather than enhance patient understanding. Potential volunteers for drug trials need to be presented with realistic and candid assessments of risk from the people most likely to know. This is generally the medical researchers and practitioners involved in the development and delivery of the drug. A candid but digestible assessment of drug novelty, and an indication of how far it stretches current understanding, could help volunteers make an informed choice as to the risk they are taking.

Inevitably, though, there are limits to the process of informed consent, which cannot be expected to fully bridge the gap of expertise between volunteers and the scientists who have dedicated years of study to drug development. Volunteering will always involve a certain element of recklessness with a leap of faith. Providing realistic remuneration for this leap is not unreasonable. It is argued, particularly in the UK, that large payments encourage volunteers to ignore risk and are therefore unethical. It is precisely the point that payment encourages risk taking, and it is generally an accepted ethical practice when applied to occupations that have a higher than usual risk – these occupations provide a higher than usual pay. It is also unrealistic to believe that a volunteer will knowingly opt for a more risky trial in the absence of any financial reward. A flat payment across all drug trials provides the false impression of equal risk and undermines informed consent.

Finally, volunteers for drug trials should be guaranteed effective compensation mechanisms in the event of serious injury or death. These mechanisms will enhance the consent by highlighting the real possibility of a negative reaction while also protecting companies from litigation, and the volunteers and their families from financial ruin, when adverse reactions occur.

A grown-up society understands that risk cannot be completely eliminated if innovation is to continue in any field. Extensive animal testing and innovative techniques of drug delivery can minimise the risk to volunteers but, eventually, the full dose of any new drug must be tested on human beings. This process will always include a leap into the unknown; if it were otherwise then drugs could simply be taken directly to patients with perfect effectiveness and zero side effects. Given the reality of phase 1 trials with highly novel drugs, future adverse events are bound to happen occasionally. Volunteers should be aware of this with payment for their bravery and compensation for any tragedy.

Stuart Derbyshire is a senior lecturer in psychology at the University of Birmingham, England.

(1) Trial and terror, Robin McKie and Jo Revill, Observer, 19 March 2006

(2) Violent reaction to monoclonal antibody therapy remains a mystery. Science 2006; 311: 1688-9

(3) Scientists are ‘puzzled’ by this kind of drug: so should last week’s trial have taken place?, Robert Matthews, Telegraph, 19 March 2006

(4) Hunig T, Dennehy K. CD28 superagonists: Mode of action and therapeutic potential. Immunology Letters 2005; 100: 21-28.

(5) Derbyshire S. Time to abandon the three Rs. The Scientist 2006; 20(2): 23.

(6) Wilding AR, Bell JA. Improved early clinical development through human microdosing studies. Drug Discovery Today 2005; 10: 890-4

To enquire about republishing spiked’s content, a right to reply or to request a correction, please contact the managing editor, Viv Regan.



Leave a comment

You must be logged in to comment. Log in or Register now.