The Ethics Of Phase 1 Research In Patients

Introduction

Phase 1 trials are the gatekeeper of clinical research as they are often the first time a potential new drug treatment enters humans ( First-in-Man or FIM trials). In the past FIM trials usually enrolled between 20 and 80 healthy volunteers carefully screened for the absence of disease and in part paid for trial participation. However, this is no longer the case. Due to recent concerns over drug safety – for example the Northwick Park experience with TGN1412 causing serious adverse reactions in healthy volunteers - and with the advance of medical science leading to the development of treatments with greater specificity Phase 1 research now increasingly considers patients.

The involvement of patients brings a new dimension to research; patients can conceivability benefit from clinical trials and in considering entering a clinical trial both the patients and the attending doctor must consider the alternatives of treatment. For this latter reason, perhaps, many clinical trials involving patients only consider the terminally ill as candidates for the clinical trials. The perception of benefit also means that patients in Phase 1 clinical trials are not paid. There are also ethical considerations when considering the enrolment of patients. In the conduct of medical care the free will or autonomy of the patient is seen as prime. Failure to respect the wish of patients in deciding their treatment is seen as morally wrong, and in most countries subject to legal sanction. Coupled with the demands of autonomy, and in support of it, are the requirements to fully inform patients about the potential effects of treatment and the alternatives available. Balancing these roles and responsibilities on the medical profession are the requirements that doctors should do no harm and also ensure the just use of resources. These ethical principles are enshrined in the ethical principles of deontology. In this school of ethical thought it is the intentions not the outcomes that determine morally acceptable behaviour. For example if you have acted in a way that respected patient autonomy, and fully informed them of the outcomes of choices, and the patient has a bad outcome, you have still acted in a morally acceptable manner. In contrast scientific research has a different ethical perspective. It is the magnitude of the outcome that matters in Phase 1 research in order to select a dose for a subsequent trial or the determination of a toxic dose (Shamoo, 2008). Another factor in the conduct of research is that the results should be generalisable to a population of patients in order to extrapolate the meaning of results. In this instance the intention is not to select a treatment for an individual patient. The focus on the outcome of the clinical trial, and the intention to gather generalisable information gives the ethics of clinical research – and Phase 1 clinical research in particular – a so called consequentialist approach. Consequentialist ethics are concerned with the outcome of an action, rather than the intention of the action. Perhaps the best known consquentialist approach is utilitarianism – acting in a way that leads to the benefit for the greatest number. Phase 1 research is often said to be utilitarian in nature as it seeks generalisable knowledge.

The deontological and consequentialist and approaches often lead to different definitions of what is morally correct and acceptable. Consider this. You are standing at a bus stop when a man comes up to you and says he is being chased by an axe murderer. He says he is going to hide in the bushes till the person passes and does so. A few minutes later a man carrying an axe runs up and asks if you have seen anybody, what do you do?

As a consequentialist, you would say you saw nobody as the consequences of your action would lead to murder.

As a deontologist in a strict sense you would say he is in the bushes as it is always morally wrong to deceive anybody!

The conflict between the deontological approaches of medical practice and the utilitarian perspective of clinical research in Phase 1 has lead some ethicists to conclude that it is not ethical to enrol patients in Phase 1 studies. The foundation of this claim is that as these are patients their care is paramount, and patients cannot exercise autonomy in their choice as

1. Phase 1 clinical trials causes more harm to patients than potential benefits.

2. Patients cannot provide a valid informed consent, as they are not consenting for themselves but the collection of generalisable information.

3. Patients whom are terminally ill are vulnerable and so are liable to be manipulated.

Taking these factors in turn:

1. Phase 1 clinical trials causes more harm to patients than potential benefits.

This statement leads to another question – more harm that what or more benefit than what? Both the deontological and consequentialist approaches to medical ethics requires careful consideration of the risks and potential benefits of a medical intervention be it an operation or a clinical trial. In making an assessment of the intervention it is usual to have some form of comparator. In Phase 1 clinical trials three methods have been used. The first is to compare it to the alternative of palliative care – particularly in cancer, but also in inflammatory diseases. The second way of comparing risks and benefits is to look at the approved cancer treatments and draw parallels with the currently licensed medications available from the FDA or the European regulators. The third is to draw parallels with the use of off label medications in cancer (Joffe & Miller, 2006).

The use of a comparison to palliative care is not valid. The patients whom enter a Phase 1 clinical trial have already rejected it as an option at the moment they enrol in a clinical trial. In addition they have not rejected it in the future by enrolling in a trial.

A comparison to existing treatments is also misleading. Many Phase 1 patients have exhausted treatment options so marketed treatments are not a valid comparison. In addition drugs on the marketplace are often registered on the basis of different clinical outcomes than those being measured in clinical trials. Finally as 90% of drugs fail in the development process, the use of a benchmark to existing drugs sets a very high barrier.

Instead perhaps it is more valid to consider the risks and benefits of enrolling in a clinical trial against the use of off label medication by doctors and patients. In this case the treatment population is the same – both patients enrolling in clinical trials and those wanting to use off label treatments have rejected palliative care in the first instance. In addition of label use outcomes are often expressed in terms of anecdotal outcomes rather than registration endpoints. The problem is that this data is hard to obtain. However, what is known in the field of oncology is that approximately 4% of patients have meaningful clinical endpoint outcomes in Phase 1 trials (Joffe & Miller, 2006). This contrasts with the reported rate of toxicity death rates of 0.5% (Addoler, Taylor, & Wendler, 2008). These figures argue against the proposal that Phase 1 clinical trials in patients cause harm, as in the population that enrol in clinical trials have risks and benefits comparable to off label use of existing drugs (Joffe & Miller, 2006).

A second argument for the fact that patients suffer harms in Phase 1 clinical trials is that the usual outcomes cannot measure harm, and that the harms are best expressed in terms of Quality of Life (Qol). The harm it is argues is a decline in Qol due to clinic visits and procedures. There is little data available on this aspect, but what does exist suggests that many patients Qol is not changed by enrolment in a Phase 1 clinical trial as the majority of these trials are short in duration (Addoler, Taylor, & Wendler, 2008).

Based on these findings it is difficult to argue convincingly that the risks and benefit balance of Phase 1 clinical trials in patients are better or worse than the alternatives, and therefore to argue that their exclusion from Phase 1 trials can be justified on the trials cause harm.

2. Patients cannot provide a valid informed consent, as they are not consenting for themselves but the collection of generalisable information.

The foundation to this objection is that patients do not appreciate or understand that they are participating in clinical research, and do not receive or perceive the information they are given and therefore informed consent is not valid.

Empirical studies cast doubt on this assertion (Addoler, Taylor, & Wendler, 2008). Interviews with patients enrolling into clinical trials have shown that the majority of subjects do understand the nature of the trial into which they are enrolling, albeit the primary motivation for enrolment was the hope of some benefit.

Of greater concern in the process of informed consent is the issue of voluntariness. In other words that the participation in a clinical trial stems from pressure from a physician whom is also acting as an investigator. There is some evidence that doctors over-estimate the chances of benefit from clinical trials mostly unintentionally (Addoler, Taylor, & Wendler, 2008). However, in interviews with patients, the role of physician advice in enrolling in trials appeared to be low.

On the other hand the fact that patients cannot give informed consent to enter into clinical trials also seems weak.

3. Patients whom are terminally ill are vulnerable and so are liable to be manipulated.