Clinical trials1,2 have four phases, which proceed sequentially. Not all drugs make it to the end of the process (if a drug ‘fails’ a phase 1 trial, there is little point in running it through phase 2!).
Preclinical trials: These happen in a lab, and whilst they almost invariably involve testing on animals, do not involve human testing. The purpose of these trials is to establish whether the drug is safe and effective.
Phase 1 trials:
The purpose of phase 1 trials is to discover possible side-effects, establish efficacy in humans and calculate what dosage is needed.
These trials are the first to involve humans, and so involve a degree of risk. To minimise this, researchers test on small groups (normally giving progressively greater doses to successive groups) using small doses. Sometimes healthy volunteers are used in these trials, and other times people who are very ill are recruited (e.g. the trial of a new cancer drug). Whether healthy or unwell volunteers are used depends on factors like whether the treatment involves a level of risk unacceptable to the healthy (e.g. a new form of chemotherapy would normally be trialled in an advanced cancer patient, because the level of risk and severity of expected side-effects are not acceptable to those who have nothing to gain. Further relevant factors include whether or not the treatment is feasible in a healthy volunteer.
Phase 2 trials:
Phase 2 trials test the treatment on larger groups of people with the relevant illness. The purpose is to find out more about the efficacy of the drug (including which specific thing it treats best, e.g. a new form of chemotherapy might work better on some forms of cancer than others) and more about side-effects.
These trials are tested against a placebo (e.g. an identical pill that doesn’t contain any medication). That is, those recruited into the trial will be split into two groups, one of which receives the medicine, the other of which receives a placebo. These are normally conducted under ‘double blind’ conditions: whilst everyone involved knows that there is the possibility of receiving a placebo, neither the patient nor the doctor giving the pill knows who is receiving what (obviously who gets what is nevertheless recorded). The purpose of this is to find out whether the effect the treatment has is due to the drug, or whether it’s just due to the expectation of getting better (placebo effect).
Phase 3 trials:
The purpose of phase 3 trials is to determine whether or not the new drug/treatment is better than the best currently available treatment.
Because this is a difficult thing to test, and may involve the measurement of small differences between treatments, a large sample size is needed (many people must be tested on). For example, if we were testing a new drug for cancer (let’s call it Newafex) we would get a big group of people, give half of them Newafex, and the other half would receive the best current treatement (whatever drug they would have had if Newafex didn’t exist; let’s call it Oldafex). Then we look at remission rates (the percentage of people who get a bit better) and compare.
A new drug is unlikely to get a license if it is not more effective than other drugs, although there are exceptions (if, for example, the side-effects are less severe, or variety is needed).
Phase 4 trials:
These occur after a drug has been granted a licence (and is allowed to be prescribed by doctors). Their purpose is to determine the long term risks and benefits of the treatment, and monitor its performance when used more widely. These go on for many years (and severe long term side-effects are rare), and it would consequently be impractical to perform this sort of long term testing before licensing
Getting on to a clinical trial:
Patients who may benefit from a treatment that is being trialled can be referred to that trial by their doctors; the company trialling the drug will have criteria that that patient must meet in order to qualify for the trail and it is ultimately at their discretion, not the doctor’s, whether someone is accepted.
Healthy volunteers participating in phase 1 trials tend to be recruited through adverts, and are normally paid well for their participation.
Some countries do already allow the provision of unapproved drugs that are still going through trials on compassionate grounds through programs such as Expanded Access to Investigational Drugs for Treatment Use (United States), Special Access Program (Canada), and Temporary Authorisation for Use (France).3
1 ‘Clinical trials and medical research – Phases of trials’, NHS Choices, 5 September 2011, http://www.nhs.uk/Conditions/Clinical-trials/Pages/Phasesoftrials.aspx
2 ‘Phase 1, 2, 3 and 4 trials’, Cancer Research UK, 6 September 2012, http://cancerhelp.cancerresearchuk.org/trials/types-of-trials/phase-1-2-3-and-4-trials
3 ‘Compassionate Use of Unapproved Investigational Product’, Pfizer, http://www.pfizer.com/research/research_clinical_trials/compassionate_use_policy.jsp
We live in a free society, and accept that people have the right to do as they please; including exposing themselves to risk, as long as in so doing they do not harm others (Mill’s harm principle).1
To deny people the right to choose to take drugs that are still undergoing testing that, whilst they are risky, may save their lives, is a violation of this principle: in choosing to expose themselves to that risk, no-one else is harmed and indeed in the long run others may be saved as a result of the tests. Further, with the assistance of medical professionals and other support, the decision to take this risk can easily be well-informed and consensual.
1 Wilson, Fred, "John Stuart Mill", in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring 2012 Edition), http://plato.stanford.edu/archives/spr2012/entries/mill/
It is not the case that this is a policy with no harms other than to the person with a terminal illness (see opposition arguments).
Second, it seems unreasonable to suggest people are making a free and informed choice in this instance: no-one has sufficient information for taking the drug to represent anything but a gamble; this is why there is a need for tests.
At a point when all ordinary medical avenues have been expended, and the outcome appears bleak, new treatments still undergoing trials can be seen as the last hope.
People are often aware of the existence of currently experimental drugs, they are likely to research into possible cures, and indeed there may have been attempts by their doctor to get the patient onto the trial.
However, not everyone who could benefit from treatment is accepted onto a clinical trial: some trials, at some stages, restrict their recruitment to, for example, patients with no complicating factors or other illnesses.
It is unethical and cruel to make people live out their last days knowing that there was something that could have helped, but to which access was restricted through no fault of their own: thus, you should allow anyone with a terminal illness access to such treatments.
It is not cruel if it can be shown that this restriction is in the patient’s own interest. The status quo prevents patients from living out their last days on a stream of experimental drugs. We prevent drug companies from using them as risk-free testing (under your policy drug companies would presumably be able to shrug off any responsibility for adverse consequences by saying that it was the patient’s choice to try an experimental drug), and allow them instead to receive the appropriate support for someone at the end of their life, and come to terms with that.
Further, it is important to remember that drugs at this stage are not necessarily miracle cures! If someone is refused access to a trial this is normally to reduce the risk of adverse consequences: it is wrong to give someone an experimental drug that could negatively impact the quality of their final days.
Under the status quo, someone with a terminal illness is offered two choices: death, or to join a trial (where such trials exist). However, when they join a trial they face the possibility that they will be given a placebo, not the drug.
Whilst this is probably in the best interest of future patients (a good clinical trial will determine the efficacy of the new treatment), it rides roughshod over the rights of the current patients (not to be sacrificed for future generations) and the duty of physicians to act in the best interests of their present patients.
There are two consequences here: the first is that it is morally dubious to use the present patients as mere means to an end, rather than acting in their own best interests, especially where, if randomized to the placebo arm the outcome of death is a certainty.
The second consequence is a practical one: compliance with the trial is lessened at the point at which patients can take alternative measures to increase their chance of survival. This was best documented during the early stages of the AIDS crisis in the 1980s, where there was evidence of ‘cheating’ during the trials1. People lied or bribed their way into clinical studies; and shared drugs to dilute the ‘risk’ of being on placebo.
This has the obvious impact of casting doubt on the scientific results of the trials: you can no longer be sure who has taken what, and what other conditions they may have.
1 Schüklenk, Udo, and Lowry, Christopher, ‘Terminal illness and access to Phase 1 experimental agents, surgeries and devices: reviewing the ethical arguments’, British Medical Bulletin, Vol.89, 2009, pp.7-22, http://bmb.oxfordjournals.org/content/89/1/7.full.pdf?keytype=ref&ijkey=Ip64J5FUXThKvP0
First, note that the reason for the existence of the placebo arm is to determine if the drug is more effective than placebo, so in some cases the drug will not be, and nothing will have been lost!
Second, for this point to stand, it has to be shown why the present generation should be prioritised above all future ones: the consequences of giving the present patients a slightly increased chance of survival is to negatively impact patients in the future in a myriad ways (see opposition arguments).
Third, there are a number of reasons to doubt that this is, in fact in the present patient’s best interest: it is not the case that terminally ill people have ‘nothing to lose’ and can therefore be used as human guinea pigs (providing there is an, as yet unspecified, probability of survival). The large-scale provision of un-trialled drugs may well result in side-effects denigrating the quality of the patient’s remaining years.
Finally, the practical consequence considered can be sidestepped through a) better supervision of trials and b) improved doctor-patient relationships (a particular problem during the AIDS crisis). Further, note that the case of AIDS is something of an anomalous one: AIDS patients were more numerous and politicised than any other group before or since, thus enabling this sort of trial-breaking behaviour.
We need medical trials. It is important to have large groups of recruits, which can often be very difficult: a problem with the speed at which new treatments for rare diseases is the rate of recruitment (and therefore the length of time taken to complete the trial)1.
If you pass this motion, trials will face large problems with recruitment, an area where there are already sometimes shortages2: if people can get access to the drug without a) the possibility of being placed in the placebo arm or b) inconveniences of being part of a trial, there is a reduced chance of them choosing to enter the trial.
Consequently, the sample size in trials will be decreased. This will have a couple of outcomes:
First trials will take a longer time to be completed as a result of fewer volunteers and this is bad for patients currently taking the new drug as well as for future patients. This is because it will take longer to determine the safety of the drug meaning if it is dangerous those taking the drug will be taking it for longer before the danger is fully appreciated and if safe then the drug will have taken longer to get to the market than it could have. The longer the trials take to complete, the more people are forced to decide whether to take the drug in the absence of reliable information. This means that, at such a stressful time, people are effectively forced to gamble the quality of their remaining years with the hope of gaining a few more (new drugs are unlikely to be ‘miracle cures’. Rather, they are likely to extend life by driving the disease into remission).
It is important to remember that, at this stage, it has yet to be determined whether new drugs are more effective than old ones, and second, that the sorts of drugs used to treat terminal illnesses tend to come with substantial side effects. As a consequence, if many people are using a new treatment before trialling has been completed, they may be using something that is not effective and has side-effects that significantly impact the quality of the last years of their life.
Finally, the longer trials are delayed, the greater the chance that future trials will be biased by media hype ad speculation. It is both easy and profitable for media outlets to exaggerate early successes of a drug with claims and headlines such as “wonder drug”.
This is problematic because of the tendency towards confirmation bias on the part of researchers: the greater their expectation of a positive result, the more likely they are to alter data to receive that result. Note that this is not as a result of deliberate fraud or deception, but rather, the result of any number of small decisions that, cumulatively, create a large result.
1 Jenkins, John, ‘Considerations for Clinical Trial Designs’, U.S. Food and Drug Administration, https://rarediseases.info.nih.gov/files/Jenkins.pdf
2 ‘Volunteer for research at UNClinicalStudies.org’, University of Michigan Comprehensive Cancer Center, http://www.cancer.med.umich.edu/living/clinical-trials.html
First, this may well be overridden by the individual rights of present patients (see proposition arguments).
Second, the greater time taken to recruit is one that may be offset by greater numbers: whilst the trial will be of a lower quality (no control group, etc.) there will nevertheless be a greater number of people willing to take the drug (people who wouldn’t have wanted to be part of a trial, but are willing to try the new treatment). Consequently, it may well be possible to compensate for the other problems with the trial. Further, alternative trialling models can be employed, for example using patients who choose not to take the drug as the control group. Whilst you lose the benefit here of having a double-blind trial (as under the status quo), you gain in terms of the benefits to current patients.
If these drugs are made available, you risk giving many people false hope in the last days of their lives. People, particularly when in desperate situations, tend to overestimate a treatment’s efficacy.
Given that these treatments are still undergoing the trial process, it is possible that they are ineffective, or have side-effects that outweigh any benefits.
Thus, to allow such drugs and treatments to be handed out during the testing process, there is a great risk of giving people false hope. This is especially the case given the compromised role of the physician in this scenario: ordinarily, if a patient wants an experimental drug, they can have a discussion with their physician that stresses the ‘in trial’ nature of the drug, and thus the uncertainty of it working. Subsequent experiences (the inconveniences of trials; filling in forms and receiving expenses) reinforce the idea that these drugs were experimental, and that the bulk of the benefit from the trial accrues for future patients.
Consequently, in that scenario it is easier for the physician to help the patient to come to terms with the end of life; to deal with this and to realise that any trial drugs give only a slim chance of improvement. In the scenario envisaged by this proposition, experimental drugs can be acquired as easily as licensed ones, and therefore there is no longer that clear distinction for the patient between ‘doing all you can’ in the ordinary sense, (trying every treatment that is known to be effective) and trying ‘one more (experimental) drug’.
Therefore, the patient is less likely to be able to come to terms with their own condition, and therefore less likely to be able to deal with the emotional trauma inflicted not only upon them, but on close family and loved ones.
Doctors are trained in the presentation of news to their patients. This includes the delivery of bad news, and the dispelling of media-myths. Patients with terminal illnesses are often well-informed about their disease, and (in particular those with chronic conditions) often gain a good understanding of the possibilities of future treatments.
The risk that they may all get carried away on a wave of false hope is, consequently, minimal. Patients in this circumstance are more than capable of reaching, in conjunction with their physician, an informed decision regarding experimental drugs, and make a choice accordingly. The moderate risk of someone making an error in no way outweighs the chance of giving someone some more time with their family.
Countries that already allow access to treatments that have not completed trials do not just allow the doctor to simply proscribe the drug as with any other. Rather the doctor will need to apply for access to the drug.1 In addition the drugs company will also have to give its approval.2 As a result it is unlikely that the patient will consider this the same way as they do normal drugs.
1 ‘Special Access Programme – Drugs’, Health Canada, 15 August 2005, http://www.hc-sc.gc.ca/dhp-mps/acces/drugs-drogues/sapfs_pasfd_2002-eng.php
2 ‘Compassionate Use of Unapproved Investigational Product’, Pfizer, http://www.pfizer.com/research/research_clinical_trials/compassionate_use_policy.jsp
Testing new drugs is a very expensive process, in 2000 the average cost was estimated at around 86 million for the large scale phase III tests1 however this is contested and it could be much higher it represents 40% of pharmaceutical companies R&D expenditures, which since a recent estimated the development cost of a drug can be up to $5.8billion (due to including failures) the cost of trials would in some cases then be $2billion,2 which is currently funded by pharmaceutical companies. They fund these tests because it is either impossible, very difficult or very risky to access large markets before testing has been completed (e.g. in the USA companies are only allowed to sell new drugs “off-study”, i.e. during trials, at cost3)
If you allow all terminally ill patients access to experimental drugs, you reduce the incentive for companies to continue testing their products: they will have access to a large market prior to the completion of testing, and will therefore have no incentive to complete trials, which are expensive and risk finding the product ineffective.
1 DiMasi, Joseph A. et al., ‘The price of innovation: new estimates of drug development costs’, Journal of Health Economics, Vol.22, 2003, pp.151-185, p.162 http://moglen.law.columbia.edu/twiki/pub/LawNetSoc/BahradSokhansanjFirstPaper/22JHealthEcon151_drug_development_costs_2003.pdf
2 Roy, Avik S. A., ‘Stifling New Cures: The True Cost of Lengthy Clinical Drug Trials’, Project FDA Report, No. 5, April 2012, http://www.manhattan-institute.org/html/fda_05.htm
3 Schüklenk, Udo, and Lowry, Christopher, ‘Terminal illness and access to Phase 1 experimental agents, surgeries and devices: reviewing the ethical arguments’, British Medical Bulletin, Vol.89, 2009, pp.7-22, http://bmb.oxfordjournals.org/content/89/1/7.full.pdf?keytype=ref&ijkey=Ip64J5FUXThKvP0
There is no reason why, under this model, you could not retain suitable precautions to ensure that this doesn’t happen.
For example, you could continue the obligation for companies to sell the drug at cost (they would, naturally, continue to be incentivised by the huge profits they expect to make when the drug is licensed). Further, you could impose financial sanctions on companies that refused to take appropriate action towards completing testing.
By allowing anyone who is critically ill to use experimental drugs you enhance the already dubious role of drug company reps: especially in the USA, (where doctors do not operate under the NHS guidance found in the UK), there is already a problem of patterns of prescription being altered by the techniques of drug reps, rather than by evidence1. Where drugs are for sale before they have completed testing, there is even less evidence available, and therefore less ability for physicians to contest the claims of either reps or their own patients (who may have heard of the drug during their own research). Hence you magnify the problem of potentially ineffective of even harmful prescription.
1 Harris, Gwyn, ‘Pharmaceutical representatives do influence physician behaviour’, Family Practice, Vol.26 2009, pp.169-70, http://fampra.oxfordjournals.org/content/26/3/169.full
Drugs that are still undergoing clinical trials do not have a complete void of information about them. Presumably this policy covers drugs that have completed at least some testing in humans (say, phase one of the trials), and therefore at least some information would be available on which doctors and patients could base their decisions.
Further, it is implausible to suggest that doctors are entirely under the sway of advertisers: whilst drug reps under the status quo have some influence in getting a doctor to use one drug rather than another, this is in instances where there is little to choose between those products, and (importantly!) both are licensed, safe and effective. They would clearly not be so reckless as to blindly follow a drugs rep and prescribe an untested product to their patient.
‘Phase 1, 2, 3 and 4 trials’, Cancer Research UK, 6 September 2012, http://cancerhelp.cancerresearchuk.org/trials/types-of-trials/phase-1-2-3-and-4-trials
DiMasi, Joseph A. et al., ‘The price of innovation: new estimates of drug development costs’, Journal of Health Economics, Vol.22, 2003, pp.151-185, http://moglen.law.columbia.edu/twiki/pub/LawNetSoc/BahradSokhansanjFirstPaper/22JHealthEcon151_drug_development_costs_2003.pdf
Harris, Gwyn, ‘Pharmaceutical representatives do influence physician behaviour’, Family Practice, Vol.26 2009, pp.169-70, http://fampra.oxfordjournals.org/content/26/3/169.full
‘Special Access Programme – Drugs’, Health Canada, 15 August 2005, http://www.hc-sc.gc.ca/dhp-mps/acces/drugs-drogues/sapfs_pasfd_2002-eng.php
Jenkins, John, ‘Considerations for Clinical Trial Designs’, U.S. Food and Drug Administration, https://rarediseases.info.nih.gov/files/Jenkins.pdf
‘Clinical trials and medical research – Phases of trials’, NHS Choices, 5 September 2011, http://www.nhs.uk/Conditions/Clinical-trials/Pages/Phasesoftrials.aspx
‘Compassionate Use of Unapproved Investigational Product’, Pfizer, http://www.pfizer.com/research/research_clinical_trials/compassionate_use_policy.jsp
Roy, Avik S. A., ‘Stifling New Cures: The True Cost of Lengthy Clinical Drug Trials’, Project FDA Report, No. 5, April 2012, http://www.manhattan-institute.org/html/fda_05.htm
Schüklenk, Udo, and Lowry, Christopher, ‘Terminal illness and access to Phase 1 experimental agents, surgeries and devices: reviewing the ethical arguments’, British Medical Bulletin, Vol.89, 2009, pp.7-22, http://bmb.oxfordjournals.org/content/89/1/7.full.pdf?keytype=ref&ijkey=Ip64J5FUXThKvP0
‘Volunteer for research at UNClinicalStudies.org’, University of Michigan Comprehensive Cancer Center, http://www.cancer.med.umich.edu/living/clinical-trials.html
Wilson, Fred, "John Stuart Mill", in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring 2012 Edition), http://plato.stanford.edu/archives/spr2012/entries/mill/