Trial and Error


Amidst a wave of public misunderstanding of scientific method and the rise of alternative therapies, Emer Sugrue examines how scientific research and clinical trials operate

The human brain is an amazing thing, but it is easily fooled. We are conditioned to search for patterns in life but we often see patterns where they don’t exist. Things can look like they are related when they are not. An intervention can look effective while being useless, or worse, causing harm.

The way we can tell whether things really work or whether it’s our minds tricking us is to have a controlled trial. This writer can feel science students rolling their eyes from here, but it’s something that is poorly understood, and not just by those of us who frequent the Newman building. Measurable outcomes are frequently dismissed by those promoting alternative therapies claiming that “science doesn’t have all the answers”. The media bizarrely portrays scientists as unelected authority figures, dictating conflicting decrees on how we must live our lives. The only science news most people see is the Daily Mail’s campaign to report every substance known to man as either a cause or cure for cancer. Scientific research is not a proclamation from on high, and science certainly does not have all the answers; but with patience, it can ask the right questions.

To help me understand more about the role of clinical research and how it works in Ireland, I spoke to Dr Peter Doran, Scientific Director for the UCD Clinical Research Centre and Professor Pat Murray, Professor of Clinical Pharmacology.

Clinical research centres are a fairly new institution in this county, with the very first opening in the Royal College of Surgeons in 2001, and the UCD CRC opening in 2007. Dr Doran explains the role of the CRC in the scientific community.

“The objective was really to create a core infrastructure to allow any patient-orientated research to happen. So whereas before that might have been done in outpatient departments and the hospital, the idea was to create a proper infrastructure.

“Since the CRC opened we’ve had about 20,000 research patient visits. Our investigators have leveraged about €12 million in funding so it’s made a significant impression on the research landscape. Very importantly, what it’s allowed us to do is make sure that new treatments, new interventions and cutting edge programs are there so Irish patients can benefit from them. Ultimately that’s been the major objective, how Irish patients can get access to the best emerging care.”

Most clinical research that involves the testing of a new drug progresses in an orderly series of steps, known as phases. This allows researchers to gain reliable information about the intervention and protects the patients. A new intervention starts off as a hypothesis for a how a particular drug might work. If it seems promising in theory, you try it on animals to see if it kills them. It probably will, and it’s probably back to the drawing board at this point. But should Fievel survive, then it’s time to test it on people.

Phase I trials are the first studies done in small groups of healthy humans to evaluate how a new drug should be given, how often, and what dose is safe, how quickly it is excreted from the body and so on, and to see if it kills them, of course. In Phase II the treatment is given to a larger group of a few hundred people with the relevant ailment, to see if it is effective and to further evaluate its safety. Phase III trials test the new treatment on hundreds or even thousands of people, comparing the new treatment to the existing standard or a placebo.

Bringing a treatment to trial is a lengthy and costly process. It costs around $500 million to bring a new drug to market, and even getting as far as initial trials is a huge endeavour for a research scientist.

“Assuming you have a completed protocol, and that’s a big assumption,” laughs Dr Doran, “the Irish Medicines Board approval is a maximum of 90 days, and that’s assuming that at your 45 day review you’ve to go back with a lot of information so that’s the maximum. Then the ethics committee turnaround is probably 60-90 days as well for most protocols. I mean, that’s assuming that the protocol is appropriate, because you may have a situation where somebody writes a protocol and it’s rejected and then they have to go back and do a large amount of work to bring it up to the standard that is required.” Altogether it takes at least a year, if not several, to bring a hypothesis to trial.

The reason trials take so long to get approved is a rigorous adherence to ethics. A proposal will have to go through several medical boards and ethics boards before patients are even approached. Ethics in science has advanced dramatically in the past few decades. Horrifying stories emerge every so often about patients purposely infected with diseases or left without treatment, but such events would be impossible today. Ethical science today doesn’t just take into account the basic health and rights of the subjects but the design of the experiment in regards to how useful the research is.

Professor Murray explains that “the one thing that everyone agrees on, and the ethics committees make a big point of this, they’ll look at not only the risks to patients and any potential benefits when deciding whether the trial is ethical, they’ll also look at if the trial design is adequate and are you studying the right number of the right kind of patients to actually get an answer. If they think your design is a mess they’ll reject it and say that’s not an adequate trial, you’ll end up wasting several hundred people’s time and spend a lot of money and you won’t actually answer the question.

“The scientific integrity is just as important in many ways as the ethics and the protection of subjects, because if you do any one of them wrong you’re not doing valid clinical research.”

However, various aspects of clinical trials have been criticised. There are frequent claims that pharmaceutical or industry-funded trials are biased. In 2003, a systematic review scrutinized thirty separate studies regarding whether funding affected findings and overall, studies funded by the drug company were four times as likely to give results that were positive and favourable to the company than independent studies. The UCD CRC itself is partly funded through the industry, but Dr. Doran does not believe that this is an issue:

“Industry protocols are all approved by what’s known as the competent authority; in Ireland the competent authority is the Irish Medicines Board so the protocol is very clear in terms of what’s being done. Before we get involved in them at the CRC they have to be approved by the ethics committee so there’s a very clear line in terms of what the protocol is.

“Everything in the Clinical Research Centre is done in accordance with a set of standards called ICH GCP, the International Committee for Harmonisation Good Clinical Practise guidelines. They’re a global set of guidelines which tell us how to do clinical research and everything is done with GCP in mind, regardless of how it’s funded or where it originates from.”

There are also allegations of publication bias, particularly associated with industry-funded research.  Publication bias refers to the practise of positive trials being much more likely to be published than negative ones. In an anonymous survey published by the scientific journal Nature in 2005, 6% of scientists admitted failing to present data that contradicted their previous work. Another paper, published in the New England Journal of Medicine went through all known trials on SSRIs (Selective Serotonin Re-uptake Inhibitors, a class of drugs used widely as antidepressants) registered with the Food and Drug Administration and attempted to explore possible correlations between published results and industry interests. There were 37 studies which were assessed by the FDA as positive and, with a single exception, every one of those positive trials was published in full. There were also 33 studies which had negative or unclear results; 22 of those were not published at all while the remaining 11 were written in a way that showed them as having a positive outcome. Professor Murray assures me however, that safeguards are in place to prevent this sort of misinformation.

“There’s now a need for clinical trials to be registered on a international electronic database so that even if your trial is negative and you don’t like the result, you still have to post the results on to that resource so people will know. They’ll know not only about the one that works, which is what you want them to hear about, but they’ll hear about the other nine that didn’t work and they can make an informed decision about whether that represents how the literature should be.

“Whether the answer is what you would like it to be or whether is isn’t, it doesn’t differ between industry and academics. Industry may want it to turn out a certain way because there are financial implications, academics may want it to turn out a certain way because of academic implications, but in both cases you actually have this very rigorous design so you end up publishing the truth, whatever it turns out to be.”

It is the ideal of truth that is so fascinating. In the arts anyone’s opinion is as good as another and entire careers are made from revisionism, post-revisionism and post-modern revisionism where you just deconstruct your own life choices. The UCD CRC and all other research centres attempt to reduce human error in scientific research and despite the complexity of ethics, money, reports and dead mice it all comes back to one basic question; does it work?