Trial and Error

In John le Carré’s 2001 novel “The Constant Gardener,” a villainous drug company uses quicklime and unmarked graves to obliterate evidence that a clinical drug trial had gone wrong. In Britain five years later, a real-life trial for a drug designed to interact with the immune system became the stuff of horror movies: healthy volunteers’ brains on fire, fingers falling off, toxin-stained eyes that felt as if they’d pop out of the subject’s head.


So disfigured was one participant that the case acquired a memorable nickname: the “Elephant Man” trial. As one volunteer put it, “Everybody thought we were toxic waste.”

Now, in France, another clinical trial has gone awry, with one subject dead and five hospitalized in a clinical trial in January by the company Biotrial. Small wonder, then, that many consumers are suspicious of these trials, in which human beings serve as guinea pigs for companies with a profit motive. And there’s little question, on examination, that these trials were handled badly, and that stricter oversight is needed.

But people who mistrust clinical trials and their Big Pharma sponsors need to keep a few things in mind.

First, shocking abuses like the ones in Britain and France are exceptions, not the rule. As events unfolded in France, The Independent newspaper carried a feature on “The Troubled History of Clinical Drug Trials,” listing just three, from 1999 to 2006. During that time, thousands of trials were carried out — many of them successfully — that never made the news.

Second, the need for new drugs is growing exponentially. With improved diagnosis, longer life expectancy, and sedentary lifestyles that have led to higher rates of obesity, many diseases are on the rise. For example, the number of Americans with Alzheimer’s disease could reach 7.1 million by 2025, a 40 percent increase in just a decade. At the moment, Alzheimer’s has no effective treatment. But as our understanding grows for this and other killers, we can design more specific drugs with better results and fewer side effects.

Of course, the precise trajectory for getting a drug all the way from initial concept to market varies widely, and the pharmaceutical industry laments what it characterizes as high costs and endless delays. One recent study estimated it took more than a decade and $2.6 billion to bring a prescription medicine to market, on average. Consumer advocacy groups and other researchers paint a much more nuanced picture.

What’s clear, though is that it takes a good deal of time, effort, and expense to approve new medicines, and patients with terminal illnesses often can’t wait that long. In that sense, taking part in clinical trials can literally be a matter of life and death.

Third: The absence of clinical trials would do little to guarantee safety or eliminate the bald pursuit of profit. Purported miracle cures, after all, spread like wildfire through social media, often appealing to the desperation of sick people and their families by promising privileged knowledge. (“What no oncologist will tell you!”) To quote Britain’s Motor Neurone Disease Association, “Allowing treatments to be made available which have not been through the established clinical trials process encourages unscrupulous individuals to market ineffective treatments for monetary gain.”

And finally, clinical trials are — or should be — subject to rigorous regulation. In the United States and Europe, new chemical entities (NCE’s) are optimized in the lab for years before their Phase I safety trials. Healthy volunteers are used, to avoid giving an untested drug to those in fragile health. Only after a drug is deemed safe for humans does it move on to Phase II, in which it is tested for efficacy against a placebo. Patients are under the care of two physicians, and they can discontinue the trial whenever they wish. Phase III is a repeat of Phase II in a much larger group, and successful Phase III results enable a company to apply to the Food and Drug Administration for permission to market the drug. For as long as the drug is on the market, it is monitored (Phase IV), though the process can be sporadic.

In the U.S., moreover, the National Institutes of Health is required to maintain a database of FDA-approved clinical trials, online at clinicaltrials.gov. The site, which went live in 2000, offers information on trials, including patient eligibility and study design, so that patients interested in signing up for one can discuss the pros and cons with their doctor. (In 2006, the World Health Organization called for all trials worldwide to be registered on the site. Although clinical trial disclosures are below ideal standards, the situation is improving. Readers can sign a petition for greater disclosure here.)

None of this is to say that the system is foolproof.

The Elephant Man trial in Britain and the recent events in France raise crucial questions about the regulation of clinical trials. (Visual by iStock.com)

The Elephant Man trial in Britain and the recent events in France raise crucial questions about the regulation of clinical trials. Lessons learned about the trial of the British drug, code-named TGN1412, weren’t applied in France; for example, a key recommendation following the British debacle was that Phase I volunteers be tested one at a time until a new dose is proved safe. In France, Biotrial simultaneously administered multiple high doses of the painkiller — developed by a Portuguese pharmaceutical company, Bial — to multiple volunteers. (This did not violate any regulation: The Bial drug fulfilled none of the criteria for “at risk” drugs developed by the European Medicines Agency after the British incident.)

Parexel, the British company that conducted the clinical trial of TGN1412, didn’t medically examine participants beforehand, and the briefing for volunteers was rushed, with no time for questions. In France, the authorities weren’t told of the problems with Bial’s drug until three days after the trial was stopped. Both testing companies continued administering the compounds to volunteers after they knew others had suffered severe side effects or had been hospitalized. Biotrial’s representative told me that “at the time the sixth dosing occurred, we had no further information that [the] subject was not going to return to the unit,” but the company did know that the “subject” – a human being – had been hospitalized.

In short, the private companies made up their own rules, and there is little to suggest they’ve been sanctioned for it. An official report on the British incident effectively absolved the company; as of March, the French authorities were still investigating the drug company and the operator of the trial there. Parexel told me that “patient safety is our top priority” but that it couldn’t discuss “the specifics of any one patient-related matter.”


What’s needed is a set of universal guidelines, stringently enforced. An international expert panel to continually review best practices and devise protocols in line with the most up-to-date information would be a step in the right direction, and official bodies should more closely supervise the trials they approve. The French health minister, Marisol Touraine, has called for any hospitalization during a clinical trial to “lead to an immediate suspension of the trial until the safety of volunteers is guaranteed.”

Because a drug’s structure is commercially sensitive and a closely guarded secret, the French authorities were left in the dark about the painkiller under study, having to resort to online databases for clues to its identity. That’s unacceptable: In the future, all available information on a test compound should be released on demand in an emergency.

Greater transparency must also apply to regulators. The Duff report, the work of a panel of experts assigned by the British government to investigate the TGN1412 incident, baffliingly concluded that there was “no evidence of overdose …  or [errors in] administration” ¾ despite abundant evidence that the drug was administered too quickly, in doses that were far too high. (Parexel’s own literature conceded that it had an “unprecedented biological effect,” though the company downplayed the effect as nothing more than headaches and nausea.)

Since what can heal can also harm, as the old saying has it, more work is needed on the feasibility of microdosing. Starting doses for Phase I trials are traditionally based on animal studies. But now, radiolabeling technology allows scientists to determine how a drug behaves in the human body even in tiny doses — typically, one-hundredth the expected effective dose — and then extrapolate for effect, rather than extrapolating between species.

Would I take part in a clinical trial? First, I’d check to make sure it was registered with clinicaltrials.gov, and I’d consult an independent medical professional. If alarm bells sounded for any reason, I’d discontinue. But if I were seriously ill, I’d definitely give it a shot.

Mel Mann, a long-term leukemia survivor and a well-known advocate for cancer patients, got access to a game-changing cancer drug years before it was on the market. As he put it on Twitter: “Gleevec saved my life, as all the other treatments had failed. I was blessed to be part of the trial and would recommend clinical trials to other patients who are seeking better outcomes.”

Naomi Elster is a writer, scientist, and PhD candidate in cancer medicine at the Royal College of Surgeons, Ireland, supported by the Irish Cancer Society.