Wednesday, May 10, 2006

The New England Journal of Medicine Weighs in on the TGN 1412 Trial Debacle, But With a Twist

The New England Journal of Medicine published a commentary on May 4, 2006, on the ill-fated trial of TGN 1412. (Wood AJJ, Derbyshire J. Injury to research volunteers - the clinical-research nightmare. N Engl J Med 2006; 354: 1869-1871.) The first author of the study, Dr Alastair JJ Wood, will be joining Symphony Capital as a managing director in August, according to a New York Times article published May 2, 2006. Per the Times, "Dr Wood has been working as a consultant to Symphony since the fund's inception in 2002.... Symphony, which has $315 million under management, invests in particular drugs. It buys the rights to a drug from a biotechnology company and then works with the company to conduct clinical trials. The biotech company has the right to buy back the drug later at a specified price."

We have previously posted, most recently here, about the disastrous trial, implemented by Parexel International , of a new monoclonal antibody designated TGN 1412, manufactured by TeGenero AG. All six healthy volunteers who got the antibody soon became critically ill.

Unlike some other cases discussed on Health Care Renewal, this one had already received attention in prominent UK medical and scientific journals, including the British Medical Journal, the Lancet, and Nature. It's good to see that the most prominent US medical journal has weighed in.

The article by Wood and Derbyshire focused on the following points, in the article's own words (with emphasis added):
At some point in the development of every drug, the drug must be given for the first time to humans in a phase 1 trial. Until now, such trials have had a remarkably good safety record....
It is standard practice to begin with very small doses, often orders of magnitude below those determined to be nontoxic in animals and those expected to produce any effect in humans. Doses are then increased slowly, as the experience at lower doses is continually evaluated.
System or human failures — such as errors in dosage, manufacturing, or administration — are usually prevented by rigorous procedures for drug preparation and administration.
Toxic effects such as acute liver injury, leukopenia, cardiac arrhythmia, or rash may be related to the new drug molecule itself but unrelated to its intended mechanism of action. Considerable efforts are made to identify these types of toxicity in vitro and through studies in animals. However, our incomplete understanding of the mechanisms underlying such toxicity and the limitations of animal models inevitably mean that some potentially serious toxic effects go undetected in preclinical screening....
The second type of toxic effect results from the action of the drug on its intended biologic target. Such effects are always unknown when a target is 'drugged' for the first time — and there must always be a first time.
When a compound addressing a new biologic target is tested for the first time in humans, much greater caution must be exercised. Such caution should include avoidance of treating multiple volunteers simultaneously or without a reasonably long interval between them.
In some cases, a phase 1 trial does not, in fact, represent the first attempt to manipulate a particular biologic target — though the researchers may be unaware of previous efforts. Clearly, we should not be exposing people to such manipulation if it has been shown, in studies in either humans or animals, to carry serious risks outweighing any potential benefits.
Unfortunately, the companies that generate early safety data consider them proprietary — a concern that must somehow be reconciled with patients' safety. Volunteers rightly expect that we put their safety before competitive advantage, and researchers have an ethical obligation to prevent the exposure of additional volunteers to previously identified risks.
How can we improve the knowledge base for designing trials of new drugs directed at novel targets and make it available to developers and regulators when they are considering the safety of such trials? One approach would be to ensure that all data from preclinical drug research are held in a secure database, indexed by biologic target, and accessible only by major regulatory authorities, which are used to handling confidential data.
There are fundamental questions about which, if any, details of a clinical trial involving volunteers should ever be confidential or whether safety and ethics principles can be ensured only by an open, transparent process in which such trials and protocols are registered in a public database.

The commentary presented a reasonble summary, but it seemed to tread lightly in certain areas.

In particular, it side-stepped the question of whether that there was particular information available before the TGN 1412 trial that would have suggested that affecting the CD28 receptor targeted by TGN 1412 might be dangerous. There is at least a question about this. An article in the (UK) Times quoted Angus Dalgleish, "the previous studies which caused similar side effects were in patients already suffering from cancer, but [the researchers] should have known they would get a meltdown because the drug was hitting exactly the same target." An article in Nature (Hopkin M. Can super-antibody drugs be tamed?) stated, "with hindsight, it might be no surprise that the compound, dubbed a 'superagonist' antibody by its creators, could run amok in the immune system." Furthermore, research on drugs that target the CD3 receptors in mice, which have a similar function to the CD28 receptors, showed, "uncontrolled cytokine release was a problem - albeit not a large one because the mice were already immunodepleted...."

Furthermore, the New England Journal commentary did not address questions about whether the research subjects in this trial really gave informed consent. This question was raised in, among other places, a commentary in the British Medical Journal (Goodyear M. Learning from the TGN 1412 trial: this experience should foster an open culture in medical research. Brit Med J 2006; 332: 677-678.) See also our posts here and here.

Finally, it did not challenge the notion that safety data about particular drugs should be kept confidential during drug development. In constrast, Goodyear (see above), wrote, "this tragedy creates one more imperative for an open culture in medical research, a culture that many fear is increasingly losing its way."

Given that Dr Wood, the first author of this article, already consults for and will soon become a top leader in an organization that buys rights for particular drugs from biotechnology companies and then helps to run clinical trials on them, his choice of emphasis is understandable. Perhaps someone from a different background would have placed emphasis elsewhere.

But it is time for the final twist. The New England Journal only identified Dr Wood as "a professor of medicine and pharmacology at Vanderbilt University School of Medicine." It did not mention Dr Wood's current or future positions with Symphony Capital. Yet Dr Wood's new position at Symphony Capital had been made public by the New York Times two days before the article was published.

It seems that prominent medical journals still have difficulties disclosing financial and other interests of authors that might meaningfully affect their opinions, and might be relevant to readers who want to understand where these authors are coming from.

No comments: