Cancer Trials Stopped Early For The Wrong Reason
6 CommentsBy Ed Silverman // April 9th, 2008 // 6:40 am
The usefulness of some cancer meds is being exaggerated because drugmakers are stopping trials the moment they find a benefit, according to a study that will be published in Annals of Oncology, The Telegraph reports.
Researchers have discovered a sharp increase in the number of studies that are terminated early and argue that drugmakers are halting trials whenever benefits are discovered - and use the premature results as the basis for license applications. As a result, the researchers claim, drugs that are hailed as breakthroughs could be of less benefit or even cause more harm, the paper writes.
A team from the Mario Negri Institute for Pharmacological Research, in Milan, Italy, found that of 14 trials stopped early between 2005 and 2007, 11 were used to support a license application. Trials on well-known drugs regarded as breakthroughs, are among those which stopped early, The Telegraph continues. These included Herceptin and Tyverb for breast cancer and Avastin for bowel and renal cancer.
Writing in the online journal Annals of Oncology, Giovanni Apolone, maintained that “this suggests a commercial component in stopping trials prematurely. We are aware that trials stopped early because they are showing benefit may result in the identification of promising new treatments for patients. However, findings obtained in this way require subsequent confirmation. Without such evidence, unsafe and ineffective drugs could be marketed and patients’ health could well be jeopardized.”
Commenting on the results, David Kerr, an Oxford University cancer expert, says the process for evaluating new drugs had been greatly strengthened in the past two decades, but the trend of stopping trials early was worrying. “There’s great pressure on the cancer community to get drugs out as soon as possible,” he tells the paper. “Everyone wants new and powerful therapies available quickly. But we need to be sure we understand the evidence first.”
Dan
Or, such a trial is discontinued before a significant risk is discovered following any benefit that stops the study.
Nathan
But remember that (in theory) part of the reason for stopping the trial early is out of compassion and ethics. If a stongly drug looks like it is saving lives, why withhold it’s benifits from a population that is dying? You can always add a few more warning labels later on. You can’t bring a dead cancer victim back to life in order to be treated with your lifesaving drug.
Jack2
It’s statistically HIGHLY improbable that a drug that works early in the trial (with a small number of treated patients at an interim analysis) would then be found to not work later (when a larger number of patients are treated). The ability to detect a significant difference increases the larger your sample.
Look at the AIDS vaccine Merck attempted. When they did an interim analysis they expected to find no difference (because it was too early), but hoped to find the vaccine worked. The interim analysis showed the vaccine was worse than comparison (placebo? I don’t remember). It’s not like Merck kept going with the trial because it hoped they would turn things around and find something different at the end- why put more patients at risk. If you see statistically significant changes at an interim analysis, you can be confident those changes are real - since they needed to be so dramatic to get detected at all.
Furthermore, many drugs include open-label extension trials, specifically designed to detect additional adverse events. Once you know it works you can then start to look exclusively at safety.
Former Pharma Marketing Executive
Jack2, you make an excellent point. However, as we can see in this case it seems that pharma was not willing to keep the extended access trials open.
Certainly, once the drug has been found to show real benefit early on, everything must be done to get it to the patients quicker. The FDA here in the US allows for the Fast Track program. However, this system also puts the impetus on pharma to continue to provide and accrue data over the longer term as part of the early approval process. Failure to comply includes the levying of fines, if I am not mistaken.
I think this is where patient groups play a key role in monitoring the long term data.
Above all, we need to remember that even though a drug saves lives, it may not be in the long run what turns out to be best for the patient. It may cause problems in the not too long term that prevent the patient from seeking other types of treatments. Specifically, if you look at some of the drugs like sutent, gleevec and others and the risk (albeit it small) for heart attacks. This suggess that patients will be in less than optimal condition to try other newer treatments.
However, with cancer you have to outweight the risks versus benefits. Missing from the equation are issues the patients need to understand with regards to quality of life prolonged and what complications, further medical treatment and financial burdens this may represent.
I can’t imagine finding answers to these issues is easy at all.
M Helm, MD
It can be hard to get to the relevant end point (survival) if a trial ends early because there is tumor regression or non-progression or some other surrogate end-point. If the best you can hope for is a holding action on the tumor, and you don’t change the prognosis, is there a benefit to the patient?
Sadly, even with cancer medications, it can not be assumed that just because there is some repsonse that this corresponds to a “saved life.”
Licensing/approval should not be the only rationale for studying a drug. Perhaps reviewing agencies (or the “cancer community”) should insist that cancer treatment trials be designed in such a way that if an interim analysis shows a benefit there is some ethical and appropriate modification of the protocol allowing continuation of trial to meet the original goals and establishing whether or not there is a benefit in days of quality life added, if not cure?
ol cranky
Correct me if I’m wrong but weren’t most of these trials halted on the recommendation of independent DSMBs?