Campaign 2008

More
medical examiner
columns

Waiting Doom
How hospitals are killing E.R. patients.
Zachary F. Meisel
posted July 24, 2008
Your Health This Week
Are C-section rates rising because mothers are getting older? And why cholesterol drugs for kids are such a bad idea.
Sydney Spiesel
posted July 14, 2008
The Nose That Never Knows
The miseries of losing one's sense of smell.
Elizabeth Zierah
posted July 8, 2008
The Sex Difference Evangelists
The next best-seller.
Amanda Schaffer
posted July 7, 2008
Dr. R2-D2
The invasion of the surgeon robots.
Kent Sepkowitz
posted June 27, 2008
Search for more medical examiner articles
Subscribe to the medical examiner RSS feed
View our complete medical examiner archive

medical examiner: Health and medicine explained.

Group TherapyHow poorly designed trials for cancer drugs are hurting patients.

By Jessica Wapner
Posted Thursday, Jan. 10, 2008, at 12:02 PM ET

(Continued from page 1)

All of these factors mean that many drugs make it through Phase 2, only to fail in Phase 3. The poor predictive value of Phase 2 cancer trials is well-documented. One study found that 81 percent of Phase 3 cancer studies had worse outcomes than their Phase 2 precursors, meaning that the drug turned out to help fewer people than expected. And not one of the 181 Phase 2 studies included in this report reliably predicted the drug's effectiveness. By contrast, the standard randomized and blind Phase 2 design does far more to eliminate all variables except the drug and produce results that are more likely to be replicated at Phase 3.

So why do drug companies continue to do single-group studies? Many of the small biotechs developing today's cancer treatments have only one or two drugs in the pipeline. Wooing investors depends on the quality of those drugs, and quality is gauged by advancement through clinical trials. Even if the drug turns out to be a dud, touting positive results along the way to Phase 3 can keep a company afloat while its scientists generate new drug candidates.

Other incentives also come into play. Phase 2 cancer studies are often run by younger doctors for whom a successful study means a prominent publication, which leads to grant funding and job offers. In fact, many scientists do Phase 2 trials with no intention of ever proceeding to Phase 3; they just want the good results. Academic institutions benefit, too: Patient recruitment is a lucrative business for medical centers, and any positive trial is an opportunity to enhance its reputation. These incentives coupled with inertia from having done studies the same way for so long are a recipe for inefficiency. There are more Phase 2 trials in cancer than in any other area of medicine; almost all new cancer drugs look good at that stage, and very few of them end up helping patients.

But so what? What does it matter if companies use Phase 2 studies to lure investors, and doctors use them to lure job offers? Shouldn't good drugs make it through clinical trials anyway? As it turns out, the answer is no. Fewer than 5 percent of cancer patients enroll in clinical trials, far too few for testing all the drugs in clinical development. As Dr. Glassman notes, these volunteers are wasted on the majority of drugs that turn out to be no good, while promising drugs languish because there is no one left to test them on. So not only do the cancer patients willing to participate in clinical trials lose out, so do all the other patients anxiously awaiting better medicines. The wastefulness of the Phase 2 cancer study design is its own source of harm.