What happens when a clinical trial fails? This year, the Food and Drug Administration approved some 40 new medicines to treat human illnesses, including 13 for cancer, three for heart and blood diseases and one for Parkinson’s. We can argue about which of these drugs represent transformative advances (a new medicine for breast cancer, tested on women with relapsed or refractory disease, increased survival by just a few months; a drug for a type of leukemia had a more lasting impact), but we know, roughly, the chain of events that unfolds when a trial is positive. The drug is approved for human use; “postapproval marketing” is deployed to commercialize the treatments; slick ads materialize on TV; fortunes are built. Yet the vastly more common experience in the life of a clinical scientist is failure: A pivotal trial does not meet its expected outcome. What happens then?
A few years ago, I was a lead investigator in a study for a drug for blood cancer. Let’s call the medicine O. The compound, designed to kill leukemia cells, had shown efficacy on cancer cells in Petri dishes. The trial was backed by a small company with just a handful of employees, many of whom had invested their lives, and their life savings, in the company.
The first patient to enroll was a 60-something woman whom my colleagues and I had been treating with other medicines. Unfortunately, all the other drugs had stopped working. Her illness caused her bone marrow — the body’s nursery for the genesis of blood cells — to malfunction, and so her blood counts would collapse every two weeks. She would be back at the hospital, awaiting a blood transfusion. The repeated transfusions, in turn, provoked immune responses, making it hard to find a match for her. She would wait for hours, or even days, before we could find a subtype of blood that would not be rejected by her body.
We started the experimental medicine on a Monday. When she returned to the clinic two weeks after, I could see her face illuminated with the warm flush of color that is instantly recognizable to a hematologist. Her blood counts were up. We bumped fists for the first time in our lives — she was usually more formal — and I sent her home with a congratulatory nod. But then the response flickered off. Her blood counts sank again. A few weeks passed, and the drug stopped working altogether.
In 74 hospitals, where other trials were ongoing, patients were also struggling with strange, flickering responses. From 2003 to 2006, nearly 300 patients were enrolled in a randomized trial — with one group receiving the medicine and the other a placebo. The overall survival rates showed no difference.