New Cancer Approach: Fast, Cheap, and Radical

Frustrated by the cost and the dead-slow pace of drug development, a group of researchers today launched a unique new strategy for running clinical trials. Their aims are bold. They want to bring numerous test products under a single umbrella, validate new biomarkers, share data publicly, maintain uniform standards across a wide range of candidate drugs, and quickly weed out losers. Predicted to cost $26 million over 5 years, the project will begin by testing new therapies for breast cancer. Leaders say they hope that clinicians working on other diseases will copy their experiment-and melanoma may be next. The just-launched experiment is about 60% funded, according to its sponsor, the independent Foundation for the National Institutes of Health.

The project is called I-SPY 2, which, just barely, stands for "Investigation of Serial studies to Predict Your therapeutic response with imaging and molecular analysis 2." (It follows a pilot test called I-SPY 1.) The first patients will be enrolled next week, said Laura Esserman, a surgeon at the University of California, San Francisco, who briefed reporters at the National Press Club in Washington, D.C. She's the trial's clinical leader. She said five therapeutic agents have been selected for initial testing.

Anna Barker, deputy director of the National Cancer Institute, said a systemic reform of the clinical trial system is "long overdue."

One of the project's promised payoffs is a coherent body of knowledge, said Janet Woodcock, director of the Center for Drug Evaluation and Research at the U.S. Food and Drug Administration. She called it an "experiment with a high likelihood of success." Senator Arlen Specter of Pennsylvania (D) appeared at the briefing to give I-SPY 2 his blessing and remind people that he's pushing to increase the annual budget of the National Institutes of Health from $30 billion to $40 billion.

The chief aim of I-SPY 2, according to Esserman and others, is to collapse the time it takes to get a green or red light for a potential new breast cancer drug from 10 years to 1 year. Other goals are to identify which drugs work best for which cancer patients, using novel biomarkers to describe more precisely than before the match between disease subtypes and treatments. The new strategy calls for administering drugs before surgery—usually doctors do surgery first—and precisely imaging the tumors' response before moving on to extended drug therapy.

The approach also requires a new suite of statistical methods to make it possible to obtain credible results from small test groups. The plan is to predict very early which drugs are likely to succeed and which are not, and promote the winners rapidly to larger (phase III) trials, which most likely would be run by private companies, without data sharing. Biostatistician Don Berry of the M.D.Anderson Cancer Center in Houston, Texas, is overseeing I-SPY 2's data analysis. The concept, he said, is to continuously add new drugs to multiple arms within the trial, promote a few, and drop the rest. Berry concluded: "We hope this trial will go on forever."