In the push to match medical therapies to the genetic underpinnings of disease, lung-cancer treatments have been at the frontier. But the 1.6 million people diagnosed with this cancer every year will take scant comfort in knowing that of the past 20 late-stage trials of drugs to treat it, only two yielded positive results. And in only one of those 20 were patients chosen systematically by screening for biomarkers such as relevant blood proteins or DNA sequences.
Now, an ambitious project aims to improve those success rates and speed new treatments to market by matching companies with the patients whose tumours are most genetically relevant to the therapies they are trying to develop. The project is slated to launch next year and, if successful, could be expanded to other cancers.
The project was spearheaded by the Friends of Cancer Research, a think tank and advocacy group in Washington DC, and has won the support of the US National Cancer Institute and the US Food and Drug Administration (FDA). The idea is to streamline the drug-approval process by bringing pharmaceutical companies together to test multiple experimental drugs in late-stage clinical trials under a single, ‘master’ protocol. “The drive is to make the whole process of personalized medicine more efficient,” says Eric Rubin, vice-president of oncology clinical research at Merck, a pharmaceutical firm based in Whitehouse Station, New Jersey.
Plug and play
Launching a large, late-stage clinical trial typically takes more than two years and requires some three dozen administrative and regulatory approvals. To simplify this tangle, the master protocol will create an experimental plan to test several candidate drugs in hundreds of clinics across the United States. The initial protocol is expected to include up to six drugs; others may be added later, without the need for fresh protocol approval each time. “It’s like a Plug and Play,” says David Gandara, an oncologist at the University of California, Davis, who is in charge of drafting the plan. “So you don’t waste time over and over.”
Gandara has advocated this approach for the past decade, but the FDA and the pharmaceutical industry voiced support only recently — swayed by a growing body of data revealing that cancers are, in effect, many rare diseases with different genetic roots (see Nature 455, 148; 2008). A genetically targeted drug may work, but only in a fraction of cases. Such rare effects could easily be overlooked in a trial that contains a mix of patients whose cancers have heterogeneous causes, and the costs for drug companies to sort them all and run scores of separate trials are prohibitive.
Under the master protocol, by contrast, patients will be screened for various biomarkers and assigned to trials for drugs that are most likely to be effective. The approach does away with the need for patients to undergo multiple screenings: participating companies could enrol them from a large, central pool. It also eases pressure on the (often minute) tissue samples taken during lung biopsies, because many tests can be done at the same time, says Rubin.
A similar model is already being tested in two smaller clinical trials for breast and lung cancers (see Nature 464, 1258; 2010). Both trials involve multiple biomarkers, drugs and clinics, and both won support from pharmaceutical companies. But that does not mean that drug companies will embrace a larger, more developed venture, says Roy Herbst, an oncologist at the Yale School of Medicine in New Haven, Connecticut, who chairs the steering committee of the master-protocol project. It is much easier to coax a company into a group effort for a small, early trial than to persuade it to give up any measure of control over a late-stage one crucial for gaining regulatory approval.
Companies also prefer to maintain control of proprietary information rather than deposit early results into centralized databases. “It’s a challenge,” says Herbst. “Many of them might think they can do it alone, and may worry about losing autonomy.”
The project’s organizers tried to address industry concerns early on, says Ellen Sigal, founder and chairwoman of Friends of Cancer Research. At a planning meeting in March, representatives from more than 20 drug companies were assured that the FDA supports the protocol and has statisticians working to help shape it — making the agency more likely to feel comfortable basing approval decisions on data from the trial. Organizers also pledged to have a neutral third party monitor the trial, to ensure that drugs made by competing companies would not be directly compared.
Gandara hopes that the speed and lower costs will also draw industry partners. Late-stage clinical trials can cost between US$50 million and $100 million; Gandara estimates that the master protocol could cut that to $25 million or less.
Companies might also be wooed by easy access to the National Cancer Institute’s vast network of treatment centres and clinicians who are experienced in conducting clinical trials. That network will allow the trial to be conducted at 500 sites in the United States and Canada and enable it to enrol up to 1,000 patients a year.
Thus far, the downside of participating seems minimal, says Richard Gaynor, head of oncology-product development at Eli Lilly, a pharmaceutical firm based in Indianapolis, Indiana. “It will be an interesting experiment,” he says.
- Nature 498,146–147 () doi:10.1038/498146a