In the search for cancer drugs, mice get new role

Amid a glut of treatments in need of testing, a researcher is giving them simultaneously to rodents and humans

By Carolyn Y. Johnson
Globe Staff / April 5, 2010

E-mail this article

Invalid E-mail address
Invalid E-mail address

Sending your article

Your article has been sent.

  • E-mail|
  • Print|
  • Reprints|
  • |
Text size +

Dr. Pier Paolo Pandolfi has begun an experiment that could radically change the way new cancer drugs are tested: Not far from where his colleagues give trial treatments to patients at Beth Israel Deaconess Medical Center, he’s giving the same regimen to mice.

Normally, mice are test subjects for drugs long before they make it into the clinic. Animal models of disease are important to better understand cancer and to determine which of myriad compounds are safe and promising enough to be tried in people.

But in the new work, funded with a $4.2 million economic stimulus grant from the National Cancer Institute, mice are receiving treatment in tandem with humans, in the hope that what is learned at the bedside can be integrated with results from the lab bench to speed up and streamline the development of cancer drugs.

The novel strategy is an attempt to address a bottleneck that has emerged in the cancer drug pipeline because of a proliferation of drugs in development.

“There are 800-plus drugs in the pharmaceutical pipeline, and . . . we already don’t have enough people going into clinical trials to begin with,’’ said Cheryl Marks, associate director in the division of cancer biology at the National Cancer Institute. “Our traditional clinical trial structure . . . is simply not serving us very well.’’

The realization that clinical trials may need a new approach stems from research in recent years that has revealed the tremendous complexity of the disease.

Cancers found in one organ — breast, prostate, or lung, for example — can actually be multiple diseases, with different underlying genetic causes. So scientists are developing targeted therapies that fight cancer by blocking or interfering with specific molecules involved in tumor growth, a shift from the old approach of using one-size-fits-all chemotherapy.

There are about 850 cancer drugs in clinical development, according to the Tufts Center for the Study of Drug Development, a 70 percent increase in three years.

With so many options, it’s often not clear to clinicians which drug — or combination of drugs — will work best for which types of patients. And there are not enough patients, time, or money to find out. That’s where the mice come in.

Mice can be engineered to carry one or more faulty genes known to cause human cancers, and in the pilot project, various human prostate- and lung-cancer gene mutations are being inserted into mice. Those animals are given identical treatments to people in clinical trials, like a shadow group of patients. Pandolfi and collaborators from the Dana-Farber/Harvard Cancer Center hope the mice, with simpler genetics than the humans, will reveal more about what is happening in the patients.

For example, if a drug fails in six of the mice but works in one with a particular genetic mutation, it could help scientists better understand which patients it might work for. By going back and forth between the genetic information from the mice and the patients, researchers could design the next stage of a clinical trial to test only the subset of patients they expect to respond.

If the mice develop resistance to the treatment, as often happens in cancer, researchers can begin to look for combinations of drugs that might be useful that could help inform future clinical trials.

“The real goal of our effort is to understand why one patient responded and why the other didn’t, so that then the next round of trials, we give the drug only to that patient who will respond,’’ said Pandolfi, director of the cancer genetics program at Beth Israel Deaconess.

In addition to the stimulus grant, he has also received a five-year, $3.75 million grant to create a blueprint for how to best use these co-clinical trials.

Pandolfi first used this strategy 15 years ago on rare acute promyelocytic leukemia. By using genetically engineered mice, he and others were able to understand how different forms of the disease responded to treatment. They used mice to identify experimental combinations of drugs that might work. By alternating back and forth between the clinic and the lab, the strategy turned a once-fatal cancer into a treatable disease.

Dr. Derek Raghavan, director of the Taussig Cancer Institute at the Cleveland Clinic, said that moving back and forth from a human clinical trial to look at genetically engineered mouse models is one interesting approach to the bigger problem of refining and improving the design of clinical trials. In bladder cancer, Raghavan plans to use a model created by taking patient tissue and putting it into a mouse to better determine the sequence and combination in which drugs should be given.

“I would say there’s a bunch of people around the country who are all sitting back and saying, ‘The model is kind of broken,’ ’’ Raghavan said. “We don’t have enough patients going into clinical trials; we have to be smarter about how to work up the first stages of clinical research.’’

One of the patients whose treatment is being mirrored through Pandolfi’s experiment is Ben Gilmore, a 62-year-old civil engineer from Rochester. Gilmore has aggressive prostate cancer and has been coming to Boston regularly for treatment, including chemotherapy, hormones, and a drug called Avastin, which is already approved for other cancers.

Not far from the ninth-floor doctor’s office where Gilmore goes for treatment, groups of mice are going through the same regimen.

“Whatever I participate in is obviously going to be another step along the road to finding the best approach to treating cancer,’’ Gilmore said.

For doctors and scientists, there is hope that the combination of data from Gilmore and from dozens of mice might give them critical clues.

If Pandolfi “comes up and tells us while we’re doing the trial a particular group of patients is more likely to respond to a drug, we can think about the trial differently and go after more of those patients,’’ said Dr. Glenn Bubley, Gilmore’s doctor and director of the genitourinary medical oncology program at Beth Israel Deaconess.

“The thing about the animal, you can kind of do nine trials at once . . . so that in this decade we can start making faster progress.’’

Carolyn Y. Johnson can be reached at