|PETER McLAREN BLACK, M.D., Ph.D.|
Harvard Medical School
Children’s Hospital, Boston and Brigham and Women’s Hospital
Q: As both a basic scientist and a practicing neurosurgeon who sees the worst kinds of brain tumors on the operating table, what do you think are the biggest obstacles to finding novel therapies for brain tumors?
Black: For one thing, brain tumors are still considered an orphan disease, and have not traditionally been part of what neuroscience thinks about. At the same time, I don’t think cancer research thinks of brain tumors very often, because they’re less common than lung, breast, or prostate cancer. Still, they are the third leading cause of cancer-related deaths in men 15 to 54, the fourth leading cause of death overall in women 15 to 54, and the most common solid tumor in childhood. So these are important tumors.
Another major problem has been that there hasn’t been a cadre of scientists that has worked for a generation or two on brain tumors, because these tumors have generally been considered rather rare and untreatable. I think they’re neither now. So we haven’t had the funding, and we haven’t had the scientists needed to be able to make major inroads. I think that is changing now, partly due to work done by the Dana Alliance. The idea is that as you get more neurobiologists and more neuro people thinking about brain problems, some of them are going to think about brain tumors, which are very interesting in terms of how they relate to brain development.
“...there hasn’t been a cadre of scientists that has worked for a generation or two on brain tumors, because these tumors have generally been considered rather rare and untreatable.”
Q: Your lab has been involved in developing “anti-angiogenic drugs,” which block blood vessel growth to starve tumors. Where is this research heading?
A: There are two main continuing challenges: 1) finding new anti-angiogenic drugs, and 2) thinking about how to deliver them. Right now our lab is very much involved in these studies, and one of the particular problems we’re working on is the question of delivery systems for these drugs. I think it was Judah Folkman who said that the major toxicity for many of these molecules is not a biological toxicity, it’s financial. That’s because the anti-angiogenic molecules are so expensive that it’s very hard to use them in large quantities—a treatment costs many tens of thousands of dollars.
We’ve had a couple of clinical trials with anti-angiogenic drugs, and it’s clear that they have to be used in conjunction with other drugs as well. And so far, they have not been hugely effective against brain tumors. The problem is getting the drugs, which are taken orally, to cross the blood-brain barrier in a strong enough dose to do what we hope they would do in a tumor.
On the other hand, these drugs do have some effect on other organs, and may start blood vessel formation in areas that you might worry about. So, if you get them directly into the brain, this is not such a concern. I think this is still an area of considerable interest.
To solve the delivery problem, one of the things we’re thinking might be possible is to give anti-angiogenic drugs directly into the tumor after it has been surgically resected. A paper in Clinical Cancer Research in November pointed out that, using this approach, you can get better effect with much less quantity of drug. We are also looking at injections or slow-release systems composed of polymers or nanoparticles that might be loaded with drug that would keep the tumor from getting a blood supply, or from growing. The idea, someday, would be to take the tumor out and leave these small, sort of “tapioca” in the cavity, similar to what has been done with radiation pellets. I think we’re at least many years away from such a scenario, because we not only need better drugs but also a better way to get them into the brain.
Q: Neurosurgery might be the best example of how brain-imaging technology has directly impacted patient care. How are these technologies changing neurosurgery practice today?
A: I think neurosurgery is a pretty good example of a clinical application for brain-imaging technology, particularly for making a difference in the actual management of brain tumors. Imaging has allowed us to detect tumors earlier, to follow them more accurately, and to guide us so we end up with better localization of the tumor and more complete, safer removal. It is really very, very powerful in terms of manipulating the brain. The technologies are changing, and are being applied to neurosurgery in increasingly innovative ways. One example is intraoperative MRI, in which neurosurgical procedures are performed within a specialized MR scan, enabling minute-by-minute planning and assessment of the surgery progress and precise localization of lesions.
“Imaging has allowed us to detect tumors earlier, to follow them more accurately, and to guide us so we end up with better localization of the tumor and more complete, safer removal.”
Q: What do you think the role should be of the clinician-scientist in bridging gaps between patient care and basic science?
A: The clinician scientist has an absolutely crucial role, and it goes both ways, translating from the bench to the bedside and vice versa. Number one, the clinician scientist can think about advances that are accruing in basic oncology or basic neurobiology and how they apply to brain tumors. An example is somebody thinking about the linkage between development and glioblastoma: Does a glioblastoma come from a multipotential precursor? Is it a cancer of stem cells? So the clinician scientist can start to think about how basic findings might be applied, how one might then get started toward clinical trials, and then actually do clinical trials of drugs.
But the other direction is just as important. The clinician scientist can frame the questions for the basic scientist that are clinically relevant, for instance, by saying, “We want a drug that will stop the blood vessel formation for this particular kind of tumor because that’s what it relies on to grow.” That becomes very important for the basic scientist because he or she can start to think about what applications there may be for new discoveries. This can provide a kind of motivation for continuing to find out about the tumor.
EVAN SNYDER, M.D., Ph.D.
The Burnham Institute La Jolla,California
Q: You’re a stem cell biologist. How did you get involved in brain tumors?
Snyder: It’s really a perfect marriage, when you think about it. The qualities that the stem cell possesses are exactly the reasons brain tumors are so elusive to therapy. Brain tumor cells migrate, sometimes great distances. They invade normal tissue. They may break off from the main tumor mass as satellite cells and crawl along blood vessels or white-matter tracts, seeding new little tumors all along the way. We had already established that stem cells were also migratory, that they could be engineered to express therapeutic molecules, and that they had a propensity for “homing in” on pathology, even when widely disseminated. Indeed, we had come to believe that cancer cells and stem cells were really just two sides of the same coin, sharing a number of similar properties. It simply made sense that stem cells would be a great way to deliver therapeutic agents to heretofore elusive brain tumors, in a way superior to anything we have now.
Beginning more than a dozen years ago, I would hear researchers say time and time again, “We did gene therapy for a brain tumor model in the mouse, but in humans it didn’t work, because we couldn’t get to all of the tumor cells.” After a while, it seemed to me that the stumbling block was not so much a failure to find effective genes but rather a problem with delivering those genes effectively to the most virulent of tumor cells, a virulence based on the intriguingly stem-like behavior of tumor cells. I began doing Gedankenexperiment—little “thought experiments”— thinking, “I bet stem cells could deal with this problem.” But I had no intention of ever going into the brain tumor field, a field that was new to me and, I perceived, fairly crowded and competitive. Helping to establish and legitimize the fledgling field of stem cell biology was already occupying every aspect of my life. I was content to do these brain tumor “thought experiments,” and not much more.
Then, in the mid-90s, my best friend, Dr. Jim Galambos, was diagnosed with a brain tumor. The irony was that he was a linguist, and the inoperable tumor started in and infiltrated his left temporal lobe, the language center. His kids came to me and said, “We read that you’re doing this stuff with stem cells. Is there anything you can do for our father?” As a clinician, I knew there wasn’t a lot that could be done for my friend. But I promised them—as a neurobiologist—I would dedicate my work to eradicating this scourge for the sake of future families. So, with that impetus, the “thought experiments” eventually became real experiments.
Q: You ended up collaborating with Xandra Breakefield at Massachusetts General Hospital to prove the concept that neural stem cells could be used to deliver gene therapy to a tumor, and more recently with your former trainee, now a scientist at Yonsei University in South Korea. What did you find?
A: It turned out to work. The stem cells really did exactly what I hoped they would do, constituting a novel phenomenon we could document thanks to careful observations by a post-doc in my lab, Dr. Karen Aboody. The findings actually took on even greater interest and significance because they also served to unveil some new fundamental aspects of stem cell biology. They became the first published demonstration (in Proceedings of the National Academy of Science in 2000) that stem cells could home in on pathology, even in the adult brain, along huge distances, and along non-stereotypical pathways. We were also able to show that one could put stem cells (even neural stem cells) in the bloodstream and they could still find pathology, even in the brain. These phenomena have been replicated by many others and applied to other brain tumors, as well as to other cancers (using a variety of stem cell types). The approach has even been used for treating multiple sclerosis and stroke with stem cells, as well as trying to treat metastatic cancers of non-neural origin. This gave rise to the notion of the “homing effect” of stem cells, and had a broad impact on the stem cell field. In the paper’s Acknowledgment, I dedicated the research to my friend Jim.
In the most recent work, we genetically engineered human neural stem cells (cells suited for clinical use) to deliver a gene that produces “tumor necrosis factor-related apoptosis-inducing ligand,” or TRAIL [a protein that induces tumor cells to die]. They effectively traveled throughout the main tumor site and to the metastases, attacked the cancer, and reduced tumor size dramatically.
Q: What would this therapy look like in a clinical trial?
A: In the animal models, we’ve done it a few ways. You can target the tumor directly and let the stem cells find any cells that break away from the mass. Or, you can stick the stem cells somewhere else, even on the opposite side of the brain, and they will find the cells. The fear with glioblastoma is that there may be individual cells migrating microscopically away from the tumor. The stem cells will find these cells. The third route that we have used is to put the stem cells into the [cerebral] ventricles, which is also a way of making sure they distribute widely throughout the brain.
In brain tumor patients, we may want to utilize all three of these approaches. For example, a patient newly diagnosed with a brain tumor typically has to have it excised as much as possible. At the time of surgery, it may be possible to also instill stem cells bearing a tumor-destroying (oncolytic) gene into the cavity to target any cells left behind at the margins or that have already broken away from the main tumor mass. Brain tumor patients often have reservoirs put into their ventricles for receiving chemotherapy; this could be a route through which one could re-dose the patient with stem cells. If a patient has a recurrence, or even if something suspicious is detected on CT scan or MRI, the patient is typically re-biopsied; this procedure would be an opportunity to implant the stem cells into the area of the recurrence. And, because one visible region of recurrence probably means there are others that can’t be seen on scans, the homing ability of the stem cells becomes advantageous, because they will seek out these potential seeds for new masses. In other words, a clinical trial could be conducted without subjecting patients to additional invasiveness.
Q: When might we see human trials of stem cell-based gene therapy in glioblastoma?
A: We are ready to go, and interested in hearing from any company or foundation or academic institution that wishes to partner with us to get this going quickly. Much of the private sector has not jumped into the stem cell field, for various reasons—some because of the politics, others because clinical translation seems so distant. But gaps in present brain tumor therapy are so consistent with the known properties, capabilities, and techniques of stem cell biology, that there is no question in my mind that this is the low-hanging fruit in the field. We have the potential for helping to satisfy—if even in a small measure—an acute medical need. All we lack now is the funding to give it a shot in a clinical trial.