Why is brain cancer so hard to treat?

[THEME NOTES] FLORA LICHTMAN: Hey, it’s Flora, and you’re listening to Science Friday. Brain cancers are rare, but devastating. Patients face short life expectancies after diagnosis. Brain tumors are often impossible to remove completely, and the typical course of therapy, radiation and chemotherapy, doesn’t always stop them from spreading. It’s a daunting class of disease to study, but that didn’t dissuade Dr. Michelle Monje from trying. She’s spent two decades working on understanding how brain tumors form and interact with healthy brain cells, and pioneered the field of cancer neuroscience. Dr. Michelle Monje is a Professor of Pediatric Neuro-oncology at Stanford Medicine. Michelle, welcome to Science Friday.

MICHELLE MONJE: Thank you so much for having me.

FLORA LICHTMAN: Thanks for being here. Let’s go back in time. When you started 20 years ago, how much was understood about brain cancers?

MICHELLE MONJE: 20 years ago, there was really very little understood about the fundamental biology of brain cancers, and this wasn’t for lack of effort in the field. But the approach to brain cancers was, I think, missing some really important components and also limited by just technological abilities to study this really difficult disease and difficult to access disease. This was especially true about childhood brain cancers. When I was a medical student, I remember seeing the very first patient that I ever saw with diffuse intrinsic pontine glioma, which we now call diffuse midline glioma of the pons.

This is one of the most common aggressive cancers in pediatric neuro-oncology. It is the leading cause of cancer-related death in children. And there was just nothing known about it. And really, at the time, no way to study it. And seeing this disease unfold in the first patient that I cared for, watching this beautiful young girl who was a bright light coming into the clinic with just minimal symptoms of crossed eyes and a little bit of weakness on one side, watching her lose all ability to move, to speak, to communicate, and ultimately pass within six months felt like something I just couldn’t turn away from.

And so I decided at the time to really focus on these diseases and to approach them from the perspective of neurobiology. It really seemed to be cancers that happened in the childhood nervous system at particular points of development, at particular ages, in specific locations for those ages. And that really spoke to the idea that somehow these childhood brain cancers were diseases of brain development and childhood gone wrong. And I thought that perhaps I could approach it from that perspective.

FLORA LICHTMAN: I mean, it’s interesting that it was such a black box and so devastating and that you thought, I have to take this on. Why did you think you might be able to make some progress?

MICHELLE MONJE: Optimism?

[LAUGHTER]

And maybe a little bit of hubris. At the time I was a M.D. PhD student at Stanford, and I was studying neural stem and precursor cell biology. And one of the things that we had to do to study neural stem and precursor cells was to culture them from the brains, either of experimental animals or in the early postmortem hours after passing from humans. So I thought that we could apply that same technique to brain cancers.

One of the real problems with these diffusely infiltrative, kind of intrinsic diseases, like diffuse intrinsic pontine glioma, is that there isn’t resection tissue to study in the laboratory. You can’t surgically remove these cancers. And at the time, they weren’t even biopsied. That has changed. Pediatric neurosurgeons have been trained to do very– expertly skilled neurosurgeons have been trained to do biopsies. But at the time, there was not even biopsy tissue of diffuse midline gliomas. And so I thought maybe we could apply those techniques from neurobiology to culture, perhaps in the early postmortem period, these tumor cells.

FLORA LICHTMAN: So you’d have a model to work off of?

MICHELLE MONJE: Exactly. Because we had no way to study this in the laboratory. There was no model. This was just at the time– this is in the early 2000s to 2010s, and next generation sequencing had just come to be. And then very early in my clinical fellowship as a pediatric neuro-oncology fellow, one of my young patients with DIPG, his family, asked if they could donate his organs after death. There’s limited organ donation that can happen in children with cancer. We did donate what we could, but I told them that they could donate his tumor if they were interested, and they were very interested in that. And so when he passed, they donated his brain in the early hours after his passing, and I was able to culture his tumor.

And that became the first cell culture and xenograft mouse model of DIPG. And that was really a breakthrough that that child’s donation allowed us, for the very first time, to look at how these cells behave, try to understand what normal cell types they resemble, try to understand what potential therapies they might respond to and collaboratively with many others engaged in broad drug screening efforts using these patient-derived pediatric brain cancer models, trying to identify therapies that might target the cancer.

FLORA LICHTMAN: Why is brain cancer so hard to treat?

MICHELLE MONJE: Part of the challenge is the blood-brain barrier and getting sufficient quantities of the drug to the tumor. Part of the problem is treating enough of the brain. When you can’t get through the blood-brain barrier, you can deliver the drug right to the tumor. But these are not localized diseases. These are cancers that really spread very far and wide throughout the nervous system very early in the disease course. And so you have to treat a large volume of brain. You have to get enough of the drug to the cancer cell.

And much of cancer biology has thought about cancer in isolation. A lot of drug development is focused on screening cancer cells alone in a Petri dish. What kills them? What do they need to grow that we can take away? What are the mutations that cause them to occur in the first place that we can disrupt pharmacologically and then disrupt their growth-driving engine? That’s what most of oncology does. But things are very different when you take a cancer cell and don’t study it in a Petri dish, when you put it into the brain itself. Suddenly, those medicines that work very well when the cancers are alone in a Petri dish don’t work very well in the context of either the mouse brain or the patient’s brain.

But there’s much more going on than just the difficulty of getting the cancer cells exposed to helpful medicines when we think about brain cancer. And what has become clear over the last decade or so is that the cancer cells’ interactions with the nervous system protect it from these therapies. They protect it from the therapies and really are the fundamental driver of both growth and invasion.

FLORA LICHTMAN: Wow. Is the nervous system like a shield? How is it protecting these cancer cells?

MICHELLE MONJE: So maybe I can take a step back and say that we talked earlier about how these cancers emerge from brain development gone wrong in children and brain plasticity gone wrong in adults. There are ongoing processes by which your brain changes and adapts in response to experience and activity in adulthood. And so when I started my laboratory as an independent investigator in 2011, essentially the first set of experiments that I did was to test how neuronal activity– the activity of the neural circuits that the cancer is invading– might influence their pathology, might influence their growth, their invasion, and ultimately their resistance to therapy. And what we found was very striking, and we’re still studying it today. The activity of neurons of the patient’s brain very powerfully promotes the growth of a wide range of brain cancers.

FLORA LICHTMAN: How? How do they do it?

MICHELLE MONJE: So the activity of the nervous system very, very strongly promotes the growth of a wide range of these glial cancers, of gliomas, of diffuse intrinsic pontine glioma, of glioblastoma, of various forms of these glial malignancies in both kids and in adults. And so digging a little deeper and through more research, we ultimately realized that the cancer cells were forming real electrophysiologically functional synapses with the neurons and that that synaptic communication, the electrochemical communication, the electrical currents in the cancer cells was the fundamental driver of the cancer cell growth and invasion.

FLORA LICHTMAN: So the electrical signaling was actually driving the growth?

MICHELLE MONJE: Yeah, that’s right.

FLORA LICHTMAN: We have to take a break. But when we come back, I want to talk about how these advances in our understanding could improve treatment.

[THEME NOTES]

FLORA LICHTMAN: Can you interrupt those electrical signals?

MICHELLE MONJE: Yeah. What’s really exciting about this is that it’s a whole new avenue of therapeutic opportunities. So the electrical signaling between the normal brain and the cancer cells is really driving growth and progression. And we actually know that this is important, even in patients, because the degree to which the cancer is functionally connected to the rest of the brain robustly predicts survival, both in glioblastoma and in diffuse midline gliomas. So we know that this is important.

And what’s exciting is we can disrupt it. The way that the cancer cells are taking advantage of these signals, the way that the cancer cells are connecting to the neural circuits that they are invading, is through mechanisms that we’ve already developed medicines to target. Certain anti-seizure medicines, medicines of psychiatry, medicines of cardiology even, are designed to target the exact neurotransmitter receptors and ion channels that the cancer is subverting for its own nefarious purposes. And so when we test some of those medicines that target a particular mechanism we’ve uncovered that the cancer is using to grow, it does slow things down– in mice, at least– and it’s something that we can test in prospective trials in patients.

FLORA LICHTMAN: Have you started trials on this?

MICHELLE MONJE: We’re gearing up to begin trials. One of the medicines that we identified that disrupts one particularly important interaction in diffuse midline gliomas, we were able to look back at patient databases and ask if a patient was on this particular anti-seizure medicine, did they do better? Did they live longer? And this retrospective data analysis did support the idea that this was a useful therapy, that kids who were on this particular anti-seizure medicine, in fact, had better outcomes. And so we have both laboratory data, we have retrospective clinical data, and we’re gearing up right now to begin prospective clinical trials.

FLORA LICHTMAN: I’m guessing this will be one piece of a treatment puzzle. Are there other promising approaches that you would pair with this?

MICHELLE MONJE: Absolutely. And I think it’s really important to think about targeting the way that the cancer takes advantage of the nervous system as one crucially important pillar of a multi-pronged approach. Disrupting these interactions is going to slow down the cancer growth. It’s going to potentially make it more vulnerable to other therapies. But for these very difficult and very aggressive cancers, we need multimodal approaches. We have the opportunity to leverage the immune system that has been very effective in some cancer types for treatment, and that we can hopefully effectively leverage in both pediatric and adult brain cancers to help clear the tumor from the nervous system.

FLORA LICHTMAN: How far are we along with that, with the immunotherapy approach?

MICHELLE MONJE: So there’s been a great deal of research. There are a number of what appear to be really promising approaches. One that I began working on about a decade ago, we wondered if we could leverage CAR T-cell therapy to treat these diffusely infiltrative gliomas, to sort of send in engineered immune cells that are designed to specifically target and remove cancer cells. And the first step in immunotherapy is to identify a good target for that immune cell, that engineered immune cell, to go after.

And so screening those patient-derived cell cultures that we had from my patients, we identified one particular target that was very, very highly expressed in diffuse midline gliomas. It’s called GD2. And then after identifying that, I reached out to a colleague of mine at Stanford, and together we started testing her CAR T-cells, targeting GD2 in our models of diffuse midline glioma, and found, now nearly 10 years ago, that in the mouse models, the CAR T-cells cured the mice, cleared all the tumors. It was really remarkably effective.

We brought that to clinical trial six years ago, beginning in the spring of 2020, and we’ve been treating children with GD2-targeting CAR T-cell therapies over the last six years. And we’ve seen really promising signs of efficacy. We’ve seen kids who’ve gotten better, at least for a while. We’ve seen tumors shrink. We really are heartened that this is an approach that is going to be part of a solution.

It’s always harder to treat people than laboratory models, and we’re not seeing the rate of cures that I wish we were seeing. But I believe that as we learn more and combine this approach with the neuroscience approaches, with other medications, we’ll start to really make transformative progress. So as a field, everyone is working on different aspects. And the challenge now is to bring all of those different potential and promising therapeutic approaches together in the right combination.

FLORA LICHTMAN: What keeps you going?

MICHELLE MONJE: My patients. We have to find a solution to these terrible cancers. They deserve better. They deserve effective therapy. And I do believe that the answer is now within reach. It felt very far from reach when I started this 20 years ago. But now, I really believe that each of these different lines of inquiry are going to lead us to truly effective therapy. And one thing that is incredibly heartening, and that also helps keep me going, is that when we discover something in diffuse intrinsic pontine glioma and other diffuse midline gliomas or in glioblastoma, we find that what we’ve discovered is applicable to other cancers in ways that we couldn’t have predicted. And so the work feels very meaningful from that perspective because some common principles are really emerging.

It makes some sense that cancers of the nervous system would take advantage of nervous system activity. But we’ve also found that brain metastases, for example, from lung cancer, do the same thing when they go to the brain. It’s also now clear that peripheral nerves throughout the body are interacting with peripheral cancers in ways that are growth-promoting, metastasis-promoting, and similarly targetable when you disrupt these connections. So this whole field of cancer neuroscience has emerged largely from this initial work in childhood brain cancers. And that’s very satisfying. Seeing these common principles that have the potential to really help millions of people facing cancer is very promising and very encouraging to keep going down these paths.

FLORA LICHTMAN: Thank you, Michelle, and good luck with this research.

MICHELLE MONJE: Thank you.

FLORA LICHTMAN: Dr. Michelle Monje is a professor of pediatric neuro-oncology at Stanford Medicine. Shoshannah Buxbaum produced this episode. And if you have questions you want us to look into or comments about the show, give us a call– 877-4-SCIFRI is our number. That’s 877-4-SCIFRI. The listener line is always open. I’m Flora Lichtman. We’ll catch you next time.

[THEME NOTES]

Copyright © 2026 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/