Dr. Sandeep Kunwar Outlines Blood-Brain Barrier Solution for Glioblastoma

Drug delivery remains one of the biggest barriers to treating glioblastoma, particularly because the blood-brain barrier prevents most therapies from reaching tumors in the brain. According to Dr. Sandeep Kunwar, only about 0.01% of many biologic therapies, such as antibodies or gene-based treatments, can cross into brain tissue, limiting their effectiveness.

Kunwar — a neurosurgeon and professor at UCSF, and also the co-founder of Precision NeuroMed — explained that convection-enhanced delivery (CED) offers a different approach by bypassing the blood-brain barrier entirely. The technique uses a small catheter placed directly into the brain to slowly infuse treatment into the surrounding tissue, allowing doctors to deliver therapeutic concentrations to tumors while reducing effects on the rest of the body. With newer imaging and AI-based planning tools, researchers hope this approach could more precisely target migrating glioblastoma cells and expand access to treatment in the future.

CURE: For patients with recurrent glioblastoma, what is the biggest challenge today when it comes to getting drugs into the brain?

Kunwar: This challenge applies not just to recurrent glioblastoma, but really to most diseases of the central nervous system. We understand a lot of the biology of these diseases and we’ve developed some remarkable drugs that cured animals. But the barrier between those results and treating humans has often been drug delivery. Glioblastoma also has additional hurdles compared with many other cancers.

The key issue is the blood-brain barrier. I think of it like Saran Wrap around every blood vessel in the brain. It’s the result of about 100 million years of evolution designed to keep foreign molecules out of the brain. That protection is important, but it also makes treatment difficult. Small molecules like Temodar (temozolomide) can cross the barrier somewhat and that drug has been FDA approved for many years. But small molecules tend to slow the disease rather than stop it.

The therapies with the greatest potential for long-term control are biologics — proteins such as antibodies, gene therapies or cellular therapies. The problem is that only about 0.01% of those drugs actually reach the brain. So we either affect the rest of the body or fail to get enough drug into the tumor to reach therapeutic levels. The challenge is delivering a sufficient concentration of drug into the brain while still protecting healthy brain tissue.

Many patients hear about the blood-brain barrier. Can you explain in simple terms why it makes treating glioblastoma so difficult?

The blood-brain barrier is essentially the brain’s protective filter. Every blood vessel in the brain is tightly regulated to prevent most substances from entering brain tissue. We’re exposed to many chemicals every day through the air we breathe and the food we eat, and the brain is such a delicate organ that it needs that protection.

But the same system that protects the brain also blocks many therapies. Even with small molecules, only a very small percentage actually enter the brain. For larger therapies like antibodies or cellular treatments, almost none of the drug crosses the barrier. About 99.9% remains in the bloodstream instead of reaching the tumor.

How does convection-enhanced delivery differ from the way chemotherapy or other systemic treatments are typically given?

For decades we’ve relied on treatments given through the bloodstream, either by IV or pills. That approach is convenient, but for glioblastoma it hasn’t produced many advances. The last approved drug for this disease came about 20 years ago.

Convection-enhanced delivery, or CED, takes a different approach. Instead of trying to force drugs through the blood-brain barrier, we go around it. We place a small catheter directly into the brain and slowly infuse the drug into the surrounding tissue.

There are several advantages. First, the drug stays primarily in the brain, so we avoid many systemic toxicities like liver or kidney damage. Second, we can deliver therapeutic concentrations directly to the tumor area. If we know the dose needed to affect tumor cells, we can actually get that amount into the brain tissue.

Another important advantage is that this approach may reach the migrating tumor cells that spread into nearby brain tissue. I often call them the “queen bees,” because each of those cells can eventually form a new tumor. If we can target those cells without harming healthy brain tissue, then we have a real opportunity to control the disease.

CED has existed for about 20 years, but the technology has gradually improved. Now we’re seeing the ability to deliver more complex therapies, including protein-based and gene-based treatments, directly into the brain.

From a patient’s perspective, what would treatment with this type of delivery system involve?

There are two main parts. The first is planning. I often compare CED to an HVAC system. If you pump 70-degree air through one vent in a room, eventually the entire room reaches that temperature. In a similar way, we deliver drug through a pressure gradient so it spreads through a region of brain tissue.

The brain isn’t uniform, though. Different areas have different tissue densities, so we need to plan carefully where the catheter should go. In the past, surgeons had to estimate this using two-dimensional scans, which could be complicated.

Now we can use a patient’s MRI to create a three-dimensional model of their brain and simulate how fluid will move through the tissue. In effect, it becomes a GPS system that helps determine the best location for the catheter.

The second part is the procedure itself. It’s minimally invasive and similar to procedures used for brain biopsies or deep brain stimulation. A very small incision and tiny drill hole are used to place the catheter. The surgery can take about an hour.

After the catheter is placed, the drug is infused slowly while the patient rests, reads or watches television. Infusions can last several hours, and when they are complete the catheter is removed and the patient can go home. Our goal is to eventually make this an outpatient procedure that is well tolerated and not painful.

Where does PNM-201 currently stand in clinical development?

PNM-201 is a drug I helped co-develop at the NIH more than 20 years ago. What’s interesting is that with advances in delivery technology, we’ve come back to this molecule because of its biological effectiveness.

Glioblastoma appears to contain at least two main populations of cells. One group grows quickly, while another group divides slowly but migrates into surrounding brain tissue. Those migrating cells — the “queen bees” — are a major reason tumors return.

PNM-201 targets those cells. The molecule attaches to the tumor cell and is internalized. We sometimes describe it as a “smart bomb” because it selectively enters those cells and can destroy them with only a small number of molecules. Unlike standard chemotherapy, it does not rely on cells actively dividing.

The drug previously went through a phase 1 trial and was even fast-tracked toward phase 3 because the early results were promising. However, those trials happened when convection-enhanced delivery technology was still very early. Drug delivery was inconsistent and sometimes the drug didn’t reach the intended tissue.

Even so, we saw some remarkable outcomes, including long-term survivors. One patient has now lived about 20 years after treatment. When we reviewed the data, it became clear the challenge was not the drug itself but the delivery technology.

Today we believe we can deliver the drug much more reliably. However, we still need to remanufacture the drug and go through the FDA process before testing it again in patients. Realistically that may take a couple of years, although we hope to shorten the timeline to around 14 to 16 months before returning to clinical trials.

If the AI-enabled planning platform works as intended, what could it mean for patients?

After treating patients with glioblastoma for more than 20 years, I’ve learned that quality of life is incredibly important. Some of our long-term survivors have lived active lives for years after treatment.

Our goal isn’t just extending survival by a few weeks. We want to see meaningful benefit, potentially delaying death by years or even preventing it in some patients. We already have a patient about 20 years out from treatment who is working and traveling.

Another important goal is accessibility. We don’t want this treatment limited to a small number of major medical centers where patients must travel long distances. The AI planning platform is designed to standardize the procedure, similar to how GPS made navigation easier. Our hope is that this technology will allow more surgeons to perform the treatment safely so patients can receive care closer to home.

Transcript has been edited for clarity and conciseness.

For more news on cancer updates, research and education, don’t forget to subscribe to CURE®’s newsletters here.