Virginia Tech Is Engineering Viruses to Kill Brain Tumors. Dogs Will Help Test Them.

Doges Editorial · 2026-05-09 · 9 min read

Researchers at Virginia Tech are combining two experimental approaches to fight glioblastoma — and their road to human trials runs through dogs. What they've built, and what the dog connection means for both species.

James Weger-Lucarelli was at a cancer research conference at the Fralin Biomedical Research Institute when he started talking to Samy Lamouille. Weger-Lucarelli works with viruses. Lamouille works with brain cancer. They were at the same conference for different reasons — Weger-Lucarelli for personal ones, as it turned out — and they started describing their respective problems to each other. By the end of the conversation, they had the outline of something that had not existed before.

The result of that conversation, funded by a $414,482 National Institutes of Health grant, is a novel approach to treating glioblastoma — one of the most aggressive and treatment-resistant cancers known to medicine — using engineered viruses that can cross the blood-brain barrier, locate cancer cells, and deliver a molecule that kills the tumor cells that chemotherapy cannot reach. And when the time comes to test this in a clinical context before human trials, dogs will be involved.

## A Disease That Doesn't Respond

Glioblastoma is brutal in a specific way: it doesn't quit. Even aggressive treatment — surgery to remove as much of the tumor as possible, followed by radiation and chemotherapy — only extends the typical patient's life by twelve to fifteen months. The cancer returns, usually from cells that survived the initial onslaught and reactivated. The cells that come back are often the most resistant to treatment in the first place.

> "It's a very complex disease with significant heterogeneity. There are many variations of this disease from one patient to another." — Samy Lamouille, Virginia Tech Fralin Biomedical Research Institute

The specific culprit is a population of cells called glioblastoma stem cells — highly resilient, capable of dormancy, and able to evade conventional chemotherapy because they stop dividing during treatment and remain alive in a quiescent state. When treatment ends, these cells reactivate. The tumor returns. This is the cycle that Lamouille has spent years trying to interrupt.

## The Peptide That Works When Chemo Doesn't

Lamouille's contribution to the collaboration is a peptide called JM2. He discovered it while studying how glioblastoma stem cells communicate internally — specifically, noticing that a protein called connexin43, which normally sits at the cell surface to facilitate communication between cells, was appearing inside glioblastoma stem cells in an unusual location. He investigated whether blocking that internal protein interaction would affect the cancer cells. The results were striking.

While glioblastoma stem cells proved resistant to standard chemotherapy in lab conditions, JM2 killed them consistently across samples taken from different patients. The peptide had found a vulnerability that chemotherapy couldn't touch. The problem was delivery: peptides break down rapidly in the bloodstream and cannot easily cross the blood-brain barrier — the protective shield that keeps most substances out of brain tissue.

## The Virus That Goes Where Drugs Can't

This is where Weger-Lucarelli's contribution arrives. He works with alphaviruses — a family of viruses with a natural property that is, from a cancer researcher's perspective, remarkable: they can cross the blood-brain barrier when delivered into the bloodstream. The same characteristic that makes certain viruses neurologically dangerous can, in theory, be redirected toward a therapeutic purpose.

> "The virus has a natural property — it's able to get across the blood-brain barrier and get into the brain when it's delivered through the bloodstream. So that's the origin of those conversations." — James Weger-Lucarelli, Virginia-Maryland College of Veterinary Medicine

The engineered virus, as Lamouille and Weger-Lucarelli are developing it, would serve as a delivery vehicle for JM2 — using the virus's natural brain-seeking ability to carry the cancer-killing peptide to exactly where it needs to go. The concept builds on a century of research into oncolytic viruses, first observed to shrink tumors in the 1950s when cancer patients experienced unexpected remissions after viral infections. That line of research was set aside when chemotherapy became dominant; it has been returning to attention as chemotherapy-resistant cancers have demanded new approaches.

## Directed Evolution — Like Dog Breeding, but for Viruses

To make the delivery virus even more effective, the researchers are using a technique called directed evolution — harnessing the virus's natural tendency to evolve rapidly in order to select for strains that more precisely target cancer cells while leaving healthy tissue alone. Weger-Lucarelli offered an analogy that lands differently than the usual laboratory language: the process is similar, in principle, to how wolves were selectively bred over thousands of years to become the extraordinarily diverse set of dog breeds we have today. The same fundamental biology — genetic variation, selection pressure, heritable change — applied at a much faster timescale.

## The Dog Connection

The research team is currently in the preclinical phase, testing the engineered viruses in mouse models. If those results support the approach, the next planned step is clinical trials in dogs — a decision that reflects both the opportunity and the logic of comparative oncology. Dogs develop glioblastoma naturally. It is one of the more common brain tumors in dogs, particularly large breeds, and their shorter lifespan accelerates the research timeline: a five-year clinical trial in dogs generates data that might take fifteen years to gather in humans.

A treatment that works in dogs with naturally occurring glioblastoma is substantially more likely to translate to humans than a treatment that works only in mice with induced tumors. The dog trials, if they proceed, will potentially offer new treatment options for canine patients at the same time they generate data relevant to human clinical trials. This is the promise of comparative oncology: the path to helping one species runs directly through helping another.

> "My mom, Leila, passed away from lung cancer, and I've always wanted to help prevent that from happening to other people, which is why I was at the cancer conference. So this project represents a significant shift in research direction for me." — James Weger-Lucarelli

## A Long Road, a Real Destination

Lamouille is direct about the timeline. From current preclinical work to a phase three human clinical trial, the path typically takes more than ten years. The regulatory hurdles, the clinical trial phases, the data requirements — none of it is fast. What is true now is that the approach exists, that it is funded, that it is in active development, and that the logic connecting the peptide to the virus to the tumor has been scientifically articulated in ways that have attracted institutional support.

For the dogs who will eventually be recruited into clinical trials, if the research proceeds as the team hopes, the benefit will be immediate and real: a new treatment option for a cancer with very few. For the humans who might eventually follow — patients with glioblastoma who currently face a prognosis measured in months — the path goes through dogs. It nearly always does, when the destination is somewhere the medicine has never been before.