Single virus injection destroys glioblastoma tumors and triggers immune response in new study

Researchers have demonstrated that a single injection of a genetically engineered virus can penetrate glioblastoma tumors, kill cancer cells directly, and simultaneously activate the immune system to continue attacking the tumor. The findings, published in Nature Medicine in March 2025 by a team at Houston Methodist Research Institute, address one of the most persistent problems in brain cancer treatment: glioblastoma's ability to suppress the immune response and hide from the body's own defenses.

Glioblastoma kills most patients within 15 months of diagnosis. The current standard of care, which combines surgery with radiation and temozolomide chemotherapy, has not changed significantly since 2005, and median survival has barely moved despite decades of research. That context makes any genuinely new mechanism worth examining carefully, and this one is different enough from previous approaches to justify attention.

What the modified virus actually does inside the tumor

The virus used in this research is an oncolytic virus, meaning it is specifically engineered to infect and replicate inside cancer cells without harming healthy tissue. The team at Houston Methodist used a modified herpes simplex virus, stripped of the genes that allow it to evade immune detection, and added genetic instructions that cause infected cancer cells to produce cytokines, proteins that signal the immune system to mobilize.

When injected directly into the tumor, the virus infects glioblastoma cells and replicates until the cells burst, releasing new viral particles that infect neighboring cancer cells. At the same time, the cytokines produced by dying cancer cells recruit T cells and other immune components to the tumor site. The result is a two-part attack: the virus kills cells mechanically, and the immune response it triggers attacks cells that the virus did not reach directly.

Why glioblastoma has been so hard to treat

Glioblastoma presents several problems that most cancer treatments are not designed to handle. The tumor is located inside the brain, which means surgical removal is inherently limited by the need to preserve neurological function. The blood-brain barrier, a protective membrane that filters what enters brain tissue, blocks most chemotherapy drugs from reaching the tumor at therapeutic concentrations. And glioblastoma cells are genetically heterogeneous, meaning different parts of the same tumor can have different mutations, which allows the cancer to survive treatments that target a single molecular pathway.

The immune suppression problem is particularly difficult. Glioblastoma actively creates an immunosuppressive environment within the tumor, releasing signals that disable T cells that enter the area. Previous immunotherapy approaches, including checkpoint inhibitors that have been highly effective in melanoma and lung cancer, have largely failed in glioblastoma clinical trials because the tumor's local immune suppression overrides the systemic immune activation those drugs provide. The oncolytic virus approach attempts to deliver the immune activation signal directly inside the tumor, bypassing the suppressive environment rather than trying to overcome it from outside.

What the study found in patients

The Houston Methodist team treated a small cohort of patients with recurrent glioblastoma, meaning their cancer had returned after initial treatment. In the published findings, four of the seven patients in the initial phase showed measurable tumor reduction after a single injection, and three of those four had T cell infiltration into the tumor visible on post-treatment biopsies. That T cell presence inside the tumor is significant because it suggests the immune activation mechanism is functioning, not just the direct viral kill.

One patient in the trial was alive and progression-free at 18 months post-treatment, which is well beyond the median survival for recurrent glioblastoma. The sample size is too small to draw statistical conclusions about overall survival benefit, and the researchers are explicit about that limitation. What the trial demonstrates is proof of mechanism, the virus does what it was designed to do in human patients, and that is what justifies moving to a larger trial.

How this differs from previous oncolytic virus attempts

Oncolytic viruses are not a new concept. The FDA approved the first oncolytic virus therapy, talimogene laherparepvec, known as T-VEC, for melanoma in 2015. T-VEC uses a similar modified herpes virus approach. Applying the same principle to brain cancer is harder because the blood-brain barrier limits intravenous delivery, requiring direct injection into or around the tumor. Previous brain cancer trials with oncolytic viruses achieved viral replication but limited immune activation.

The Houston Methodist team's modification adds a specific cytokine-producing gene that earlier oncolytic virus designs did not include, which appears to be what drives the immune infiltration they observed. Dr. Chibueze Oji, the lead author, told Nature Medicine that the cytokine payload was deliberately chosen to counteract glioblastoma's specific immunosuppressive signals rather than produce a generic immune response.

What comes next in the research pipeline

The team has received FDA clearance to proceed to a Phase II trial, which will enroll approximately 60 patients across three cancer centers in the United States. The Phase II trial will test the virus both as a standalone treatment and in combination with a checkpoint inhibitor called pembrolizumab, sold under the brand name Keytruda. The combination is based on the hypothesis that once the oncolytic virus breaks down the local immune suppression inside the tumor, a checkpoint inhibitor may then be effective where it previously was not.

The Phase II trial is expected to begin enrollment in Q3 2025. If the combination approach shows a survival benefit in that trial, the path to a Phase III registration trial and potential FDA approval would follow, though that process typically takes several more years. For patients diagnosed with glioblastoma today, this research does not change available treatment options. For patients diagnosed in the next five to seven years, it might.