Glioblastoma, a highly aggressive brain cancer with a five‑year survival rate below five percent, has not responded well to conventional therapies, including immunotherapy. Carriers of chimeric antigen receptor (CAR) technology are genetically reprogrammed immune cells engineered in the laboratory to recognize and destroy cancer cells.
A collaborative group from the University of Geneva (UNIGE) and the Geneva University Hospital (HUG) has engineered CAR‑T cells that attack glioblastoma by targeting proteins found in the tumour micro‑environment, demonstrating effectiveness in an animal model and laying the groundwork for human clinical trials.
The findings were published in the Journal for ImmunoTherapy of Cancer.
Glioblastoma forms a mass in the brain composed of malignant cells and surrounding non‑cancerous cells, as in most tumours. However, it is distinctive in that it contains very few naturally occurring T lymphocytes, the immune cells capable of recognizing and eliminating cancerous cells.
“Unlike melanoma or some lung cancers, glioblastoma does not respond to standard immunotherapies because it lacks T cells,” said Valérie Dutoit, researcher in the Department of Medicine and the Translational Research Center in Onco‑Hematology (CRTOH) at UNIGE.
“Therefore, our strategy is to provide the missing T cells by producing them in the laboratory,” she added.
The creation of CAR‑T cells involves harvesting patient T cells, genetically modifying them in vitro so they can locate and kill tumour cells, and then reinfusing them into the patient.
“Identifying tumour‑specific proteins that can be targeted by T cells without harming healthy tissue is especially challenging in glioblastoma, which is highly heterogeneous,” explained Denis Migliorini, professor in the Department of Medicine and head of the neuro‑oncology unit at HUG.
In earlier work, the team identified PTPRZ1, a marker expressed on the surface of some glioblastoma cells, as an important target. However, attacking a single antigen leaves the tumour vulnerable to relapse.
Building on this, they added a second target, Tenascin‑C (TNC), a protein secreted into the tumour’s extracellular matrix. By directing CAR‑T cells against TNC, the cells trigger a pro‑inflammatory cascade that kills the producer cells.
“Moreover, we have shown that CAR‑T cells can eliminate glioblastoma cells that do not express TNC, broadening their effect while sparing healthy tissue,” Migliorini noted.
Scientists have also faced the problem of rapidly emerging resistance mechanisms that exhaust CAR‑T cells. By identifying three markers of cellular exhaustion and counteracting their influence, Dutoit’s group succeeded in significantly prolonging CAR‑T cell activity in mice bearing human‑derived glioblastoma.
This encouraging data now supports the transition to a clinical trial. “Our aim is to generate genetically modified T cells against multiple targets simultaneously,” said Migliorini. “The trial is planned to commence in about a year, with sites in Geneva and Lausanne.”
“Personalised adjustment of CAR‑T cells will be essential to eliminate as many tumour cells as possible in the face of heterogeneity,” added Dutoit.