Neural Stem Cell Treatment May Improve Malignant Glioma Survival Rates

It can be devastating diagnosing a patient with a high-grade malignant glioma knowing what’s around the bend—the overall two-year malignant glioma survival rate is only 20%. Oncolytic virotherapy, however, has shown promise given its potential to be safe and tolerable for malignant glioma patients as well as trigger anti-tumor immune responses. But oncolytic adenoviruses, one of the most commonly used oncolytic viruses, cannot always cross the blood-brain barrier effectively. That’s where neural stem cells come in.

Clinical trials have shown that neural stem cells can not only cross the blood-brain barrier but can also enter the brain parenchyma to specifically target malignant glioma cells. In fact, Fares et al. found that using neural stem cells to deliver an oncolytic adenovirus to glioblastoma cells in mice can even improve survival rates. Now, they are hoping to achieve similar results in humans.

Published in The Lancet Oncology, their recent phase one in-human trial examined the use of neural stem cell delivery of an engineered oncolytic adenovirus for the treatment of newly diagnosed high-grade malignant gliomas. Their study, to their knowledge, is the first in-human trial of its kind.

Delivering Oncolytic Virotherapy Through Neural Stem Cells: A Novel Approach

Fares et al.’s phase one dose-escalation trial consisted of 12 patients who had been newly diagnosed with a high-grade malignant glioma between April 2017 and November 2019, and whose tumors were accessible for injection. Three patients had a tumor with a methylated MGMT gene promotor and the remaining nine patients had a tumor with an unmethylated the MGMT gene promotor.

Researchers used their previously engineered oncolytic adenovirus, CRAd-S-pk7, which had been injected into neural stem cells prior to the trial to create NSC-CRAd-S-pk7. Patients received the injection of the adenovirus into the tumor bed after neurosurgical resection, and injections were given in up to 10 sites in the wall of the resection cavity.

Three patients received a low dose of neural stem cells loaded with the oncolytic virotherapy particles, three a moderate dose and six a high dose. Ten to 14 days after the injection, patients underwent standard rounds of radiotherapy and chemotherapy.

The median progression-free survival rate was found to be 9.1 months, and the median overall survival rate was 18.4 months. For patients who had a glioma with an unmethylated MGMT promoter, the median progression-free survival was 8.8 month, and the median overall survival was 18 months. These numbers rose to 24.2 months and 36.4 months respectively in patients with gliomas with methylated MGMT promotors.

These results are thought to be promising for the future of high-grade malignant glioma survival rates, according to the study authors, given that all patients had favorable survival outcomes. The survival benefit for the unmethylated MGMT promotor patients is especially worth noting, as these patients do not typically respond to temozolomide treatment alone and are thought to only have a median progression-free survival rate of 5.3 months, with an overall survival rate of 12.7 months.

However, in order to validate these outcomes, Fares et al. say they will need to progress to their phase two and three trials, during which they will be able to study a larger cohort in controlled conditions. They hope that during these phases they will be able to further demonstrate the ability of their engineered oncolytic adenovirus to prolong survival rates in malignant glioma patients.

Innovations in Malignant Glioma Treatments

Prior to Fares et al.’s use of oncolytic virotherapy with neural stem cells, there have been several recent leaps in innovation in high-grade glioma treatment.

This past August, researchers at Tel Aviv University in Israel created one of the first perfusable 3D-bioprinted glioblastomas that can be used for reproducible drug testing. Their model not only replicates the cancerous cells but their entire microenvironment inside of the brain. Researchers believe that this will allow them to better predict how a clinical treatment will react inside of a patient’s body, potentially leading to personalized treatments in the future.

Also this year, researchers created a deep learning model that is believed to be one of the first to be able to classify the six most common brain tumor types using just one 3D MRI brain scan. Using intracranial 3D MRI scans from multi-institutional datasets, they were able to develop what they called a convolutional neural network that was then trained to distinguish between images depicting healthy tissue and images depicting tumors. From there, the model could then classify the tumor by type.

Their model will not only help lead to a completely automated workflow for tumor classification but can also potentially be extended to other brain tumor types or neurological disorders in the future.