An Experimental T-Cell Therapy Bought Years for Four Children With Terminal Brain Cancer

A child with an aggressive brain tumor had already been through seventeen rounds of chemotherapy and radiation before doctors at Children's National Hospital took T cells from the child's own blood, exposed them to tumor proteins in a lab, multiplied them, and infused them back. More than two years later, that child and three others with terminal diagnoses remain alive, according to Phase 1 results published in Nature Medicine last week and summarized in Scientific American.

The trial enrolled 33 children and young adults with either newly diagnosed diffuse intrinsic pontine glioma (DIPG) or recurrent pediatric brain tumors, including glioblastoma, medulloblastoma, ependymoma, and astroblastoma. Its primary endpoint was safety and dose-finding, not efficacy. The durable responders sit inside that small, uncontrolled safety study: three children with recurrent disease, all heavily pre-treated, alive with no evidence of disease between two and five years after therapy, plus one DIPG child alive more than two years out.

The therapy, called TAA-T, takes T cells from the patient's blood, exposes them to antigen-presenting cells loaded with three tumor-associated antigens common across pediatric brain cancers, and expands them outside the body before IV reinfusion. There is no genetic engineering. The targeting is multi-antigen rather than the single synthetic receptor that defines CAR-T therapy. CAR-T cells are engineered to bind one specific surface marker, which works well when a tumor reliably expresses that marker. Pediatric brain tumors are hard to resect because they infiltrate eloquent brain regions, and the blood-brain barrier limits systemic drug delivery. A T cell that recognizes several antigens at once has more options when a tumor has learned to hide one.

One child with a brain-stem tumor died from complications investigators attributed to the treatment at the highest dose tested. Two other patients developed serious tumor swelling. The most common adverse events were fatigue and headache.

Catherine Bollard, chief research officer at Children's National Hospital and co-senior author of the study, called the results "an important step toward developing safer and more effective T cell therapies for children with devastating brain cancers." Tim Hassall at Queensland Children's Hospital, who was not involved in the study, was supportive but measured. In coverage by New Scientist, he said no one in the field is "jumping up and down" yet. The result is a durable signal in a small subgroup, not a survival claim.

Two follow-on trials are already in motion. The first combines TAA-T with focused ultrasound to temporarily open the blood-brain barrier, letting more T cells reach the tumor. The second uses each patient's tumor sequencing to design a personalized antigen target, rather than relying on three shared antigens. Both are early-stage. The trial is registered as ReMIND, NCT03652545 on ClinicalTrials.gov.

For a field that has spent a decade watching CAR-T struggle against solid pediatric brain tumors, the question is whether a non-engineered, multi-antigen, autologous approach can do what engineered cells could not. The durable signal in three of 33 patients is the starting point, not the answer. Larger cohorts and longer follow-up are the next tests.