Lymph node–like immune hubs in the skull respond to brain tumors before the rest of the body does, study finds

Lymph node–like immune structures inside the skull's bone marrow mount a faster, local attack on brain tumors than the body's wider lymph nodes do, at least in mice, according to a Nature paper published Aug. 19, 2026 by researchers at Washington University School of Medicine in St. Louis.

The structures are not the brain's own tissue. They are small immune hubs nested in the marrow of the skull, the flat bones that wrap the head. From that perch, they sample the surface of the dura mater, the tough outer membrane that lines the inside of the skull, and they fire before immune cells in the neck or the rest of the body do. Kipnis called the hubs "security stations" in the institutional release: local outposts that flag trouble for the rest of the body before distant lymph nodes catch the signal.

The team disrupted the hubs pharmacologically in mice with glioblastoma, the most aggressive human brain cancer. Tumors in those animals grew faster and the mice died sooner than controls whose hubs were left alone. The direction of the effect is the point: removing the hubs made the cancer worse.

The same paper describes a therapeutic proof of concept. The team loaded a gel with three immune-boosting proteins and placed it under the scalp of glioblastoma-bearing mice. The gel drove antibody responses that began in the skull hubs and then spread outward to lymph nodes in the neck and beyond. Treated mice rejected tumors more effectively and lived longer than controls.

The skull-side location of the hubs matters for treatment design. Glioblastoma is hard to treat because the blood-brain barrier, the tight layer of cells that lines the brain's blood vessels, keeps most drugs out. Reaching the immune system through the marrow sidesteps that barrier, which is why a scalp-applied gel can drive a tumor response in mice.

The work extends a line of discoveries from senior author Jonathan Kipnis's lab. In 2015 the group identified dural lymphatic vessels, the brain's physical connection to the wider lymphatic system, overturning a long-held belief that the brain was sealed off from the body's immune surveillance. More recently the lab described microchannels through the skull bone that let antigens and immune cells move between the marrow and the surface of the brain. The new paper adds the missing third piece: lymph node–like tissue that organizes that traffic. First author Jang Hyun Park, PhD, a postdoctoral fellow, will open his own lab at KAIST in fall 2026.

Kipnis argues the hubs matter beyond glioblastoma. He lists Alzheimer's disease, Parkinson's disease, schizophrenia, and long COVID as conditions where the brain's immune environment shapes disease. He frames reaching the brain through the skull as a way to avoid the systemic side effects of drugs given through the bloodstream. None of those claims is supported by the current data. They are translational hypotheses, not treatment promises.

The scope is narrow. The functional evidence comes from mice. The human data in the paper is structural: bone-marrow samples from human skulls contain immune cells of the same types, not proof that the same anti-tumor response happens inside a living human brain. The new paper is one Nature study. It opens a question, not a cure.

The next step is to map the human hubs in detail and to test whether drugs delivered through the skull or the scalp can reach the same targets that the gel reached in mice. Watch for follow-up from the Kipnis lab and from the new Park lab at KAIST.