Pilocytic astrocytoma (PA) is a benign tumor with a low risk of high-grade transformation. High-grade astrocytoma with piloid features (HGAP) is a recently recognized tumor entity that may arise de novo or from low-grade gliomas. Here, we report a patient harboring both a fourth ventricular PA and a spinal cord tumor. Methylation profiling classified the brain tumor as PA and the spinal tumor as HGAP. Both tumors shared a KIAA1549::BRAF fusion with identical DNA breakpoints, confirming a common clonal origin. The spinal tumor also showed complex chromosome copy number aberrations, including homozygous deletion of CDKN2A/B. These findings suggest additional molecular alterations associated with tumor progression within the piloid lineage. To our knowledge, this is the first report providing comprehensive, paired molecular characterization of both PA and HGAP in the same patient, offering insight into their potential evolutionary relationship.

A 2-year-old Boxer dog was examined for a 2-week history of progressive, painful right-lateralized paraparesis. Magnetic resonance imaging showed a focal ovoid mass with suspected intramedullary, or less likely, intradural location at the level of L3 vertebral body. Based on the imaging features, neoplasia, and specifically, a nephroblastoma, was suspected. Surgical excision found a difficult to excise mass with indistinct borders. The dog showed marked deterioration postoperatively and was euthanized 3 days later due to suspected myelomalacia. Histopathology and immunohistochemistry of the biopsy revealed that the spinal cord was replaced by a poorly organized proliferation of cells that consisted of both neuronal and glial components, embedded within an edematous neuropil. Immunohistochemistry of the lesion showed positivity for neuronal and astrocytic markers. These findings were consistent with a ganglioglioma, a rare neoplasm in dogs.

Neurosurgical diseases-including brain tumors, hemorrhage/trauma, ischemia, infection, epilepsy, and spinal cord injury-share convergent neuro-immune mechanisms. In the acute phase, sterile inflammation and barrier disruption trigger innate immune cascades. During the subacute phase, immune resolution and clearance determine the quality of tissue repair. In the chronic phase, persistent immune-glia interactions and synaptic remodeling influence epileptogenesis and long-term cognitive outcomes. Recent discoveries-such as the meningeal immune niche, meningeal lymphatic system, and glymphatic clearance pathways-have redefined the classical concept of "CNS immune privilege." The central nervous system is no longer viewed as immune-isolated, but rather as a compartment whose immunity can be directly modulated by surgical intervention and perioperative management. This review proposes a conceptual framework in which neurosurgery serves as a programmable "immune anchor point." By integrating knowledge of neuro-immune interface architecture and temporal dynamics, we establish a closed-loop model encompassing structural pathways, immune dynamics, delivery/timing, and efficacy/toxicity. This paradigm shift aims to accelerate breakthroughs in CNS immunotherapies. The article unfolds along three main themes: (1) the structural foundations of neuro-immune communication-including barrier systems, the meningeal immune niche, and meningeal lymphatic-glymphatic coupling; (2) temporal immune dynamics across acute, subacute, and chronic phases, and their roles in edema, secondary injury, and failed resolution; and (3) the brain tumor immune microenvironment, with a focus on surgical synergy and analysis of why immunotherapies (checkpoint inhibitors, vaccines, oncolytic viruses, cell therapies) have largely failed in glioblastoma. Finally, we propose a translational roadmap integrating perioperative immune management, spatial omics stratification, and local immunoengineering.

Relapsed acute lymphoblastic leukemia (ALL), particularly with central nervous system (CNS) involvement, remains a major cause of treatment failure and is inadequately controlled by existing antibody-drug conjugates (ADCs) with tubulin inhibitors. To address this limitation, we developed IL-7Rα-targeted monoclonal antibodies and identified clone 577 as the lead candidate. Using this antibody, we generated ADCs conjugated with either monomethyl auristatin E (MMAE) or the highly potent DNA-damaging payload-PNU-159682 (PNU). In head-to-head comparisons, 577-PNU showed >50-fold greater potency than 577-MMAE in vitro and induced complete tumor regression in xenografts at a 20-fold lower dose. Additionally, 577-PNU provided durable systemic disease control and markedly reduced leukemic infiltration in the brain and spinal cord in both preventive and established murine CNS disease models, offering direct evidence of effective CNS penetration. Safety assessments demonstrated stable body weight, normal hematology and serum biochemistry, and no treatment-related pathologies. Collectively, these findings provide the first preclinical evidence that IL-7Rα-directed ADCs armed with DNA-targeting payload PNU-159682 can achieve durable elimination of systemic and CNS leukemia at tolerable doses, demonstrating both clinical feasibility and CNS disease control, and establishing a compelling rationale for their translational and clinical development in relapsed and refractory ALL.