Radiobiology and Radioresistance in High-Dose Radiosurgery for Brain Tumors: A Hypothesis-Generating Study Using an Intracranial Glioma Mouse Model.

PURPOSE: Stereotactic radiosurgery (SRS) is a precise, noninvasive treatment for brain tumors, yet underlying radiobiological mechanisms remain unclear. This study explored long-term dose-dependent tumor response to high-dose SRS in a murine glioma model, focusing on tumor-associated macrophages (TAMs) as key regulators of tumor microenvironment and immune modulation.

METHODS AND MATERIALS: Using the intracranial GL261-glioma mouse model, single-dose SRS was administered at either 20 Gy (clinically prescribed dose) or 40 Gy as a dose escalation approach (n = 24/dose). Tumor response was assessed longitudinally using 7T magnetic resonance imaging at predefined intervals (d7, d30, d90, and d180) post-SRS or earlier upon symptom onset. Histological analyses performed at each timepoint evaluated cell proliferation, apoptosis, vascular morphology, blood-brain/tumor-barrier integrity, hypoxia, TAM recruitment, and polarization. Immune cell populations within tumor microenvironment were characterized using flow cytometry. Statistical analyses included a t-test and one-way analysis of variance with Bonferroni or Dunnett correction.

RESULTS: SRS efficacy was dose-dependent: 40 Gy suppressed tumor growth, whereas 20 Gy led to regrowth in 29% of cases between d30-51 post-SRS, necessitating stratification into responders and nonresponders. Responders demonstrated reduced cell proliferation, sustained apoptosis, and vascular remodeling indicative of vessel normalization. Nonresponders exhibited ≤94.5% increased hypoxia and ≤300-fold increased CXCR4 expression compared with responders. TAM recruitment inversely correlated with tumor volume (r = -0.8619; P = .0056). The M1/M2 ratio in nonresponders was similar to that of matched controls, but 3.8-fold and 4.8-fold lower than in responders at d30 and d90 post-SRS, respectively. Fluorescence-activated cell sorting analysis confirmed an increased M1/M2 ratio by d30 in responders.

CONCLUSIONS: This study offers key insights into longitudinal SRS radiobiology, highlighting the dynamic role of TAMs in sustaining long-term tumor control. Our findings support an association between hypoxia, CXCL12/CXCR4 signaling, and treatment resistance, and suggest a potential SRS-induced vascular normalization that may support improved therapeutic outcomes. Although not designed to establish causality, the data provide a spatially and temporally resolved framework to guide future mechanistic studies and inform more effective SRS-based combination strategies.