KEY POINTS
- This narrative review examines radiotherapy for IDH-wildtype glioblastoma and IDH-mutant grade 3–4 astrocytoma as biologically distinct diseases. The authors integrate clinical and mechanistic evidence published between 1980 and 2026, focusing on DNA-damage response, radioresistance, tumor microenvironment, molecularly targeted radiosensitizers, immunotherapy, and normal-brain protection. The review was not based on a formal systematic search or meta-analysis and does not introduce new patient-level clinical results.
- Established treatment standards remain supported by randomized evidence. For fit patients with newly diagnosed glioblastoma, 60 Gy in 30 fractions with concurrent and adjuvant temozolomide remains a reference regimen; the original Stupp trial improved median overall survival from 12.1 to 14.6 months (HR 0.63; p<0.001). In older patients, hypofractionated RT with temozolomide improved survival compared with RT alone (9.3 versus 7.6 months; HR 0.67; p<0.001). For IDH-mutant grade 3 astrocytoma, the final CATNON analysis supports 12 cycles of adjuvant temozolomide (HR 0.54; p<0.0001), whereas concurrent temozolomide did not demonstrate a significant overall survival benefit.
- Molecularly distinct gliomas present different opportunities and risks for radiosensitization. Glioblastoma stem-like cells demonstrate enhanced DNA repair, replication-stress responses, hypoxia-associated resistance, and adaptive interactions with the tumor microenvironment. These mechanisms motivate investigational combinations involving ATR, ATM, DNA-PK, PARP, and other DNA-damage-response inhibitors. Conversely, IDH-mutant gliomas may exhibit impaired homologous-recombination repair, raising a theoretical concern that pharmacologic IDH inhibition could restore DNA repair capacity and alter radiosensitivity. Clinical evidence defining the optimal timing of IDH inhibitors relative to RT remains insufficient.
- Randomized dose-escalation and reirradiation trials demonstrate why biological rationale does not guarantee survival benefit. In SPECTRO GLIO, MRSI-guided dose escalation to 72 Gy did not improve median overall survival (22.2 versus 22.6 months). In recurrent glioblastoma, NRG/RTOG 1205 found no overall survival improvement from adding 35 Gy in ten fractions to bevacizumab (10.1 versus 9.7 months; HR 0.98), despite improved median progression-free survival (7.1 versus 3.8 months). These results support continued investigation of imaging-guided intensification while emphasizing the need for clinically meaningful endpoints.
- Proton therapy and adaptive treatment offer technical advantages without established survival superiority. A secondary analysis of a randomized proton-versus-IMRT study found less grade ≥3 lymphopenia with protons (14% versus 39%; p=0.024), but improved immune preservation has not been shown to increase survival. Other prospective comparisons reported favorable toxicity or fatigue signals without demonstrated superiority for cognition or survival. MRI-guided adaptation, tighter target margins, and multimodal imaging may reduce normal-brain exposure, but prospective evidence linking these changes to better long-term function remains limited.
- FLASH radiotherapy and radioimmunotherapy combinations remain investigational in malignant glioma. Preclinical FLASH studies have reported normal-brain protection while maintaining tumor growth delay, but the effect depends on pulse structure, total dose, and biological context rather than average dose rate alone. Radiation-induced immune signaling through cGAS–STING and related mechanisms provides a rationale for immunotherapy combinations, while the immunosuppressive glioblastoma microenvironment remains a major barrier. No validated clinical FLASH regimen or radiotherapy–checkpoint inhibitor combination has established a survival advantage in glioblastoma.
- Long-term neurocognitive protection is particularly important in IDH-mutant disease, but proposed constraints remain uncertain. Mechanistic evidence implicates hippocampal progenitor injury, microvascular dysfunction, chronic inflammation, and oligodendrocyte injury in radiation-associated cognitive decline. The review discusses a proposed mean left-hippocampal dose objective ≤30 Gy, but explicitly notes that this is not a validated universal constraint for adult high-grade glioma. Similarly, published cumulative EQD2 values for reirradiation are heterogeneous and should not be interpreted as interchangeable tolerance limits. Tumor coverage and individualized risk assessment must remain central.
CLINICAL TAKEAWAY
Glioma radiotherapy is increasingly informed by molecular biology, imaging, and improved dose delivery, but most emerging strategies have not demonstrated clinically meaningful superiority over established approaches. Distinguishing IDH-wildtype glioblastoma from IDH-mutant astrocytoma is essential when interpreting treatment evidence and balancing tumor control against late neurological toxicity. The review provides a useful biological framework, not new evidence supporting immediate changes in radiotherapy prescriptions.