BACKGROUND Brain metastases develop in 44-63% of patients with EGFR-mutant non-small cell lung cancer (NSCLC) during the course of disease and are associated with neurological morbidity and reduced quality of life. Osimertinib, a third-generation EGFR tyrosine kinase inhibitor (TKI) with good central nervous system penetration, is the standard first-line treatment based on the FLAURA trial. However, among osimertinib-treated patients who do not receive local brain therapy, the brain is frequently the first site of progression, and intracranial relapse often arises from initially involved or residual lesions. Retrospective series suggest that adding upfront stereotactic radiation to EGFR-TKI improves intracranial and overall outcomes, and a pooled analysis of two randomized phase II trials (OUTRUN/TROG 17.02 and LUOSICNS) has reported encouraging results; prospective data remain limited, particularly in Southeast Asia. In Vietnam, Gamma Knife is available at few centres whereas LINAC is widely available. No prospective study has reported outcomes of osimertinib combined with LINAC-based SRS/SRT in Vietnamese patients.

OBJECTIVES

(1) To evaluate intracranial (RANO-BM) and extracranial (RECIST 1.1) response to osimertinib combined with LINAC-based SRS/SRT in EGFR-mutant NSCLC with brain metastases. (2) To evaluate survival outcomes (intracranial progression-free survival, progression-free survival, overall survival) and adverse events.

DESIGN Prospective, single-centre, single-arm, open-label interventional study conducted at the Department of Medical and Radiation Oncology, Oncology Center, Cho Ray Hospital, Ho Chi Minh City, Vietnam. The study population is patients with symptomatic brain metastases, i.e., brain metastases accompanied by neurological symptoms, either at diagnosis or arising while on osimertinib; symptoms are classified as raised intracranial pressure (headache, nausea/vomiting), focal neurological deficits (limb weakness, sensory disturbance, visual or language disturbance, seizure, imbalance) or cognitive-behavioural changes (memory/concentration decline, personality or mood change). Eligible patients are identified consecutively through the multidisciplinary oncology tumour board and enrolled after written informed consent. Time zero for all time-to-event endpoints is the first day of SRS/SRT.

INTERVENTION Systemic therapy: osimertinib 80 mg orally once daily. Patients not yet on systemic therapy start osimertinib concurrently with SRS/SRT; patients already receiving osimertinib continue it without interruption during SRS/SRT. Osimertinib is continued until progression, unacceptable toxicity or withdrawal; treatment after progression follows institutional practice.

Stereotactic radiation: frameless immobilization with a thermoplastic mask (Fraxion stereotactic system); CT simulation with slice thickness of 1 mm or less (Siemens Somatom Definition AS) fused with contrast-enhanced MRI (or PET); planning on Monaco TPS (VMAT/DCAT, flattening-filter-free beams); delivery on an Elekta Versa HD LINAC with cone-beam CT image guidance. GTV = contrast-enhancing lesion; CTV = GTV; PTV = GTV + 0-2 mm. Dose: lesions of 2 cm or smaller, 18 Gy in 1 fraction; lesions larger than 2 cm, 27 Gy in 3 fractions or 30 Gy in 5 fractions. Planning goals: GTV D100% at least 100% and Dmax at most 135% of prescription; PTV D98% at least 100%; Paddick conformity index close to 1; gradient index approximately 3-5. Normal-tissue constraints (1/3/5 fractions): normal brain V12 \< 5-10 cc / V20 \< 20 cc / V24 \< 20 cc; brainstem Dmax 8-12 / 21-24 / 25-30 Gy; optic nerves/chiasm Dmax 8-10 / 15-18 / 25-30 Gy. Planning is completed within 3 days of simulation and treatment is delivered within approximately 1 week.

ASSESSMENTS Baseline: history and examination, ECOG performance status, neurological symptoms, contrast-enhanced chest-abdomen CT (PET-CT or bone scan as indicated), contrast-enhanced brain MRI (CT if MRI is contraindicated), AJCC 8th edition TNM staging, complete blood count, liver and renal function, electrolytes, hepatitis B and C serology, ECG and echocardiography. Response: brain MRI (RANO-BM) and chest-abdomen CT (RECIST 1.1) every 12 weeks (3 osimertinib cycles) or when clinically indicated; images are reviewed by radiologists not involved in treatment. Safety (CTCAE v5.0): acute radiation toxicity recorded at each fraction by radiation therapists and physicians; clinic visits every 4 weeks; blood tests, ECG and echocardiography every 12 weeks; late radiation toxicity (radiation necrosis, intracranial haemorrhage, seizure) recorded throughout follow-up. After progression, participants are followed for subsequent therapy, intracranial progression (if extracranial progression occurred first), survival and late toxicity through clinic visits and telephone contact.

SAMPLE SIZE Sample size was calculated with the single-proportion formula n = Z\^2 p(1-p)/d\^2, with Z = 1.96 (95% confidence), an expected 6-month intracranial PFS rate p = 0.88 (based on OUTRUN/TROG 17.02) and precision d = 0.10, giving a minimum of 41 participants.

STATISTICAL ANALYSIS Continuous variables are summarized as mean (SD) or median (IQR); categorical variables as counts and percentages, compared with the chi-square or Fisher exact test. Response rates are reported with 95% confidence intervals. Time-to-event endpoints are estimated with the Kaplan-Meier method and compared with the log-rank test; multivariable Cox regression is used to explore factors associated with intracranial PFS and PFS (e.g., EGFR mutation subtype, number and size of brain metastases, T and N status). Two-sided p \< 0.05 is considered significant. Analyses are performed in Stata.

ETHICS The protocol was approved by the Ethics Committee in Biomedical Research, Cho Ray Hospital (approval No. 46-26/CN-HDDD, 27 February 2026; full-board review). All participants provide written informed consent; personal data are coded and kept confidential. The study forms part of a doctoral thesis at Hanoi Medical University.