Main Session
Sep 27
PQA 01 - Gastrointestinal Cancer and Central Nervous System

2277 - Evaluating Stereotactic Radiosurgery as an Alternative to Whole Brain Radiotherapy for NSCLC Patients with 15 or More Brain Metastases

03:00pm - 04:00pm ET
Poster Hall - Exhibit Hall A
Screen: 6
POSTER

Presenter(s)

Wen-Chi Yang, MD, PhD - National Taiwan University Hospital, Taipei, 100

Y. N. Lim1,2, W. C. Huang3, F. M. Hsu3,4, and W. C. Yang3,4; 1Department of Radiotherapy and Oncology, Pantai Hospital Penang, Penang, Malaysia, 2Department of Radiotherapy and Oncology, Sarawak General Hospital, Kuching, Malaysia, 3Division of Radiation Oncology, Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan, 4Department of Radiation Oncology, National Taiwan University Cancer Center, Taipei, Taiwan

Purpose/Objective(s):

Brain metastasis is common in non-small cell lung cancer (NSCLC). While whole-brain radiotherapy (WBRT) is the standard treatment for multiple metastases, it causes significant neurocognitive toxicity. As systemic breakthroughs improve NSCLC survival, mitigating WBRT-related neurotoxicity is a priority for preserving quality of life. Stereotactic radiosurgery (SRS) better preserves neurological function and is effective for multiple brain metastases up to 20 lesions. However, its safety and efficacy for a very high disease burden remain insufficiently explored.

Materials/Methods:

We retrospectively evaluated NSCLC patients with 15 or more brain metastases treated with CyberKnife or LINAC HyperArc SRS between September 2021 and April 2025. Baseline characteristics, molecular profiles, dosimetric parameters, and adverse events (AEs) were collected from medical records. Endpoints included brain progression-free survival (bPFS) and overall survival (OS), AEs, time to salvage WBRT, and leptomeningeal metastasis (LM) incidence. Survival was estimated using the Kaplan-Meier method, and time-to-event endpoints used cumulative incidence (CI) curves with death as a competing risk. Spearman correlation was used in dosimetry correlation to other clinical factors.

Results:

Fifty-seven patients (median age 66, range 36–87) were analyzed; 50 (87.7%) had actionable driver mutations and 28 (49.1%) had prior brain radiotherapy. Median lesion number was 25 (range 15–126) and median cumulative tumor volume was 4.6 cm3 (range 0.5–39.8). Median prescription dose was 1800 cGy (range 1200–2400 cGy). At median follow-up 9.8 months, 35 (61.4%) had intracranial progression and 29 (51%) died. Median bPFS and OS were 4.9 and 13.8 months; bPFS was 44.7% at 6 months and 24.1% at 12 months. Salvage WBRT was given to 10 (17.5%) patients (CI 8.7% at 6 months; 15.2% at 12 months). Among those without LM before SRS (n=46), 9 (19.6%) developed LM (CI 5.6% at 6 months; 15.5% at 12 months). Early AEs occurred in 14%; late AEs in 22.8% (Grade 1: 19.3%), most commonly fatigue and dizziness. Radionecrosis (RN) occurred in 7 patients (12.3%), with only one experiencing Grade 3 RN. Dosimetric analysis showed tumor volume and lesion number positively correlated with mean brain dose and brain V12Gy (both p < 0.001). CyberKnife showed better conformity by Paddick gradient index (p < 0.001) and brain V12Gy normalized to tumor volume (p = 0.006) compared to HyperArc. Patients lacking actionable mutations (p < 0.001) or receiving multiple prior systemic treatments (p = 0.039) had worse bPFS.

Conclusion:

These findings suggest that SRS is a safe and viable alternative to WBRT for NSCLC patients presenting with 15 or more brain metastases, offering intracranial control with limited toxicity.