P17.05.A CLINICAL OUTCOMES FOLLOWING SERIAL STEREOTACTIC RADIOSURGERY VERSUS WHOLE BRAIN RADIOTHERAPY FOR MULTIPLE BRAIN METASTASES

• Published in: Neuro-oncology (Charlottesville, Va.) Vol. 26; no. Supplement_5; p. v88
• Main Authors: Chen, J, Ziemer, B P, Witztum, A, Ni, L, Cui, K L, Boreta, L, Nakamura, J L, Sneed, P K, Morin, O, Capaldi, D P I, Braunstein, S E
• Format: Journal Article
• Language: English
• Published: 17.10.2024
• ISSN: 1522-8517; 1523-5866
• DOI: 10.1093/neuonc/noae144.292

Abstract BACKGROUND In the setting of multiple brain metastases (BM), several factors may impact decision-making for management with whole brain radiotherapy (WBRT) versus (vs) stereotactic radiosurgery (SRS). We aimed to evaluate clinical outcomes among selected patients with ≥10 BM treated with serial SRS vs WBRT. MATERIAL AND METHODS We retrospectively reviewed 150 patients with ≥10 BM treated at first brain RT with either WBRT (3/2016-4/2023) or Gamma Knife-based SRS offered serially at intracranial (CNS) progression (12/2007-4/2023). Patients with leptomeningeal disease (LMD) at first brain RT were excluded. Data were obtained from the electronic medical record or GammaPlan software. Multivariable logistic regression analyses determined predictors of receiving upfront SRS vs WBRT. The Kaplan-Meier method determined overall survival (OS), time from first SRS to WBRT, and time to LMD development. The log-rank test compared differences in OS among patients who received upfront SRS vs WBRT. Multivariable Cox regression models were used to examine predictors of OS and LMD development. RESULTS Median follow up was 12.4 months (IQR: 3.7-29.8). Most patients had lung (40.7%) or breast (28.7%) histology; few (16.0%) had surgical resection of BM. Predictors of receiving upfront SRS (n=73) vs WBRT (n=77) included Karnofsky Performance Status (KPS) ≥90 (OR 6.54, 95% CI 2.49-18.89, p<0.001) and melanoma in comparison to breast histology (OR 14.13, 95% CI 1.48-344.55, p=0.041). Median time from first SRS to WBRT was 10.0 months (IQR: 5.1-22.6). Median time from first brain RT to LMD development (n=35) was 15.3 months (IQR: 8.7-29.9). The 6- and 12-month probabilities of OS from first brain RT were 69.0% and 55.1%, while the 6- and 12-month probabilities of freedom from LMD were 93.2% and 86.3%, respectively. Patients who received upfront serial SRS vs WBRT had improved 6-month OS (94.5% vs 44.3%, p<0.001), with a similar finding in favor of serial SRS on multivariable analysis with covariates including KPS and others (HR 0.25, 95% CI 0.15-0.41, p<0.001). Predictors of worse OS included histology other than breast, lung, or melanoma (HR 2.96, 95% CI 1.64-5.32, p<0.001) and CNS progression ≤6 months (HR 1.64, 95% CI 1.05-2.58, p=0.030), but not first brain RT type (p=0.583). There was a trend for worse OS with extracranial progression ≤6 months (HR 1.51, 95 % CI 0.99-2.29, p=0.053). CNS progression ≤6 months after first brain RT predicted for LMD (HR 5.15, 95% CI 1.95-13.62, p<0.001), while age ≥60 was protective against LMD (HR 0.44, 95% CI 0.20-0.99, p=0.046). CONCLUSION Independent of receiving upfront serial SRS or WBRT, patients with markers of more aggressive tumor biology, including CNS progression ≤6 months and age <60, had increased risk for LMD development. Along with histology and CNS progression ≤6 months, patient selection for serial SRS vs WBRT may impact OS and should be thoughtfully implemented.