Clinicogenomic predictors of survival and intracranial progression after stereotactic radiosurgery for colorectal cancer brain metastases
Chengcheng Gui,Henry S Walch,Kirin D Mueller,Lillian A Boe,A Turan Ilica,James Strong,Jordan E Eichholz,Kenny K H Yu,Jessica A Wilcox,Paolo Manca,Yao Yu,Yoshiya Yamada,Brandon S Imber,Steven B Maron,Michael B Foote,Rona Yaeger,Nikolaus Schultz,Luke R G Pike
DOI: https://doi.org/10.3171/2024.5.JNS24534
2024-08-30
Abstract:Objective: Brain metastases (BM) from colorectal cancer (CRC) are associated with dismal prognosis. When BM-directed therapy is considered, better methods are needed to identify patients at risk of poor oncological outcomes in order to optimize patient selection for closer surveillance or escalated therapy. The authors sought to identify clinicogenomic predictors of survival and intracranial disease progression after CRC BM have been treated with stereotactic radiosurgery (SRS). Methods: Patients with newly diagnosed CRC BM treated with SRS between 2009 and 2022 who had next-generation genomic sequencing data available were included. Frameless SRS was delivered in 1-5 fractions, alone or after neurosurgical resection. Outcomes included overall survival (OS) and intracranial progression (IP), evaluated per patient treated with SRS, and local progression (LP), evaluated per BM. Associations between baseline clinicogenomic features and outcomes were evaluated with Cox regression and competing risk regression, with death as a competing risk. Results: This analysis included 123 patients with 299 BM. At BM diagnosis, 111 patients (90%) had progressive extracranial disease, and 79 patients (64%) had ≥ 3 sites of extracranial metastasis. The median (IQR) number of BM was 2 (1-3) per patient. The median (IQR) biologically effective dose (BED) was 51.3 (51.3-65.1) Gy, corresponding to a prescription of 27 Gy in 3 fractions. OS, IP, and LP estimates at 1 year after SRS were 36%, 55%, and 12%, respectively. OS was independently associated with progressive extracranial disease (HR 4.26, 95% CI 1.63-11.2, p = 0.003) and ≥ 3 extracranial metastatic sites (HR 1.84, 95% CI 1.12-3.01, p = 0.02). LP was less likely when BM received BED ≥ 51.3 Gy (HR 0.24, 95% CI 0.07-0.78, p = 0.02), independent of BM diameter (HR 1.21/cm, 95% CI 0.8-1.84, p = 0.4). IP was independently associated with genomic alterations; TP53 driver alterations were associated with higher risk of IP (HR 2.71, 95% CI 1.26-5.79, p = 0.01), whereas MYC pathway alterations were associated with lower risk (HR 0.15, 95% CI 0.03-0.68, p = 0.01). Conclusions: The authors identified clinicogenomic features associated with adverse outcomes after SRS for CRC BM. Progressive and extensive extracranial metastases predicted worse OS. Insufficient SRS doses predicted greater risk of LP. Wild-type TP53 and alterations in the MYC pathway were independently associated with lower risk of IP. Patients at high risk of IP may be considered for closer surveillance or escalated therapy.