Predictors of Adverse Radiation Effect in Brain Metastasis Patients Treated With Stereotactic Radiosurgery and Immune Checkpoint Inhibitor Therapy.

PURPOSE: Immune checkpoint inhibitor (ICI) therapy has recently been found to improve survival in patients with a number of cancers, including those with metastatic disease. There is an association of adverse radiation effect (ARE) in patients with brain metastases who have been treated with stereotactic radiosurgery (SRS) and ICIs. METHODS AND MATERIALS: Single-institution retrospective review identified 1118 brain metastases treated with SRS between 2013 and 2018 that had received ICI therapy and 886 metastases that did not receive ICI. Toxicity grading was done via the Common Terminology Criteria for Adverse Events v4.0 grading criteria. Cumulative incidence of ARE was estimated using competing risks methodology; univariate and multivariable regression models were generated to estimate the subdistribution hazard (sHR) of ARE. RESULTS: Two-year cumulative incidence of ARE was 4.5% and 2.1% in patients treated with and without ICI, respectively (Gray's P = .004). Of the 52 metastases exhibiting ARE during the follow-up period, ARE severity by Common Terminology Criteria for Adverse Events v4 was grade 1 in 14 patients, grade 2 in 15, grade 3 in 9, and grade 4 in 14. There were no grade 5 events. Factors associated with an increased sHR of ARE on univariate analysis included ICI, metastasis volume, SRS dose, prescription isodose line, cavity-directed SRS, and V12. Multivariable analysis revealed prescription isodose line (sHR 0.95, P < .01) and ICI (sHR 2.58, P < .01) as significant predictors of ARE. Increasing V12 was associated with a rapidly increasing risk of adverse radiation effect in patients who received ICI. CONCLUSIONS: Our findings suggest that patients receiving ICI have an increased risk of ARE after radiosurgery for brain metastases, with large metastases being at particularly high risk.

Linear accelerator-based radiosurgery is associated with lower incidence of radionecrosis compared with gamma knife for treatment of multiple brain metastases.

BACKGROUND: Gamma knife (GK) and linear accelerator (LINAC)-based stereotactic radiosurgery (SRS) both offer excellent local control in the management of multiple brain metastases. The efficacy and toxicity of LINAC and GK SRS have not been directly compared in the modern era. We studied outcomes in patients treated with LINAC SRS and GK at two separate institutions. METHODS: We identified patients treated with either LINAC or GK who were treated to ≥2 lesions and had available follow up. LINAC patients were treated using single-isocenter multitarget technique. We used Cox regression, Fine and Gray competing risks regression, and nearest neighbor propensity score matching to account for confounders and imbalance between cohorts. Kaplan-Meier curves were used to estimate overall survival and rates of radionecrosis. RESULTS: We identified 391 patients who were treated in 537 courses to a total 2699 lesions (LINAC: 1014, GK: 1685). After propensity score matching, GK was associated with similar overall survival (HR = 0.86; 95% CI 0.59-1.24; p = 0.41) and higher rate of radionecrosis (HR = 3.83; 95% CI 1.66-8.84; p = 0.002) compared to LINAC. In a secondary propensity score matched analysis comparing radionecrosis in single-fraction LINAC and GK, GK remained associated with higher incidence of radionecrosis (HR = 4.42; 95% CI 1.28-15.29; p = 0.019). CONCLUSIONS: In this multi-institutional study, we found similar overall survival with lower incidence of radionecrosis in patients treated with LINAC compared to GK SRS. These findings are hypothesis generating and should be validated in an independent cohort.