Resection plus stereotactic radiosurgery versus stereotactic radiosurgery alone and control of brain...

Seizures cause significant morbidity in patients with brain metastasis (BrM). Local therapies for BrM, including resection with adjuvant stereotactic radiosurgery and stereotactic radiosurgery (SRS) alone, have undefined seizure outcomes. The authors sought to compare seizure control in patients with BrM-induced seizures treated with either modality.

Patients who received resection surgery plus adjuvant SRS (S+SRS) or SRS for BrM at a National Cancer Institute (NCI)-designated Comprehensive Cancer Center between January 2015 and December 2023 were retrospectively reviewed. Patients with pretreatment seizure and semiology attributable to an untreated metastasis were included. The cumulative incidence rates of first posttreatment seizure were estimated for both treatment groups. Multivariable analyses identified risk of posttreatment seizure under subdistribution hazards modeling in the competing risk setting to account for death. The authors secondarily assessed overall survival (OS) and local and distal progression rates between treatments.

Two hundred fifty-five of 5284 patients treated for BrM had pretreatment seizures, and 190 patients met the inclusion criteria: 76 (40%) treated with S+SRS and 114 (60%) treated with SRS. One hundred eighty-five (97.4%) patients were taking an antiseizure medication at the end of treatment. The 6-month rate (95% CI) of seizure freedom was 86.71% (77.97%-93.23%) for patients treated with S+SRS and 69.18% (60.54%-77.46%) for patients treated with SRS (p = 0.003). The 12-month rates of seizure freedom were 74.40% (64.01%-83.74%) and 65.59% (56.80%-74.24%) for the S+SRS and SRS groups, respectively (p = 0.091). In the multivariable models, S+SRS was associated with 56% and 27% reductions in seizure risk relative to SRS at 6 and 12 months, respectively, though this was only statistically significant at 6 months. Median OS (95% CI) was 2.62 (1.52-6.52) years for S+SRS patients and 0.83 (0.62-1.54) years for SRS patients (p < 0.001). Treatments did not differ in terms of the rates of local failure (p = 0.52) or distal intracranial progression (p = 0.26) among patients with radiographic follow-up. In subgroups stratified by maximum tumor diameter, patients treated with S+SRS had a lower 6-month cumulative incidence rate of recurrent seizure than patients treated with SRS, though this was only statistically significant for patients with tumors > 3 cm.

Patients treated with S+SRS experienced fewer posttreatment seizures at 6 months than patients treated with SRS, despite larger index lesions in patients treated with S+SRS and prevalent antiseizure medication use in the entire study population. Although multivariable models showed no significant associations at 12 months or beyond, the role of resection in controlling BrM-induced seizures warrants further investigation.