2196 - Stereotactic Radiosurgery and Immunotherapy Treatment Timing for Brain Metastases

Purpose/Objective(s):

Brain metastases (BM) are a major driver of cancer-related mortality. Stereotactic radiosurgery (SRS) and immunotherapy (IT) treatment have arisen as standard treatments for BM, but there is no consensus regarding the synergistic efficacy of concurrent treatment (CT) vs. receiving sequential treatment (ST) of upfront SRS followed by IT. Prior studies suggesting a survival benefit with concurrent therapy also may be confounded by immortal time bias, which occurs when patients in one group must survive a certain period to receive or be classified into a treatment, creating a guaranteed survival advantage unrelated to the treatment effect itself. To address the gap in CT vs. ST efficacy, we used the National Cancer Database (NCDB) to provide large-scale, population-level evidence on optimal SRS-IT timing while accounting for this methodological survival bias.

Materials/Methods:

We evaluated the 2025 release of the NCDB for patients who had undergone single fraction SRS and IT as first therapies for the most common types of solid cancers that caused brain metastases. Patients diagnosed from 2016-2021 with complete radiation and systemic therapy information were included. CT was defined as IT treatment within 30 days of SRS. To account for immortal time bias, we required survival of 3 months for all patients. We evaluated all primary tumor types as well as subgroups of non-small cell lung cancer (NSCLC) and melanoma. Survival was estimated via Kaplan-Meier curves and compared via Cox proportional hazards regression.

Results:

Among 2,181 patients with brain metastases treated with SRS and IT, 72.9% (n=1,590) received concurrent therapy (IT administered within ±30 days of SRS) and 27.1% (n=591) received ST. Median follow-up was 19.1 months. For all primary tumor types considered together, 2-year overall survival (OS) was 47.5% with concurrent versus 46.6% with non-concurrent therapy (p=0.75). For NSCLC, 2-year overall survival (OS) was 47.3% for the 1,260 CT patients versus 48.0% for the 485 ST pts (p=0.85). For melanoma, 2-year overall survival (OS) was 60.8% for 148 CT pts versus 52.0% for 30 ST pts (p=0.57).

Conclusion:

After accounting for immortal time bias by requiring 3-month survival for all patients, concurrent SRS-IT had no difference in survival compared to sequential therapy for brain metastases in a non-randomized, observational cohort of patients. This finding was consistent across all primary tumor types and in subgroup analyses of NSCLC and melanoma. These results suggest that the survival benefit previously attributed to concurrent therapy in retrospective studies may be substantially influenced by immortal time bias and the optimal timing of SRS and IT may be more flexible than previously reported.