Clinical Factors Associated With 30-Day Mortality Among Patients Undergoing Radiation Therapy for Brain Metastases

Divya Natesan; David J Carpenter; Will Giles; Taofik Oyekunle; Donna Niedzwiecki; Zachary J Reitman; John P Kirkpatrick; Scott R Floyd

doi:10.1016/j.adro.2023.101211

Clinical Factors Associated With 30-Day Mortality Among Patients Undergoing Radiation Therapy for Brain Metastases

Adv Radiat Oncol. 2023 Mar 8;8(4):101211. doi: 10.1016/j.adro.2023.101211. eCollection 2023 Jul-Aug.

Authors

Divya Natesan¹, David J Carpenter¹, Will Giles¹, Taofik Oyekunle², Donna Niedzwiecki², Zachary J Reitman¹, John P Kirkpatrick¹, Scott R Floyd¹

Affiliations

¹ Departments of Radiation Oncology.
² Biostatistics and Bioinformatics, Duke University Medical Center, Durham, North Carolina.

Abstract

Purpose: Existing brain metastasis prognostic models do not identify patients at risk of very poor survival after radiation therapy (RT). Identifying patient and disease risk factors for 30-day mortality (30-DM) after RT may help identify patients who would not benefit from RT.

Methods and materials: All patients who received stereotactic radiosurgery (SRS) or whole-brain RT (WBRT) for brain metastases from January 1, 2017, to September 30, 2020, at a single tertiary care center were included. Variables regarding demographics, systemic and intracranial disease characteristics, symptoms, RT, palliative care, and death were recorded. Thirty-day mortality was defined as death within 30 days of RT completion. The Kaplan-Meier method was used to estimate median overall survival. Univariate and multivariable logistic regression models were used to assess associations between demographic, tumor, and treatment factors and 30-DM.

Results: A total of 636 patients with brain metastases were treated with either WBRT (n = 117) or SRS (n = 519). The most common primary disease types were non-small cell lung (46.7%) and breast (19.8%) cancer. Median survival time was 6 months (95% CI, 5-7 months). Of the 636 patients, 75 (11.7%) died within 30 days of RT. On multivariable analysis, progressive intrathoracic disease (hazard ratio [HR], 4.67; 95% CI, 2.06-10.60; P = .002), progressive liver and/or adrenal metastases (HR, 2.20; 95% CI, 1.16-3.68; P = .02), and inpatient status (HR, 4.51; 95% CI, 1.78-11.42; P = .002) were associated with dying within 30 days of RT. A higher Karnofsky Performance Status (KPS) score (HR, 0.95; 95% CI, 0.93-0.97; P < .001), synchronous brain metastases at time of initial diagnosis (HR, 0.45; 95% CI, 0.21-0.96; P = .04), and outpatient palliative care utilization (HR, 0.45; 95% CI, 0.20-1.00; P = .05) were associated with surviving more than 30 days after RT.

Conclusions: Multiple factors including a lower KPS, progressive intrathoracic disease, progressive liver and/or adrenal metastases, and inpatient status were associated with 30-DM after RT. A higher KPS, brain metastases at initial diagnosis, and outpatient palliative care utilization were associated with survival beyond 30 days. These data may aid in identifying which patients may benefit from brain metastasis-directed RT.