Clinical and Radiographic Outcomes From Repeat Whole-brain Radiation Therapy for Brain Metastases in the Age of Stereotactic Radiosurgery Susan Guo, MD,* Ehsan H. Balagamwala, MD,* Chandana Reddy, MS,* Paul Elson, ScD,w John H. Suh, MD,*z and Samuel T. Chao, MD*z Objectives: Repeating whole-brain radiation therapy (WBRT) in patients with progressive/recurrent brain metastases is controversial. We retrospectively reviewed our experience of repeat WBRT in an era where stereotactic radiosurgery was also available. Methods: In our IRB-approved database, 49 patients received repeat WBRT from 1996 to 2011. Median initial dose of WBRT was 30 Gy in 10 fractions (range, 27 to 37.5 Gy); median reirradiation dose was 20 Gy in 10 fractions (range, 14 to 30 Gy). Median Karnofsky per- formance status (KPS) at reirradiation was 70 (range, 40 to 90). Median number of discrete lesions at reirradiation was 6 (range, 1 to 30). Median interval between initial diagnosis of brain metastases and relapse requiring repeat WBRT was 11.5 months (range, 1.5 to 49.2 mo). Overall survival and relapse-free survival were summarized using the Kaplan-Meier method. The log-rank test was used to com- pare outcomes between groups. Results: Ninety percent of patients completed repeat WBRT. Median survival after repeat WBRT was 3 months (95% CI, 1.9-4.0). Thirteen patients had improved neurological symptoms (27%), 12 were sta- ble (24%), and 14 had worsening symptoms (29%). On radiographic follow-up of 22 patients, 10 (46%) were improved, 4 (18%) were stable, and 8 (36%) progressed. Improved neurological symptoms after repeat WBRT and higher KPS at first follow-up were associated with improved survival (P = 0.05 and 0.02). Conclusions: Repeat WBRT was well tolerated. Modest survival times are seen. Prognostic factors for survival include improved neurological symptoms after repeat WBRT and higher KPS at first follow-up. Repeat WBRT may be useful to improve neurological symptoms in patients with limited treatment options, especially those who are not appropriate stereotactic radiosurgery candidates. Key Words: brain metastases, reirradiation, whole-brain radiation therapy, retreatment, salvage treatment (Am J Clin Oncol 2016;39:288–293) B rain metastases are diagnosed in 20% to 40% of cancer patients, with rising incidence due to improving detection and treatment of systemic malignancy. 1–3 Historically, brain metastases portended a rapidly fatal course in cancer patients. As systemic treatments have advanced, there have been reports of long-term survivors with brain metastases. One series of 1300 patients showed that 2.5% of patients survived Z5 years, and 15 of these 32 patients had recurrence of local or distant brain disease. 4 Randomized data have shown that the combination of surgery and whole-brain radiation therapy (WBRT) improves outcomes over either modality alone. 5,6 Stereotactic radio- surgery (SRS) is being used more frequently due to reports of potentially improved local control when combined with WBRT and potentially decreased neurocognitive toxicities compared with WBRT. 7–9 The treatment paradigm is unclear for patients who have relapsed brain metastases after prior WBRT. Options include surgery, SRS, repeat WBRT, or palliative care. SRS is fre- quently used for salvage therapy; however, whether patients with high volume of brain metastases and poor performance status truly benefit from SRS has not been well documented. Repeat WBRT is not frequently used due to concerns of potential toxicities. 9 We retrospectively reviewed our single institutional experience of repeat whole-brain irradiation in an era where SRS was also available for treatment of brain metastases. MATERIALS AND METHODS Using our IRB-approved database of brain metastases patients, we identified 49 consecutive patients who received repeat WBRT for progressive or recurrent brain metastases from 1996 to 2011 at our main campus and satellite facilities. Only patients who underwent 2 courses of radiation with tra- ditional WBRT fields were included; patients who underwent reirradiation with limited fields not encompassing the whole brain were excluded from our analysis. We also excluded patients who received prior WBRT at other institutions. WBRT was performed using opposed laterals on a linear accelerator with 6 MV photons. Clinical information was extracted from the paper and electronic medical records in this retrospective review. The date of death was obtained from the medical record; when this was not available, the date of death was obtained from the Social Security Death Index. We defined clinical response as follows: complete response refers to disappearance of neurological symptoms, partial response refers to alleviation of neurological symptoms, stable refers to no change in neurological symptoms, pro- gressive disease refers to worsening of neurological symptoms, and asymptomatic refers to lack of neurological symptoms at the beginning and end of treatment. Clinical and radiographic evaluation did not account for differences in steroid use and dosing. Patients were categorized as improved (complete or partial response), stable, or progressive after repeat WBRT. Radiographic response was defined as follows: complete response refers to disappearance of radiographic brain From the Departments of *Radiation Oncology; wQuantitative Health Sciences, Cleveland Clinic; and zRose Ella Burkhardt Brain Tumor and Neuro-Oncology Center, Cleveland, OH. The authors declare no conflicts of interest. Reprints: Samuel T. Chao, MD, Department of Radiation Oncology, Cleveland Clinic, 9500 Euclid Ave, Desk T28, Cleveland, OH 44195. E-mail: chaos@ccf.org. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0277-3732/16/3903-0288 DOI: 10.1097/COC.0000000000000051 ORIGINAL ARTICLE 288 | www.amjclinicaloncology.com American Journal of Clinical Oncology  Volume 39, Number 3, June 2016 Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.