Improved Outcomes with Adjuvant Stereotactic Radiosurgery Compared with Fractionated Stereotactic Radiation Therapy for Resected Intracranial Metastases

Abstract

Post-operative stereotactic radiosurgery (SRS) is increasingly implemented for patients with resected intracranial metastases, with recent randomized data demonstrating improved cognition and equivalent overall survival compared with adjuvant whole brain radiotherapy (WBRT). Several institutional series have also demonstrated promising local control rates utilizing adjuvant fractionated stereotactic radiotherapy (FSRT) to cavities. Our purpose was to compare our institutional outcomes of linac-based SRS with FSRT for patients with resected brain metastases. We reviewed the medical charts and dosimetric data of 68 patients, treated with either SRS or FSRT to the resection cavity, from February 2002 to May 2017. SRS was delivered in a single fraction of 15 - 21 Gy (BED10 = 37.5 – 65.1 Gy), prescribed to the 60 – 85% isodose line. FSRT was delivered in 5 - 6 fractions of 25 - 36 Gy (BED10 = 37.5 – 57.6 Gy), typically prescribed to the 95% isodose line. Planning target volume was defined on MRI as resection cavity plus 0 - 3 mm. Local control and survival outcomes were estimated using the Kaplan-Meier method and analyzed using the log-rank test and Cox proportional hazards models. Chi-square test was performed to identify factors predicting local failure. Seventy-five intracranial lesions were identified. Median age was 60 years with a median follow-up of 11 months (range: 1-119 months). The majority (92%) of lesions received a gross total resection, with 47% originating from a lung primary. Forty-nine cavities received FSRT, while 26 were treated with linac-based SRS. The median diameter of lesions treated with FSRT and SRS was 2.9 cm and 2.3 cm, respectively (p = 0.3). One-year local control for patients who received SRS compared with FSRT was 91% vs 66%, respectively (HR 5.0 [1.1 - 22.1], p = 0.03). The 1-year local control for patients who received SRS, high dose (≥ 30 Gy) FSRT, and low dose (< 30 Gy) FSRT was 91%, 84%, and 52% (p = 0.007). The majority of recurrences were in-field (88%) for patients treated with FSRT, compared to 50% in-field failure rate for those who received SRS. Resection cavity location (p = 0.9), tumor histology (p = 0.8), and pre-operative tumor size (p = 0.85) did not predict for local failure. The median PTV for cavities that recurred was 22.5 cc compared with 13.0 cc for those that did not (p = 0.16). The 1-year distant brain control with SRS compared with FSRT was 61% vs 41% (p = 0.06). There was no significant difference in overall survival (79% vs 62%, p = 0.2). We observed a significantly higher rate of local control in patients who received adjuvant cavity linac-based SRS compared with FSRT, suggesting that increased biologic effective dose may correspond with improved control for resected brain metastases. These treatments need to be evaluated prospectively to validate the above findings.