Stereotactic reirradiation for local failure of brain metastases following previous radiosurgery: Systematic review and meta-analysis

Mauro Loi*, Saverio Caini, Silvia Scoccianti, Pierluigi Bonomo, Kim De Vries, Giulio Francolini, Gabriele Simontacchi, Daniela Greto, Isacco Desideri, Icro Meattini, Joost Nuyttens, Lorenzo Livi

*Corresponding author for this work

Research output: Contribution to journalReview articleAcademicpeer-review

8 Citations (Scopus)

Abstract

Introduction: Local failure (LF) following stereotactic radiosurgery (SRS) of brain metastases (BM) may be treated with a second course of SRS (SRS2), though this procedure may increase the risk of symptomatic radionecrosis (RN). Methods: A literature search was conducted according to PRISMA to identify studies reporting LF, overall survival (OS) and RN rates following SRS2. Meta-analysis was performed to identify predictors of RN. Results: Analysis included 11 studies (335 patients,389 metastases). Pooled 1-year LF was 24 %(CI95 % 19–30 %): heterogeneity was acceptable (I2 = 21.4 %). Median pooled OS was 14 months (Confidence Interval 95 %, CI95 % 8.8–22.0 months). Cumulative crude RN rate was 13 % (95 %CI 8 %–19 %), with acceptable heterogeneity (I2 = 40.3 %). Subgroup analysis showed higher RN incidence in studies with median patient age ≥59 years (13 % [95 %CI 8 %–19 %] vs 7 %[95 %CI 3 %–12 %], p = 0.004) and lower incidence following prior Whole Brain Radiotherapy (WBRT, 19 %[95 %CI 13 %–25 %] vs 7%[95 %CI 3 %–13 %], p = 0.004). Conclusions: SRS2 is an effective strategy for in-site recurrence of BM previously treated with SRS.

Original languageEnglish
Article number103043
JournalCritical Reviews in Oncology/Hematology
Volume153
DOIs
Publication statusPublished - Sep 2020

Bibliographical note

Publisher Copyright:
© 2020 Elsevier B.V.

Fingerprint

Dive into the research topics of 'Stereotactic reirradiation for local failure of brain metastases following previous radiosurgery: Systematic review and meta-analysis'. Together they form a unique fingerprint.

Cite this