Stereotactic Radiosurgery Versus Reoperation in Small Surgically Accessible Recurrent Glioblastoma

Antonio Dono; Pavel Pichardo-Rojas; Sigmund Hsu; Jay-Jiguang Zhu; Sunil Krishnan; Mark Amsbaugh; Angel I Blanco; Nitin Tandon; Yoshua Esquenazi

doi:10.1227/neu.0000000000004091

Stereotactic Radiosurgery Versus Reoperation in Small Surgically Accessible Recurrent Glioblastoma

Neurosurgery. 2026 May 27. doi: 10.1227/neu.0000000000004091. Online ahead of print.

Authors

Antonio Dono¹, Pavel Pichardo-Rojas¹, Sigmund Hsu^{1

2}, Jay-Jiguang Zhu^{1

2}, Sunil Krishnan^{1

2}, Mark Amsbaugh^{1

2}, Angel I Blanco^{1

2}, Nitin Tandon^{1

2}, Yoshua Esquenazi^{1

2

3}

Affiliations

¹ Vivian L. Smith Department of Neurosurgery, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, Texas, USA.
² Memorial Hermann Hospital-TMC, Houston, Texas, USA.
³ Center for Precision Health, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, Texas, USA.

PMID: 42200661
DOI: 10.1227/neu.0000000000004091

Abstract

Background and objectives: Even with multimodal therapy, glioblastoma invariably recurs. Reirradiation with stereotactic radiosurgery (SRS) and reoperation are frequent salvage treatment options for recurrent glioblastoma (rGBM) isocitrate dehydrogenase-wildtype. No study has compared the safety and efficacy of these treatments in a homogeneous rGBM population. In this study, we evaluate gamma knife (GK)-SRS vs reoperation in rGBM.

Methods: This retrospective study evaluated surgically accessible rGBM between 2005 and 2022. All patients received adjuvant radiotherapy and were evaluated for GK-SRS or reoperation on recurrence. Cox multivariable analysis and propensity-score matching were performed to address confounders in outcomes. Post-recurrence survival (PRS) was the primary endpoint.

Results: We identified 119 patients. Among these, 38 underwent GK-SRS and 81 underwent reoperation. Although patients undergoing GK-SRS had improved PRS and overall survival compared with reoperation, they had significant differences in baseline characteristics, particularly in tumor volume (reoperation group: 8.7 cm3 [IQR = 3.4-17.8 cm3] vs GK-SRS 1.7 cm3 [IQR = 0.2-6 cm3], P < .001). Given these significant differences, a propensity-score matching accounting for tumor volume and Karnofsky performance status was performed, comparing patients in which equipoise between GK-SRS and resection existed (GK-SRS 2.5 cm3 vs reoperation 2.6 cm3 and Karnofsky performance status 80 for both). Among these patients (23 in each group), PRS (19.2 months vs 15.1 months, P = .617) and overall survival (31.9 months vs 25.5 months, P = .176) were similar. Nevertheless, complications remained higher in the reoperation group (26.1% vs 4.3%, P = .040).

Conclusion: In patients with small, surgically accessible first recurrent glioblastoma, GK-SRS achieved survival outcomes comparable to reoperation after matching for key baseline differences, while demonstrating a lower complication rate. These findings support GK-SRS as a reasonable local salvage option in carefully selected low-volume recurrences.

Keywords: Glioblastoma; IDH-WT; Recurrence treatment; Reoperation; SRS.

Grants and funding

Dr Marnie Rose Foundation