Background: Glioblastomas (GBM) are typically treated with surgery and radio-chemotherapy, with patient survival often depending on the extent of tumor resection. This study compares outcomes of GBM surgery using 5-ALA, intraoperative neuroelectrophysiology, and neuro-navigation, either in a standard setting (STD) or enhanced by mixed reality (MR) guidance.

Methods: This retrospective study included GBM patients who underwent resection at Geneva University Hospitals between 2015 and mid-2022, excluding biopsies and partial debulking. Primary outcomes included postoperative residual tumor volume (RV) based on postoperative contrast uptake on the MRI, while secondary outcomes were gross total resection (GTR), extent of resection (EOR), new postoperative deficits, overall survival (OS), progression-free survival (PFS), and Karnofsky performance scores. Confounding factors such as intraoperative monitoring and use of fluorescence were analyzed.

Results: Of 115 patients, 76 were in the STD group and 39 in the MR group, with comparable demographics. The MR group had significantly lower RV (median 0.01 cm³ vs. 0.34 cm³, p=0.008) and higher GTR rates (median 50% vs. 26.7%). EOR was also superior in the MR group (median 99.9% vs. 98.2%, p=0.002). New focal deficits occurred in 39% (STD) and 36% (MR) of cases (p=0.84). While median OS was not significantly different (475 vs. 375 days, p=0.63), median PFS was longer in the MR group (147 vs. 100 days, p=0.004).

Conclusion: MR guidance improves the quality of tumor resection and enhances progression-free survival without increasing postoperative deficits, although it does not significantly impact overall survival.

Title	Published in	Year	Views	Downloads

Surgical and clinical impacts of mixed reality-guided glioblastoma resection versus standard neuronavigation: improving tumor surgery	Frontiers in oncology	2025	17	11
Robustness assessment of regressions using cluster analysis typologies: a bootstrap procedure with application in state sequence analysis	BMC Medical research methodology	2024	55	18

Archive ouverte UNIGE

Roth, Léonard