Background Patients with recurrent glioblastoma (rGB) have a poor prognosis with a median overall survival (OS) of 30–39 weeks in prospective clinical trials. Intravenous administration of programmed cell death protein 1 and cytotoxic T-lymphocyte-associated antigen 4 inhibitors has low activity in patients with rGB. In this phase I clinical trial, intracerebral (IC) administration of ipilimumab (IPI) and nivolumab (NIVO) in combination with intravenous administration of NIVO was investigated. Methods Within 24 hours following the intravenous administration of a fixed dose (10 mg) of NIVO, patients underwent a maximal safe resection, followed by injection of IPI (10 mg; cohort-1), or IPI (5 mg) plus NIVO (10 mg; cohort-2) in the brain tissue lining the resection cavity. Intravenous administration of NIVO (10 mg) was repeated every 2 weeks (max. five administrations). Next generation sequencing and RNA gene expression profiling was performed on resected tumor tissue. Results Twenty-seven patients were enrolled (cohort-1: n=3; cohort-2: n=24). All patients underwent maximal safe resection and planned IC administrations and preoperative NIVO. Thirteen patients (cohort-1: n=3; cohort-2: n=10) received all five postoperative intravenous doses of NIVO. In cohort-2, 14 patients received a median of 3 (range 1–4) intravenous doses. Subacute postoperative neurological deterioration (n=2) was reversible on steroid treatment; no other central nervous system toxicity was observed. Immune-related adverse events were infrequent and mild. GB recurrence was diagnosed in 26 patients (median progression-free survival (PFS) is 11.7 weeks (range 2–152)); 21 patients have died due to progression. Median OS is 38 weeks (95% CI: 27 to 49) with a 6-month, 1-year, and 2-year OS-rate of, respectively, 74.1% (95% CI: 57 to 90), 40.7% (95% CI: 22 to 59), and 27% (95% CI: 9 to 44). OS compares favorable against a historical cohort (descriptive Log-Rank p>0.003). No significant difference was found with respect to PFS (descriptive Log-Rank test p>0.05). A higher tumor mRNA expression level of B7-H3 was associated with a significantly worse survival (multivariate Cox logistic regression, p>0.029). Conclusion IC administration of NIVO and IPI following maximal safe resection of rGB was feasible, safe, and associated with encouraging OS. Trial registration NCT03233152.

Extent of resection (EOR) is a well-known prognostic factor in patients with newly diagnosed IDH-wildtype glioblastoma. However, reported survival times across resection categories vary between reports, and outcomes of submaximal or supramaximal resection remain less well defined. We conducted a systematic review and meta-analysis on the association between EOR and overall survival (OS) in patients with newly diagnosed IDH-wildtype glioblastoma treated with chemoradiotherapy. Studies were included if OS was reported by EOR category. Risk ratios (RRs) for 1- and 2-year survival were pooled using a random-effects model. Study quality was assessed using the Newcastle-Ottawa Scale. Thirty-one studies involving 26,167 patients were included. Supramaximal resection (SupraMR) was associated with significantly improved 2-year survival compared to maximal CE resection (MR) (RR 0.70, 95 % CI 0.55–0.88). Compared to submaximal resection (subMR), MR was associated with higher 1-year survival (RR 0.59, 95 % CI 0.53–0.67) and 2-year survival (RR 0.82, 95 % CI 0.77–0.87). Biopsy alone was associated with the poorest outcome. Findings remained robust in sensitivity analyses excluding SEER and RTOG cohorts. Increasing EOR seems to be associated with improved survival in newly diagnosed IDH-wildtype glioblastoma. SupraMR offers the greatest benefit, while submaximal resection appears to be more favorable than biopsy. These findings support the prognostic relevance of EOR and underscore the need for prospective studies with standardized resection classifications. The balanced summary of survival data for each resection class provided in this review can serve as a basis for effect estimation and sample size calculations in future trials.

Purpose: Automated glioblastoma segmentation from magnetic resonance imaging is generally performed on a four-modality input, including T1, contrast T1, T2 and FLAIR. We hypothesize that information redundancy is present within these image combinations, which can possibly reduce a model’s performance. Moreover, for clinical applications, the risk of encountering missing data rises as the number of required input modalities increases. Therefore, this study aimed to explore the relevance and influence of the different modalities used for MRI-based glioblastoma segmentation. Methods: After the training of multiple segmentation models based on nnU-Net and SwinUNETR architectures, differing only in their amount and combinations of input modalities, each model was evaluated with regard to segmentation accuracy and epistemic uncertainty. Results: Results show that T1CE-based segmentation (for enhanced tumor and tumor core) and T1CE-FLAIR-based segmentation (for whole tumor and overall segmentation) can reach segmentation accuracies comparable to the full-input version. Notably, the highest segmentation accuracy for nnU-Net was found for a three-input configuration of T1CE-FLAIR-T1, suggesting the confounding effect of redundant input modalities. The SwinUNETR architecture appears to suffer less from this, where said three-input and the full-input model yielded statistically equal results. Conclusion: The T1CE-FLAIR-based model can therefore be considered as a minimal-input alternative to the full-input configuration. Addition of modalities beyond this does not statistically improve and can even deteriorate accuracy, but does lower the segmentation uncertainty.

Purpose: Evidence suggests that MAPK pathway activation, as measured by ERK1/2 phosphorylation (p-ERK), predicts overall survival (OS) in patients with recurrent glioblastoma receiving anti-PD-1 therapy. We aimed to validate these findings in independent cohorts. Experimental Design: In a 24-patient clinical trial on recurrent glioblastoma and high-grade gliomas, we examined the link between p-ERK levels and OS. Patients received intravenous nivolumab, followed by maximal safe resection and an intracerebral injection of either ipilimumab alone or combined with nivolumab. Biweekly adjuvant nivolumab was then administered up to five times (NCT03233152). Using REporting recommendations for tumor MARKER prognostic studies (REMARK) criteria, we conducted independent analyses for p-ERK quantification and statistical evaluations. Additional comparative analysis included prior cohorts, totaling 65 patients. Cox proportional hazards models and meta-analysis were employed to assess p-ERK as a predictive biomarker after immunotherapy. Results: Lower median p-ERK+ cell density was observed compared with prior studies, likely due to variable tissue processing across cohorts. Nonetheless, high p-ERK was associated with prolonged OS, particularly in isocitrate dehydrogenase wild-type glioblastomas (P = 0.036). Median OS for high and low p-ERK patients were 55.6 and 30 weeks, respectively. Multivariable analysis reinforced p-ERK's significance in survival prediction (P = 0.011). Upon p-ERK normalization across cohorts (n = 65), meta-analysis supported the survival benefit of elevated tumor p-ERK levels (P = 0.0424). Conclusions: This study strengthens the role of p-ERK as a predictive biomarker for OS in patients with glioblastoma on immune checkpoint blockade. Future research should focus on further validation in prospective trials and the standardization of preanalytical variables influencing p-ERK quantification.

BACKGROUND AND OBJECTIVES: External ventricular drain (EVD) placement is often performed freehand, a technique subpar to accurate yet impractical image-guided methods, yielding optimal placement in only 70%. The aim of this study was to address shortcomings in EVD placement and image guidance technologies by implementing high-accuracy augmented reality (AR) guidance. METHODS: We conducted a prospective clinical pilot study to assess feasibility, safety, and clinical performance of EVD placement using a standalone AR headset equipped with high-accuracy inside-out infrared tracking and software addressing EVD placement. Placement quality was reported using a newly defined extended modified Kakarla scale, and dichotomized into clinically relevant outcome parameters. Results were compared with a nonconcurrent freehand control group using one-sided Fisher exact tests. RESULTS: Eleven AR-guided EVD placements were performed, achieving functional placement in all cases on the first attempt, vs 7 (64%) in the control group (P = .045); successful placement in 9 (82%) vs 5 (45%); optimal in 8 (73%) vs 3 (27%) (P = .043); suboptimal in 2 (18%) vs 5 (45%); and failed in 0 vs 1 (9%). No AR-guided placements required revision, whereas the freehand group had a 36% reintervention rate (P = .045). Procedure-related complications occurred in 2 AR-guided cases (18%), vs 5 (45%) freehand (all post-reintervention). CONCLUSION: This study presents the first clinical use case of EVD placement using high-accuracy AR guidance contained in a standalone head-worn navigation system. Safe and reliable outcomes using a validated pipeline were demonstrated, eliminating stick-and-poke attempts and resulting in improved quality, increased single attempt success rates, and reduced revision and complication rates. Based on these results, a multicenter randomized controlled trial will be initiated.

Positron emission tomography (PET) with a [18F]fluoroethyl)-L-tyrosine ([18F]FET) tracer is of growing importance in the management of glioblastoma for the estimation of tumor extent and extraction of diagnostic and prognostic parameters. Robust and accurate glioblastoma segmentation methods are essential to maximize the benefits of this imaging modality. Given the importance of setting the foreground threshold during manual tumor delineation, this study investigates the added value of incorporating such prior knowledge to guide the automated segmentation and improve performance. Two segmentation networks were trained based on the nnU-Net guidelines: one with the [18F]FET PET image as sole input, and one with an additional input channel for the threshold map. For the latter, we investigate the benefit of manually obtained thresholds and explore automated prediction and generation of such maps. A fully automated pipeline was constructed by selecting the best performing threshold prediction approach and cascading this with the tumor segmentation model. The proposed two-channel network shows increased performance with guidance of threshold maps originating from the same reader whose ground-truth tumor label the prediction is compared to (DSC = 0.901). When threshold maps were generated by a different reader, performance reverted to levels comparable to the one-channel network and inter-reader variability. The proposed full pipeline achieves results on par with current state of the art (DSC = 0.807). Incorporating a threshold map can significantly improve tumor segmentation performance when it aligns well with the ground-truth label. However, the current inability to reliably reproduce these maps—both manually and automatically—or the ground-truth tumor labels, restricts the achievable accuracy for automated glioblastoma segmentation on [18F]FET PET, highlighting the need for more consistent definitions of such ground-truth delineations.