BACKGROUND:Grade 2 isocitrate dehydrogenase (IDH)-mutant (mIDH) glioma imposes significant health-related quality of life (HRQoL) and economic burden. OBJECTIVE:The aim of this study was to explore patient preferences and socioeconomic burden in grade 2 mIDH glioma and its treatment. METHODS:A discrete choice experiment (DCE) was administered to patients across the US, UK, Canada and Germany. Attributes included life expectancy, time until tumour progression, side effects, future risk of challenges performing work/usual activities and treatment modality. DCE data were analysed using a mixed-effects logit model. Relative attribute importance scores and trade-offs between attributes were estimated. RESULTS:A total of 118 participants (56% male; mean age 41.9 years; 67.8% employed) were included in the final analysis. Four of the five attributes were independent drivers of patients' treatment preferences. Participants preferred treatments with additional years of life expectancy and tumour-free progression and were averse to treatments with higher risks of side effects and future challenges performing work/usual activities. Treatment modality did not significantly influence preferences. Based on 10% risk increments, relative attribute importance scores showed life expectancy was the most important driver of choice (26.5%), followed by side effects (21.6%). Side effects (31.9-37.9%) and future issues with work/usual activities (20.6-24.4%) became relatively more important when expressed as 20-30% increments. Patients were willing to trade life expectancy to avoid increased risk of side effects. CONCLUSIONS:Patients placed the greatest value on gains in survival. Avoidance of risks of side effects or impacts to daily activities were also significant predictors of choice, with patients willing to trade life expectancy to avoid risk attributes. Incorporating patient preferences is essential to support shared treatment decision making and ensure treatment strategies align with patients' values and expectations.

The clonal glioma driver mutation IDH1R132H gives rise to a major histocompatibility class II-restricted neoepitope. A multicenter, first-in-human phase 1 trial met its prespecified primary endpoints by demonstrating safety and immunogenicity of an IDH1-R132H peptide vaccine (IDH1-vac) integrated into standard of care in 33 participants with newly diagnosed grade III and IV (World Health Organization classification 2007) IDH1-R132H+ astrocytomas (NOA16). Here we report on the clinical and immunological long-term follow-up of this trial as secondary and translational endpoints. The 8-year progression-free and overall survival (OS) rates were 0.42 (confidence interval (CI): 0.24-0.59) and 0.66 (CI: 0.46-0.79), respectively. For participants with grade IV astrocytoma, median OS was 106.1 months (CI: 39.6-not estimable (NE)), comparing favorably to the published median OS in this population ranging from 31.6-56.4 months. Within the responder group, sustained antibody responses to IDH1-R132H were associated with a favorable long-term clinical course. IDH1-vac-induced T cell responses were detected in the inflamed brain lesion of an IDH1-vac-associated pseudoprogression, whereas no IDH1-vac-induced T cells were found in participants with early progressive disease. The favorable long-term outcome of the NOA16 cohort supports investigating IDH1-vac in persons with newly diagnosed grade 3 and 4 (World Health Organization classification 2021) IDH-mutant astrocytomas in a randomized phase 2 trial (ClinicalTrials.gov identifier: NCT02454634 ).

Abstract Introduction: HPV-negative head and neck squamous cell carcinoma (HNSCC) remains highly lethal, and even the recent KEYNOTE-689 trial showed limited response to PD-1 blockade. To clarify mechanisms of immunoresistance, we performed multi-omic single-cell profiling of HPV-negative HNSCC to define the states, clonotypes, and spatial niches of tumor-reactive T cells. Methods: Two patients with untreated HPV-negative HNSCC underwent surgical tumor sampling. Biopsies were allocated for dissociative single-cell RNA and TCR sequencing, for single-cell spatial whole transcriptome imaging with TCR add-in on the Bruker CosMx platform, and for functional co-culture of autologous tumor cells and tumor-infiltrating lymphocytes (TILs) on the Bruker Beacon system with robotic retrieval of reactive clones followed by single-cell TCR sequencing. Results: Dissociative and spatial datasets were high quality, and dissociative clustering guided spatial annotation. Two tumor-reactive TIL populations emerged: a transcriptionally reactive population marked by terminal exhaustion and clonal hyper-expansion, and a functionally reactive population identified on the Beacon platform through tumor-killing and IFNγ secretion. Neither appeared in adjacent normal mucosa. Spatial projection showed functionally reactive clonotypes progressing along a pseudotime continuum from minimally exhausted stromal states toward a deeply exhausted yet proliferative state within malignant epithelial islands. Transcriptionally reactive TILs shared this exhausted-proliferative phenotype and primarily localized within tumor islands. Conversely, Tregs and immunosuppressive macrophages were confined to stromal regions and showed minimal spatial overlap with tumor-reactive T cells. Malignant epithelium strongly enriched CXCL14 (> ten-fold over stroma), yet tumor-reactive TILs remained rare, with a tumor-to-reactive T cell ratio greater than 20 to 1. Conclusions: This is the first study to integrate dissociative, spatial, and functional single-cell analytics to define tumor-reactive TIL biology in solid tumors. We find that tumor-reactive T cells can infiltrate malignant epithelial islands, and that their deepest infiltration corresponds to the most exhausted states, indicating that chronic antigen stimulation is the primary driver of dysfunction. The confinement of suppressive cells to stromal niches suggests that their influence is exerted through maintenance of a restrictive barrier stroma rather than through proximity-based inhibition. Together, these findings highlight insufficient chemotactic recruitment of tumor-reactive T cells as a central barrier to effective antitumor immunity. The strong epithelial enrichment of CXCL14 positions it as a promising candidate for enhancing tumor-directed T cell migration. Citation Format: Joseph Zenga, Musaddiq Awan, Fumou Sun, Stuart J. Wong, Abdullah Memon, Tyce Kearl, Ean Norenberg, Margaret L. Hoang, Joseph M. Beechem, Eric Hobbs, Ashley Heck, Rachel Liu, Daniel McGuire, Erin Piazza, Peiman Hematti, Heather Himburg. Dissociative, spatial, and functional single-cell profiling reveals antigen-driven exhaustion and chemotactic failure of tumor-reactive T cells in HPV-negative HNSCC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 4984.

Abstract Adult gliomas exhibit substantial molecular and pathological variation. Prior single-cell studies have characterized intratumoral heterogeneity; in glioblastoma, tumor cells adopt lineage-associated programs—such as oligodendrocyte- and neuro- progenitor (OPC, NPC), astrocyte (AC) and mesenchymal (MES) like states. However, how these states recur across other gliomas remains unclear. Given their shared environment, defining conserved versus entity-specific tumor characteristics is essential for understanding glioma heterogeneity. Here, we analyzed 310 tumor cores from 284 patients spanning oligodendroglioma, astrocytoma, glioblastoma, and ependymoma, using single-cell spatial transcriptomics with a 344-gene panel (10x Xenium). This approach yielded a dataset of 2.8 million cells, including 18 tumor microenvironment (TME) cell types. To further probe local interactions, we profiled a validation cohort of astrocytomas using a 5,096-genes panel and multiplexed immunofluorescence (57 proteins). Our data revealed 9 recurrent tumor states across gliomas, organized into neighborhoods shaped by local cellular interactions. These formed structured layers, with OPC-/NPC-like tumors near cortical interfaces and AC-like regions forming a continuum with MES1-gliosis. Neighborhoods were largely subtype-specific suggesting that tissue organization and not cell composition alone, underlies glioma heterogeneity. Consistently, methylation classification—central to subtype definition in gliomas—stratified the cohort into 13 classes with distinct neighborhood profiles, including characteristic architectures in RTK1, RTK2, and MES glioblastomas. Neighborhoods also showed distinct TME variation: P2RY12+ microglia were enriched in AC-like and MES1-gliosis regions, while CD163+ myeloid cells concentrated in MES1-gliosis and MES2-hypoxia niches. These immune patterns corresponded to neighborhood-specific inflammatory programs and at the entity level to myeloid-dominated inflammation in ependymoma and low inflammation in oligodendroglioma. Beyond neighborhood definition, we evaluated their diagnostic potential. In astrocytoma, AC-like enrichment marked favorable prognosis, whereas MES1-like prevalence defined high-risk patients. Given this, we asked whether neighborhoods show histopathological correlates in routine H&E. Morphology embeddings extracted with histopathology foundation models recovered clusters mirroring niche composition, indicating that neighborhoods can be inferred from H&E morphology and incorporated into diagnostic workflows. Collectively, these findings link glioma microanatomy heterogeneity by connecting tumor states to their TME. By capturing spatially dependent programs, our work provides a framework for integrating spatial architecture into biological understanding, diagnosis, and risk stratification. Citation Format: Anna Mathioudaki, Zaira Seferbekova, Samuel P. Rutz, Michael Ritter, Domenico Calafato, Gleb Rukhovich, Felix Hinz, Philipp Mahlknecht, Franziska M. Ippen, Ekaterina Popova, Sophia Schinkewitsch, Nela Runa Koeberer, Nina Wilhelm, Pooja Sant, Jan-Philipp Malm, Sascha Dietrich, Christel Herold-Mende, Nima Etminan, Antje Wick, Sandro Krieg, Michael Platten, Andreas von Deimling, Felix Sahm, Abigail Suwala, Moritz Gerstung. Spatial tissue architecture as a unifying principle of tumor and microenvironmental states in adult gliomas [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 4953.

Advances in molecular understanding and diagnostic precision of glioblastoma enable the identification of key genetic alterations in a timely manner and, in principle, allow treatments with targeted compounds based on molecular markers. Here we report the results of the phase 1/2 umbrella trial NCT Neuro Master Match (N2M2), which evaluated targeted treatments in 228 patients with newly diagnosed glioblastoma without O6-methylguanine DNA-methyltransferase promoter hypermethylation. Stratification for treatment was conducted by a trial-specific molecular tumor board across five subtrials, each evaluating a targeted therapy-alectinib, idasanutlin, palbociclib, vismodegib or temsirolimus-selected according to the best-matching molecular alteration. Patients without matching alterations were randomized between subtrials without strong biomarkers using atezolizumab and asunercept, and the standard of care (SOC), temozolomide. All received radiotherapy. The primary endpoints were dose-limiting toxicities (phase 1) and progression-free survival at 6 months (PFS-6; phase 2). Secondary endpoints included safety and tolerability, as well as overall survival (OS). The subtrials for alectinib and vismodegib did not open as they did not have matching patients. The idasanutlin subtrial (n = 9) was terminated early at the discretion of the manufacturing company. The temsirolimus subtrial (n = 46) demonstrated a PFS-6 of 39.1% and median OS of 15.4 months in patients with activated mammalian target of rapamycin (mTOR) signaling compared to a PFS-6 at 18.5% in the SOC group (n = 54), meeting the primary endpoint. The atezolizumab (n = 42), asunercept (n = 26) and palbociclib (n = 41) subtrials did not meet the primary endpoint for efficacy. The safety signals of N2M2 match prior experiences with the drugs in quality and quantity; no relevant negative interaction with the parallel radiotherapy was noted. The results of the N2M2 trial support further investigation of temsirolimus in addition to radiotherapy in patients with newly diagnosed glioblastoma with activated mTOR signaling. ClinicalTrials.gov registration: NCT03158389.

Supplementary Table 1: Representativeness of study participants.

Supplementary Figure 3: Characteristics of RTK1 tumors in comparison to RTK2 and MES. A Distribution of MGMT promotor methylation in main glioblastoma methylation subgroups. Classifier assignment is based on the v12.5 version (n = 295). Mutational profiles in B RTK1 tumors (n = 89), C RTK2 tumors (n = 140) and D MES tumors (n = 66).

BACKGROUND:Meningiomas represent the most common primary intracranial tumors in adults, with World Health Organization (WHO) grade 1 typically associated with favorable outcomes following gross total resection (GTR). METHODS:This retrospective study included patients with CNS WHO grade 1 meningioma and available DNA methylation profiles (n = 210). Clinical tumor characteristics and treatment course (eg, surgical resection, extent of resection, radiotherapy [RT]) were evaluated. Integrated Scores (InS) were calculated based on methylation family using the DKFZ brain tumor classifier, CNS WHO grading, and chromosomal losses, categorized as low, intermediate, or high. Survival analyses employed Kaplan-Meier and Cox regression methods, with local PFS defined as the primary endpoint. RESULTS:In newly diagnosed cases, GTR was associated with a 93.0% 3-year progression-free survival (PFS), compared to 69.3% following subtotal resection (STR). Stratification by IntS showed that patients in the IntS-low group had superior outcomes: 3-year PFS of 93.4 after GTR and 77.4% after STR. In contrast, patients with IntS-intermediate/high profiles showed significantly worse outcomes, with PFS of 85.9% after GTR and 40.0% after STR. Following tumor recurrence, particularly those with IntS-intermediate/high, postoperative RT after STR may improve 3-year PFS to 88.9%, compared to much lower PFS rates in newly diagnosed cases managed without adjuvant RT after STR (3-year PFS: 40.0%). CONCLUSIONS:Our findings highlight the combined impact of both the extent of resection and molecular risk profile on prognosis in newly diagnosed cases. While conservative management is feasible in lower-risk primary cases, recurrent or higher-risk patients may benefit from early postoperative RT.

Lisavanbulin is a prodrug of the microtubule-targeting agent avanbulin. Both avanbulin and lisavanbulin have demonstrated significant antitumor activity in several preclinical tumor models including glioblastoma. Previous human studies demonstrated that 48-h infusions of intravenous lisavanbulin were well tolerated with preliminary activity in recurrent glioblastoma. The current phase 1/2a study evaluates the safety and tolerability of once-daily oral lisavanbulin in patients with solid tumors or recurrent glioblastoma or high-grade glioma. Lisavanbulin is associated with profound, durable responses in a subset of patients with recurrent refractory grade 4 astrocytoma or glioblastoma. We present here the clinical and translational results from this trial, including a description of a response-predictive molecular signature that warrants further exploration in these tumor types of significant unmet need. The study is registered at ClinicalTrials.gov (NCT02490800).

EO2401, composed of peptide sequences from gut bacteria that cross-react with glioblastoma tumor-associated antigens (IL13Rα2, BIRC5/survivin, and FOXM1) plus a helper peptide, was administered to HLA-A2 positive patients with first progression of a glioblastoma requiring ≤ 2 mg dexamethasone/day. Mature data from cohorts-2a/2b (EO2401/nivolumab, n=44) and 3 (EO2401/nivolumab/bevacizumab, n=26) are provided. Ex vivo tetramer-staining on PBMCs demonstrated expansion of CD8 T-cells specific for EO2401 TAAs in 84% of tested patients for cohort-2a/2b and in 92% for cohort-3. Adverse events were limited to those expected for nivolumab and bevacizumab, as well as local site reactions to EO2401 (96% events grade 1-2). Improved outcomes were observed for cohort-3 (median follow-up 33.5 months) versus 2a/2b (median follow-up 25.0 months) including: time of study therapy (4.8 vs 2.3 months), median PFS (4.8 vs 1.9 months), ORR (38% vs 14%) and median OS (12.1 vs 10.8 months). Median OS for patients who underwent second surgery prior to EO2401 start vs no surgery were 18.5 vs 8.5 months for cohort-3 and 11.1 vs 10.2 months for cohort-2a/2b. Multivariable Cox regression analyses revealed that OS for cohort-3, but not cohort 2a/2b was associated with second surgery (HR=0.08, 95%-CI 0.01–0.75, p=0.027), gender (HR=10.2, 95% CI 2.17–48.3, p=0.003) and baseline dexamethasone use (HR=9.35, 95%-CI 2.03–43.1, p=0.004) whereas MGMT methylation status and tumor size was not. For external validation, application of the same model to the CM-143 dataset (1st glioblastoma progression randomized to nivolumab [n=83] vs bevacizumab [n=77]) revealed no associations with OS for bevacizumab while MGMT status (HR=1.95, 95%-CI 1.12–3.17, p=0.007) and baseline corticosteroid use (HR=2.30, 95%-CI 1.42–3.71, p=0.001) associated with OS for nivolumab. Our data indicate that a randomized trial further evaluating EO2401 is warranted and also suggest that surgery prior to triplet therapy, along with baseline dexamethasone use, should be further investigated.

Supplementary Table 2: Characteristics of patients with RTK1 tumors compared to RTK2 and MES according to classifier version v12.5. P-value was calculated for comparison between RTK1 and both other groups (RTK2 and MES) combined. NA, not available.

Background:This first-in-human study evaluated EGFRvIII × CD3 TCB, a novel T cell bispecific antibody, in patients with newly diagnosed EGFRvIII-positive glioblastoma. Methods:Patients with newly diagnosed glioblastoma received escalating doses of EGFRvIII × CD3 TCB following chemoradiation. The primary objectives were to evaluate safety/tolerability and define the maximum tolerated dose (MTD); secondary objectives included pharmacokinetics (PK), immunogenicity, pharmacodynamics, and clinical activity. Results:Thirty-six patients were enrolled, 32 with unmethylated and 4 with methylated MGMT promoter. EGFRvIII × CD3 TCB doses ranged from 0.004 to 10 mg Q3W, administered either on a flat or step-up dose schedule. One DLT occurred (grade 3 seizure). The MTD was not reached. Most adverse events (AEs) were of grade 1-2 severity, with headache being the most common treatment-related AE (22%). EGFRvIII × CD3 TCB showed dose-proportional PK in serum and cerebrospinal fluid (CSF), with a CSF/serum ratio of 0.08. At the highest dose tested, 10 mg Q3W, maximum serum concentrations remained 6-fold below the lower boundary of the predicted anticipated therapeutic dose. Conclusions:The administration of EGFRvIII × CD3 TCB in a maintenance setting, following standard of care treatment, was safe and well tolerated up to the highest tested dose of 10 mg Q3W. However, evidence of efficacy was not observed at the evaluated doses, suggesting that a study of higher dose levels may be warranted.

Supplementary Figure 1: Overview of the clinical trial and biomarker analysis. Out of 596 patients in the EORTC-26101 trial, patient tumor tissue to perform DNA methylation (Meth) and NGS sequencing analysis was available from 380 patients (63.8%). Of these, 99 patients treated with lomustine and 189 patients treated with lomustine and bevacizumab were eligible to detect biomarkers for bevacizumab response.

Supplementary Figure 6: Overall Survival according to PTEN mutation status in the group of IDH-wildtype glioblastomas (n = 362).

Background Diffuse hemispheric glioma, histone 3 (H3) G34-mutant, has been newly defined in the 2021 World Health Organization (WHO) classification of central nervous system tumors. Here we sought to define the prognostic roles of clinical, neuroimaging, pathological, and molecular features of these tumors.Methods We retrospectively assembled a cohort of 114 patients (median age 22 years) with diffuse hemispheric glioma, H3 G34-mutant, central nervous system WHO grade 4, and profiled the imaging, histological, and molecular landscape of their tumors.Results Compared with glioblastoma, H3 G34-mutant diffuse hemispheric gliomas exhibited less avid contrast enhancement, necrosis, and edema on MRI. Comprehensive analyses of mutational and DNA copy number profiles revealed recurrent mutations in TP53 and ATRX, homozygous deletions of CDKN2A/B, and amplifications of PDGFRA, EGFR, CCND2, and MYCN. MGMT promoter methylation was detected in 79 tumors (75%); 11 tumors (13%) showed DNA copy number profiles suggestive of circumscribed deletions on 10q26.3 involving the MGMT locus. Median survival was 21.5 months. Female sex, gross total resection, and MGMT promoter methylation were positive prognostic factors on univariate analysis. Among radiological, pathological, and molecular features, the absence of pial invasion and the presence of microvascular proliferation and CDK6 amplification were positive prognostic factors on univariate analyses.Conclusions This study refines the clinical and molecular landscape of H3 G34-mutant diffuse hemispheric gliomas. Dedicated trials for this novel tumor type are urgently needed.

Supplementary Figure 4: Multivariate analysis. Baselines were MGMT = “methylated” and Classifier = “GBM MES”. Classifier assignment is based on the v12.5 version (n = 380).

Supplementary Figure 7: Exceptional survivors of the EORTC-26101 study (n = 78).

Supplementary Figure 2: Overall (A) and progression-free (B) survival in the EORTC-26101 clinical study (n = 596) and EORTC-26101 biomarker cohort (n = 380).