Christian Elias Badr

The human genome is pervasively transcribed into protein-coding and regulatory non-coding RNAs whose functions are coordinated within higher-order nuclear architectures. However, direct mapping of RNA–RNA and DNA–DNA interaction networks in complex human tissues at single-cell resolution has remained a major challenge. Here, we present SCIENCE-seq, a multimodal single-cell interactomics platform enabling simultaneous detection of RNA–RNA interactions and chromatin DNA–DNA contacts within individual cells. Applied to primary glioma specimens, SCIENCE-seq reveals cancer-specific interactions organized by lncRNAs and centered on key oncogenic drivers, including EGFR, hTERT, SOX2, CDK6, CDC42, CD47, and HOX loci. These datasets uncover previously unrecognized molecular relationships between premature lncRNAs and pre-mRNAs that regulate transcription, alternative splicing, and polyadenylation. Notably, we identify a glioma-specific trans-chromosomal HOX hub driven by five interacting lncRNAs. Targeting specific RNA and DNA interactions using steric antisense oligonucleotides (ASOs) or CRISPRi selectively suppresses oncogenic programs in malignant cells while sparing normal tissue, establishing chromatin interactomes as actionable therapeutic targets.

Brain tumors in older adults are difficult to diagnose and manage due to nonspecific symptoms, overlapping with neurodegenerative diseases such as dementia, and significant tumor heterogeneity. Although molecular markers such as IDH1/2 mutations, MGMT promoter methylation, TERT alterations, and 1p/19q co-deletion have improved glioma classification and prognostic assessment, current care still relies on invasive tissue biopsies, which limit longitudinal monitoring and may not fully capture tumor complexity because of sampling bias, assay variability, and limited accessibility. Liquid biopsy offers a promising alternative that enables the detection of tumor-derived DNA, RNA, proteins, and extracellular vesicles, supporting earlier diagnosis and real-time monitoring of disease progression and treatment response. However, liquid biopsy for brain tumors is not yet clinically definitive due to low biomarker abundance, lack of standardization, and limited validation, and therefore, it cannot replace tissue diagnosis. Ongoing research focuses on multi-analyte biomarker panels, improved assay standardization, and integration with imaging and tissue-based data. In parallel, artificial intelligence and machine learning are advancing the field by integrating multi-omics and radiomic data to enhance detection, classify tumors, and predict key molecular alterations, supporting the emerging framework of radiogenomics. Together, these developments are driving a shift toward more precise and dynamic approaches to brain tumor diagnosis and management, with relevance for improving outcomes in older adults with brain cancer.

Maximal safe surgical resection remains a critical component of glioblastoma (GBM) management, improving both survival and quality of life. However, complete tumor removal is hindered by the infiltrative nature of GBM and its proximity to eloquent brain regions. Fluorescence-guided surgery (FGS) has emerged as a valuable tool to enhance intraoperative tumor visualization and optimize resection outcomes. Currently used fluorophores such as 5-aminolevulinic acid (5-ALA), fluorescein sodium (FS), and indocyanine green (ICG) have distinct advantages but are limited by suboptimal specificity, shallow tissue penetration, and technical constraints. 5-ALA and SF often yield unreliable signals in low-grade tumors or infiltrative regions and also pose challenges such as phototoxicity and poor depth resolution. In contrast, near-infrared (NIR) fluorescence imaging represents a promising next-generation approach, providing superior tissue penetration, reduced autofluorescence, and real-time delineation of tumor margins. This review explores the mechanisms, clinical applications, and limitations of currently approved FGS agents and highlights future directions in image-guided neurosurgery.

Glioblastoma (GBM) relies on fatty acid metabolism to sustain its aggressive growth. While the role of stearoyl-CoA desaturase-1 (SCD1) in GBM is established, the function of its brain-enriched isoform, SCD5, remains unexplored. Here, we demonstrate that SCD5 is essential for glioblastoma stem cell (GSC) maintenance and genomic stability, with elevated expression in GSCs that declines upon differentiation, underscoring its role in tumor initiation. Through shotgun lipidomics, ¹³ C metabolic flux analysis, and functional genomics, we reveal that SCD1 and SCD5 play non-redundant roles in fatty acid desaturation, with SCD5 preferentially desaturating C18:0 and uniquely contributing to sphingolipid remodeling. Genetic silencing of either isoform disrupts cell cycle progression, impairs DNA damage repair, and reduces GSC viability, while SCD5 knockdown significantly extends survival in orthotopic GBM models. Mechanistically, loss of SCD activity or saturated fatty acid accumulation triggers PARP1 hyperactivation and subsequent degradation, depleting RAD51 to compromise homologous recombination and induce parthanatos. These findings uncover a lipid-mediated vulnerability in GBM, linking fatty acid desaturation to PARP1-dependent genome integrity. Targeting SCD5 may offer a novel therapeutic strategy to eliminate therapy-resistant GSCs and enhance the efficacy of genotoxic or immunotherapeutic interventions.

Glioblastoma (GBM) is the most aggressive primary brain tumor in adults, characterized by a dismal prognosis and limited therapeutic options. Its highly invasive nature and pronounced intratumoral heterogeneity underscores the urgent need for innovative and targeted therapeutic strategies. One promising approach is synthetic lethality, which exploits cancer-specific genetic vulnerabilities to selectively eliminate tumor cells. A well-characterized example involves the deletion of methylthioadenosine phosphorylase (MTAP), commonly observed in GBM and other malignancies. This review focuses on synthetic lethality targeting protein arginine methyltransferase 5 (PRMT5) in MTAP-deleted GBM. Loss of MTAP leads to the accumulation of methylthioadenosine (MTA), a metabolite that partially inhibits PRMT5, thereby creating a selective vulnerability to PRMT5 inhibition which is used to inhibit the residual function of PRMT5. We critically evaluate preclinical and clinical data on both first- and second-generation PRMT5 inhibitors, with particular emphasis on MTA-cooperative compounds that selectively exploit MTAP deficiency. Despite promising anti-tumor activity in vitro, the clinical efficacy of PRMT5 inhibitors is often limited by the tumor microenvironment, particularly the impact of non-malignant cells that attenuate drug activity. Finally, we explore rational combination strategies that integrate PRMT5 inhibition with existing therapies to enhance clinical outcomes in GBM.

Breast cancer brain metastases (BCBM) are incurable, and new therapies are urgently needed. BCBM upregulates stearoyl-CoA desaturase (SCD), an enzyme that catalyzes the synthesis of monounsaturated fatty acids, suggesting a potential metabolic vulnerability. Here, we test the effect of a brain-penetrant, clinical-stage SCD inhibitor (SCDi) on breast cancer cells and mouse models of BCBM. We show that SCDi markedly reshapes the lipidome of breast cancer cells, resulting in endoplasmic reticulum stress, DNA damage, impaired DNA damage repair, and cytotoxicity. Importantly, SCDi alone or combined with a PARP inhibitor prolongs the survival of BCBM-bearing mice. Furthermore, pharmacological inhibition of SCD enhances antigen presentation by dendritic cells, increases interferon signaling, promotes the infiltration of cytotoxic T cells, and decreases the proportion of exhausted T cells and regulatory T cells (Tregs) in the tumor microenvironment (TME) in a syngeneic mouse model of BCBM. Additionally, SCDi reduces the engagement of immunosuppressive pathways, including the PD-1:PD-L1/PD-L2 and PVR/TIGIT axes in the TME. These findings suggest that SCD inhibition could be an effective strategy to both intrinsically reduce tumor growth and reprogram anti-tumor immunity in the brain microenvironment to treat BCBM.

Brain organoid technology has revolutionized our ability to model human neurodevelopment in vitro. However, current techniques remain limited by their reliance on simplified endothelial cell populations rather than a complete endothelial system. We engineered Multi-Region Brain Organoids (MRBOs) that integrate cerebral, mid/hindbrain, and complex endothelial organoids into one structure. Different from the earlier approaches based on isolated HUVECs, our endothelial organoids contain diverse vascular cell types, including vascular progenitors, mature endothelial cells, pericytes, proliferating angiogenic cells and stromal cells. Our strategy employs a sequential modulation of key developmental pathways to generate individual organoids, followed by optimized fusion conditions that maintain regional identities while supporting cellular integration. Single-nucleus RNA sequencing shows that MRBOs develop discrete neural populations specific to each brain region alongside specialized endothelial populations that establish paracrine signalling networks. Integration analysis with human fetal brain data shows that MRBOs contribute to 80% of cellular clusters found in human fetal brain tissue (Carnegie stages 12-16), whereas CellChat analysis identifies 13 previously uncharacterized endothelial-neural signalling interactions. Notably, we uncover endothelial-derived factors that support the persistence of intermediate progenitor populations during hindbrain development, but not during the cerebral one thereby revealing a new role for complex endothelial populations in regional brain patterning. This platform enables matching multiple developmental regions, fully incorporating the endothelial nature at the same time providing opportunities for studying neurodevelopmental disorders in which neural-endothelial interactions are disrupted. Our engineered MRBO system establishes a foundation for investigating complex neurodevelopmental processes, providing an enabling context closer to physiological relevance.

Microglia are critical innate immune cells of the brain. In vivo targeting of microglia using gene-delivery systems is crucial for studying brain physiology and developing gene therapies for neurodegenerative diseases and other brain disorders such as NeuroAIDS. Historically, microglia have been extremely resistant to transduction by viral vectors, including adeno-associated virus (AAV) vectors. Recently, there has been some progress demonstrating the feasibility and potential of using AAV to transduce microglia after direct intraparenchymal vector injection. Data suggests that combining specific AAV capsids with microglia-specific gene expression cassettes to reduce neuron off-targeting will be key. However, no groups have developed AAV capsids for microglia transduction after intracerebroventricular (ICV) injection. The ICV route of administration has advantages such as increased brain biodistribution while avoiding issues related to systemic injection. Here, we performed an in vivo selection using an AAV peptide display library that enables recovery of capsids that mediate transgene expression in microglia. Using this approach, we identified a capsid, MC5, which mediated enhanced transduction of microglia after ICV injection compared to AAV9. Furthermore, MC5 enhanced both the efficiency (85%) and specificity (93%) of transduction compared to a recently described evolved AAV9 capsid for microglia targeting after direct injection into the brain parenchyma. Exploration of the use of MC5 in a mouse models of Alzheimer’s disease revealed transduced microglia surrounding and within plaques. Overall, our results demonstrate that the MC5 capsid is a useful gene transfer tool to target microglia in vivo by direct and ICV routes of administration.