PiC was described as a key protein in maintaining mitochondrial homeostasis, and earlier work had suggested that it may help regulate MPTP opening. The researchers reported that irradiation specifically reduced PiC expression in glioma cells. This suppression was followed by failure of mitochondrial energy metabolism and loss of mitochondrial membrane potential, providing evidence that radiation disrupts several linked features of mitochondrial function.

Experiments examining the underlying sequence placed mitochondrial DNA replication and MPTP activity at the center of the response. Radiation-induced PiC suppression was associated with dysregulation of mtDNA replication and significantly increased, sustained MPTP opening. Taken together, the reported results support a cascade in which radiation suppresses PiC, promotes persistent pore opening and mitochondrial dysfunction, and engages the mitochondrial apoptotic pathway that drives glioma cells toward programmed death.

Concordant changes in apoptotic markers were observed in murine tumor models, extending the cell-based findings to an intact biological setting. The in vivo results were consistent with the mechanistic observations made in glioma cells, although the supplied study summary does not provide numerical effect sizes, radiation doses, model details, or statistical estimates. The work therefore supports the relevance of the PiC–MPTP pathway in experimental glioma systems, but it does not by itself establish how much the pathway accounts for treatment response in people.

Beyond the reported findings, the authors proposed that targeting the PiC–MPTP axis could represent a potential strategy for overcoming radioresistance and enhancing radiosensitivity in glioma. That proposal remains a therapeutic hypothesis rather than a demonstrated clinical intervention. The evidence presented comes from glioma cells and murine tumor models, not a human trial, and does not show that manipulating PiC or MPTP would be safe or effective in patients.

Subject of Research: Radiation-induced PiC suppression, mitochondrial permeability transition, and apoptosis in glioma models

Article Title: The role of the mitochondrial phosphate carrier in modulating mitochondrial homeostasis and enhancing radiosensitivity in tumor cells

Article References: Shen, L., Liu, X., Lu, Y., Wang, K., Gu, H., Yuan, S., & Qian, J. (2026). The role of the mitochondrial phosphate carrier in modulating mitochondrial homeostasis and enhancing radiosensitivity in tumor cells. Molecular Biology Reports, 53(1), Article 1767. https://doi.org/10.1007/s11033-026-12779-7

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12779-7

Keywords: glioma, radiotherapy, mitochondrial phosphate carrier, PiC, mitochondrial homeostasis, permeability transition pore, apoptosis, radiosensitivity, mitochondrial DNA, tumor models, role, mitochondrial

The paired analysis classified 135 tumors, or 61.1%, as IE1-high and 124, or 56.1%, as UL123-positive. IE1-high status was observed in 74.8% of grade III–IV gliomas compared with 31.4% of grade I–II tumors. The corresponding UL123-positive proportions were 70.2% and 25.7%. Both assays also identified markers more frequently in patients aged 50 years or older, in tumors lacking the IDH1 R132H mutation, and among those with a preoperative Karnofsky Performance Status below 80. Neither marker showed a significant association with sex. These findings extend earlier reports that HCMV proteins and nucleic acids can be detected in some gliomas, although other studies have reported little or no signal using different methods.

Immunohistochemistry detected IE1 in the nuclei and cytoplasm of tumor cells. Investigators combined staining intensity, scored from zero for no staining to three for strong staining, with the proportion of stained cells to produce a final score of zero to 12. Tumors scoring four or higher were classified as IE1-high, while those scoring below four were IE1-low. Three pathologists, blinded to the PCR results and patient outcomes, evaluated the sections and resolved disagreements by joint review. A nonimmune mouse immunoglobulin G control was used in place of the primary antibody. Positive cells were generally sparse and weakly stained in low-grade tumors, whereas staining was more widely distributed and conspicuous in high-grade tumors.

UL123 DNA was measured in triplicate by quantitative real-time PCR, with the mean cycle threshold value used to classify each sample. A mean Ct below 35 was considered UL123-positive, and a value of 35 or above was considered negative. Melting-curve analysis was used to assess amplification specificity. The assay documentation did not establish sample-specific DNA amplifiability in UL123-negative specimens, so a negative result could reflect low viral DNA, DNA degradation, or PCR inhibition rather than complete absence of the virus. The operational cutoff was not analytically validated, an important qualification because threshold choice can materially affect detection rates and group assignments.

The protein and DNA assays produced the same binary classification for 162 of the 221 paired specimens, corresponding to 73.3% agreement. Cohen’s kappa was 0.451, indicating an association beyond chance but not enough agreement for the tests to be treated as interchangeable. There were 100 double-positive and 62 double-negative samples, while 35 tumors were positive only by immunohistochemistry and 24 only by PCR. The paired positive rates did not differ significantly overall, with a McNemar P value of 0.193, and the assays remained associated when results were separated by tumor grade, age, or sex. Possible explanations for discordance include uneven distribution of viral material within a tumor and the detection of protein when DNA levels fall below the PCR threshold, but neither mechanism was directly tested.

Kaplan–Meier analysis showed substantially lower survival estimates among patients whose tumors were positive for either marker. At 36 months, estimated overall survival was 25.9% in the IE1-high group and 54.7% in the IE1-low group. Median overall survival was 18.2 months for IE1-high patients, while the median was not reached in the IE1-low group; the log-rank P value was below 0.001. For UL123, 36-month survival was 25.0% among positive cases and 52.6% among negative cases, with medians of 16.7 months and not reached, respectively, and again a P value below 0.001. These were unadjusted comparisons, and the study did not report hazard ratios or confidence intervals.

Patients whose tumors were positive by both assays had the poorest observed outcomes. Among the 162 concordant cases, 36-month overall survival was 21.0% for the 100 double-positive patients and 59.7% for the 62 double-negative patients, with a log-rank P value below 0.001. Median survival was 15.6 months in the double-positive group and was not reached in the double-negative group. The same pattern appeared within both grade categories. In grade I–II disease, estimated 36-month survival was 9.1% for double-positive cases and 56.1% for double-negative cases; in grade III–IV disease, the figures were 22.5% and 66.7%. The combined survival analysis excluded 59 patients with discordant results, so its estimates apply specifically to patients with matching assay classifications.

Several factors prevent the observed survival differences from being interpreted as an independent prognostic effect of HCMV. Older age, lower preoperative performance status, higher grade, and IDH1 wild-type status were themselves associated with worse outcomes and were unevenly distributed between marker groups. The analysis did not use multivariable survival adjustment, did not include postoperative treatment or extent of resection, and used a Ct threshold without analytical validation. It also grouped tumors mainly by histologic grade rather than by fully integrated contemporary molecular classification, while IDH1 R132H testing alone does not identify every IDH1 or IDH2 mutation. More aggressive tumors may make viral markers easier to detect, and marker status could partly reflect tumor biology rather than drive progression.

The study’s principal strength is its paired testing of protein and DNA in a relatively large, single-center glioma cohort, including grade-stratified clinicopathological and survival analyses. Its main limitations are the need for independent replication, potentially influential tissue-selection and assay procedures, incomplete control documentation for negative PCR results, and the absence of adjusted survival models. Future studies should use blinded testing with predefined thresholds, validate PCR amplifiability, examine additional viral DNA targets and multiple regions of each tumor, apply current integrated tumor classifications, and report adjusted hazard estimates and confidence intervals. Until such validation is available, IE1 protein and UL123 DNA should be regarded as investigational markers associated with aggressive glioma features and poorer observed survival, rather than proven causes of disease or clinically ready prognostic tools.

Subject of Research: HCMV IE1 protein and UL123 DNA associations with glioma features and overall survival

Article Title: Association of HCMV IE1 protein and UL123 DNA with clinicopathological features and survival in patients with glioma

Article References: Hao, Z., Liu, Q., Guo, Y., Zhang, J., Ding, R., Chen, L., & Yin, X. (2026). Association of HCMV IE1 protein and UL123 DNA with clinicopathological features and survival in patients with glioma. Frontiers in Cellular and Infection Microbiology, 16, Article 1975136. https://doi.org/10.3389/fcimb.2026.1975136

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1975136

Keywords: human cytomegalovirus, glioma, IE1 protein, UL123 DNA, immunohistochemistry, quantitative PCR, overall survival, IDH1 status, tumor grade, Association, HCMV, protein

The appeal of these vesicles lies in their biology. Because they carry a selective cargo of proteins, lipids and non-coding RNAs that mirror their parent cell, vesicles shed by diseased neurons act as molecular messengers of brain pathology. Their lipid bilayer protects this cargo from enzymatic degradation in the bloodstream, allowing long-term biobanking, while their membranes can fuse with biological barriers and slip across the blood–brain barrier into peripheral circulation. Once there, they can be harvested from a simple blood draw, cerebrospinal fluid, urine or saliva, turning what was once inaccessible central pathology into a minimally invasive liquid biopsy.

The review, which searched PubMed, Web of Science and Embase from database inception to January 2026, distilled roughly 26 studies into primary clinical biomarker evidence, supported by about 15 mechanistic investigations. Alzheimer’s disease features most prominently. Pathological amyloid-beta plaques and phosphorylated tau tangles begin accumulating 15 to 20 years before symptoms appear, making early detection critical. Longitudinal work by Kapogiannis and colleagues in the Baltimore Longitudinal Study of Aging showed that neuron-derived vesicles from people who later developed Alzheimer’s carried elevated p-tau181, p-tau231 and phosphorylated IRS-1, with a multiplexed model achieving an area under the curve of 0.896 in training and 0.800 in testing cohorts.

Subsequent studies strengthened the case. Jia and colleagues found that Aβ42, total tau and p-tau181 in immunoenriched plasma vesicles performed comparably to established cerebrospinal fluid biomarkers, with a combined validation AUC of 0.980 for distinguishing Alzheimer’s patients from healthy controls. Beyond the canonical toxic proteins, vesicle-borne synaptic proteins such as synaptopodin and synaptophysin decreased in both Alzheimer’s and frontotemporal dementia, and importantly correlated with cognitive scores, whereas Aβ42 and p-tau181 did not. A p-tau396 signal that rises in Alzheimer’s but stays at baseline in frontotemporal dementia offers a potential differential marker, while declines in the complement regulator CD59 in astrocyte-derived vesicles and shifts in lysosomal proteins such as cathepsin D point to neuroinflammation and endolysosomal dysfunction years before clinical onset.

Parkinson’s disease tells a parallel story built around α-synuclein, the protein whose intracellular aggregation defines the disorder. Jiang and colleagues reported that α-synuclein in serum neuron-derived vesicles was approximately two-fold higher in Parkinson’s patients than in controls, achieving an AUC of 0.86, and, strikingly, was already elevated in people with rapid eye movement sleep behavior disorder, a prodromal condition. Adding clusterin to form an α-synuclein-to-clusterin ratio pushed differential diagnostic accuracy to 0.980 against other proteinopathies and 0.940 against multiple system atrophy. A later meta-analysis confirmed elevated total α-synuclein across plasma vesicle studies, and proteomic profiling by Kitamura and colleagues identified vesicle-associated apolipoprotein A1 as strongly and negatively correlated with Hoehn and Yahr disease stage.

Non-coding RNAs add a second analytical dimension. In cerebrospinal fluid vesicles, a miR-153 and miR-409-3p combination reached an AUC of 0.990 for Parkinson’s discrimination, while a serum panel of miR-19b, miR-24 and miR-195 achieved 0.946. Bioinformatic screening by He and colleagues went further, mapping serum vesicle microRNAs onto Hoehn–Yahr stages: miR-374a-5p and miR-374b-5p rose across all active phases, miR-199a-3p fell specifically at stage II, miR-28-5p peaked at stage III, and miR-22-5p and miR-151a-5p diverged at stage IV, hinting at non-invasive molecular staging of a disease currently assessed largely by clinical examination.

The review extends this logic across the wider neurological spectrum. In acute ischemic stroke, vesicle SIRT2 achieved an AUC of 0.838, miR-134 correlated with both stroke severity scores and infarct volume, and miR-422a distinguished acute from subacute phases with an AUC of 0.957, while a five-circular-RNA brain-vesicle panel reached 0.921 with 95 percent sensitivity. In glioma, where tissue diagnosis currently requires surgery, vesicle EGFRvIII transcripts detected high-grade tumors with roughly 80 percent accuracy, and a three-circular-RNA signature distinguished high-grade astrocytoma from controls with perfect accuracy in a small single-center cohort. In multiple sclerosis, vesicle myelin oligodendrocyte glycoprotein rose during relapses, and a miR-122-5p and miR-196b-5p panel separated relapsing patients from remission with an AUC of 0.866. In Huntington’s disease, mutant huntingtin was detectable in plasma neuron-derived vesicles, and a combined model of tRF-Gly-GCC, miR-27a-3p and neurofilament light chain distinguished premanifest carriers with an AUC of 0.908. In amyotrophic lateral sclerosis, vesicle TDP-43 separated patients from controls with AUCs of 0.94 to 0.99 across two independent cohorts.

Yet the review is notably candid about the field’s unresolved problems. Signals overlap across diseases: miR-21 rises both in glioma and in the subacute phase of stroke, and neurofilament light chain climbs in Alzheimer’s, ALS, Huntington’s and stroke alike, suggesting these molecules may reflect shared processes such as neuroinflammation and axonal injury rather than specific entities. Attribution is equally thorny. The L1CAM immunocapture technique widely used to isolate neuron-derived vesicles from blood was challenged by Norman and colleagues, who found L1CAM mostly in soluble protein fractions, though a later single-vesicle study found L1CAM-positive vesicles co-expressing neuronal markers in 30 to 63 percent of cases. Blood vesicles are also overwhelmingly derived from blood cells and endothelium, so CNS signals may be diluted or confounded by peripheral contributions.

Standardization and validation gaps loom just as large. Pre-analytical factors such as fasting, platelet activation during venipuncture, hemolysis, lipoprotein contamination and freeze–thaw cycles can all distort vesicle measurements, and isolation techniques trade purity against yield. Although a systematic review of 471 vesicle-related clinical trials found that nearly half assess diagnostic applications, only 11 percent examined vesicle subpopulations, and reporting of isolation methods remains incomplete. Exceptional results, such as 100 percent accuracy in small cohorts, may reflect overfitting rather than clinical readiness. The authors point to the urinary ExoDx Prostate assay, which earned FDA Breakthrough Device Designation after prospective validation, as a template, and call for standardized MISEV2023-compliant workflows, covariate adjustment, and prospective multicenter validation before brain-derived vesicle panels can move from promising research candidates to clinically deployable diagnostics.

Subject of Research: Extracellular vesicles as liquid biopsy biomarkers for diagnosing central nervous system diseases

Article Title: Utility of extracellular vesicles in the diagnosis of central nervous system diseases

Article References: Wang, M., Yang, J., Gao, C., Qu, S., Mi, H., Guo, Y., Gui, T., & Gai, Z. (2026). Utility of extracellular vesicles in the diagnosis of central nervous system diseases. Frontiers in Neurology, 17, Article 1959745. https://doi.org/10.3389/fneur.2026.1959745

Image Credits: AI Generated

DOI: 10.3389/fneur.2026.1959745

Keywords: extracellular vesicles, liquid biopsy, blood-brain barrier, Alzheimer's disease, Parkinson's disease, biomarkers, neurodegenerative disease, stroke, glioma, multiple sclerosis, amyotrophic lateral sclerosis, exosomes

The trial focuses on solid tumours for which CAR T-cell treatment remains far less developed than it is for leukaemia, lymphoma and multiple myeloma. Among the representative cancers named in the protocol are neuroblastoma, glioma and osteosarcoma. These diseases can carry aggressive biological features, respond poorly to conventional treatment and frequently relapse despite intensive therapy. Patients whose disease returns or proves resistant to existing approaches have few effective options, particularly when standard agents have been exhausted. TARGETS therefore addresses both a scientific problem—the limited success of CAR T cells against solid tumours—and a clinical one—the shortage of promising treatments for patients with advanced refractory disease.

GD2 is a ganglioside, a class of molecules composed of sugars linked to a fatty component, and it is expressed by a variety of tumours. Its presence can make it useful as a target, but the effectiveness of an immune treatment depends on more than simply recognising the molecule. A CAR is a synthetic receptor that gives a T cell the ability to detect a specified surface antigen and trigger an immune attack when that antigen is encountered. The investigational product described in TARGETS uses a newly engineered second-generation CAR directed against GD2, combining the antigen-binding machinery of the receptor with intracellular signalling designed to activate the modified T cell.

All participants will receive autologous CAR T cells, meaning that the therapeutic cells are manufactured from each patient’s own immune cells rather than supplied as an off-the-shelf donor product. Before treatment, investigators will collect peripheral blood through leukapheresis, isolate the relevant cells and engineer them to express the anti-GD2 receptor. Manufacturing will follow Good Manufacturing Practice requirements, while broader trial procedures will adhere to Good Clinical Practice principles and the Declaration of Helsinki. The approach is personal to each participant but also complex: it requires coordinated blood collection, specialist cell processing, quality testing and transport before the finished product can be returned for administration.

Before the CAR T cells are infused, patients will receive lymphodepleting chemotherapy. This preparatory treatment reduces the number of lymphocytes in the blood and creates space for the engineered cells to expand, while modifying the immune environment to support their persistence and activity. The CAR T product is then administered intravenously. Because the trial involves a newly engineered construct and is the first clinical evaluation of this particular product, investigators will combine intensive inpatient observation after infusion with longer follow-up. The study plans to monitor participants for up to four years, allowing researchers to capture both immediate toxicities and potential longer-term immune effects.

Safety is the primary focus of the first part of TARGETS. The investigators will use a conventional 3+3 dose-escalation scheme across three planned dose levels. In this design, small groups of participants receive an initial dose, and additional patients are enrolled only when the observed toxicities permit escalation. The process helps identify a dose-limiting toxicity and establish the maximum tolerated dose while balancing treatment exposure against risk. A Data and Safety Monitoring Board will oversee the study, and the researchers will record adverse events, serious adverse events and other safety signals using recognised clinical-trial criteria.

A basket design can accelerate the evaluation of a targeted therapy across several tumour types because the defining feature is shared biology rather than a single organ. Here, eligibility centres on tumours that express GD2 and have relapsed or refractory disease. The approach allows the same CAR T-cell product to be tested in neuroblastoma, glioma, osteosarcoma and other qualifying solid tumours, while also revealing whether biological responses differ markedly between diseases. At the same time, the small planned cohort of roughly 20 participants means the trial is not designed to provide definitive comparative evidence. Its role is early-stage: establish tolerability, characterize dose-related effects and look for signals that justify larger, disease-focused studies.

The protocol has passed ethics review by the Human Research Ethics Committees of the Faculty of Medicine at Chiang Mai University, the Faculty of Medicine Ramathibodi Hospital at Mahidol University, and the Faculty of Medicine at Prince of Songkla University. Written informed consent will be obtained from participants or, where applicable, their parents or legally authorized representatives before enrollment, and assent will be sought from children in line with local regulations and institutional policies. Several investigators involved in the study also report a Thai petty patent covering a GD2-specific CAR T-cell receptor. The work was supported by Genomics Thailand and the Health Systems Research Institute, with partial support from Chiang Mai University.

CAR T cells are not a routine treatment for relapsed GD2-positive solid tumours, and the publication describes a study protocol rather than clinical results. No efficacy figures or conclusions about tumour shrinkage are therefore available from this research. The central scientific question is whether a second-generation anti-GD2 receptor can produce a tolerable and potentially useful immune response in a difficult group of cancers. TARGETS will provide early safety data, define a workable dose and offer a first look at whether engineered T cells can shrink established tumours. If those early signals prove encouraging, they could provide the foundation for larger trials designed around the tumour types most likely to benefit.

Editorial clarification: This article describes a published trial protocol, not treatment results or independent confirmation that recruitment has started. “First-in-human” refers to this particular investigational product.

Subject of Research: A phase I/IIa basket trial of autologous anti-GD2 CAR T cells for relapsed or refractory GD2-positive solid tumours

Article Title: A basket trial of autologous Anti-GD2 chimeric antigen receptor (CAR) T Cells in patients with relapsed or refractory GD2-positive solid tumors (TARGETS): study protocol for a phase I/IIa multicenter open-label trial

Article References: Koonrungsesomboon, N., Anurathapan, U., Chaiyawat, P., Dukaew, N., Kittilukkana, A., Orrapin, S., Choed-amphai, C., Sathitsamitphong, L., Charoentum, C., Hongeng, S., Sripornsawan, P., Sangkhathat, S., Preedagasamzin, S., Atjanasuppat, K., Jittorntrum, B., Suanpan, K., Suthapot, P., & Pruksakorn, D. (2026). A basket trial of autologous Anti-GD2 chimeric antigen receptor (CAR) T Cells in patients with relapsed or refractory GD2-positive solid tumors (TARGETS): study protocol for a phase I/IIa multicenter open-label trial. BMC Cancer. https://doi.org/10.1186/s12885-026-17149-x

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17149-x

Keywords: CAR T cells, GD2, solid tumours, neuroblastoma, glioma, osteosarcoma, immunotherapy, dose escalation, clinical trial, Thailand, basket, trial

The clinical motivation behind the work is striking. Epidemiological and clinical studies suggest that patients with a history of stroke or traumatic brain injury face an increased risk of developing brain tumors, a risk that can be roughly three-fold to seven-fold depending on the age and sex of the patient. Yet despite these observations, the mechanism linking brain injury and cancer has remained unclear. The researchers set out to determine whether stroke actively promotes glioma growth and to uncover the factors mediating the connection, working with both human and mouse models.

First authors Dr. Qi Ye and graduate student Christine Madamba from the laboratory of corresponding author Dr. Hyun Kyoung Lee focused on stroke as the injury and glioma as the tumor. Glioma carries a devastating five-year survival rate, and the team found that stroke made its course even worse. In both human and mouse glioma models, stroke promoted tumor infiltration into injured brain regions and reduced overall survival. When the researchers examined the cellular and molecular makeup of the tumors more closely, they discovered that stroke triggers a remodeling of the tumor microenvironment, the complex ecosystem of cells and signals that surrounds and supports cancer cells.

At the heart of this remodeling was the emergence of a distinct population of brain cells called tumor-associated astrocytes, or TAAs. Astrocytes are star-shaped glial cells that normally support neurons by regulating blood flow, neurotransmitter uptake and the blood-brain barrier. In the stroke-injured tumor environment, however, astrocytes took on new physiological and molecular characteristics, including diminished calcium activity. Calcium signaling in astrocytes is a key communication channel that governs how these cells interact with their surroundings, and its reduction marked the TAAs as functionally distinct from their normal counterparts.

The TAAs did not act alone. The researchers observed that they were accompanied by the accumulation of two other cell types: remodeled tumor-associated microglia, the brain’s resident immune cells, and immune cells called macrophages, together known as tumor-associated macrophages or TAMs. Microglia and macrophages are known players in the brain’s response to injury, clearing debris and coordinating inflammation, but in this context they appeared to be recruited into a tumor-promoting alliance alongside the astrocytes.

The most consequential findings came from experiments that tested whether these cell populations were merely bystanders or active drivers of cancer progression. When the team restored calcium signaling in the tumor-associated astrocytes, or removed the TAMs, stroke-induced glioma progression was suppressed. These results identify both populations as critical mediators of the stroke response, proving that the injury-induced changes in the brain actively communicate with cancer cells and influence tumor behavior rather than simply marking time after damage.

The implications reach beyond glioma. Neuron-tumor interactions have previously been shown to contribute to cancer growth, but this study demonstrates that other brain cells, astrocytes, also seem to communicate with brain cancer cells and shape their behavior. As Lee notes, astrocytes should be considered when studying cancer mechanisms and therapies, adding to growing evidence pointing at a role for these glial cells in brain tumor growth. The work reframes the brain’s non-neuronal cells as active participants in cancer biology, not passive scaffolding.

Therapeutically, the findings suggest that the injury-induced, tumor-promoting pathways uncovered in this study are potential targets. If the astrocyte and macrophage responses that follow brain injury can be modulated, the risk of glioma growth in patients with a history of stroke or traumatic brain injury might be reduced. The researchers emphasize that this supports continuing research into strategies that could intervene in these pathways before tumors take advantage of the injured brain environment.

For patients and clinicians, the study provides a biological explanation for a clinical pattern that has long been observed but never mechanistically understood. Brain injury can now be considered a risk factor for brain cancer, and the cellular mediators of that risk, tumor-associated astrocytes with reduced calcium activity and tumor-associated macrophages, offer concrete starting points for prevention and treatment research. The work was a broad collaboration involving investigators from Baylor College of Medicine, Duncan NRI at Texas Children’s Hospital, Virginia Tech, Pennsylvania State University, Houston Methodist Research Institute, Weill Cornell Medical College, the University of Rhode Island and the University of Texas Health Science Center at Houston, among others, with first authors Qi Ye and Christine Madamba leading the experimental effort under Lee’s direction.

Subject of Research: Neuroscience and Cancer Research

Article Title: Researchers link brain injury to brain tumors

Article References: Researchers link brain injury to brain tumors. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: glioma, stroke, brain injury, brain tumor, astrocytes, tumor microenvironment, Nature Cancer, neuroscience, cancer research, Texas Children's Hospital, Researchers, link

Gliomas are the most common primary brain tumors in adults, and their precise segmentation on MRI is far more than an academic exercise. Radiologists and neurosurgeons rely on the boundaries of three distinct sub-regions, the peritumoral edema that swells around the tumor mass, the necrotic core where tissue has died, and the active tumor where proliferating cells drive disease progression. Treatment monitoring and surgical planning both hinge on how accurately these compartments are outlined. The gold standard in machine learning research, the BraTS challenge dataset, has fueled years of progress in automated segmentation, yet the field continues to search for input representations that expose tumor biology more clearly to the algorithms doing the work.

The idea behind rT1/T2 maps is rooted in quantitative MRI physics. T1 and T2 are intrinsic relaxation properties of tissue, describing how quickly hydrogen nuclei recover their magnetization after excitation and how rapidly the signal decays, respectively. Different tissues relax at different rates, which is why a glioma looks different from healthy white matter on a scanner. Rather than feeding the network the raw weighted images, whose intensities depend on scanner settings and acquisition protocols, the study reconstructed a pixel-wise ratio of the two relaxation properties. This relative map is a synthetic modality, computed from preprocessed T1-weighted and T2-weighted sequences, and it condenses complementary information from both acquisitions into a single normalized channel.

To test whether this synthetic input actually helps, Khorasani trained backbone U-Nets, the workhorse architecture of medical image segmentation, on 381 cases drawn from the BraTS 2023 dataset. The U-Net design pairs an encoder that compresses spatial information with a decoder that reconstructs fine-grained masks, and its performance can be tuned by swapping the backbone network that performs the feature extraction. Six well-known convolutional backbones were put to the test: ResNet34, ResNet152, InceptionV3, VGG16, DenseNet121, and MobileNet. Each model was trained in a single-channel configuration, meaning the network saw only the rT1/T2 map rather than the multi-sequence stacks common in clinical pipelines, and evaluation used 5-fold cross-validation with patient-wise train-test splitting to ensure that no patient’s scans appeared in both training and testing data.

Performance was measured with three complementary metrics. The Dice Similarity Coefficient quantifies the overlap between the predicted mask and the ground truth, ranging from zero for no agreement to one for perfect overlap. Intersection over Union penalizes both false positives and false negatives in a single ratio. The 95th percentile Hausdorff Distance, abbreviated HD95, captures how far the predicted boundary strays from the true boundary in millimeters, focusing on the worst 5 percent of surface points to avoid being fooled by a few outlier pixels. Across the peritumoral edema, necrotic core, and active tumor regions, these metrics together paint a picture of both volumetric accuracy and geometric precision.

The results favored the synthetic maps. A ResNet34 backbone U-Net trained on rT1/T2 maps achieved the best performance for edema segmentation, with a Dice score of 0.901, an HD95 of 1.57 millimeters, and a correlation coefficient of 0.997. The same architecture led on the necrotic core, reaching a Dice of 0.816, an HD95 of 2.43 millimeters, and a correlation of 0.994. For the hardest target, the actively proliferating tumor, an InceptionV3 backbone U-Net took the lead with a Dice of 0.837, an HD95 of 1.66 millimeters, and a correlation of 0.975. These numbers indicate that the networks were not merely capturing coarse tumor extent but resolving biologically meaningful internal structure.

To rank the many combinations of backbone and input modality objectively, the study applied TOPSIS, the Technique for Order of Preference by Similarity to Ideal Solution, a multi-criteria decision analysis method that scores each candidate by its geometric distance to an ideal and an anti-ideal solution. In this ranking, rT1/T2 maps placed first overall with a closeness coefficient of 0.794, and InceptionV3 emerged as the top-ranked backbone with a closeness coefficient of 0.938. The analysis provides a systematic answer to a question that single-metric comparisons often leave ambiguous, namely which pairing of architecture and input delivers the best balance across all three tumor sub-regions and all evaluation metrics simultaneously.

The author is careful to frame the finding with appropriate scientific caution. Within this single-channel experimental setting, the rT1/T2 maps showed what the study describes as a modest but consistent average advantage over conventional T1-weighted and T2-weighted sequences across the backbone architectures tested. Crucially, the study leaves open whether that advantage stems from the ratio transformation itself, which may highlight tissue contrast in a way the raw sequences cannot, or simply from the fact that the synthetic map draws on information contained in both parent sequences. The appropriate control, a multi-channel input combining T1 and T2 side by side, was not tested here and is flagged as a priority for future work.

The implications for clinical practice are nonetheless intriguing. Single-channel inputs reduce memory demands and computational cost, which matters for deploying segmentation models in hospitals with limited hardware or for integrating them into real-time surgical navigation. A synthetic modality derived from sequences that every glioma patient already receives adds no scanning burden, no contrast agent, and no acquisition time. If the advantage holds up against multi-channel baselines and external datasets, ratio-based inputs could become a standard preprocessing step rather than a research curiosity.

The work also fits into a broader trend in medical AI, where the spotlight is shifting from ever-larger networks toward smarter representations of the data themselves. Quantitative and semi-quantitative maps, including relaxation ratios, diffusion-derived metrics, and perfusion parameters, encode tissue physiology more directly than conventional weighted images. As segmentation models are increasingly asked to support treatment monitoring and surgical planning for glioblastoma and other gliomas, the choice of input modality may prove as consequential as the choice of architecture. This study, supported by Isfahan University of Medical Sciences and approved under the Declaration of Helsinki, offers a measured but compelling data point in that evolving conversation, and its open-access publication ensures that other teams can immediately test the idea on their own data.

Subject of Research: Synthetic rT1/T2 maps as a novel input modality for backbone U-Net glioma sub-region segmentation

Article Title: Synthetic rT1/T2 maps as a novel input modality for backbone U-Net glioma sub-region segmentation

Article References: Khorasani, A. (2026). Synthetic rT1/T2 maps as a novel input modality for backbone U-Net glioma sub-region segmentation. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02782-6

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02782-6

Keywords: glioma, brain tumor segmentation, deep learning, U-Net, MRI, rT1/T2 maps, BraTS dataset, ResNet, InceptionV3, TOPSIS, medical imaging, artificial intelligence

18F-fluciclovine, chemically anti-1-amino-3-[18F]fluorocyclobutane-1-carboxylic acid, is a synthetic amino acid analogue designed to exploit the amino acid transporters that tumor cells overexpress. It crosses the blood-brain barrier and is taken up primarily through the sodium-dependent alanine-serine-cysteine transporter 2, known as ASCT2, with a supporting role for the sodium-independent L-type amino acid transporter 1, LAT1. Crucially, unlike natural amino acids, fluciclovine is not metabolized inside cells and is not incorporated into protein synthesis, so the signal it generates reflects transporter-dependent accumulation rather than metabolic processing. That property makes it attractive for imaging glioma, but it also means the tracer lingers in the bloodstream longer than tracers that are rapidly cleared into protein, which can contaminate measurements in blood-rich structures.

The researchers retrospectively analyzed 18F-fluciclovine PET/CT images from 20 consecutive patients scanned at Nagoya University Hospital between July 2024 and June 2025. All were treatment-naïve, with no history of neurosurgical resection, radiotherapy, or chemotherapy, and none were taking corticosteroids, sedatives, or antiepileptic drugs at the time of imaging. The cohort had a mean age of 49.6 years, ranging from 6 to 77, and included 14 patients with glioma and 2 with primary central nervous system lymphoma. Tracer was administered intravenously at a mean dose of about 231 MBq, and PET acquisition began 10 minutes after injection, running in list mode for 20 minutes with dynamic images reconstructed every 2 minutes alongside a static acquisition on a Biograph 16 Horizon scanner.

Two board-certified nuclear medicine physicians measured standardized uptake values in 24 predefined head regions, 12 intracerebral and 12 extracerebral. The intracerebral set covered the four cortical lobes, white matter, thalamus, basal ganglia, midbrain, pons, medulla oblongata, cerebellum, and cerebellar vermis. The extracerebral set spanned the pituitary and pineal glands, choroid plexus, venous and cavernous sinuses, lacrimal and parotid glands, pharynx, nasal cavity, skeletal muscle, bone marrow, and skin. Volumes of interest were placed as spherical regions on axial images with CT for anatomical reference, and cortical measurements were taken in tumor-free contralateral gray matter, a standardized approach borrowed from 11C-methionine PET studies to minimize measurement error and improve comparability across subjects.

The headline finding is that fluciclovine uptake in the brain is remarkably low and remarkably uniform. White matter showed the lowest SUVmean of any region, 0.15 ± 0.04, while the cerebral cortex averaged 0.28 ± 0.06, with no significant differences among the frontal, parietal, temporal, and occipital lobes. The cortex-to-cerebellum ratio stayed between 0.67 and 0.71 across lobes, a consistency that matters because it suggests minimal nonspecific accumulation anywhere in the cerebral hemispheres. Among the deeper brain structures, the cerebellar vermis led with an SUVmean of 0.46 ± 0.09, followed by the pons, cerebellum, thalamus, basal ganglia, midbrain, and medulla, all clustered between roughly 0.35 and 0.43.

Dynamic time-activity curves, available in 15 patients, showed that brain uptake crept upward only slightly during the 2-to-20-minute window, with absolute SUV changes of 0.05 to 0.08 in the cortex and no more than 0.15 anywhere in the parenchyma. That stability carries a practical implication: SUV measurements are unlikely to be substantially distorted by minor timing variations within this interval. The picture outside the brain was entirely different. The pituitary gland and parotid gland peaked at the first 2-minute frame and then washed out markedly, declining 41.9 percent and 35.1 percent respectively over the acquisition period, while the choroid plexus, nasal cavity, lacrimal gland, venous sinus, bone marrow, and cavernous sinus showed moderate declines of 22.6 to 31.2 percent.

The extracerebral biodistribution is where the baseline becomes genuinely useful for diagnosis. The parotid gland showed the highest SUVmean of any head region, 3.39 ± 0.82, roughly twelve times the cortical value, followed by the pharynx at 2.51, the pituitary gland at 2.32, the nasal cavity at 2.30, skeletal muscle at 2.23, bone marrow at 1.99, and skin at 1.88. Uptake in muscle, marrow, and skin was heterogeneous: the posterior neck, lateral pterygoid, and extraocular muscles lit up, as did the marrow of the clivus and occipital bone and the skin of the vertex. These hot spots likely reflect high ASCT2 expression in metabolically active cell populations, abundant regional blood flow, and, in the marrow, hematopoietic activity concentrated where red marrow persists in adults.

Eight patients also underwent 11C-methionine PET/CT within three weeks, allowing a direct head-to-head comparison of the two amino acid tracers. Methionine showed significantly higher uptake in every brain region, all p ≤ 0.004, and cortical SUVpeak values of 1.15 to 1.32 against fluciclovine’s 0.30 to 0.31. No significant correlations between the two tracers were observed in the brain, which the authors attribute to differences in transporter distribution: methionine rides mainly on LAT1 and LAT2, while fluciclovine depends chiefly on ASCT2, whose expression in brain endothelial cells is lower than LAT1’s. Outside the brain, the relationship reversed in places. Fluciclovine was significantly higher in the venous sinuses, skeletal muscle, and nasal cavity, possibly because it is not incorporated into protein synthesis and remains in the blood pool, whereas methionine is rapidly cleared from the bloodstream.

Age left its fingerprints on the extracerebral map. Pineal gland and bone marrow uptake both declined significantly with age, with correlation coefficients of −0.64 and −0.63 respectively, likely reflecting the progressive calcification of the pineal gland and the physiological shrinkage of hematopoietic red marrow. Cavernous sinus uptake, by contrast, rose with age. Intracerebral uptake showed no significant age-related changes, with correlation coefficients between −0.16 and 0.20. Incidental findings reinforced the caution needed in interpretation: a subcutaneous neurofibroma in a patient with neurofibromatosis reached an SUVmax of 3.37, and mild uptake appeared in the mucosal thickening of three cases of maxillary sinusitis, with a mean SUVmax of 1.10. Fluciclovine uptake is not tumor-specific, and benign lesions with high cellular proliferation or metabolic activity can light up.

The authors acknowledge the limits of their study. The participants were scanned for clinical indications rather than being healthy volunteers, so subtle metabolic alterations in normal-appearing tissue cannot be excluded, and the small, heterogeneous cohort with incomplete histological confirmation limits generalizability. Fasting conditions were not standardized, dynamic imaging began only at 10 minutes post-injection so early kinetics were not captured, and manual volume-of-interest placement introduces observer variability that was not formally assessed. Still, the contralateral measurements broadly matched previous reports, supporting the validity of the findings. For clinicians, the payoff is concrete: the very low, uniform brain uptake should enhance contrast with tumor tissue and aid preoperative margin delineation, while awareness of the hot physiological zones, particularly the venous sinuses and bone marrow near the skull base and tentorium, should help avoid false-positive readings where such uptake can mimic tumor. The team calls for multicenter validation in larger cohorts with standardized protocols, but the baseline they have drawn gives interpreters of fluciclovine head PET a reference framework they have never had before.

Subject of Research: Physiological biodistribution of 18F-fluciclovine PET in the head: as a baseline for interpretation in patients with suspected glioma

Article Title: Physiological biodistribution of 18F-fluciclovine PET in the head: as a baseline for interpretation in patients with suspected glioma

Article References: Abe, Y., Minamimoto, R., Fujita, N., Higashi, M., Ito, R., Kato, K., Deguchi, S., Ohka, F., Saito, R., & Naganawa, S. (2026). Physiological biodistribution of 18F-fluciclovine PET in the head: as a baseline for interpretation in patients with suspected glioma. Frontiers in Nuclear Medicine, 6, Article 1850254. https://doi.org/10.3389/fnume.2026.1850254

Image Credits: AI Generated

DOI: 10.3389/fnume.2026.1850254

Keywords: Physiological, biodistribution, F-fluciclovine, head, baseline, interpretation, patients, suspected, glioma, scientific research

Gadolinium-based contrast agents have been a cornerstone of magnetic resonance imaging since the late 1980s. By highlighting tissues with enhanced perfusion and permeability, they allow radiologists to spot tumors, map their boundaries, distinguish malignant from benign lesions, and monitor how treatment is working. But the agents carry baggage. In patients with severe kidney impairment they have been linked to nephrogenic systemic fibrosis, a rare but debilitating condition. Studies have also documented gadolinium deposits in the brain, bones, and other tissues even in people with normal kidney function, and the metal — which is not biodegradable — is increasingly turning up in rivers and groundwater after passing through wastewater treatment plants ill-equipped to remove it.

Those concerns have fueled a wave of research into AI methods that could reduce or eliminate the need for contrast injection altogether. One particularly ambitious approach is the synthesis of so-called T1ce images — T1-weighted scans taken after contrast administration — from a set of non-contrast scans such as T1, T2, and T2-FLAIR. Previous studies have generally judged these synthetic images by pixel-level similarity metrics like peak signal-to-noise ratio and structural similarity, or by qualitative review from radiologists. The GE HealthCare team, led by Dimosthenis Gkotsis and Hongxu Yang, argued that such measures are not enough. If synthetic images are ever to inform consequential decisions — tumor grading, radiotherapy planning, surgical margins, or distinguishing recurrence from radiation necrosis — they must be tested on the downstream tasks that actually matter.

The researchers built three state-of-the-art generative models trained on the publicly available BraTS19 dataset, which comprises 335 glioma patients, 259 with high-grade tumors and 76 with low-grade ones. The first model, called CoNeS, is a conditional neural field with shift modulation that encodes positional information to guide image generation. The second is a denoising diffusion probabilistic model, or DDPM, conditioned on the non-contrast sequences. The third was a hybrid that fed the CoNeS output into the diffusion model as an additional conditioning channel. The diffusion model also employed a coarse-to-fine sampling strategy that cut the effective number of sampling steps from 1,000 to 120, achieving roughly a ninefold speedup in image synthesis.

On standard image quality metrics, the hybrid CoNeS + DDPM model came out on top. On the test set it achieved a peak signal-to-noise ratio of 27.27 dB and a mean aggregate error of 6.81 percent on tumor-containing slices, comfortably ahead of the standalone models. But the picture grew murkier when the team zoomed into the tumors themselves. Within segmented tumor volumes, performance dropped for all models, and it was worst in the enhancing tumor region — precisely the tissue that gadolinium is supposed to illuminate. The mean error in that region reached nearly 18 percent even for the best model, suggesting that the intensity distributions of synthetic tumors are shifted relative to the real thing.

The segmentation experiments drove the point home. A 3D ResNet-10 UNet, trained from scratch under five-fold cross-validation, segmented whole tumors from synthetic images with Dice scores reaching 97 to 100 percent of what real contrast images achieved. But for the enhancing tumor component — the most clinically consequential target — Dice scores with synthetic inputs ranged from just 58 to 98 percent of the real-image benchmark, with the best synthetic configuration scoring 0.739 against 0.745 for real contrast images, and the pre-contrast baseline languishing at 0.434. Wilcoxon tests confirmed that the differences were statistically significant for nearly every comparison involving the enhancing region.

The grading task told a more nuanced story. The team evaluated two families of classifiers: a radiomics pipeline that extracted eighteen handcrafted features — from sphericity and entropy to texture measures like GLCM contrast — and fed them to a random forest, and a deep learning classifier that consumed the full brain volume end-to-end. Radiomics-based models achieved higher accuracy and area under the curve, with the real-contrast configuration reaching an AUC of 0.884, while deep learning models showed higher specificity and, in several synthetic settings, higher precision. Bootstrap analysis with 10,000 iterations revealed that neither family was uniformly superior, and — crucially — that no statistically significant AUC differences separated synthetic from real contrast images for either classifier family. In other words, the synthetic images largely preserved the information needed to distinguish high-grade from low-grade gliomas.

Yet the qualitative findings exposed the limits of those aggregate statistics. In several cases the synthetic images looked superb, with the generative models’ intrinsic denoising actually producing cleaner grey-white matter differentiation than the real scans. But in other cases the models hallucinated. A ring-enhancing pattern — a hallmark of high-grade glioma — appeared around a lesion that was in fact a low-grade tumor, a fabrication that could flip a grading decision. More alarming still, in one case a small contrast-enhancing region near a surgical resection border, the very finding that would determine whether a patient’s tumor had recurred and how treatment should proceed, was missed by every model. The anatomical structures — ventricles, midline, cortex — were faithfully reproduced, but the clinically decisive enhancement patterns were not.

The authors are careful about what their results permit. The study relied on a single, class-imbalanced public dataset, lacked external validation, and did not include a formal blinded review by radiologists. Genomic and histopathological information, increasingly central to glioma management, was absent from the synthesis inputs. Still, the conclusion is unambiguous: while AI-generated contrast-enhanced MRI has advanced dramatically — particularly through the marriage of conditional neural fields and diffusion models — the synthetic images cannot yet stand in for real gadolinium-enhanced scans in clinical practice. The researchers suggest that training on larger, more diverse datasets and incorporating additional input sequences such as diffusion-weighted and susceptibility-weighted imaging may close the gap. Until then, the contrast injection — with all its costs and controversies — remains irreplaceable at the moment when it matters most.

Subject of Research: Synthetic AI-generated contrast-enhanced MRI for glioma segmentation and grading

Article Title: Downstream evaluation of synthetic AI-generated T1-weighted contrast-enhanced MR images in glioma segmentation and grading

Article References: Gkotsis, D. E., Schwarz, A. J., Wolber, J., Ferenczi, L. M., & Yang, H. (2026). Downstream evaluation of synthetic AI-generated T1-weighted contrast-enhanced MR images in glioma segmentation and grading. Frontiers in Radiology, 6, Article 1931738. https://doi.org/10.3389/fradi.2026.1931738

Image Credits: AI Generated

DOI: 10.3389/fradi.2026.1931738

Keywords: glioma, MRI, contrast agents, gadolinium, deep learning, image synthesis, diffusion models, tumor segmentation, radiomics, brain tumors, medical imaging AI, BraTS dataset

The BBB is a specialised neurovascular interface that guards the central nervous system, formed primarily by brain capillary endothelial cells welded together by tight junctions and buttressed by basement membranes, pericytes and astrocytic end-feet. Its restriction operates through distinct phenotypes. The physical barrier, built from tight junctions, limits the paracellular diffusion of ions and small molecules. The selective barrier, comprising influx transporters such as GLUT1, efflux pumps such as P-glycoprotein and the transcytosis machinery, governs transcellular traffic and actively expels many therapeutic compounds. A metabolic barrier can further degrade compounds in transit, although the review notes this aspect has not been systematically examined in the borneol literature and was left outside its scope. Because the desired direction of barrier modulation differs radically by disease, any compound that pushes permeability both ways demands careful scrutiny.

That scrutiny is exactly what the conflicting literature seemed to require. In ischaemic stroke, excessive barrier disruption fuels secondary injury, oedema and inflammatory cell infiltration, so restoring integrity is beneficial. In glioma, by contrast, the problem is the opposite: the intact barrier and blood-tumour barrier (BTB) block drug access, and transient permeability enhancement could improve delivery of chemotherapy. Borneol, remarkably, has been associated with both outcomes. It remains an essential component of classical anti-stroke formulas such as the Angong Niuhuang Pill, a Qing-dynasty preparation still used for acute cerebrovascular disease, and is classified in traditional frameworks as an ‘ushering drug’ that guides co-administered herbs toward the brain. The key question, the review argues, is not whether borneol affects the BBB but how its net effect becomes context-dependent.

The evidence from stroke models points consistently toward restoration. Ischaemia dismantles the barrier through multiple pathways: ATP depletion destabilises ion pumps and tight junctions; calcium overload activates proteases such as calpain that cleave junctional proteins; reactive oxygen species damage junctional components and activate matrix metalloproteinases that degrade the basal lamina; inflammatory cytokines alter permeability; and disruption of the MFSD2A-caveolin-1 axis increases endothelial endocytosis. A review of fifteen preclinical studies found that borneol intervenes at several of these points, restoring ATPase activity, reducing calcium load, boosting antioxidant defences and suppressing inflammatory mediators. In a rat model of middle cerebral artery occlusion, borneol treatment reduced albumin extravasation, and in an oxygen-glucose deprivation co-culture model of human cerebral microvascular endothelial cells with pericytes it reduced horseradish peroxidase transfer, two readouts of macromolecular leakage.

The most mechanistically defined pathway in the stroke context links borneol to the suppression of caveolae-mediated transcytosis via the MFSD2A and caveolin-1 (Cav-1) signalling axis. MFSD2A is a transporter for docosahexaenoic acid that maintains the distinctive lipid composition of brain endothelial membranes and thereby suppresses caveolae formation, while Cav-1 is the structural coat protein required for assembling caveolar vesicles. Under ischaemic conditions, MFSD2A is downregulated and Cav-1 upregulated, driving vesicular transcytosis and macromolecular leakage. Borneol treatment reversed both expression changes and markedly reduced endothelial vesicle counts on transmission electron microscopy. Complementary work in a permanent occlusion model showed reduced serum MMP-9, pointing to parallel protection of the extracellular matrix. The authors caution, however, that the MFSD2A/Cav-1 pathway was first identified through network pharmacology prediction and still awaits direct molecular-level confirmation.

In drug-delivery models, the picture flips. A preclinical pharmacokinetic review found that borneol co-administration increased maximum brain concentrations of twelve co-delivered compounds by 26 to 155 percent and area-under-the-curve values by 38 to 197 percent across intragastric, intranasal, intraperitoneal and intravenous routes. Critically, the effect is dose-dependent, transient and reversible within four to twenty-four hours. In rats given synthetic borneol, electron microscopy showed widening and discontinuity of endothelial tight junctions at four hours, peaking at eight, with junctional structure returning to control appearance by forty-eight hours. Junctional changes in cell monolayers resolved within twenty-four hours of removing the agent, indicating that the loosening depends on continued exposure rather than structural injury.

The clearest molecular insight into this opening mechanism comes from tight-junction dynamics. When mouse brain endothelial cells were treated with borneol at 10 micrograms per millilitre for six hours, transendothelial electrical resistance (TEER) dropped rapidly, most markedly at one hour, then began recovering by four hours. Crucially, total claudin-5 protein levels remained unchanged, but cytoplasmic claudin-5 increased significantly, and immunofluorescence confirmed redistribution of the protein from cell-cell junctions into the cytoplasm. Borneol, in other words, does not destroy a key junctional component; it displaces it from its functional position, weakening the barrier through reorganisation rather than degradation. In blood-tumour barrier studies, borneol lowered TEER, enhanced cisplatin delivery through reversible ZO-1 downregulation, and did not inhibit P-glycoprotein activity when used alone, though one study associated its barrier-opening effect with increased ICAM-1 expression, a marker that promotes leukocyte adhesion and endothelial damage.

The translational stakes are vividly illustrated by nanoparticle work. Borneol co-treatment increased the four-hour transport ratio of paclitaxel-loaded nanoparticles across an in vitro BBB model from 13.11 to 19.33 percent for PEG-coated particles and from 18.38 to 23.85 percent for particles modified with the CGKRK tumour-targeting peptide. In U87MG glioma-bearing mice, fluorescence imaging showed greater brain and tumour accumulation, and mean survival time rose from 17.4 days with saline to 38.9 days with the combination of CGKRK-modified nanoparticles and borneol. L-borneol has also been shown to transiently open the mouse BBB across doses of 100 to 900 milligrams per kilogram, enhance cisplatin brain penetration and extend survival in glioma-bearing animals, establishing the adjuvant concept in vivo.

The review’s decisive analytical move is to compare matched readouts across the two contexts. On TEER, borneol increased resistance in oxygen-glucose-deprived human endothelial monolayers but decreased it rapidly in intact mouse endothelial monolayers. On junctional structure, electron microscopy showed preservation of tight-junction ultrastructure in ischaemic models, whereas intact barriers showed claudin-5 redistribution into the cytoplasm. On macromolecular passage, albumin leakage fell in stroke models while Evans blue albumin entry rose in healthy rats, and endothelial vesicle numbers fell in stroke while nanoparticle transport ratios rose in delivery settings. Although these experiments were not conducted in a single directly comparable design, the directional contrast across matched measures is strikingly consistent: borneol shifts injured barriers toward restoration and intact restrictive barriers toward transient passage.

What remains unresolved, the authors stress, is the upstream regulatory basis of this bidirectionality. Whether the trajectory is governed by inflammatory state, endothelial metabolic stress, baseline barrier integrity or a combination of factors is not yet determined, and no single study has examined borneol under both injured and intact barrier conditions within one design. Substantial heterogeneity in model, species, dose, formulation and timing further complicates interpretation, and risks of transient BBB opening, including neuroinflammation, leukocyte trafficking and oedema, cannot be ignored. Yet this apparent flexibility may be borneol’s most interesting pharmacological property. A barrier-active compound that drives the endothelium in a therapeutically relevant direction depending on pathological context could, if its responsiveness can be brought under sufficient control, offer advantages over rigidly unidirectional agents. The next step is to identify the conditions under which its adaptability becomes a predictable and governable therapeutic property, transforming a two-thousand-year-old adjuvant into a pharmacologically controllable instrument of brain access.

Subject of Research: Context-dependent effects of the monoterpene borneol on blood-brain barrier integrity and permeability

Article Title: Comparative evaluation of borneol as a context‐dependent modulator of blood-brain barrier function: barrier-restorative or permeability-enhancing agent?

Article References: Yin, X., & Thomas, S. A. (2026). Comparative evaluation of borneol as a context‐dependent modulator of blood-brain barrier function: barrier-restorative or permeability-enhancing agent?. Frontiers in Drug Delivery, 6, Article 1934025. https://doi.org/10.3389/fddev.2026.1934025

Image Credits: AI Generated

DOI: 10.3389/fddev.2026.1934025

Keywords: blood-brain barrier, borneol, ischaemic stroke, drug delivery, tight junctions, claudin-5, MFSD2A, caveolin-1, transcytosis, TEER, glioma, traditional Chinese medicine

The research, published in Nature, was led by scientists from The Jackson Laboratory, Yale University, and collaborators spanning ten countries. Drawing on data from the international Glioma Longitudinal Analysis Consortium, known as GLASS, the team analyzed tumors from 183 patients whose cancers were sampled both before treatment and again after recurrence. The samples were collected through more than 500 surgeries performed across North America, Europe, Asia, Australia and South America, allowing researchers to directly compare how the same tumors changed over time within individual patients.

The scale of the effort matters because IDH-mutant gliomas are rare. Individual institutions rarely accumulate enough paired samples to draw statistically meaningful conclusions about how these tumors evolve. “IDH-mutant gliomas are rare, which makes them challenging to study. Through this international collaboration, we were able to gather enough samples to characterize the most common ways each type of IDH-mutant glioma changes when it recurs,” said study author Frederick Varn, a JAX assistant professor who co-led the research. By pooling tumors from dozens of centers, the consortium could identify recurring patterns of evolution rather than isolated case anecdotes.

IDH-mutant gliomas primarily strike patients in their 30s and 40s, an age group in which a brain cancer diagnosis is particularly devastating. Although many patients live for years after diagnosis, these tumors almost always return, and when they do, they tend to stop responding to the treatments that initially held them in check. Understanding the genetic events that accompany that transition has been a central goal of the neuro-oncology field, because it may reveal vulnerabilities that could be targeted before resistance becomes entrenched.

The most striking findings emerged from comparing the two main subtypes of IDH-mutant glioma: oligodendroglioma and astrocytoma. Although both are defined by the same IDH mutation, the study showed that they follow sharply divergent evolutionary paths after treatment. Oligodendrogliomas tended to acquire new point mutations in genes that help drive tumor growth, adding fresh genetic damage one alteration at a time. Astrocytomas, by contrast, were far more likely to undergo large-scale chromosomal events, gaining and losing entire stretches of DNA as the disease progressed.

“Oligodendroglioma and astrocytoma are distinct tumor types, despite sharing the IDH mutation. This study shows that differences that exist when the tumor is first detected can set the stage for what happens after they are treated,” Varn said. In other words, the biology a tumor carries at diagnosis appears to shape the menu of evolutionary strategies available to it later. A tumor’s starting genome, in this view, is not merely a snapshot of its present state but a forecast of how it will respond to the selective pressures of therapy.

The treatment itself also left measurable marks. The chemotherapy drug temozolomide, a standard therapy for these tumors, appeared to leave distinct genetic fingerprints in recurrent tumors, a signature pattern of mutations that researchers could detect in the post-treatment samples. This observation underscores a growing theme in cancer genomics: therapy is not a neutral event but an evolutionary force, and the drugs used to control a tumor can actively shape the genetic changes it acquires as it fights back.

Crucially, some of the genetic changes observed at recurrence were associated with worse outcomes for patients. In astrocytomas, large-scale chromosomal abnormalities, including the duplication of entire sets of chromosomes and catastrophic events in which chromosomes broke apart and reassembled, were linked to shorter survival after recurrence. In oligodendrogliomas, new genetic changes that emerged as tumors returned, including alterations linked to uncontrolled cell growth, were likewise associated with poorer prognosis. These associations suggest that the mode of evolution is not just biologically interesting but clinically meaningful, potentially helping physicians identify which patients face the greatest risk once their disease returns.

The findings arrive at a moment of real momentum for the field. In 2024, the FDA approved vorasidenib, an IDH inhibitor, for certain patients with these tumors, marking a major therapeutic advance. While the new study focused specifically on the changes tumors undergo after radiation and chemotherapy rather than after IDH inhibition, its results can be applied to tailor treatment approaches to specific glioma types and to the genetic changes each type is likely to acquire as it evolves. A clinician who knows that an astrocytoma is prone to wholesale chromosomal upheaval, for example, may monitor and treat it differently than one facing an oligodendroglioma that accumulates targeted mutations.

“Our results and datasets provide a reference that helps understand how current and new therapies alter tumor behavior. Ultimately, we hope this will enable us to not just observe but to direct tumor behavior and make therapies more effective that way,” said Roel Verhaak, Ph.D., vice chair for research and professor of neurosurgery at Yale School of Medicine. That ambition, steering tumor evolution rather than merely documenting it, represents a shift in how researchers think about treating cancers that recur. The study was conducted through the GLASS Consortium with researchers from institutions across North America, Europe, Asia and Australia, and was funded by the National Institutes of Health along with additional public, philanthropic and institutional sources. For patients facing a disease that almost always returns, the work offers something concrete: a genetic map of what recurrence actually looks like, and a foundation for anticipating it.

Subject of Research: Genomic evolution of recurrent IDH-mutant gliomas

Article Title: What happens when brain tumors return? International study tracks genetic changes behind recurring cancers

Article References: What happens when brain tumors return? International study tracks genetic changes behind recurring cancers. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: glioma, IDH mutation, astrocytoma, oligodendroglioma, tumor recurrence, genomic evolution, temozolomide, GLASS Consortium, Jackson Laboratory, Yale University, Nature, treatment resistance