The concept, which the team calls LUA-ECG for left upper arm ECG, rests on a simple physiological observation. The electrical activity of the heart does not stop at the torso; it propagates through the body as volume-conducted fields that can be detected wherever conductive tissue meets an electrode. Clinicians have long known that arm-based recordings show recognizable cardiac waveforms, and commercial armband monitors have exploited this for basic rhythm detection. What has been missing is diagnostic depth. A single bipolar arm lead can tell you whether the heart is beating regularly, but the standard 12-lead ECG earns its clinical authority by viewing the heart’s electrical vector from twelve different angles, revealing the spatial signatures of infarction, hypertrophy, conduction block, and dangerous repolarization abnormalities.

The Taiwanese team’s answer is to multiply the viewing angles on the arm itself. Their prototype wraps a ring of electrodes around the circumference of the left upper arm, capturing 28 differential lead signals simultaneously. Because each pair of electrodes samples the cardiac field from a slightly different orientation around the limb, the ensemble of 28 signals encodes a rich, multi-directional picture of the heart’s electrical activity. From this pool, an algorithmic selection procedure identifies eight key differential leads whose waveforms best correspond to the eight independent components of the conventional 12-lead configuration, allowing the system to reconstruct the familiar clinical leads without any torso placement. The approach was refined with a matching strategy that optimizes the assignment between arm-derived signals and standard leads, drawing on established assignment-algorithm techniques from the engineering literature.

To test whether the reconstructed signals could stand up to clinical scrutiny, the researchers ran a two-phase validation in thirty healthy adults. In the first phase, each volunteer wore the arm-based system while a standard 12-lead ECG was recorded simultaneously, providing a beat-by-beat gold standard for comparison. In the second phase, the team quantified how faithfully the eight selected LUA-ECG signals reproduced the diagnostic features that cardiologists actually measure: the RR interval governing heart rate, the PR interval reflecting conduction through the atrioventricular node, the QRS duration marking the speed of ventricular depolarization, and the QT and corrected QT intervals that gauge the heart’s electrical recovery and flag risk of lethal arrhythmias when prolonged.

The results were strikingly consistent. Waveform correlations between the arm-derived signals and their standard counterparts ranged from 0.72 to 0.90 on average, indicating substantial to strong agreement in morphological shape across the reconstructed leads. More important for clinical use, the mean absolute errors in the timing measurements were tight: between 4.57 and 7.77 milliseconds for all five interval measurements, recorded at a sampling rate of 500 hertz. To put those numbers in context, cardiologists typically regard interval differences on the order of 10 to 20 milliseconds as measurement noise, and QTc prolongation thresholds that trigger drug-safety concern sit hundreds of milliseconds away from these error bars. An armband that keeps interval errors under eight milliseconds is operating well within the tolerance that matters for screening and long-term trend monitoring.

The significance of this precision becomes clear when you consider what current wearables actually deliver. Consumer smartwatches typically record a single lead, enough to detect atrial fibrillation but blind to the spatial patterns that localize a heart attack or reveal inherited conduction disease. Researchers have tried to bridge the deficit with artificial intelligence, training neural networks to reconstruct 12-lead waveforms from one, two, or three patch leads, with recent work using masked autoencoders and LSTM architectures. Those approaches show promise but inherit a fundamental limitation: information that was never recorded must be inferred, and the inferences can fail precisely in the abnormal hearts where diagnostic accuracy matters most. The LUA-ECG strategy takes a different path, capturing genuinely independent multi-orientation signals in hardware rather than hallucinating missing leads in software.

The left upper arm is also a deliberately practical choice of location. Unlike the chest, the arm is accessible, comfortable, and compatible with clothing, making it feasible for a device to be worn for days rather than minutes. The upper arm sits far enough from the powerful skeletal muscle of the forearm to reduce motion artifact, yet close enough to the torso that the cardiac field remains strong. Prior studies from other groups have mapped how bipolar lead quality varies around the mid-upper-arm circumference and have demonstrated that armband devices can sustain usable ECG recordings during daily life. The new work extends that foundation from single-lead rhythm monitoring to multi-lead morphology, which is the real dividing line between fitness gadgets and diagnostic instruments.

Long-duration monitoring is where the clinical payoff could be largest. The classic Holter monitor, introduced in the 1960s, still relies on chest electrodes and wires, and patient adherence degrades quickly over multi-day recordings. Intermittent arrhythmias, QT prolongation under psychotropic medication, and silent ischemic episodes all demand exactly the kind of continuous, unobtrusive capture that a comfortable armband enables. Because the LUA-ECG system reproduces the temporal intervals of the standard ECG with millisecond-level accuracy, it could in principle support not just rhythm surveillance but drug-safety monitoring for QTc prolongation, a use case where regulatory agencies already accept ambulatory ECG evidence. The researchers frame their system as a reliable approach for extended-duration monitoring applications, and the healthy-subject validation is the necessary first proof of concept.

Cautions remain, and the authors are appropriately measured. The study enrolled thirty healthy adults, whose clean signals and normal anatomy represent the easiest possible test case. Patients with myocardial infarction, bundle branch block, ventricular hypertrophy, or distorted torso geometry may shift the cardiac vector in ways that alter how standard leads map onto arm orientations, and the lead-selection algorithm will need validation in these populations before any clinical claim can be made. Motion artifact, sweat, electrode-skin impedance, and day-long wear stability, the perennial enemies of wearable ECG, were not the focus of this controlled recording session. The team also holds a pending U.S. patent on the physiological signal measuring method, suggesting a commercialization pathway, but regulatory clearance for diagnostic use would require substantially larger and more diverse trials.

Even with those caveats, the study marks a genuine engineering milestone: the demonstration that the informational content of the 12-lead ECG, long considered inseparable from the ten-electrode torso ritual, can be substantially recovered from a single band on one arm. The work was funded by Taiwan’s National Science and Technology Council and conducted with cardiologists at Cathay General Hospital in Taipei under institutional ethics approval. If subsequent studies in cardiac patients confirm what healthy volunteers have shown, the familiar image of the wired ECG patient could give way to something far simpler: a discreet armband, worn through ordinary life, quietly recording the heart from every angle a cardiologist would want to see.

Subject of Research: Wearable multi-electrode left upper arm system for 12-lead ECG measurement in healthy subjects

Article Title: Measurement of 12-Lead ECG Using Multi-Orientation Electrodes on the Left Upper Arm in Healthy Subjects

Article References: Wu, S.-Y., Lu, S.-H., Lin, W.-C., Lin, W.-C., Chen, M.-F., Tsai, C.-L., Ko, W.-C., & Lin, K.-P. (2026). Measurement of 12-Lead ECG Using Multi-Orientation Electrodes on the Left Upper Arm in Healthy Subjects. Annals of Biomedical Engineering. https://doi.org/10.1007/s10439-026-04390-5

Image Credits: AI Generated

DOI: 10.1007/s10439-026-04390-5

Keywords: wearable ECG, 12-lead electrocardiogram, left upper arm ECG, cardiac monitoring, biomedical engineering, arrhythmia detection, QT interval, Holter monitoring, electrodes, signal processing, cardiology, health technology

Diminished ovarian reserve describes a situation in which the ovary holds fewer remaining follicles than expected for a woman’s age, often accompanied by reduced levels of anti-Müllerian hormone, or AMH, in the blood and a disappointing yield of eggs during in vitro fertilization cycles. For the millions of women who face this diagnosis, treatment options remain limited largely because the underlying biology has stayed elusive. While chromosomes, genetics and blood flow to the ovary have all been implicated, researchers have increasingly turned their attention to the follicular microenvironment, the local soup of cells, fluids and signaling molecules in which each egg matures. It is here, the new study suggests, that a specific and potentially targetable form of cellular damage may be at work.

Ferroptosis is not ordinary cell death. Unlike apoptosis, the tidy, programmed dismantling of a cell, ferroptosis is a violent chemical cascade in which iron catalyzes the peroxidation of lipids in cell membranes, literally rusting them from within. The process is held in check by glutathione peroxidase 4, or GPX4, an enzyme that repairs oxidized lipids and is considered the central guardian against ferroptotic collapse. When GPX4 activity falters or oxidative pressure overwhelms it, membranes rupture and the cell dies in a way that floods surrounding tissue with inflammatory signals. Because the ovary is rich in iron and because developing follicles are metabolically demanding, ferroptosis has emerged as a compelling suspect in ovarian dysfunction, but its role in DOR had never been directly examined in human follicular cells until now.

The research team, led by Nadiye Koroglu of Acibadem Mehmet Ali Aydinlar University and Aylin Yaba of Yeditepe University Faculty of Medicine, recruited 81 women undergoing intracytoplasmic sperm injection, a form of IVF in which a single sperm is injected directly into an egg. Forty-six of the participants had diminished ovarian reserve while 35 had normal ovarian reserve, serving as controls. During egg retrieval, the researchers collected cumulus cells, the specialized support cells that cling to the oocyte and supply it with nutrients, metabolic intermediates and developmental signals. They also sampled follicular fluid, the liquid that bathes the growing egg inside its follicle. Because cumulus cells share a intimate metabolic dialogue with the oocyte, damage to these cells can translate directly into compromised egg quality, making them an ideal window into follicular health.

The molecular readouts were revealing. Using quantitative reverse transcription polymerase chain reaction, the team measured the expression of ferroptosis-associated genes and found that both GPX4 and EMP1 were significantly elevated in the cumulus cells of DOR patients compared with controls. At first glance, higher GPX4 might seem protective, but the authors interpret this upregulation as a compensatory response: the cells appear to be mounting a defense against rising lipid peroxidation pressure, a molecular cry for help that indicates the ferroptosis machinery has been activated. Consistent with this interpretation, measurements of reactive oxygen species, or ROS, in the follicular fluid showed significantly higher concentrations in the DOR group, confirming an oxidatively stressed environment around the developing eggs. Notably, ferritin levels, a marker of iron storage, did not differ between the groups, suggesting that the oxidative damage in DOR is not simply a story of iron overload but of a broader redox imbalance.

The oocytes themselves told a parallel story of energetic decline. Using MitoTracker fluorescence, a dye that accumulates in active mitochondria in proportion to the electrical charge across their membranes, the researchers assessed germinal vesicle-stage oocytes, the immature eggs whose nuclear material is still enclosed. In eggs from women with DOR, mitochondrial membrane potential-related fluorescence was significantly reduced. This matters because the mitochondrial membrane potential is the engine of cellular energy production; a weakened potential means less ATP generation, poorer calcium handling and impaired completion of meiosis, all of which compromise the egg’s ability to be fertilized and develop into a viable embryo. Mitochondrial dysfunction has long been associated with reproductive aging, and this study provides direct evidence that it accompanies diminished ovarian reserve in human eggs retrieved during treatment.

Intriguingly, the study also probed the Hippo signaling pathway, an ancient growth-control network whose components, including MST1, LATS2 and YAP1, regulate organ size, cell proliferation and follicle activation. At the level of gene transcription, the Hippo pathway appeared unchanged: messenger RNA levels of MST1, LATS2 and YAP1 in cumulus cells did not differ between the DOR and control groups, whether measured directly or after the researchers manipulated ferroptosis in laboratory culture. But when the team turned to immunofluorescence staining to visualize the proteins themselves, a different picture emerged. Cumulus cells from DOR patients showed a significantly increased nuclear ratio of phosphorylated YAP to total YAP, along with elevated phosphorylated LATS1/2, while phosphorylated MST1 trended downward. These post-translational modifications indicate that DOR may modulate Hippo and YAP signaling not by changing how much of the pathway is produced, but by chemically altering the proteins after they are made, shifting their location and activity within the cell.

The technical achievement of the study lies in this multi-layered approach. By combining gene expression analysis, protein localization through immunofluorescence with DAPI-stained nuclei, biochemical assays of ROS and ferritin in follicular fluid, and live-cell fluorescence imaging of oocyte mitochondria, the researchers built a converging line of evidence from independent angles. Each measurement on its own could be dismissed as noise, but together they sketch a coherent mechanism: oxidative stress rises in the follicular fluid of DOR patients, cumulus cells respond by upregulating ferroptosis-defense genes, the Hippo pathway is re-tuned at the protein level, and the oocytes they support suffer measurable mitochondrial weakening. The slight, non-significant rise in intracellular ROS within cumulus cells themselves hints that the cells are under strain but have not yet crossed the threshold of overt damage, a snapshot of a process caught in progress.

The clinical implications are tantalizing, though the authors are careful to frame their findings as a foundation for further work rather than a treatment blueprint. If ferroptosis-associated oxidative stress genuinely contributes to the decline of egg quality in DOR, then interventions that shore up antioxidant defenses, such as GPX4-supporting compounds, iron chelators or lipid peroxidation inhibitors, could in principle protect the follicular microenvironment. The post-translational changes in Hippo signaling add a second potential lever, since YAP activity is known to influence follicle growth and activation, and pharmacological modulation of this pathway is an active area of reproductive research. Before any of that becomes reality, however, the findings will need to be replicated in larger and more diverse cohorts, and the causal direction will need to be established: whether ferroptotic stress drives diminished ovarian reserve or is merely a consequence of it remains the pivotal open question.

What the study undeniably delivers is a molecular portrait of a condition that has long been defined only by numbers, fewer follicles, lower AMH, fewer eggs retrieved. Behind those numbers, the research reveals a follicular ecosystem under oxidative siege, its support cells activating ancient cell-death defenses and its eggs running low on mitochondrial power. Part of this work was presented at the 41st Annual Meeting of the European Society of Human Reproduction and Embryology in Paris in 2025, and the full study, funded by the Health Institutes of Turkey, is now open access, allowing clinicians and researchers worldwide to scrutinize the data. For women facing a DOR diagnosis, the research does not yet offer a therapy, but it does offer something arguably just as valuable: a specific, testable biological mechanism, and with it, a genuine target for the next generation of fertility research.

Subject of Research: Ferroptosis-associated oxidative stress and mitochondrial alterations in the follicles of women with diminished ovarian reserve

Article Title: Ferroptosis-associated oxidative stress in cumulus cells and mitochondrial alterations in oocytes of women with diminished ovarian reserve

Article References: Koroglu, N., Dogan, S., Aydin, T., Bican, G., Kilic, E., & Yaba, A. (2026). Ferroptosis-associated oxidative stress in cumulus cells and mitochondrial alterations in oocytes of women with diminished ovarian reserve. Journal of Ovarian Research. https://doi.org/10.1186/s13048-026-02265-w

Image Credits: AI Generated

DOI: 10.1186/s13048-026-02265-w

Keywords: ferroptosis, diminished ovarian reserve, cumulus cells, oxidative stress, oocyte, mitochondria, Hippo signalling pathway, GPX4, YAP, reactive oxygen species, fertility, IVF

The report, led by Freya Townsend of Wear Referrals Veterinary Hospital in Stockton-on-Tees, United Kingdom, describes three dogs in which metastatic mast cell tumours produced polyostotic vertebral lesions, meaning tumour deposits scattered across multiple vertebrae rather than a single site of bone destruction. According to the authors, only one previous report in the veterinary literature has documented polyostotic mast cell tumour metastasis affecting the vertebrae. That scarcity is precisely what makes the new series noteworthy: it suggests that mast cell tumours deserve a place on the differential diagnosis list for extradural and polyostotic vertebral disease, a consideration that could influence biopsy decisions, staging protocols and prognostic conversations with owners.

The first case involved an eight-year-old neutered male small crossbreed presented originally for a small, raised mass on the upper lip. Histopathology classified it as a grade II tumour under the traditional Patnaik system and low grade under the more prognostically stringent Kiupel system, with a low mitotic count of one mitosis per ten high-power fields. The mass was excised completely, albeit with a narrow deep margin. Seven months later, the right mandibular lymph node was enlarged and biopsy confirmed metastatic mast cell disease. Fine needle aspirates of the opposite lymph node, liver and spleen were reassuringly negative at that stage, and the dog was started on lomustine chemotherapy. When a subcutaneous nodule later appeared at the surgical site and cytology again confirmed metastasis, treatment was switched to toceranib phosphate before referral to a specialist oncology service.

At the referral hospital, computed tomography revealed enlargement of the right mandibular and bilateral medial retropharyngeal lymph nodes, and surgical removal of these nodes confirmed overt metastatic mast cell infiltration, graded HN3 under the Weishaar classification system for nodal mast cell metastasis. The dog then received vinblastine chemotherapy combined with prednisolone, followed by a five-day course of radiotherapy totalling 20 Gy to the surgical site. One week after radiotherapy ended, the dog became reluctant to exercise. Neurological examination revealed pain on palpation of the thoracolumbar region, and within another week the dog had developed a hunched spine and ambulatory paraparesis, with absent postural reactions in the pelvic limbs. Blood work showed elevated C-reactive protein alongside neutropenia, leukopenia and reduced platelets, consistent with both inflammatory disease and chemotherapy side effects.

The second case was a fifteen-year-old neutered female springer spaniel whose story began with removal of a two-centimetre subcutaneous mass from the right caudal mammary gland. That tumour carried a high mitotic count exceeding ten mitoses per ten high-power fields, and the regional lymph node was already invaded by sheets of hyperchromatic, intermediately differentiated mast cells, classified HN2. The mass tested negative for a c-KIT mutation. Seven months after surgery, a new groin mass and imaging findings of enlarged abdominal lymph nodes and liver nodules confirmed widespread metastasis, and vinblastine with prednisolone was initiated. Two months into chemotherapy, the dog developed ambulatory paraparesis that progressed to non-ambulatory paraparesis by the time of specialist assessment, with pain suspected on palpation of the cervical spine and neurological signs localising to the mid-thoracic spinal cord segments.

The third patient, a thirteen-year-old neutered male Nova Scotia Duck Tolling Retriever, had a long history of mast cell tumours, including a recurrent low-grade shoulder mass and, more recently, a twenty-millimetre high-grade Kiupel tumour on the mid back with narrow surgical margins. Abdominal ultrasound had already shown nodular changes in the liver and spleen. The dog completed a twelve-week course of vinblastine and prednisolone but skipped restaging, and seven months later returned with a one-week history of non-ambulatory paraparesis, a large thoracic wall mass in the armpit region, and pain on palpation of the cranial thoracic spine. Neurological deficits again localised to the T3-L3 spinal cord segments, while reduced withdrawal reflex in the right forelimb was attributed to the large mass itself.

Magnetic resonance imaging of the vertebral column in all three dogs revealed the series’ defining finding: multiple lesions across many vertebrae, ranging from well-defined nodules to ill-defined patches, accompanied by similar lesions in the iliac bones of all three dogs and, in some, the ribs, sternebrae and scapula. Technically, every vertebral lesion appeared mildly hyperintense to isointense relative to the spinal cord on both T2-weighted and T1-weighted sequences, hyperintense on short tau inversion recovery sequences, and showed moderate homogeneous enhancement after gadolinium contrast. Vertebral shape was preserved, but cortical osteolysis was present in all cases, and several lesions breached the cortical margins to invade the extradural space and perivertebral soft tissues, compressing the spinal cord from moderately to severely. All three dogs were humanely euthanised following imaging, at the request of their caregivers.

Post-mortem histopathology cemented the diagnosis. In the first two dogs, neoplastic round cells filled the intertrabecular spaces of affected vertebrae, effacing normal haematopoietic tissue and, in the first case, extending into adjacent skeletal muscle and the extradural space, with mitotic counts of 49 and 5 per ten high-power fields respectively. Immunohistochemistry proved decisive: the neoplastic cells stained positive for CD117, a marker of mast cells, with c-KIT staining patterns II and III, while negative CD3 and negative CD20 or CD79A staining excluded both T-cell and B-cell lymphoma. In the third dog, the vertebral body itself showed only inflammatory change, but the extradural mass at T5 displayed a neoplastic round cell population with a mitotic count of 30 per ten high-power fields and confirmatory CD117 positivity. Splenic and hepatic metastasis was confirmed in the first two dogs, underscoring that none of the three had isolated spinal disease.

The imaging signature carries practical weight. In a retrospective study of sixty dogs with vertebral column tumours, preservation of vertebral shape, homogeneous contrast enhancement and lesions centred on bone were features associated with round cell neoplasms, a group that includes mast cell tumours, lymphoma, multiple myeloma, plasma cell tumours and histiocytic sarcoma. The signal intensities reported here, iso- to mildly hyperintense on T1 and T2 weighting, resemble those described for vertebral multiple myeloma and lymphoma, meaning mast cell metastasis can mimic its more familiar round cell cousins. Notably, the authors highlight that their T1 and T2 findings differ from the only prior polyostotic case, in which lesions were T1 hypointense and T2 hyperintense, adding to the recognised variability of mast cell tumour metastases.

What distinguishes this series from earlier reports is the pattern of bone involvement. Previous descriptions of mast cell tumours invading bone have mostly involved local infiltration or lysis near the primary tumour, such as a subcutaneous tumour invading the stifle joint and adjacent tibial plateau, or a disseminated tumour eroding the sphenoid bone and causing blindness. Most reported spinal mast cell tumours have been extradural masses compressing the cord without touching the vertebrae at all. By contrast, the three dogs described here developed distant metastatic deposits as multiple focal lesions in separate vertebrae, a haematogenous spread pattern more reminiscent of carcinoma or multiple myeloma than of the typical mast cell tumour. The authors conclude that metastasis should be actively considered in any dog with a previously diagnosed mast cell tumour that develops spinal pain, and that mast cell disease belongs among the differentials for polyostotic, extradural vertebral lesions, a message that may prompt earlier biopsy and more complete staging in similar patients.

Subject of Research: Metastatic mast cell tumours causing polyostotic vertebral lesions in dogs

Article Title: Metastatic mast cell tumours causing polyostotic vertebral lesions in three dogs

Article References: Metastatic mast cell tumours causing polyostotic vertebral lesions in three dogs. (n.d.). https://doi.org/10.1186/s44356-025-00050-3

Image Credits: AI Generated

DOI: 10.1186/s44356-025-00050-3

Keywords: canine, mast cell tumour, metastasis, vertebral lesions, MRI, histopathology, veterinary oncology, spinal tumour, polyostotic, CD117, round cell tumour, paraparesis

The research team, led by Marceli Pacan and Agnieszka Kuźniar of The John Paul II Catholic University of Lublin, set out to address a persistent puzzle in agricultural biotechnology: why microbial biostimulants often deliver inconsistent results in the field. Wheat is the world’s third most significant crop and accounts for roughly half of all cereal cultivation in the European Union, so even modest improvements in how beneficial microbes perform could have enormous consequences. The scientists compared two modern winter wheat cultivars, ‘Hondia’ and ‘Tytanika’, with ‘Rokosz’, an ancient spelt wheat valued for its higher protein and mineral content and its adaptation to low-input farming. Their central hypothesis was that the metabolic response to endophyte colonization would depend on the host genotype—a proposition the data dramatically confirmed.

The experimental design was meticulous. In laboratory tests, the researchers first optimized the bacterial cell density and temperature for inoculation, finding that 6.67 × 10⁴ cells per milliliter at 15 to 20 degrees Celsius produced the best seedling vigor. They also screened six biochar carriers made from pine or ash wood at different pyrolysis temperatures, ultimately selecting an ash-wood biochar designated J500, which supported normal seedling growth across all cultivars and proved compatible with commercial fungicide seed dressings. This compatibility matters because real-world farmers apply fungicides alongside seed treatments, and a biopreparation that fails under those conditions would be useless in practice.

The heart of the study was a 40-day greenhouse pot experiment in which inoculated and control seeds of all three cultivars were grown in non-sterilized soil, deliberately preserving the natural soil microbiota. Every five days, the researchers measured the activities of two key defense enzymes: peroxidase (POX), which drives cell wall cross-linking and reactive oxygen scavenging, and phenylalanine ammonia-lyase (PAL), the gateway enzyme of the phenylpropanoid pathway that produces phenolic acids, flavonoids, and lignin precursors. The patterns that emerged were anything but uniform. In ‘Hondia’, POX activity stayed suppressed for most of the experiment and then surged fifteen-fold within just five days at day 40, while PAL activity collapsed to an eleven-fold reduction below control levels. In ‘Rokosz’, the opposite occurred: PAL activity climbed to 11.81 IU per gram fresh weight at day 20—the highest value recorded across all treatments—and remained elevated through the end of the experiment, while POX stayed quietly low.

Untargeted metabolomics using liquid chromatography coupled to quadrupole time-of-flight mass spectrometry revealed the chemistry behind these enzymatic divergences. At day zero, just four days after inoculation, both cultivars shared a common set of exclusively induced metabolites pointing to rapid defense priming: oxidized glutathione, L-tryptophan, allantoin, the coumarin scopoletin, and 12-oxo-phytodienoic acid, a jasmonic acid precursor that functions as a stress-signaling molecule. This early signature demonstrates that the endophytes trigger metabolic activation almost immediately, before any visible growth differences appear. Yet even at this earliest stage, the genotypes differed in character. Spelt ‘Rokosz’ showed a balanced distribution of up- and down-regulated molecular features, whereas ‘Hondia’ displayed a pronounced wave of induction in its non-polar, lipid-rich fraction, with 980 differential features compared to 419 in ‘Rokosz’.

By day 40, the two crops had committed to fundamentally different strategies. Modern ‘Hondia’ accumulated wax esters at extraordinary levels—palmityl palmitate increased nearly forty-fold, and several other long-chain wax esters appeared exclusively in treated plants—suggesting reinforcement of the cuticle, the plant’s outer physical barrier. Alongside this, the cultivar showed extensive remodeling of membrane lipids, including galactolipids and sterol esters, broad accumulation of free amino acids such as tryptophan and N-γ-glutamyl-phenylalanine, and shifts in indole-related metabolism consistent with altered hormonal regulation. Meanwhile, several phenylpropanoid-derived flavonoids were strongly down-regulated, indicating that ‘Hondia’ had shifted away from chemical defense toward enzymatic oxidative management and structural reinforcement.

Ancient spelt ‘Rokosz’ told a different story. Treated plants showed significant enrichment of the phenylpropanoid pathway: p-coumaric acid rose 5.3-fold, trans-cinnamaldehyde 4.6-fold, and compounds such as chlorogenic acid, esculetin, and the flavonoid rutin appeared exclusively in biopreparation-treated samples. HPLC profiling confirmed that vanillic acid and total phenolic acid concentrations increased significantly in ‘Rokosz’ but not in ‘Hondia’. Remarkably, ‘Rokosz’ also accumulated ferulic acid in its leaves at levels comparable to its roots—an unusual pattern for cereals, where phenolics typically concentrate in roots and cell walls—potentially enhancing ultraviolet screening and photoprotection in photosynthetic tissue. The picture is of an ancient grain that channels the microbial partnership into building a dense chemical shield.

Perhaps the most striking statistical finding was convergence: biopreparation treatment reduced the metabolic differences between the two cultivars from 760 differential traits under control conditions to 423 after treatment, a 44.3 percent reduction. In other words, the same microbial inoculant pulled both genotypes toward a shared, microbially induced metabolic state, even as each retained its characteristic strategy. The researchers interpret this as evidence that subspecies-level genetic architecture determines whether endophyte colonization triggers metabolic enhancement, as in spelt, or metabolic moderation, as in modern bread wheat—a distinction consistent with prior work showing that endophytes from wild wheat ancestors can make drought-stressed plants metabolically resemble well-watered controls.

For agriculture, the implications are direct and potentially transformative. Microbial inoculants are increasingly promoted as environmentally friendly alternatives to synthetic fertilizers and pesticides, but their field performance is notoriously variable. This study suggests that much of that variability may stem from genotype-treatment interactions rather than product quality: the identical formulation can produce profoundly different metabolic outcomes depending on the crop’s genetic background. The authors argue for precision agrobiotechnology, in which microbial consortia are matched to individual cultivars and metabolic markers—phenylpropanoids, wax esters, and oxylipin signals such as 12-OPDA—are incorporated into breeding programs to select genotypes with enhanced microbial responsiveness. As climate pressures intensify and the need to reduce chemical inputs grows, understanding these hidden metabolic dialogues between crops and their endophytic partners may prove essential to feeding the world sustainably.

Subject of Research: Genotype-specific metabolic and enzymatic responses of wheat and spelt to an endophytic bacterial biopreparation

Article Title: Endophyte-induced metabolic divergence in wheat genotypes: implications for plant–soil interactions

Article References: Endophyte-induced metabolic divergence in wheat genotypes: implications for plant–soil interactions. (n.d.). https://doi.org/10.1007/s11104-026-09133-y

Image Credits: AI Generated

DOI: 10.1007/s11104-026-09133-y

Keywords: wheat, spelt, endophytes, metabolomics, plant–soil interactions, biostimulants, phenylpropanoid pathway, PAL, POX, biochar, Bacillus, sustainable agriculture

The meeting opens with a high-profile moment on Sunday, September 27, when John D. Halamka, the Dwight and Dian Diercks president of Mayo Clinic Platform, joins the Presidential Symposium in a session titled Data to Dialogue: Communicating Radiotherapy’s Value to Advance Care. His brief panel appearance, scheduled from 9:49 to 9:53 a.m. EDT, addresses a question that has moved from the margins to the mainstream of scientific discourse: how will science and healthcare information be disseminated tomorrow? The session examines the evolving roles of journals, websites, online platforms, and AI-powered content in shaping what both physicians and patients come to know about radiotherapy. It is a fitting opening note for a meeting in which machine intelligence threads through nearly every major Mayo Clinic presentation, from automated contouring to recurrence prediction.

Perhaps the most clinically consequential Mayo Clinic contribution comes in a Sunday afternoon clinical trials session, when Nadia Laack, a radiation oncologist at Mayo Clinic Comprehensive Cancer Center in Rochester, presents results from the Children’s Oncology Group trial AEWS1221. The study investigated stereotactic body radiation therapy, known as SBRT, combined with comprehensive metastasis-directed therapy in patients with metastatic Ewing sarcoma, an aggressive bone and soft-tissue cancer that has long defied conventional treatment when it spreads. The headline finding is remarkable: the researchers report the highest event-free survival recorded to date in cooperative-group trials of metastatic Ewing sarcoma. Equally important, the study establishes the feasibility and safety of delivering precisely targeted, ablative radiation doses to metastatic sites in this young patient population, a strategy that treats each site of spread as a targetable lesion rather than relying solely on systemic chemotherapy.

Artificial intelligence takes a starring role later that afternoon, when Andres Portocarrero Bonifaz, a radiation oncology medical physicist at Mayo Clinic in Florida, presents on AI-driven automation in pelvic radiation therapy during an education session on artificial intelligence, robotics, and emerging technologies in gynecologic radiation oncology. His talk traces the full pipeline of automation, from automated contouring of organs at risk to treatment plan optimization, and then extends into prediction-guided care. The central idea is that artificial intelligence can identify each patient’s individual risk of side effects before treatment begins, giving physicians a quantitative foundation for more personalized decisions about how aggressively to treat and how to protect healthy tissue. In gynecologic cancers, where pelvic targets sit close to bowel, bladder, and reproductive structures, that kind of predictive personalization could translate directly into fewer long-term toxicities.

Monday morning brings a rapid-fire quick pitch from Diego Santos Toesca, a radiation oncologist at Mayo Clinic in Arizona, on one of the most lethal malignancies in medicine: glioblastoma in older adults. His presentation describes patterns of radiographic failure at first progression among older patients treated with hypofractionated proton therapy guided by both MRI and 18F-DOPA PET imaging. The technical significance lies in the fusion of two imaging modalities: standard MRI delineates anatomical abnormality, while 18F-DOPA PET reveals metabolic activity that can expose tumor infiltration invisible to conventional scans. By targeting the combined imaging signature with a shortened course of proton therapy, which spares surrounding brain tissue from excess radiation dose, the team is mapping exactly where and how these tumors recur. Those failure patterns will inform the next generation of target volumes for a disease where every additional week of survival is hard-won.

Also on Monday, Michael Grams, a radiation oncology medical physicist at Mayo Clinic Comprehensive Cancer Center in Rochester, will describe Mayo Clinic’s first-in-human clinical translation of minibeam radiotherapy as part of an education session on spatially fractionated radiation therapy. Minibeam radiotherapy is one of the most conceptually radical ideas in the field: instead of delivering a broad, uniform radiation field, the beam is divided into many narrow, parallel microbeams separated by untouched tissue. The alternating pattern of dose appears to exploit a differential biological response, damaging tumor tissue while allowing normal tissue between the beams to recover and repair. Presenting the clinical workflow and translational insights from the first patients ever treated with this technique is a milestone moment, marking the journey of a physics concept from laboratory benches and animal models into the clinic for difficult-to-treat cancers.

Patient voice enters the scientific program through Minji Lee, a scientist in radiation oncology at Mayo Clinic in Rochester, who presents an oral session comparing the prognostic value of disease-specific patient-reported outcome measures in prostate cancer. Patient-reported outcomes, or PROs, capture symptoms and quality-of-life data directly from patients rather than from clinician assessments, and they have become increasingly influential in oncology. But not all PRO instruments are created equal, and the study systematically compares how well different measures predict overall survival in prostate cancer. The goal is evidence-based selection: identifying which validated instruments genuinely carry prognostic information so that clinicians and trialists can choose the tools that matter most, rather than defaulting to tradition or convenience.

Pediatric hematology-oncology is the focus of Brad Hoppe, a radiation oncologist at Mayo Clinic Comprehensive Cancer Center in Florida, who presents a post hoc safety analysis of the KEYNOTE-667 study during a scientific session on radiation and systemic therapy in hematologic malignancies. The analysis evaluates maintenance pembrolizumab, an immune checkpoint inhibitor, given concurrently with radiotherapy in pediatric patients with classic Hodgkin lymphoma who showed a slow early response to front-line chemotherapy. Combining immunotherapy with radiation raises a critical safety question, because both modalities can inflame the same normal tissues, and checkpoint inhibitors can trigger immune-related adverse events. Establishing that concurrent immunoradiotherapy is tolerable in children could expand the therapeutic arsenal for the subset of Hodgkin lymphoma patients whose disease does not respond quickly to chemotherapy alone.

The afternoon plenary on Monday features one of the most anticipated randomized trials in neuro-oncology. Paul D. Brown, a radiation oncologist at Mayo Clinic Comprehensive Cancer Center in Rochester and principal investigator of the Alliance A071801 phase III trial, will report results comparing postoperative single-fraction stereotactic radiosurgery with fractionated stereotactic radiosurgery for resected brain metastases. When a brain metastasis is surgically removed, radiation to the surgical cavity reduces the risk of local recurrence, but the optimal dosing schedule has remained contested. Single-fraction radiosurgery offers convenience and a single anesthesia session, while fractionated delivery spreads the dose across multiple sessions, potentially improving tumor control at the cavity margin and reducing radionecrosis. A definitive phase III answer will directly guide treatment for the hundreds of thousands of patients who develop brain metastases each year.

The Mayo Clinic program closes with two further presentations on Tuesday, September 29. Scott Lester, a radiation oncologist at Mayo Clinic Comprehensive Cancer Center in Rochester, delivers the first report from the Headlight trials, evaluating postoperative hypofractionated proton therapy for head and neck cancers unrelated to HPV. Shortening the course of proton therapy could reduce treatment burden while proton physics, which deposits dose with no exit radiation, offers a distinct sparing advantage for the salivary glands, swallowing structures, and spinal cord that surround head and neck tumors. Shortly afterward, in a session on the automated clinic, Mariana Borras-Osorio, a research fellow in radiation oncology at Mayo Clinic in Rochester, presents a machine-learning model designed to personalize PSA follow-up after prostate radiation therapy. The model integrates patient-specific clinical data with the kinetics of prostate-specific antigen, the standard blood-based biomarker monitored after treatment, to predict clinically meaningful recurrence. Together, the presentations sketch the field’s trajectory: imaging-guided precision, AI-augmented decision-making, radically novel beam geometries, and rigorously tested treatment schedules, all converging to make radiation therapy simultaneously more powerful and more personal.

Subject of Research: Mayo Clinic radiation oncology research presentations at the ASTRO 2026 Annual Meeting

Article Title: Mayo Clinic experts present radiation oncology research at ASTRO 2026

Article References: Mayo Clinic experts present radiation oncology research at ASTRO 2026. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: radiation oncology, ASTRO 2026, Mayo Clinic, artificial intelligence, SBRT, Ewing sarcoma, glioblastoma, minibeam radiotherapy, proton therapy, brain metastases, Hodgkin lymphoma, prostate cancer

The core problem is analytical. Microplastics vary wildly in size, shape, polymer type, and degree of weathering, and they hide inside complex environmental matrices. Conventional techniques such as microscopy, Fourier-transform infrared (FTIR) spectroscopy, Raman spectroscopy, and pyrolysis gas chromatography-mass spectrometry are labor-intensive, operator-dependent, and often poorly sensitive to weathered particles and nanoplastics. Traditional statistics like regression and principal component analysis struggle with the nonlinear, spatiotemporally dynamic behavior of these particles across coupled air, water, and land systems. The review, led by Obinna Chigoziem Akakuru of Miami University and the University of Cincinnati together with a large interdisciplinary team, synthesizes how machine learning and deep learning are dismantling these bottlenecks at every stage of the pipeline.

The most mature application is automated detection. Deep-learning computer vision has converted what was once a manual art into high-throughput quantitative analysis. Instance segmentation models such as Mask R-CNN, trained on curated microscopy images of fibers, fragments, and films, have achieved test accuracies exceeding 93 percent, generating pixel-level masks that allow simultaneous counting, sizing, and shape classification. U-Net architectures excel at semantic segmentation in cluttered scenes, preserving spatial context for particle aggregates. Yet performance degrades sharply on field images with natural backgrounds of organic debris, biofilms, and minerals, and models still stumble on weathered, faded particles, tangled fiber overlaps, and particles below 20 micrometers. Tools like Gradient-weighted Class Activation Mapping are now being used to peek inside these black boxes and reveal which visual features drive classification decisions.

Spectroscopy is undergoing a parallel revolution. FTIR and Raman remain the gold standards for polymer identification, but manual spectral library matching is slow and unreliable for aged or contaminated samples. One-dimensional convolutional neural networks applied to FTIR and Raman datasets have reported classification accuracies of at least 97 percent for selected polymer classes under cross-validation, while showing resilience to spectral noise and baseline drift. Random Forest ensembles outperform traditional library matching by learning to ignore spectral regions corrupted by additives or weathering byproducts. Hyperspectral imaging coupled with three-dimensional convolutional networks promises rapid, non-destructive mapping of particles in soil and biota, though it currently works only for particles above 100 micrometers and chokes on matrix interference and gigabyte-scale data volumes.

Perhaps the most radical innovation redefines what a sensor can be. Rather than relying on a single highly specific recognition element, researchers have coupled a broadly interactive estrogen receptor layer on a plasmonic optical fiber with a k-nearest neighbors algorithm that reads the time-resolved interaction fingerprint of each particle. Because different polymers, sizes, and surface charges produce subtly different binding kinetics, the system classified unknown particles with roughly 90 percent accuracy. The specificity lives in software rather than hardware, turning the sensor into a programmable, updatable platform. Related approaches using molecularly imprinted polymers and semi-selective peptide arrays generate multiplexed patterns that machine learning decodes.

Beyond detection, AI is moving toward forecasting where pollution comes from and where it will go. Unsupervised methods such as clustering link polymer and morphological fingerprints to land-use sources like wastewater effluent, agricultural film, and textile laundering, while supervised models quantitatively apportion pollution loads using receptor modeling adapted from air quality science. Physics-informed neural networks embed governing transport equations directly into deep-learning architectures, simulating microplastic fate in rivers, estuaries, and coastal sediments with finer spatial resolution than traditional finite-element models while needing less training data. Autonomous underwater and aerial vehicles running lightweight vision algorithms can map surface concentrations in near real time, capturing post-storm pulses and spatial patchiness that grab sampling misses.

The review uses the United States as a case study in how geography and regulation shape the AI opportunity. Hotspots cluster where intense human inputs meet retention mechanisms: the Columbia-Snake river system, where reservoirs act as temporary sinks; the Gulf of Mexico, where the Mississippi-Atchafalaya system draining nearly 41 percent of the contiguous country feeds persistent accumulation zones off Texas, the Mississippi Delta, and western Florida; and the Texas coast, scarred by the 2019 Formosa Plastics nurdle spill. Yet the regulatory landscape is a decentralized patchwork. California’s SB 1422 and AB 818 have catalyzed compliance-driven AI platforms, but the absence of a unified national protocol fragments the data ecosystem, trapping innovation in region-specific tools rather than scalable national solutions.

Significant obstacles remain. Training data are scarce, fragmented, and methodologically inconsistent, with incompatible metadata, uneven geographic coverage, and divergent sampling protocols. A pronounced lab-to-field gap degrades model performance when pristine laboratory particles give way to weathered, biofouled, mineral-coated environmental samples. Black-box opacity, reproducibility failures, and proprietary algorithms erode the transparency regulators need. The authors propose a clear burden of proof: AI models should earn regulatory adoption only by beating simple baseline models under realistic validation schemes, with stable out-of-domain gains, improved decision-relevant metrics, and fully auditable pipelines. Where those conditions are unmet, interpretable low-dimensional models remain the defensible choice.

On the remediation side, AI is beginning to optimize a portfolio of complementary technologies. Biological approaches exploit microbial enzymes and microalgae such as Scenedesmus, which remove microplastics by bio-flocculation with efficiencies above 84 percent. Physical methods deploy adsorbents like magnetic carbon nanotubes, biochar, and granular activated carbon, while chemical techniques include coagulation and advanced oxidation processes. Speculative microrobotic swarms guided by particle swarm optimization could one day target cleanup in coral reefs, though durability, recovery, and cost remain prohibitive. The authors stress that remediation alone is insufficient; source reduction and circular economy policies must carry the heavier load.

The review’s bottom line is a call for a coordinated paradigm shift. Priority directions include harmonized open-access datasets spanning particle properties and environmental matrices, hybrid models coupling AI with mechanistic transport and dose-response frameworks, explainable AI for toxicology and regulation, and explicit alignment of AI outputs with WHO, EPA, and EFSA risk guidelines. For data-limited regions especially, intelligent monitoring could deliver cost-effective early-warning systems. The technology, in other words, is racing ahead; the data standards, interdisciplinary collaboration, and governance frameworks must now catch up before the promise of predictive microplastic science can be fully realized.

Subject of Research: Artificial intelligence applications for microplastic pollution monitoring, predictive modeling, and risk assessment

Article Title: Artificial intelligence for microplastic pollution monitoring, predictive modeling, and risk assessment: advances, challenges, and future perspectives

Article References: Akakuru, O. C., Ray, P. A., Onyeanwuna, U. B., Eyankware, M. O., Aigbadon, G. O., Akingboye, A. S., Iheme, K. O., Usman, A., Amadi, C. C., Ofoh, I. J., Onyekuru, S. O., Opara, A. I., Akudinobi, B. E. B., & Algeo, T. J. (2026). Artificial intelligence for microplastic pollution monitoring, predictive modeling, and risk assessment: advances, challenges, and future perspectives. Environmental Earth Sciences, 85(15), Article 386. https://doi.org/10.1007/s12665-026-13075-0

Image Credits: AI Generated

DOI: 10.1007/s12665-026-13075-0

Keywords: microplastics, artificial intelligence, machine learning, deep learning, computer vision, spectroscopy, FTIR, Raman spectroscopy, pollution monitoring, risk assessment, environmental policy, physics-informed neural networks

The research team, led by Koshiro Inoue of Health Sciences University of Hokkaido, took advantage of a natural experiment hiding in the Japanese racing calendar. They surveyed finishers at the Tsukuba Marathon in November 2022 and 2023, and at the Hokkaido Marathon in August 2023. According to data from the Japan Meteorological Agency, the mean air temperatures during those races were 6.5, 20.6, and 28.2 degrees Celsius respectively, with estimated wet-bulb globe temperatures of 11.2, 22.2, and 31.5 degrees. In the classification scheme of earlier endurance research, those conditions correspond roughly to natural, moderate heat, and extreme heat environments, giving the scientists three distinct thermal snapshots of the same 42.195-kilometre distance.

In total, 1,938 male runners provided complete datasets, well above the minimum of 1,022 required by the team’s a priori power analysis. Each participant completed an online questionnaire before racing, reporting personal best marathon time over the preceding one to three years, age, years of running experience, height and weight, and detailed training habits over the previous three months, including weekly running distance, the longest single run, and weekly running frequency. Official net finish times were then retrieved from race websites. Crucially, the runners’ profiles differed between events: those who chose the hot Hokkaido race tended to have slower personal bests, higher body mass index, longer running careers, lower weekly mileage, and less frequent training than their counterparts at the cooler Tsukuba events, a pattern the researchers had to account for statistically.

The analytical centrepiece was a hierarchical multiple regression with interaction terms, a technique that allows scientists to ask not just whether a variable predicts finish time, but whether its predictive power changes across conditions. The baseline model, containing only the runner characteristics, explained about 71 percent of the variance in finish time. Adding the race event, which served as a proxy for temperature, boosted explained variance by a striking 10.7 percent, and the regression coefficients revealed a stepwise slowing from the coolest to the hottest race. Average finish times climbed from 223 minutes in the cool Tsukuba race of 2023, to 234 minutes in the milder 2022 edition, to 281 minutes in the August heat of Hokkaido.

But the headline finding lay in the third model, where temperature significantly moderated the relationships between finish time and four specific variables: personal best time, age, running experience, and body mass index. The associations between training volume, longest run, or training frequency and finish time, by contrast, remained stable across all three thermal conditions. In other words, heat selectively rewires the importance of who the runner is, while leaving the benefits of what the runner does largely intact.

Some of the individual interactions were genuinely counterintuitive. The link between personal best and finish time, strongly positive in both cool races, became significantly flatter in the extreme heat of Hokkaido. That pattern supports the minority view in the literature, advanced by economists Gasparetto and Nesseler, that faster runners suffer proportionally greater declines in hot conditions, perhaps because they run at higher intensities and push closer to their physiological limits, while slower runners, spending longer on the course, were already expected to fare worse. The new data suggest that under extreme heat, the usual hierarchy of ability is compressed, and a fast runner may lose more time to the mercury than a slower one.

Age told a more predictable but sobering story. Older runners were slower at every temperature, but the age-related penalty grew significantly steeper in the two warmer races. This aligns with physiological evidence that thermoregulation, particularly sweating and skin blood flow responses, declines with age, and with large-scale analyses of the New York City Marathon showing that heat amplifies age-related slowing in men between 30 and 64. Body mass index followed a similar trajectory: at the optimal cool temperature it bore no significant relationship to finish time, but its influence emerged and strengthened progressively as temperatures rose, consistent with the known advantages of smaller body size for heat dissipation during distance running and with epidemiological links between higher BMI and exertional heat illness risk.

Perhaps the most curious result concerned running experience. Years of training had no bearing on finish time in the cool race, was associated with slower times in the moderate heat of 2022, and, remarkably, predicted faster finishes in the extreme heat of Hokkaido. The authors suggest that long-accumulated adaptations and racing savvy may become decisive assets precisely when conditions are harshest, making veteran experience a potential key to surviving summer marathons. Meanwhile, the training variables behaved as conventional wisdom expects: greater weekly distance and higher training frequency were reliably associated with faster finishes, and, importantly, those benefits were conferred independently of temperature, implying that the performance payoff of training volume survives even as the thermometer climbs.

The study’s authors are careful about its limits. Participants were predominantly Japanese male recreational runners who volunteered for the survey, the hot condition rested on a single August race at one venue, and self-reported measures of height, weight, and training are known to carry small biases. Course differences between events cannot be fully excluded either. Still, the statistical framework was robust, the sample comfortably exceeded power requirements, and the central conclusion is hard to escape: temperature is not merely another variable in the marathon equation but a structural determinant that alters the weight of the other predictors.

The practical implications arrive at an opportune moment. Climate analyses suggest that opportunities to race in the optimal 3 to 11 degree Celsius range for recreational runners are shrinking, with the probability of ideal conditions at the Tokyo Marathon projected to fall over the coming decades. As more amateur runners face warmer start lines, the new findings point to concrete targets: body mass index, a modifiable factor, becomes a more meaningful lever in the heat, while experience can be deliberately banked. Age, irreversible, demands respect and adjusted expectations. For coaches and runners drafting training plans for the hot races of the future, the message is that the old prediction formulas, built in temperate conditions, may need a thermal correction.

Subject of Research: How air temperature modulates the physiological and training determinants of marathon finish time in male recreational runners

Article Title: Temperature‐driven modulation of factors influencing full‐Marathon time in male recreational runners

Article References: Inoue, K., Yamaguchi, A., Nabekura, Y., Ishihara, T., & Fukuie, T. (2026). Temperature‐driven modulation of factors influencing full‐Marathon time in male recreational runners. Physiological Reports, 14(17), Article e71096. https://doi.org/10.14814/phy2.71096

Image Credits: AI Generated

DOI: 10.14814/phy2.71096

Keywords: marathon, heat stress, thermoregulation, recreational runners, body mass index, running performance, aging athletes, training volume, wet-bulb globe temperature, climate change, exercise physiology, hierarchical regression

The study, published in The Lancet Regional Health – Europe, is a secondary analysis of the Copenhagen Prospective Studies on Asthma in Childhood 2010 (COPSAC2010) randomized clinical trial. Between March 2009 and November 2010, the team enrolled 623 pregnant women from the Greater Copenhagen area at week 24 of gestation and randomized them in a 1:1 ratio to receive either high-dose vitamin D3 or placebo. Crucially, both groups were advised to follow the Danish Health Authority’s standard recommendation of 400 IU per day, so the trial effectively compared 2800 IU per day against 400 IU per day from pregnancy week 24 until one week after birth. Adherence and biological effect were confirmed by measuring maternal serum 25-hydroxyvitamin D at the end of the supplementation period using isotope dilution liquid chromatography-tandem mass spectrometry, a gold-standard analytical method.

The original trial had already produced encouraging results. In earlier analyses, the children whose mothers received the high-dose supplement showed significantly higher whole-body bone mineral density and bone mineral content at ages three and six years, measured by dual-energy X-ray absorptiometry, or DXA. The effects were largest among children born during the dark Danish winter months, when maternal vitamin D status is naturally at its lowest, and an exploratory analysis even suggested a reduced risk of radiologically verified fractures when supplementation was combined with sufficient vitamin D levels in the child’s first year of life. Those findings fed directly into the influential hypothesis that prenatal vitamin D could be a cheap, safe lever for lifelong skeletal health, because bone mineral content tracks from childhood into adulthood and peak bone mass is considered the single most important determinant of osteoporosis risk.

The new analysis tested whether those early gains endure. At age thirteen, 416 children, roughly seventy-one percent of those randomized, returned for whole-body DXA scans performed on Lunar iDXA and Hologic densitometers, with each scan validated by two independent specialists blinded to treatment allocation. The results were unambiguous. There were no differences between the high-dose and standard-dose groups in total body bone mineral content, total body bone mineral density, or the equivalent total-body-less-head measures. The adjusted mean differences were vanishingly small: for total body bone mineral content, just 6.7 grams in favor of the supplemented group, with a confidence interval spanning zero and a p-value of 0.71. Sex-stratified analyses, adjustments for pubertal Tanner stage, stratification by maternal baseline vitamin D status, birth season, and a concurrent fish-oil trial all failed to uncover any hidden subgroup benefit.

The longitudinal picture tells the same story. Pooling DXA measurements from ages three, six, and thirteen in a random-intercept mixed-effects model, the researchers found no overall effect of the prenatal intervention across the entire follow-up period, and no statistical interaction between the intervention and the child’s age at scanning. In other words, the early advantage seen at ages three and six did not simply persist quietly below the threshold of a single timepoint; it dissipated as the children grew. This pattern is consistent with a transient effect of the intrauterine vitamin D environment on early bone mineralization that is progressively overtaken by the powerful hormonal and nutritional drivers of growth during childhood and puberty.

Fracture outcomes proved equally unpersuasive. Among 550 children with complete clinical follow-up to age thirteen, a remarkable 94 percent retention rate, 114 radiologically verified fractures were recorded in 99 children, spanning the clavicle, radius, ulna, tibia, fibula, femur, and humerus. Comparing the intervention groups, the hazard ratio for time to first fracture was 0.85 with a p-value of 0.40, and the incidence rate ratio was 0.84 with a p-value of 0.39, neither approaching statistical significance. Even the exploratory combined analysis, which contrasted children whose mothers received high-dose vitamin D and who themselves had sufficient 25-hydroxyvitamin D levels at six months against children with neither advantage, showed no significant reduction in fracture risk, in contrast to the team’s own earlier findings at younger ages.

One intriguing signal did survive, however. Within the supplemented group, children who had sufficient vitamin D levels at six months tended to show higher bone mineral outcomes at thirteen than those who were insufficient, and the interaction between the prenatal intervention and early-life vitamin D status reached significance for total body bone mineral content, with a p-value of 0.002. The authors interpret this cautiously as evidence that the vitamin D status of the child in the first months of life, not merely the prenatal dose, may modulate any potential skeletal benefit. Yet this exploratory observation stops short of a clinically actionable conclusion, and the corresponding fracture analyses in the same subgroup yielded only non-significant trends.

The study’s strengths are considerable and worth emphasizing in an era of nutrition headlines built on observational associations. Randomization eliminates the confounding that plagues cohort studies, where mothers who take supplements also tend to differ in diet, activity, and socioeconomic status. The double-blinded, placebo-controlled design, predefined bone endpoints, ninety-four percent longitudinal follow-up, radiologically verified fracture diagnoses, and sensitivity analyses across scanners, seasons, sex, and puberty stage collectively make this the most definitive test to date of prenatal vitamin D effects on offspring bone health. The authors acknowledge limitations, including that the trial was originally powered for persistent wheeze and asthma rather than bone outcomes, that twenty-nine percent of children lacked a thirteen-year DXA scan, and that interpreting areal bone density during the adolescent growth spurt is inherently challenging.

The findings also sharpen the contrast with the only comparable trial. The UK-based MAVIDOS study, which tested a lower dose of 1000 IU per day, reported higher offspring bone mineral density at ages six to seven, and its longer-term results have not yet been published. Whether MAVIDOS will show the same attenuation seen in COPSAC2010 is now one of the most consequential open questions in pediatric bone research. For now, the Danish data challenge the seductive idea that a single prenatal intervention can durably sculpt peak bone mass and defer osteoporosis by more than a decade. Mathematical models have predicted that a ten percent increase in peak bone mass around age twenty could delay osteoporosis onset by thirteen years, which is precisely why the field invested so much in the pregnancy hypothesis. The new results suggest that if such a shift is achievable, it will not come from prenatal supplementation alone, and that protecting children’s vitamin D status through infancy and beyond may matter at least as much as what happens in the womb.

Subject of Research: Prenatal high-dose vitamin D supplementation and offspring bone mineral content and density at age 13 years: a secondary analysis of a randomised clinical trial

Article Title: Prenatal high-dose vitamin D supplementation and offspring bone mineral content and density at age 13 years: a secondary analysis of a randomised clinical trial

Article References: Brustad, N., Sultan, T., Vahman, N., Jensen, S. K., Vinding, R., Aagaard, K., Gørtz, P. M., Haarmark, C., Bønnelykke, K., & Chawes, B. (2026). Prenatal high-dose vitamin D supplementation and offspring bone mineral content and density at age 13 years: a secondary analysis of a randomised clinical trial. The Lancet Regional Health – Europe, 71, Article 101870. https://doi.org/10.1016/j.lanepe.2026.101870

Image Credits: AI Generated

DOI: 10.1016/j.lanepe.2026.101870

Keywords: Prenatal, high-dose, vitamin, supplementation, offspring, bone, mineral, content, density, years, secondary, analysis

The experimental agent, known as E7386, belongs to a relatively new class of medicines called protein–protein interaction inhibitors. Rather than blocking an enzyme’s active site, it physically prevents the CREB-binding protein (CBP) from docking with β-catenin, a transcriptional co-activator that sits at the heart of the Wnt signalling pathway. When the CBP–β-catenin interaction proceeds unchecked, tumour cells receive a continuous stream of pro-growth and pro-survival instructions. Preclinical work had shown that E7386 suppressed intestinal polyp formation in genetically predisposed mice and shrank tumours in models where the Wnt pathway was overactive, and that combining the drug with lenvatinib — a multi-kinase inhibitor that starves tumours of their blood supply — produced greater antitumour activity than either drug alone in endometrial cancer models, regardless of whether tumours carried mutations in the CTNNB1 gene that encodes β-catenin.

The clinical context is stark. Endometrial cancer incidence is rising worldwide, and while the addition of immunotherapy to first-line chemotherapy has transformed outcomes for many patients, roughly 15 percent of women are diagnosed with advanced-stage disease, for whom five-year survival hovers around 15 to 17 percent. Once a patient’s tumour progresses after platinum-based chemotherapy and an anti-PD-(L)1 immune checkpoint inhibitor, second-line chemotherapy regimens deliver response rates of only about 15 percent. Antibody–drug conjugates have shown encouraging activity in this setting, but they require intravenous infusion and raise concerns about cross-resistance between agents sharing similar payloads, creating a genuine need for orally administered alternatives.

Study 102 enrolled 30 women between January 2023 and May 2024 across 12 sites in Japan, the Republic of Korea, and the United States. All had advanced, unresectable, or recurrent endometrial cancer that had progressed on or after platinum chemotherapy and an anti-PD-(L)1 agent such as pembrolizumab, durvalumab, or dostarlimab. Sixteen of the 30 had already been treated with lenvatinib itself. Participants received E7386 at 120 milligrams twice daily, with the lenvatinib starting dose later reduced from 20 to 14 milligrams once daily after non-clinical data suggested the combination retained its potency even at the lower dose — a change that also widened access for patients previously exposed to the drug.

The efficacy results exceeded what the heavily pretreated population would normally be expected to achieve. Eleven patients had confirmed responses — one complete and ten partial — for an objective response rate of 36.7 percent, roughly double the activity of conventional second-line chemotherapy. Responses proved durable, with a median duration of 9.1 months, and the disease control rate, which counts stable disease as well as shrinkage, reached 70 percent. Median progression-free survival was 5.5 months overall, and median overall survival had not been reached at the data cut-off, with 70 percent of patients alive at both six and twelve months.

The subgroup breakdown was particularly striking. Among the 14 women who had never received lenvatinib, the objective response rate climbed to 57.1 percent, with a median progression-free survival of 10.8 months — figures that compare favourably even with recent antibody–drug conjugate trials in the post-immunotherapy setting. Perhaps more surprising, three of the 16 patients who had already been treated with lenvatinib also responded, including two whose most recent prior therapy had been lenvatinib combined with pembrolizumab. This suggests that E7386 may re-sensitise tumours to lenvatinib’s mechanism, potentiating its antitumour activity rather than simply adding an independent effect.

Safety data supported the combination’s feasibility. Every patient experienced some treatment-emergent adverse event, but the most common — vomiting in 73.3 percent and nausea in 66.7 percent — were predominantly grade 1 or 2 and were largely controlled with standard anti-nausea medications such as 5-HT3 receptor antagonists. No grade 4 or 5 events occurred, and only two patients discontinued treatment because of toxicity. Dose interruptions and reductions were common, particularly for lenvatinib, reflecting the familiar fingerprint of that drug: proteinuria, diarrhoea, and hand–foot skin reactions appeared at rates consistent with its established profile. Because agents targeting the Wnt pathway have previously been linked to bone fragility, the investigators monitored bone density and bone turnover markers throughout; bone events occurred in just three patients, all manageable with supplements or antiresorptive therapy.

Biomarker analyses added a layer of molecular context. Circulating tumour DNA sequencing at baseline identified TP53 mutations as the most frequent alteration, present in 46.7 percent of patients, followed by mutations in KRAS, PIK3CA, PPP2R1A, PTEN, and FBXW7. Notably, a multivariate analysis adjusting for age, mismatch repair status, and prior lenvatinib exposure found no association between TP53 mutation status and any clinical outcome, and responses were seen across mismatch repair proficient and deficient tumours alike. With only 30 participants, however, the study was not powered to detect subtle predictive signatures, and the authors caution that the molecular findings are exploratory.

The trial’s limitations are real but typical of early-phase oncology research: a single-arm design without a control group, a small sample concentrated at Asian sites, and a population skewed toward patients with good performance status. Even so, the enrolled cohort closely mirrors the real-world post-immunotherapy population, with more than 80 percent having received at least two prior lines of systemic therapy. Against benchmarks such as the roughly 15 percent response rate of second-line chemotherapy, the 21 percent of lenvatinib monotherapy, and the 22 percent of sacituzumab govitecan in a mixed immunotherapy-exposed population, the 36.7 percent response rate — and 57.1 percent in lenvatinib-naïve patients — stands out.

The programme is already moving forward. A randomised dose-optimisation phase is now enrolling a more geographically balanced population of lenvatinib-naïve patients with advanced endometrial carcinoma, comparing two doses of E7386 plus lenvatinib against lenvatinib monotherapy or physician’s choice chemotherapy with doxorubicin or paclitaxel. If those results confirm the signals seen here, an all-oral regimen targeting the Wnt pathway and tumour vasculature could become a genuinely new option for women who currently have none — and a demonstration that drugging protein–protein interactions, long considered near-impossible, is becoming practical medicine.

Subject of Research: E7386 plus lenvatinib for advanced endometrial cancer after platinum chemotherapy and anti-PD-(L)1 immunotherapy

Article Title: E7386 plus lenvatinib in patients with advanced endometrial cancer that progressed on platinum-based chemotherapy and an anti-PD-(L)1 immunotherapy: a single-arm, open-label, global phase 1b/2 dose-expansion cohort

Article References: Lee, J.-Y., Hasegawa, K., Kim, B.-G., Berman, B., Suzuki, S., Corr, B., Yunokawa, M., Orr, D., Yamamoto, N., Soliman, P. T., Miller, D. S., Potdar, A. A., Sahara, T., Kimura, T., Dutta, L., Wu, J., McKenzie, J., & Kyi, C. (2026). E7386 plus lenvatinib in patients with advanced endometrial cancer that progressed on platinum-based chemotherapy and an anti-PD-(L)1 immunotherapy: a single-arm, open-label, global phase 1b/2 dose-expansion cohort. eClinicalMedicine, 100, Article 104206. https://doi.org/10.1016/j.eclinm.2026.104206

Image Credits: AI Generated

DOI: 10.1016/j.eclinm.2026.104206

Keywords: endometrial cancer, E7386, lenvatinib, CBP–β-catenin inhibitor, Wnt signalling, immunotherapy resistance, phase 1b/2 trial, anti-PD-(L)1, targeted therapy, protein–protein interaction inhibitor, progression-free survival, gynecologic oncology

The review, led by Danian Singh of the University of Auckland with colleagues Meg Parsons and Karen Fisher, employed an integrative critical narrative review methodology, combining a structured Boolean search of Scopus and Google Scholar with targeted retrieval of grey literature from Pacific regional agencies and national adaptation plans. After screening, the authors identified a core corpus of thirty-four empirical studies of gendered traditional knowledge, most published since 2020 and concentrated in Fiji and the Solomon Islands, which they read alongside twenty-five theoretical, comparative and policy works. The analysis was framed by feminist political ecology and decolonial scholarship, and the authors were explicit about their own positionality as partial insiders and outsiders to the communities whose knowledge they synthesised.

Geographically, the review resists treating the Pacific as an ecological monolith. Its primary evidence comes from the volcanic high islands of Fiji, Tonga and Samoa, where larger land areas, richer soils and extensive reef and mangrove systems shape one set of gendered knowledge practices. Against this, the authors set the low-lying atolls of Kiribati as a deliberate analytical counterpoint, where acute groundwater vulnerability and constrained nearshore gleaning grounds produce adaptive practices that differ in kind rather than merely in degree. This comparative design guards against the common error of generalising from Melanesian cases to atoll nations facing fundamentally different ecological constraints.

The first domain of evidence concerns agriculture and food security. Across Fiji, Tonga and Samoa, the literature documents a culturally defined division of labour in which men typically undertake land clearing and large-scale cropping while women manage household gardens, seed selection, weeding and the maintenance of crop diversity. In Fiji, women cultivate quick-maturing tuber crops to avert shortages and restore farms after cyclones, and on Totoya Island they combine traditional food-processing knowledge with solar-drying technology to produce cassava and breadfruit flour that substitutes for imported wheat. During cyclone seasons, Fijian women have been documented preserving cassava through traditional drying techniques and drawing on drought-resistant crops such as wild yams and sweet potatoes that remain available when water is scarce.

The Tongan case reveals both the depth of this expertise and its colonial reshaping. Before colonisation, Tongan agroforestry cultivated root crops, trees and shrubs together in mutually sustaining systems of nutrient exchange, pest defence and climate buffering, a system progressively displaced by commercially promoted monocropping. Within the customary arrangement, men cultivated the plants while women transformed them through cooking, medicine and the making of koloa, the valued textiles that secure social relationships, so that food security rested on the interlocking of both genders’ contributions. Development interventions, however, misread these arrangements entirely, promoting an urban horticulture that encouraged women to plant Western vegetables in house plots never organised for food provisioning, on the mistaken assumption that Tongan women gardened on a Western model.

The second domain concerns water and marine resources, where women’s knowledge is perhaps most consequential and most invisible. Women’s knowledge of traditional well locations enables communities to find potable water during droughts, while in coastal Fiji’s Bua Province, women proved more likely than men to identify medicinal uses of mangroves, including seedling preparations used to treat children’s coughs and disinfect wounds. Elder women in the villages of Navunievu and Denimanu articulated mangroves’ coastal protection role in vernacular terms, describing how they hold the soil together, and recalled which species best stabilise eroding shorelines, guiding younger women to plant them in vulnerable areas. Meanwhile, the customary tabu system of temporary closures, including a five-year closure reported to have restored sea cucumber and fish populations, fuses ecological and cultural functions in ways formal marine protected areas rarely achieve.

Gleaning, the collection of shellfish, crabs and seaweed from shallow reefs and mangroves, is predominantly practised by women and functions as a disaster-resilience mechanism. In Kiribati, women use finely tuned techniques passed down through generations, watching for telltale holes, bubbles in the sand and particular tidal patterns, and employing specialised methods such as wai ibo and the nocturnal luring of ghost crabs. Elders emphasise that gleaning becomes crucial after storms, when open-sea fishing is dangerous and marine organisms wash into the shallows. Yet a Solomon Islands study found that women’s exclusion from marine governance led them to breach local management rules, because closures had been sited on their customary fishing grounds without their input, eroding both equity and the legitimacy of conservation itself. Researchers have coined the term womangroves to describe how women’s mangrove-based food knowledge is rendered peripheral within a conservation discourse oriented towards male-dominated fisheries and timber.

The third and fourth domains cover disaster risk awareness and caregiving. Women in rural Fiji read unusual flowering times and shifts in lunar cycles as early signs of droughts, floods or cyclones, complementing men’s knowledge of offshore fishing and infrastructure repair, and sustaining the solesolevaki system of pooled labour that becomes a lifeline during crises. After Tropical Cyclone Winston, women in several Fijian villages were among the first to undertake reconstruction, with rapid recovery attributed partly to collective care and local leadership. In Samoa, fa’afafine performed both men’s and women’s roles in disaster response after the 2009 tsunami, yet binary-organised evacuation facilities excluded some from aid distribution, illustrating how gender-blind planning fails even those who contribute most.

The review’s central argument is that these patterns of exclusion are not incidental but the contemporary expression of colonial restructurings of gender and authority. Colonialism and missionary Christianity displaced arrangements in which Pacific women held authority over land, food production and ritual life, imposing patriarchal norms and undermining matrilineal inheritance. Those legacies persist in governance structures that engage male chiefs and elders as the relevant knowledge holders, systematically omitting women’s expertise in agroecology, water security and medicinal plants. The authors warn that incorporating traditional knowledge without a gender lens risks maladaptation, since locally led governance that is not gender-inclusive may simply relocate the exclusion of women’s knowledge rather than remedy it.

The policy implications are concrete. The authors call for institutionalising women’s genuine decision-making power rather than token representation, adjusting consultation processes to address practical constraints of timing, language and childcare, documenting gendered knowledge through community-led oral history and seasonal calendars under protocols that secure Indigenous ownership, and designing adaptation to empower rather than overburden women, whose unpaid labour already sustains community resilience. Knowledge transmission itself faces compound threats from commercialisation, migration and the erosion of customary leadership, meaning that marginalisation and loss reinforce each other across generations. Centring gender in Pacific climate adaptation, the review concludes, is both an equity imperative and a practical pathway to more effective and just resilience, because a community that consults only half its knowledge holders forfeits half its capacity to survive what is coming.

Subject of Research: Gendered traditional knowledge in Pacific Island climate adaptation and disaster risk reduction

Article Title: Gendering traditional knowledge in Pacific climate adaptation and disaster management: a critical narrative review

Article References: Singh, D., Parsons, M., & Fisher, K. (2026). Gendering traditional knowledge in Pacific climate adaptation and disaster management: a critical narrative review. Regional Environmental Change, 26(4), Article 204. https://doi.org/10.1007/s10113-026-02676-x

Image Credits: AI Generated

DOI: 10.1007/s10113-026-02676-x

Keywords: traditional knowledge, Pacific Islands, climate adaptation, disaster risk reduction, gender, Indigenous knowledge, food security, marine resources, mangroves, colonial legacies, maladaptation, Fiji