Browse Articles
Discover research articles across all indexed journals
Real-world outcomes and prognostic role of neutrophil-to-lymphocyte ratio (NLR) in patients with advanced urothelial carcinoma treated with enfortumab vedotin plus pembrolizumab (EVP).
187 Background: EVP has demonstrated a significant survival benefit in advanced urothelial carcinoma (aUC) in clinical trials. However, real-world data evaluating outcomes and early prognostic biomarkers in this setting is limited. We assessed real-world outcomes and explored the prognostic value of neutrophil-to-lymphocyte ratio (NLR), focusing on NLR prior to cycle 3 (NLR3) and the dynamic change in NLR (dNLR), in aUC patients treated with first-line EVP. Methods: We conducted a retrospective, multicenter analysis of patients with aUC treated with EVP in Alberta, Canada between September 2024 and January 2026, with a minimum follow-up of 3 months. NLR3 and dNLR were evaluated. High NLR3 was defined as ≥ 2.7. dNLR was defined as the relative change from baseline to pre-cycle 3 and categorized as a decrease > 10% versus stable/increase (≤10% decrease or any increase). Objective response rate (ORR), progression-free survival (PFS), and overall survival (OS) were analyzed. Results: Sixty patients were included (median age 69 years, 82% male). Primary tumor site was bladder in 41 patients and upper genitourinary tract in 14. Histology was pure urothelial carcinoma in 48 patients and mixed histology in 12. 33 patients had visceral metastases; most commonly involving lung (n = 16), bone (n = 12) and liver (n = 12). 15 patients had lymph node–only metastases and 12 had locally advanced unresectable disease. FGFR testing was performed in 31 patients, with alterations identified in 8. Median follow-up was 10.3 months, and the median number of EVP cycles administered was 8. The overall ORR was 62%, and 1-year PFS was 42%. Patients with low NLR3 had higher ORR (82% vs 55%; OR 0.26, 95% CI 0.07–0.96; p = 0.037) and superior survival, including PFS (1-yr 75% vs 27%; HR 0.16, 95% CI 0.04–0.54; p = 0.003) and OS (1-yr 100% vs 50%; p = 0.005) compared to high NLR3. NLR decrease > 10% was associated with better PFS (1-yr 70% vs 26%; HR 0.31; 95% CI 0.12–0.75 p = 0.01). OS difference was numerically higher with NLR decrease > 10% but not statistically significant (1-yr 73% vs 59%; p = 0.11). Conclusions: In this real-world cohort of patients with aUC treated with EVP, low NLR3 and early NLR decline were associated with higher response rates and improved survival. NLR3 and dynamic NLR changes are simple, readily available biomarkers that provide early prognostic information and may support risk stratification in routine clinical practice. NLR3, dynamic NLR, and impact on patient outcomes. NLR3 low (<2.7) vs high (≥2.7) dNLR decrease >10% vs dNLR stable/increase ORR (95% CI, p-value) 81.8% vs 54.5% (0.07–0.96; p = 0.037) 69% vs 61% (0.23–2.19; p = 0.56) 1 year PFS HR- 95% CI, p-value) 75% vs 27% (0.16, CI 0.04–0.54; p = 0.003) 70% vs 26% (0.31, 95% CI 0.12–0.75; p = 0.01) 1yr OS (log rank p-value) 100% vs 50% (p = 0.005) 72% vs 59% (p = 0.109)
Concave and Convex Molecular Curvature Modulates Spatial Electronic Environments for Controlled Electrocatalysis
A Neutral “Masked” Diborene and Its Reactivity Toward Metal‐Free Pyridine Homocoupling
ABSTRACT An unprecedented neutral “masked” non‐external‐donor‐stabilized diborene is generated via intramolecular arene dearomatization of an N‐heterocyclic imine (NHI)‐supported free diborene. The free diborene intermediate is photochemically accessible from the masked form, as supported experimentally through chalcogenation reactions. Computational studies elucidate an alkyne‐like bonding pattern and a triplet ground state with a Δ E S‐T = +1.2 kcal mol −1 for the NHI‐supported free diborene. Remarkably, the “masked” diborene activates pyridines, promoting transition‐metal‐free C−C bond formation with high C2 regioselectivity to afford homocoupling products under mild conditions. Mechanistic investigations reveal that stepwise coordination and B−C bond cleavage produce a bis‐pyridine diborene intermediate, which undergoes reductive coupling via intrinsically strong double single‐π‐electron transfer from the B═B unit to the intramolecular pyridine ligands. Furthermore, oxidation with p ‐benzoquinone allows the release of free 2,2′‐bipyridine derivatives. These findings provide access to insight into a hitherto elusive free diborene species, establish a previously unknown reactivity mode for diborenes, and demonstrate that this “masked” diborene functions as an exceptionally powerful double single‐electron donor with the potential to mimic transition‐metal behavior in the construction of organic molecules.
Deep learning classification of nasal anatomical variants on computed tomography for endoscopic sinus surgery
Abstract Endoscopic sinus surgery (ESS) carries risks such as vision loss and intracranial injury due to the proximity of critical structures and unrecognised anatomical variants. We developed convolutional neural networks (CNNs) to detect free-floating anterior ethmoidal arteries (FFAEA), Onodi cells, and Haller cells on coronal sinus CT, and evaluated BioMedCLIP, a biomedical vision-language model (VLM), in a few-shot setting. CT scans from 122 ESS patients were anonymised, standardised coronal CT images were captured, and variant presence was annotated. Five ImageNet-pretrained CNN backbones were assessed using repeated patient-wise five-fold cross-validation across 40 configurations. The best CNNs achieved balanced accuracies of 77.8 ± 2.3% (FFAEA), 74.6 ± 2.5% (Onodi), and 63.7 ± 6.2% (Haller). BioMedCLIP achieved 65.5 ± 3.2%, 63.8 ± 1.8%, and 73.5 ± 3.6%, respectively, outperforming CNNs for Haller cell detection while providing competitive performance for the other variants. These models demonstrate proof-of-concept performance for automated identification of selected sinonasal variants on standardised coronal CT images under internal patient-wise cross-validation.
Liquid Metals for Reconfigurable Bioelectronics (Adv. Mater. 38/2026)
Super‐Resolution Ultrasound Based Cell Tracking With Polymeric Nanobubbles
ABSTRACT Tracking the fate of transplanted cells in vivo remains a major challenge for the development and monitoring of cell‐based therapies. Here, we report a nanobubble (NB) based acoustic labeling strategy that enables ultrasound localization microscopy (ULM) to track transplanted cells in tumors with single‐cell‐scale sensitivity and super‐resolution spatial precision. We developed stable, biocompatible poly(butyl cyanoacrylate) NB that are efficiently internalized by cells and generate strong intracellular ultrasound signals. Furthermore, ULM allows the real‐time visualization of individual cells, their localization, and trajectory reconstruction in flow phantoms and murine breast cancers. In the latter, NB‐labeled bone marrow‐derived mononuclear cells were tracked in vivo following intra‐arterial injection, whereas unlabeled cells remained undetectable. By integrating cell trajectories with microbubble‐based super‐resolution vascular maps, our approach allows identification of the blood vessels that delivered the cells. This clinically compatible imaging strategy expands the application of ULM beyond vascular mapping and provides a platform for high‐resolution, dynamic monitoring of cell‐based therapies.
The multiple sclerosis drug fingolimod induces pore-like membrane defects, contributing to its antimicrobial activity
Automation of OncoKB annotation to enhance efficiency and consistency of molecular tumor board workflows.
26 Background: Molecular Tumor Boards (MTBs) play a critical role in translating next-generation sequencing (NGS) results into actionable treatment recommendations. Variant annotation is currently performed manually which is labour-intensive, difficult to scale, and subject to variability. We evaluated whether automated extraction and annotation could improve efficiency while maintaining concordance with expert-curated MTB annotations. Methods: A rule-based pipeline was developed to extract genomic events from heterogeneous clinical NGS reports, including single-nucleotide variants, copy-number alterations, gene fusions, and mutational signatures. Extracted events were normalized into a standardized data model and annotated using the OncoKB Application Programming Interface (API) to retrieve oncogenicity classifications and therapeutic levels of evidence. The pipeline was iteratively refined across four development versions and retrospectively evaluated on 32 reports comprising 497 unique genomic events including variants of unknown significance (VUS). Precision recall metrics (F1 and balanced accuracy scores) for extraction and actionability assignment steps were calculated using the manually validated final version as a reference. Results: Iterative refinement resulted in progressive improvements in extraction completeness and therapeutic annotation accuracy (Table 1). Manual review of a monthly MTB cohort requires up to 30 hours of expert effort, whereas automated processing completes in under 3 minutes. Conclusions: Automated extraction and OncoKB annotation can achieve performance comparable to expert manual curation when supported by iterative refinement and normalization rules. By version 4, the pipeline achieved complete concordance with validated MTB annotations and significantly reduced the processing time by up to 600-fold compared to manual review. This scalable approach enhances efficiency, consistency, and reproducibility in routine MTB workflows. While automated processing achieved complete overall concordance, fusions/rearrangements may still benefit from expert review due to format heterogeneity in clinical reports. Iterative pipeline performance across development versions. Version Extraction F1 score Actionability balanced accuracy score Remarks/patch notes v1 0.744 0.873 Initial rule-based extraction v2 0.984 1 Corrected tumour-type annotations; single-partner fusion fallback. v3 0.984 1 Standardized VUS and fusion logic v4 1 1 Resolved residual fusion edge cases
Multichannel Through‐Space Charge Transfer Triplet Photosensitizers for Rapid Visible‐Light Polymerization and High‐Fidelity 3D Printing
ABSTRACT Visible‐light‐driven polymerization offers unparalleled spatiotemporal control and energy efficiency, yet its practical performance is fundamentally constrained by the excited‐state dynamics of the photosensitizer. In this study, we report a class of ortho ‐positioned multi‐donor–acceptor molecules featuring multichannel through‐space charge transfer (TSCT) characteristics as highly efficient, heavy‐atom‐free triplet photosensitizers. By constructing a highly twisted three‐dimensional (3D) charge‐transfer network, these molecules achieve exceptional photophysical performance, including high intersystem crossing (ISC) quantum yields ( Φ ISC up to 0.86), microsecond‐scale triplet lifetimes, and an extraordinarily small singlet‐triplet energy gap (Δ E ST as low as 0.008 eV), while maintaining high triplet energies and strong visible‐light absorption. Benefiting from these attributes, the TSCT photosensitizers efficiently activate diphenyl ketone benzoyl oxime ester coinitiators via a triplet‐triplet energy transfer (TTEnT) mechanism, enabling rapid and well‐controlled acrylate polymerization under low‐intensity visible‐light irradiation. Importantly, this strategy enables, for the first time, the application of multichannel TSCT photosensitizers in digital light processing (DLP) 3D printing, allowing high‐resolution fabrication of complex 3D architectures under ambient conditions with excellent operational robustness and biocompatibility. This work establishes a versatile molecular platform for the rational design of organic triplet photosensitizers and advances visible‐light‐based precision manufacturing technologies.
CHDS: a boundary-oriented oversampling method for imbalanced data based on convex hull and Delaunay triangulation
A Mechanistic Study of Bio‐Based Nanotemplated Carbon Nanofibers Derived From Water Processable Lignin Blends for Sustainable Energy Storage Applications (Adv. Mater. 39/2026)
Lactate-fueled H3K18 lactylation sustains ΔNp63α-dependent self-renewal in limbal epithelial stem cells
Nodal status–guided postoperative radiotherapy after neoadjuvant chemoimmunotherapy for locally advanced esophageal squamous cell carcinoma: A multicenter cohort study.
112 Background: In locally advanced esophageal squamous cell carcinoma (ESCC) treated with neoadjuvant chemoimmunotherapy (Chemo-IO) followed by surgery, the role of postoperative radiotherapy (PORT) remains unclear. Tumor-draining lymph nodes (TDLNs) are critical sites for priming antitumor immunity; therefore, excessive elective nodal irradiation may compromise radiotherapy–immunotherapy synergy. This study evaluated the clinical benefit of PORT and examined whether radiation exposure to postoperative residual, clinically negative TDLN basins (rTDLNs) influences outcomes. We hypothesized that PORT confers benefit in selected patients and that sparing rTDLNs may preserve immune function and improve survival. Methods: We retrospectively analyzed 259 patients with locally advanced ESCC treated at six centers between January 2022 and December 2024. All patients underwent neoadjuvant Chemo-IO followed by curative esophagectomy (R0 resection) and had pathologically confirmed ypT3–4 and/or ypN+ disease. Endpoints were disease-free survival (DFS) and overall survival (OS). Multivariable Cox models evaluated the effects of PORT and adjuvant immunotherapy. Among PORT-treated patients (n=51), rTDLN basins were contoured and dose–volume histogram (DVH) metrics (V15/V30/V40; percent volume) were analyzed. Results: With a median follow-up of 26 months, 259 patients were included (115 ypN+, 164 ypT3–4). In the overall cohort, ypN+ independently predicted poorer DFS (HR=2.04, 95% CI 1.25–3.33; P=0.004) and OS (HR=2.38, 95% CI 1.23–4.76; P=0.010). Adjuvant immunotherapy was associated with improved OS (HR=0.48, 95% CI 0.25–0.91; P=0.024). In ypN+ patients (n=115), PORT was associated with improved OS compared with no PORT (median OS: 41.0 vs 32.4 months; HR=0.30, 95% CI 0.11–0.84; P=0.022), and adjuvant immunotherapy also conferred benefit (HR=0.32, 95% CI 0.14–0.75; P=0.009). No benefit of PORT was observed in ypN0 patients. In the PORT cohort (n=51), higher-dose irradiation to rTDLNs was associated with inferior survival. Cox regression showed that a greater proportion of rTDLN volume receiving higher radiation doses was a significant risk factor: V40 (HR=1.04, 95% CI 1.01–1.08; P=0.026) was associated with increased mortality, with similar trends for V30 (HR=1.03, P=0.052) and V15 (HR=1.03, P=0.093). Conclusions: In ESCC treated with neoadjuvant Chemo-IO and surgery, nodal status is the dominant prognostic factor and identifies patients most likely to benefit from PORT. However, higher dose exposure of postoperative residual, clinically negative rTDLN basins was associated with inferior survival, supporting a risk-adapted “TDLN-sparing” PORT strategy that targets high-risk regions while constraining rTDLN DVH exposure. Prospective validation is warranted.
Click-Gelable and Self-Healing Organogels Based on Inorganic Sub-Nanowires
Correction: The spatial heterogeneity of soil physical and chemical properties and comprehensive fertility in the Leizhou Peninsula coast
A Flexible Wireless Passive Platform for Decoupled Electrolyte and Temperature Sensing Toward Heat‑Stress Assessment
ABSTRACT Strenuous exercise and heat exposure can significantly elevate core body temperature and induce profuse sweating, leading to dehydration, electrolyte imbalance, and an increased risk of heat stress. Simultaneous yet decoupled monitoring of thermal and sweating‐related signals is therefore essential for reliable physiological assessment under intense motion. Here, we report a flexible, wireless, and fully passive dual‐mode sensing platform that enables independent detection of sweat accumulation and skin temperature on a single LC resonant architecture. The system integrates a sweat‐responsive electrochemical module and a thermosensitive conductive composite, both inductively coupled to a flexible serpentine coil for battery‐free near‐field interrogation. We establish an orthogonal “frequency–magnitude” decoding strategy: sweat‐induced dielectric changes dominantly drive resonance frequency shifts, while temperature‐dependent resistive losses primarily modulate the reflection coefficient magnitude ( S 11 ). To further enhance robustness against signal coupling and motion artifacts, a machine‐learning‐assisted spectral decoder is introduced, enabling high‐accuracy recognition of joint physiological states. Owing to its lightweight, maintenance‐free operation, and robust performance under vigorous motion, this platform provides a promising strategy for heat‐stress assessment and real‐time physiological monitoring in demanding scenarios such as military training and competitive sports.
A novel autopolysialylation activity of the ganglioside sialyltransferase ST8Sia5 regulates its secretion and enzyme activity
Potential overtreatment in MRI-defined low-risk locally advanced rectal cancer in the era of neoadjuvant therapy: An international cohort study.
118 Background: Multimodal neoadjuvant therapy, including chemoradiotherapy (CRT) and total neoadjuvant therapy, is the standard treatment for locally advanced rectal cancer (LARC). However, reliance solely on clinical staging may lead to overtreatment in biologically favorable tumors. Beyond oncological control, pelvic radiotherapy is associated with significant late toxicities, including chronic presacral sinus, secondary malignancies, and impaired bone marrow reserve, potentially compromising the tolerance to systemic therapy upon recurrence. For patients with favorable tumor biology, avoiding unnecessary pelvic irradiation can significantly impact on long-term outcomes and quality-of-life. From this perspective, we assessed the extent of potential overtreatment in MRI-defined low-risk LARC, using an international database from two tertiary centers in Japan and South Korea. Methods: Among 881 patients with clinically staged T2–3 node-positive or T3 node-negative LARC treated between 2013 and 2021, 298 who underwent upfront surgery were identified for this study. Clinical low-risk features were defined on preoperative MRI as the absence of N2 disease, mesorectal fascia involvement (≤1 mm), extramural venous invasion, or enlarged lateral pelvic lymph nodes. Of these, 169 patients (56.7%) were categorized as MRI-defined low risk and included in the analysis. Results: Within the cohort of 169 patients, pathological ≤T2N0 disease was identified in 24.3% (n=41), for whom no adjuvant treatment would typically be required. Pathological T3N0 was observed in 34.9% (n=59); in these cases, adjuvant therapy would generally be considered only in the presence of additional high-risk pathological features, such as lymphovascular invasion, non-well/moderately differentiated histology, or positive surgical margins. Notably, these high-risk features were absent in 45.5% of the pathological T3N0 patients. Overall, 40.2% of patients did not require adjuvant therapy based on the final pathology, indicating substantial clinical–pathological discordance. Conclusions: In addition to previously observed lower postoperative complication rates with upfront surgery than with neoadjuvant CRT, a considerable proportion of MRI-defined low-risk LARC demonstrated favorable pathological characteristics and low margin positivity despite clinically advanced staging. Therefore, routine preoperative intensive therapy may represent overtreatment in selected patients. These findings support MRI-guided, risk-adapted treatment strategies and warrant prospective validation.
Optimized Octahedral Geometry and Hidden Rigidity of a <i>mer</i> -Tridentate N-Heterocyclic Carbene Ligand Enhance Luminescence of an Fe(III) Complex
AI agency drives college students’ entrepreneurial thinking through human sense of agency in human and ai symbiosis
Abstract As artificial intelligence (AI) increasingly penetrates education and entrepreneurial practice, cultivating students’ entrepreneurial thinking in a “human-AI symbiotic” environment has become a crucial issue in entrepreneurship education. Based on symbiotic agency theory and triadic reciprocal determinism, this study constructs a theoretical model of “AI agency—human sense of agency—entrepreneurial thinking,” using opportunity recognition and creativity as the core dimensions of entrepreneurial thinking to examine how AI agency is associated with college students’ entrepreneurial cognition. Based on a sample of 972 college students, the study employs a hybrid research method combining SLR, SEM, and fsQCA. The results show that: (1) AI agency is composed of three dimensions: cognitive support, interaction support, and action support, forming a stable structure of human-AI symbiotic environment in entrepreneurial learning contexts. (2) AI agency is positively associated with entrepreneurial thinking both directly and indirectly through the human sense of agency. (3) High-level entrepreneurial thinking is not associated with a single factor alone but with configurations of multiple conditions, with interaction support and sense of agency playing stable core roles. (4) Furthermore, creativity and opportunity recognition exhibit different generative logics. The study theoretically elevates AI from a tool to a technological partner and capability community in entrepreneurial learning, and provides insights for the design of AI instructional courses in entrepreneurship education.