Gastric atrophy is a precursor of intestinal-type gastric cancer and is associated with mitochondrial and metabolic abnormalities. The malate–aspartate shuttle (MAS) couples cytosolic NADH oxidation to mitochondrial metabolism and intersects with glycolysis, the tricarboxylic acid cycle, aspartate metabolism, and biosynthesis. This review distinguishes direct gastric observations from mechanistic evidence obtained in other systems. Gastric studies report heterogeneous mitochondrial, metabolite, and component-level findings in different histological and etiological settings. Prospective tissue and plasma omics studies identify progression-associated signatures, but do not establish a longitudinal MAS trajectory, and integrated shuttle function remains unresolved in gastric atrophy and intestinal metaplasia. We consider three context-dependent interpretations: secondary remodeling, a maladaptive contribution to injury, and protective adaptation. These roles may coexist or change over time rather than form mutually exclusive categories. Distinguishing them requires defined baseline functional changes, bidirectional perturbations under matched upstream stress, compartment-resolved redox measurements, and complementary enzyme/carrier interventions linked to gastric phenotypes. Reconstituted shuttle activity must also be distinguished from endogenous flux in intact cells. The MAS provides a hypothesis-generating framework for gastric redox–bioenergetic research; its disease-stage behavior, causal significance, and therapeutic direction require direct validation.
BackgroundThis study aimed to develop a pretreatment MRI-derived radiogenomic surrogate of a prespecified hypoxia–angiogenesis–epithelial–mesenchymal transition transcriptional state and to evaluate its associations with treatment response and survival in nasopharyngeal carcinoma.MethodsThis multicenter observational study integrated patient-derived public transcriptomic datasets, experimental CNE2/CNE2-IR radioresistance models, a prospective RNA-sequencing cohort of 49 patients with paired contrast-enhanced T1-weighted MRI and RNA data, and two independent clinical cohorts (n = 532 and n = 118). HAE-CAS was the equal-weight mean of three within-cohort standardized module scores. Treatment response was assessed approximately 3 months after chemoradiotherapy according to RECIST version 1.1 using nasopharyngeal MRI, nasopharyngoscopy, and clinical evaluation. Rad-HAE used a five-feature Gaussian ridge model and was evaluated with 20 repetitions of five-fold nested cross-validation; response and survival outcomes were excluded from model development.ResultsAmong 49 response-evaluable RNA-sequenced patients, 17 had complete response (CR) and 32 had partial response (PR); no stable or progressive disease occurred. HAE-CAS was higher in PR than CR (rank-biserial correlation = 0.365, P = 0.035). In GSE102349, continuous HAE-CAS was not significantly associated with progression-free survival (HR per 1-SD increase = 1.20, 95% CI 0.74–1.95; P = 0.465), whereas the median-dichotomized comparison was exploratory (log-rank P = 0.024). Aggregated out-of-fold Rad-HAE correlated with HAE-CAS (Spearman rho = 0.637, P < 0.001) and discriminated PR from CR among 49 paired response-evaluable patients (AUC = 0.846, 95% CI 0.732–0.943). In two independent clinical cohorts, higher Rad-HAE was associated with shorter progression-free and overall survival.ConclusionPretreatment CE-T1WI radiomics was associated with interpatient variation in a prespecified composite HAE transcriptional state. Rad-HAE represents an imaging-based surrogate associated with the composite HAE transcriptional state and does not directly quantify individual biological processes.
IntroductionThe 2020 International Association for the Study of Lung Cancer (IASLC) grading system integrates the predominant histologic pattern and the proportion of high-grade patterns, but preoperative CT assessment is limited by spatial averaging of heterogeneous tumor components. We aimed to develop a CT-derived nested habitats score (NHS) that recursively identifies grade-associated intratumoral CT subregions, evaluate its cross-center performance for preoperative IASLC grading, and explore its transcriptomic correlates.MethodsThis retrospective two-center study included 650 patients with surgically resected lung adenocarcinoma. Center 1 formed the development cohort (n = 390); Center 2 was stratified 1:1 into validation (n = 130) and locked test (n = 130) cohorts, both from the same external institution. The prespecified primary task was pathologic IASLC Grade 1 versus Grade 2/3. NHS was generated using five-fold out-of-fold estimation in development data. Clinical-radiological, whole-tumor radiomics, NHS, and integrated models were compared, with algorithm selection confined to validation before one-time test evaluation. An independent TCGA-LUAD/TCIA CT-RNA cohort (n = 88) underwent exploratory WGCNA and Hallmark gene-set enrichment analysis.ResultsThe development/validation/test cohorts contained 41/14/14 Grade 1, 293/101/96 Grade 2, and 56/15/20 Grade 3 tumors. The selected random-forest integrated model yielded AUCs of 0.871 (95% CI, 0.784–0.936) in validation and 0.866 (95% CI, 0.784–0.935) in testing. NHS alone yielded AUCs of 0.828 and 0.854, compared with 0.620 and 0.497 for the clinical-radiological model and 0.703 and 0.775 for whole-tumor radiomics. NHS accounted for 76.3% of mean absolute SHAP contribution; removing NHS reduced AUC by 0.251 (95% CI, 0.079–0.439) in validation and 0.369 (95% CI, 0.200–0.553) in testing. Brier scores for NHS and the integrated model (0.074–0.077) were below the non-informative reference (0.096), but calibration slopes exceeded 1, and decision curves showed no net benefit over treating all patients as Grade 2/3 at low-to-intermediate thresholds. Exploratory radiogenomics showed nominal NHS-associated coexpression modules and enrichment of WNT/beta-catenin, E2F, G2M-checkpoint, and immune-related programs.DiscussionNHS discriminated IASLC Grade 1 from Grade 2/3 tumors, retained stable performance in two held-out cohorts from a second institution, and contributed most of the integrated model’s discrimination. The transcriptomic associations are hypothesis-generating, and recalibration and prospective multicenter evaluation are required before clinical application.
IntroductionKidney dysfunction has been associated with cerebrovascular atherosclerosis, but differences in renal measures between extracranial (ECAS) and intracranial atherosclerotic stenosis (ICAS) remain unclear.MethodsThis hospital-based cross-sectional study included consecutive patients admitted with acute ischaemic stroke within 14 days of onset between January 2006 and December 2015 who were classified as having the large-artery atherosclerotic subtype. Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation adapted for Asian populations. Binary logistic regression compared ICAS with ECAS across four eGFR categories. The adjusted category model used complete cases; a sensitivity analysis used 50 multiply imputed datasets for missing covariates.ResultsAmong 1,127 patients with large-artery atherosclerotic stroke, 1,015 had ECAS or ICAS and available eGFR for the category analysis; 755 had complete covariate data. Compared with the highest eGFR category, the lowest two categories had lower adjusted odds of ICAS relative to ECAS in complete-case analyses (OR 0.38, 95% CI 0.21–0.68; and OR 0.50, 95% CI 0.28–0.88) and multiple-imputation analyses (OR 0.35, 95% CI 0.21–0.59; and OR 0.54, 95% CI 0.33–0.88). Proteinuria did not differ significantly between the groups, and eGFR was not significantly associated with stenosis severity within either group.DiscussionLower eGFR was associated with extracranial rather than intracranial stenosis. Mechanistic explanations remain literature-derived and hypothesis-generating. The observational design and absence of diagnostic performance analyses preclude causal inference or conclusions about diagnostic utility or treatment selection.
ObjectiveThis study aimed to systematically compare the performance of U-Net and a conditional generative adversarial network (cGAN) for multi-tissue segmentation of knee magnetic resonance imaging (MRI) using the public SKI10 dataset, and to investigate the effects of loss functions, discriminator receptive fields, and normalization strategies on model performance through ablation experiments.MethodsThe segmentation targets included the femur, femoral cartilage, tibia, and tibial cartilage, with the background incorporated as an additional class to formulate a five-class segmentation task. All MRI images and labels were read from the original. mhd files and two-dimensional slices were extracted along the sagittal direction. Images were resized to 512 × 512 pixels and standardized using z-score normalization, whereas labels were resized with nearest-neighbor interpolation to preserve categorical encoding. Training and test sets were split at the MRI-volume level to prevent information leakage caused by adjacent slices from the same volume being assigned to different subsets. U-Net was used as the supervised baseline model, and a cGAN segmentation framework consisting of a U-Net-style generator and a discriminator was constructed. Evaluation metrics included the Dice similarity coefficient (DSC) and the 95th-percentile Hausdorff distance (HD95). Qualitative segmentation, error heatmaps, boundary comparison, per-class win rate, Bland-Altman agreement analysis, boundary DSC, false-positive/false-negative composition, surface-distance distributions, and training dynamics were further analyzed. A sub-selection of 10 of the 70 training volumes was used as a validation set for checkpoint selection; the 30 test volumes were accessed only at final evaluation and were never used for training, model selection, or hyperparameter optimisation.ResultsBaseline comparison showed that U-Net and cGAN exhibited complementary behavior in knee MRI multi-tissue segmentation. The mean DSC across the four anatomical structures was significantly higher for cGAN than for U-Net (0.8061 ± 0.0542 vs. 0.8035 ± 0.0520; P < 0.001). In contrast, the mean HD95 was lower for cGAN than for U-Net (4.02 ± 1.18 mm vs. 5.38 ± 1.44 mm), indicating better boundary-distance control and local contour fitting. Structure-specific analysis showed that segmentation of osseous structures was more stable than that of cartilage structures, and model differences were concentrated mainly in thin-layer tissues such as femoral and tibial cartilage. Loss-function ablation showed that L2 and Smooth L1 losses improved DSC and reduced HD95 in U-Net, with Smooth L1 being more favorable for boundary-distance control. In the cGAN framework, the L1 pixel-wise loss yielded the most stable performance. Discriminator receptive-field ablation showed that PixelGAN achieved the best overall performance, indicating that larger PatchGAN receptive fields do not necessarily improve segmentation of medical label maps. Comparison of normalization strategies showed that BatchNorm achieved higher DSC than InstanceNorm in both U-Net and cGAN PixelGAN.ConclusionU-Net and cGAN showed complementary strengths in knee MRI multi-tissue segmentation. U-Net provided better regional overlap and cartilage continuity, whereas cGAN PixelGAN achieved lower boundary errors. These findings support task-specific selection of segmentation architectures and discriminator configurations. However, as the results were based on a single predefined volume-level split of the SKI10 dataset, they require confirmation through repeated splits, cross-validation, and external multicenter validation.
