Guangdong General Hospital

BackgroundRecent studies have demonstrated an association between amino acids (AAs) and the occurrence of Type 2 diabetes mellitus (T2DM). However, whether there is an underlying causal relationship between AAs and T2DM, as well as their potential links to T2DM progression, complications, and treatment selection, still lacks sufficient clinical evidence.MethodsThis study included 205,208 participants from the UKB database, with 14,066 diagnosed with T2DM. We used LASSO regression, supplemented by Mendelian randomization (MR), to explore the causal relationship between AA levels and T2DM. Additionally, restricted cubic splines, receiver operating characteristic (ROC) curves, and multivariable‐adjusted regression models were applied to analyze the association between AA levels, insulin resistance, secondary complications, and treatment options for T2DM.ResultsAlanine and valine were positively associated with T2DM, while glutamine, glycine, and histidine were negatively associated with T2DM. In particular, the MR results indicated a causal relationship between T2DM and plasma glutamine and glycine, which were identified as protective factors. Other branched‐chain AAs, such as leucine and isoleucine, did not show significant positive associations in the regression analysis. Additionally, we integrated various AAs to develop a predictive model for secondary complications of T2DM. The model demonstrated high predictive accuracy for a range of T2DM‐related complications (all areas under the ROC curve > 0.730). In addition, specific AA profiles were related to insulin resistance and demand for insulin or oral hypoglycemic drug treatment.ConclusionOur study results demonstrate a close relationship between AA levels and the occurrence, development, and treatment demand for T2DM, offering potential biomarkers for predicting complications and guiding personalized treatment.

Alcohol Use Disorder (AUD) is a global health challenge, affecting 10–15% of the population, with significant social, health, and economic consequences. Although pharmacotherapies such as disulfiram, naltrexone, and acamprosate are available, their effectiveness is limited and patient adherence is often poor. Repetitive transcranial magnetic stimulation (rTMS), a non-invasive neuromodulation technique that targets neural circuits implicated in addiction. Emerging evidence suggests that rTMS may reduce alcohol craving and consumption, although results have been mixed. This review examines the neural mechanisms by which rTMS may influence AUD, summarizes current clinical evidence of its efficacy, and discusses future directions.

BackgroundEsophageal squamous cell carcinoma (ESCC) is a highly prevalent and aggressive malignancy associated with poor prognosis. Accurate tumor segmentation is critical for the detection and characterization of ESCC. However, manual segmentation remains labor-intensive and time-consuming. Deep learning presents a promising approach for automating this process.Materials and methodsThis retrospective study assessed the performance of three deep learning models—UNet, nnUNet-2D/3D, and UMamba—for ESCC segmentation using contrast-enhanced free-breathing 3DGRE MRI with 1 mm isotropic resolution. The dataset comprised 192 patients, divided into 171 for training and 21 for validation. Manual annotations excluded calcifications and hemorrhage, with maximum tumor cross-sections utilized for 2D model training and full volumetric tumor layers employed for 3D architectures. Segmentation performance was evaluated using the Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), and mean surface distance (MSD).ResultUMamba achieved the highest DSC of 0.764, significantly outperforming nnUNet-3D (0.738) and nnUNet-2D (0.702). Additionally, UMamba demonstrated superior boundary delineation, as evidenced by the lowest HD of 5.048 mm and MSD of 1.088 mm. These results highlight the importance of 3D contextual analysis over traditional 2D approaches, as illustrated by the 5.1% improvement in DSC observed with nnUNet-3D compared to nnUNet-2D. Furthermore, the computational efficiency of each model was evaluated, revealing that 3D models exhibited longer inference times than their 2D counterparts.ConclusionUMamba demonstrates superior performance for automated ESCC segmentation on high-resolution MRI, providing a robust tool that has the potential to reduce manual segmentation time and inter-observer variability in clinical practice.

BACKGROUNDEsophageal cancer (EC) is one of the most prevalent malignant gastrointestinal tumors; accurate prediction of EC staging has high significance before treatment.AIMTo explore a rational radiomic approach for predicting preoperative staging of EC based on magnetic resonance imaging (MRI).METHODSThis retrospective study included 210 patients with pathologically confirmed EC, randomly divided into a primary cohort (n = 147) and a validation cohort (n = 63) in a ratio of 7:3. All patients underwent a preoperative MRI scan from the neck to the abdomen. High-throughput and quantitative radiomics features were extracted from T2-weighted imaging (T2WI) and gadolinium contrast-enhanced T1-weighted imaging (T1WI)-Gd images. Radiomics signatures were selected using minimal redundancy maximal relevance and the least absolute shrinkage and selection operator. Then a logistic regression model was built to predict the EC stages. The diagnostic performance of the radiomics model for discriminating between stages I-II and III-IV was evaluated using the area under the curve (AUC), sensitivity (SEN), and specificity (SPE).RESULTSA total of 214 radiomics features were extracted. Following feature dimension reduction, the T1WI and T2WI sequences were retained, and 14 features from the T1WI sequence and 3 features from the T2WI sequence were selected to construct radiomics signatures. The radiomics signature combining T2WI with T1WI-Gd demonstrated superior discrimination of stages in the validation cohort (AUC: 0.851; SEN: 0.697; SPE: 0.793), which outperformed single-sequence models (AUC: 0.779, 0.844; SEN: 0.667, 0.636; SPE: 0.8, 0.8).CONCLUSIONMRI-based radiomics signatures could identify EC stages before treatment, which could serve as a noninvasive and quantitative approach aiding personalized treatment planning.

Background:Evidence is needed to determine the benefits and harms of screening for atrial fibrillation (AF) in stroke prevention. This meta-analysis aimed to evaluate the benefits and issues of AF screening among older adults.Methods:This systematic review and meta-analysis were conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. We systematically searched several databases from inception through 28 March 2025, selecting randomized controlled trials (RCTs) comparing AF screening, including systematic and opportunistic screening, versus routine practice or no screening. Two reviewers independently extracted the data and appraised the risks of bias of the studies.Results:Thirteen articles covering 12 RCTs were included in the meta-analysis. For routine screening, systematic screening, rather than opportunistic screening, was more effective in detecting new AF cases (relative risk (RR), 2.07; 95% CI, 1.41 to 3.04; p = 0.0002). However, no difference was observed in the effectiveness of systematic and opportunistic screening in detecting AF (RR, 1.39; 95% CI, 0.59 to 3.30; p = 0.45). Compared with no screening, single-time-point screening did not improve the AF detection rate, whereas intermittent/continuous screening was associated with a greater likelihood of detecting AF (RR, 2.40; 95% CI, 1.59 to 3.64; p < 0.0001). There were no significant differences in the anticoagulation prescription rate between patients who underwent screening and routine care (RR, 1.16; 95% CI, 0.94 to 1.44; p = 0.16). Systematic screening was associated with a lower risk for the composite endpoint (combination of thrombosis-related events and mortality; RR, 0.96; 95% CI, 0.93 to 0.99; p = 0.02) but not for the individual endpoints. Compared with routine care, systematic screening did not increase the risk of major bleeding (RR, 0.88; 95% CI, 0.72 to 1.06; p = 0.18), whereas a positive screening result could promote anxiety.Conclusions:Systematic screening outperformed routine care but was comparable to opportunistic screening in detecting undiagnosed AF. Systematic screening was related to a reduction in the composite endpoints of stroke and all-cause mortality without increasing the risk of bleeding.PROSPERO Registration:This systematic review was prospectively registered in PROSPERO, registration number: CRD42024558614, https://www.crd.york.ac.uk/PROSPERO/view/CRD42024558614.

e15550 Background: Fruquintinib (fru), regorafenib (reg), and trifluridine/tipiracil (TAS102) are 3 rd line treatments for patients with previously treated metastatic colorectal cancer (mCRC). Further evidence is required to support clinical decision-making in treating mCRC. In the absence of head-to-head studies, we performed a match-adjusted indirect comparison (MAIC) to evaluate the weighted OS and PFS of fru versus the published OS and PFS of reg and TAS102. Methods: Individual participant-level data (IPD) from FRESCO (NCT02314819) and the published aggregate data (AgD) from the Chinese Subgroup of CONCUR (CONCURsub) (NCT01584830) and TERRA (NCT01955837) were used. According to the eligible criteria of the two AgD trials, 14 IPDs with colon and rectum (6 in fru arm and 8 in placebo arm) were excluded in MAIC with TERRA trial. The baseline covariances of IPD were re-weighted to match those of the AgDs separately. Hazard ratios (HRs) were estimated to compare the weighted OS and PFS of fru with the published OS and PFS of reg or TAS102. Due to the poor overlap of the placebo arms between FRESCO and CONCURsub, unanchored MAIC was conducted to compare fru with reg. Anchored MAIC was performed to compare fru with TAS102. Results: All identified covariances of IPD were well-matched with those of AgDs. The effective sample size (ESS) of IPD after matching and the comparative effect estimates for OS and PFS are presented in the table. Fru showed a numerically favorable weighted OS and a statistically superior weighted PFS versus reg, meanwhile fru had a numerically favorable weighted OS and a statistically superior weighted PFS versus TAS102. Fru showed statistically superior weighted PFS in the male and K-Ras wild-type subgroups versus reg, and in the K-Ras wild-type subgroup versus TAS102. Fru showed favorable weighted OS in &lt; 65 years subgroup versus TAS102. Conclusions: Our analyses showed favorable OS and PFS for fru versus reg and TAS102, as well as beneficial PFS in male versus reg, in K-Ras wild-type subgroup versus reg and TAS102, along with superior OS in &lt; 65 years subgroup versus TAS102, highlighting its therapeutic potential in treating patients with previously treated mCRC. Limitations include the inability to adjust all covariances, biases due to unobserved covariances, and lacking comparison between reg and TAS102 due to unavailable IPDs. The findings should be validated through future RWS. Items MAIC with reg MAIC with TAS102 ESS Fru: 157 (56%), Placebo: 49 (36%) Fru: 218 (80%), Placebo: 81 (62%) OSHR * (95% CI) 0.87 (0.57, 1.3) 0.82 (0.57, 1.19)0.62 (0.41, 0.94), &lt; 65 years subgroup PFSHR * (95% CI) 0.72 (0.57, 0.93)0.50 (0.26, 0.95), male subgroup0.39 (0.2, 0.77), K-Ras wild-type subgroup 0.64 (0.44, 0.94)0.43 (0.26, 0.71), K-Ras wild-type subgroup * The upper limit of 95% CI of HR &lt; 1.0 indicates the comparison statistically favors fru.