OBJECTIVE To systematically evaluate the efficacy and safety of 6 kinds of GLP-1RAs in the treatment of type 2 diabetes mellitus （T2DM） patients with overweight or obesity， and to provide evidence-based reference for clinical practice. METHODS A comprehensive search was conducted in PubMed， Embase， Web of Science， the Cochrane Library， CNKI， VIP， Wanfang Data， and CBM from the inception to December 1， 2025. Randomized controlled trials （RCTs） were screened according to inclusion and exclusion criteria. Data extraction and risk of bias assessment were performed on the included studies. Network meta-analysis was conducted using Stata 17.0 software. RESULTS A total of 29 eligible RCTs were included， involving 7 404 patients. Six GLP-1RAs were evaluated： semaglutide， liraglutide， exenatide， dulaglutide， polyethylene glycol loxenatide， and beinaglutide. In terms of glycemic control， semaglutide had the highest probability of ranking first in reducing glycated hemoglobin （HbA1c） and fasting plasma glucose levels， followed by polyethylene glycol loxenatide. In terms of weight management， semaglutide showed the highest probability of ranking first， followed by liraglutide and exenatide. Regarding safety， dulaglutide had the highest probability of ranking first in reducing the incidence of gastrointestinal adverse events； none of the GLP-1RAs significantly increased the risk of severe hypoglycemia. Subgroup analysis revealed that liraglutide 1.8 mg， qd and exenatide extend-release 2.0 mg， qw demonstrated superior efficacy in reducing HbA1c and body weight compared with other doses/dosage forms of the same agents. CONCLUSIONS For T2DM patients with overweight or obesity， semaglutide offers the greatest benefits in glycemic control and weight reduction， while dulaglutide demonstrates superior gastrointestinal tolerability. Liraglutide 1.8 mg， qd and exenatide extend-release 2.0 mg， qw show relatively better overall efficacy in glycemic control and weight reduction among the same agents.

Objective:Using methylation characteristics of human genes to construct machine learning predictive models for screening cervical cancer and precancerous lesions.Methods:Human DNA methylation detection was performed on 224 cervical exfoliated cell specimens from the Cancer Hospital of the Chinese Academy of Medical Sciences, Tianjin Central Hospital of Gynecology Obstetrics, Xinmi Maternal and Child Health Hospital of Henan Province, West China Second Affiliated Hospital of Sichuan University, and Heping Hospital Affiliated to Changzhi Medical College collected during April 2014 and March 2015. The hypermethylated gene fragments related to cervical cancer were selected by high-density, high-association, and hypermethylated gene fragment screening and the LASSO regression algorithm. Taking cervical intraepithelial neoplasia grade 2 (CIN2) or more severe lesions as the research outcome, machine learning predictive models based on the random forest (RF), naive Bayes (NB), and support vector machine (SVM) algorithm, respectively, were constructed. A total of 144 outpatient specimens were used as the training set and 80 cervical exfoliated cell specimens from women participating in the cervical cancer screening program were used as the test set to verify the predictive models. Using histological diagnosis results as the gold standard, the detection efficacy for CIN2 or more severe lesions of the three machine learning predictive models were compared with that of the human papilloma virus (HPV) detection and cytological diagnosis.Results:In the training set of 144 cases, there were 34 cases of HPV positivity, with a positive rate of 23.61%. Cytologically, there were 37 cases diagnosed as no intraepithelial lesion or malignancy (NILM), and 107 cases diagnosed as atypical squamous cells of undetermined significance (ASC-US) or above. Histologically, there were 28 cases without cervical intraepithelial neoplasia or benign cervical lesions, 31 cases of CIN1, 18 cases of CIN2, 31 cases of CIN3, and 36 cases of squamous cell carcinoma. Seven hypermethylated gene fragments were selected from 45 genes, and three machine learning prediction models based on the RF, NB, and SVM algorithm, respectively, were constructed. In the validation set of 80 cases, there were 28 cases of HPV positivity, with a positive rate of 35.00%. Cytologically, there were 65 cases diagnosed as NILM and 15 cases as ASC-US or above. Histologically, there were 39 cases without cervical intraepithelial neoplasia or benign cervical lesions, 10 cases of CIN1, 10 cases of CIN2, 11 cases of CIN3, and 10 cases of squamous cell carcinoma. In the validation set, the area under the curve (AUC) values of the RF model, NB model, SVM model, HPV detection, and cytological diagnosis of CIN2 or above were 0.90, 0.88, 0.82, 0.68, and 0.45, respectively. The DeLong test showed that there was no statistically significant difference in the AUC values between the RF, NB, and SVM models (all P＞0.05), and the AUC values of the RF and NB models were higher than that of HPV detection (both P＜0.01), and the AUC values of the RF, NB, and SVM models were higher than that of cytological diagnosis (all P＜0.01). Compared with the NB model, the sensitivity of the RF model was similar (80.65% vs. 77.42%), but the specificity of the NB model was much higher than that of the RF model (93.88% vs. 73.47%). Conclusion:Among the machine learning prediction models for cervical cancer and precancerous lesions constructed based on human DNA methylation, the NB model has good predictive performance for CIN2 and above lesions, and may be used for screening of cervical cancer and precancerous lesions.

Lymphoma is a highly heterogeneous malignancy characterized by complex molecular regulatory mechanisms that result in significant differences in aggressiveness and prognosis across its subtypes.Gene copy number alteration(CNA)analysis,an emerging technology,has become a pivotal tool in the precision re-search and management of lymphoma.By detecting DNA deletions,amplifications,and chromosomal copy number changes,CNA analysis addresses the limitations of traditional cytogenetic techniques,enhances the ac-curacy of subtype classification,and aids in evaluating tumor heterogeneity and disease progression.This re-view provides a comprehensive summary of CNA detection methods and their applications in lymphoma,with a focus on recent advancements in the field.It offers a comparative analysis of CNA detection techniques and discusses their role in precision diagnosis,subtype classification,monitoring disease progression,predicting therapeutic resistance,and assessing prognosis.Additionally,the review explores the potential applications of CNA analysis in uncovering molecular regulatory mechanisms,optimizing therapeutic strategies,and impro-ving patient survival outcomes.

Objective:To construct a sizeable naive human Fab phage display antibody library to screen high-affinity specific antibodies in vitro. Methods:Total RNA was extracted from peripheral blood mononuclear cells (PBMCs) of 126 healthy individuals, subsequently reverse-transcribed into cDNA, and used as a template. PCR amplification was performed to obtain the V H from IgG, IgM and light chain κ, λ, separately, with the initial PCR products serving as templates for a second round of PCR. Overlap extension PCR was employed to generate fragments of the κ and λ light chains. These fragments were ligated with the phage vector pNC3, which harbors the variable region 1 of the heavy chain, to construct a recombinant phage plasmid. This plasmid was then electroporated into competent Escherichia Coli TG1 cells to establish a naive human Fab phage display antibody library. One hundred clones were randomly selected for identification and sequencing, and antibody gene polymorphisms were analyzed using the IMGT database and MAFFT software. Recombinant α-hemolysin from Staphylococcus aureus was utilized to screen Fab antibody fragments through biopanning of the antibody library, followed by random selection of phage ELISA-identified clones. The positive clones (antigen A450∶blank control A450≥2.1) were sequenced. Results:Two large naive Fab phage display antibody libraries were successfully constructed, in which the capacity of κ and λ chain antibody libraries were 1.25×10 11 and 1.54×10 11, respectively. The titers for two antibody libraries were 6.04×10 13 CFU/ml and 3.50×10 13 CFU/ml. The positive transformation insertion rates for κ and λ chain antibody libraries were 96% (96/100) and 100% (100/100), respectively. Sequence analysis revealed that all antibody sequences were unique. The amino acid sequences in the skeletal region were relatively conserved. In contrast, significant variations in the length of the complementarity determining region (CDR) were found, and the diversity of amino acid sequence of the complementary determining region was high, especially the CDR3. Analysis using the IMGT database indicated that the sequences exhibited a broad distribution across variable-diversity-joining gene families. After six rounds of panning, specific phage antibodies enrichment targeting α-hemolysin were achieved. A total of 142 monoclonal antibodies were sequenced, yielding 8 distinct Fab antibody sequences. Conclusion:This study successfully constructed two naive human Fab phage display antibody libraries with large capacity and good diversity, which can be used for screening human antibodies for serum epidemiology.

Objective:To construct a sizeable naive human Fab phage display antibody library to screen high-affinity specific antibodies in vitro. Methods:Total RNA was extracted from peripheral blood mononuclear cells (PBMCs) of 126 healthy individuals, subsequently reverse-transcribed into cDNA, and used as a template. PCR amplification was performed to obtain the V H from IgG, IgM and light chain κ, λ, separately, with the initial PCR products serving as templates for a second round of PCR. Overlap extension PCR was employed to generate fragments of the κ and λ light chains. These fragments were ligated with the phage vector pNC3, which harbors the variable region 1 of the heavy chain, to construct a recombinant phage plasmid. This plasmid was then electroporated into competent Escherichia Coli TG1 cells to establish a naive human Fab phage display antibody library. One hundred clones were randomly selected for identification and sequencing, and antibody gene polymorphisms were analyzed using the IMGT database and MAFFT software. Recombinant α-hemolysin from Staphylococcus aureus was utilized to screen Fab antibody fragments through biopanning of the antibody library, followed by random selection of phage ELISA-identified clones. The positive clones (antigen A450∶blank control A450≥2.1) were sequenced. Results:Two large naive Fab phage display antibody libraries were successfully constructed, in which the capacity of κ and λ chain antibody libraries were 1.25×10 11 and 1.54×10 11, respectively. The titers for two antibody libraries were 6.04×10 13 CFU/ml and 3.50×10 13 CFU/ml. The positive transformation insertion rates for κ and λ chain antibody libraries were 96% (96/100) and 100% (100/100), respectively. Sequence analysis revealed that all antibody sequences were unique. The amino acid sequences in the skeletal region were relatively conserved. In contrast, significant variations in the length of the complementarity determining region (CDR) were found, and the diversity of amino acid sequence of the complementary determining region was high, especially the CDR3. Analysis using the IMGT database indicated that the sequences exhibited a broad distribution across variable-diversity-joining gene families. After six rounds of panning, specific phage antibodies enrichment targeting α-hemolysin were achieved. A total of 142 monoclonal antibodies were sequenced, yielding 8 distinct Fab antibody sequences. Conclusion:This study successfully constructed two naive human Fab phage display antibody libraries with large capacity and good diversity, which can be used for screening human antibodies for serum epidemiology.

OBJECTIVE To explore the improvement effect and mechanism of salidroside on radiation-induced parotid gland injury in rats. METHODS Rats were randomly assigned into normal group， radiation group， salidroside low-dose （salidroside-L， 50 mg/kg） group， salidroside high-dose （salidroside-H， 100 mg/kg） group， and salidroside-H+inhibitor （100 mg/kg salidroside+0.1 µmol/kg H-89） group， with 10 rats in each group. Except for the normal group， rats in the other groups were subjected to radiation exposure to establish a model of radiation-induced parotid gland injury. Rats in each group were intraperitoneally injected with the corresponding drug or normal saline， once a day， for 40 consecutive days. After the last administration， the levels of reactive oxygen species （ROS）， cyclic adenosine monophosphate （cAMP）， superoxide dismutase （SOD）， and amylase in serum were detected； the pathological changes of parotid gland tissue were observed； the apoptosis rate of parotid gland tissue cells， the expression levels of B-cell lymphoma-2 （Bcl-2） and its associated X protein （Bax）， mRNA expression levels of interleukin-6 （IL- 6） and tumor necrosis factor-α （TNF-α）， the protein expression levels of type Ⅲ collagen （Col Ⅲ）， vasoactive intestinal peptide （VIP）， and the phosphorylation level of protein kinase A （PKA） in parotid gland tissue were determined. RESULTS Compared with normal group， the levels of ROS， amylase， apoptosis rate， Bax expression level， mRNA expression levels of IL-6 and TNF- α， and protein expression level of Col Ⅲ in the radiation group were significantly increased， while the levels of cAMP， SOD， Bcl-2 expression level， VIP protein expression level and PKA phosphorylation level were significantly decreased （P＜0.05）. Compared with radiation group， the above indicators in the salidroside-L group and salidroside-H group were significantly improved （P＜0.05）， and the improvement in the salidroside-H group was more significant （P＜0.05）； inhibitor H-89 significantly reversed the changes in the above indicators of the salidroside-H group （P＜0.05）. CONCLUSIONS Salidroside can effectively alleviate radiation-induced parotid gland injury in rats， and its mechanism may be related to the activation of the VIP-cAMP pathway.

Objective To explore the value of time-dependent diffusion MRI(td-dMRI)parameters for differentiating invasive breast cancer(IBC)with ductal carcinoma in situ(DCIS)(IBC-DCIS)from simple IBC.Methods A total of 19 patients with IBC-DCIS(IBC-DCIS group)and 53 patients with simple IBC(IBC group)confirmed by surgery and postoperation pathology were retrospectively enrolled.Breast td-dMRI acquired with oscillating gradient spin-echo(OGSE)and pulsed gradient spin-echo(PGSE)sequences before operation were interpreted,and apparent diffusion coefficient(ADC)and microstructure parameters,including OGSE-ADC value,PGSE-ADC value,cellularity,cell diameter,intracellular volume fraction and extracellular diffusion coefficient were obtained and compared between groups.Receiver operating characteristic(ROC)curves of parameters being significantly different between groups were plotted,and the area under the curve(AUC)was calculated to evaluate the efficacy of these parameters for differentiating IBC-DCIS from IBC.Results Significant differences of OGSE-ADC value,PGSE-ADC value,cellularity,cell diameter,intracellular volume fraction and extracellular diffusion coefficient were found between groups(all P＜0.05).The AUC of the above parameters for differentiating IBC-DCIS from IBC was 0.81,0.79,0.78,0.68,0.77 and 0.81,respectively.Conclusion td-dMRI parameters could be used to noninvasively and effectively differentiate IBC-DCIS from simple IBC.

Objective To explore the value of time-dependent diffusion MRI(td-dMRI)parameters for differentiating invasive breast cancer(IBC)with ductal carcinoma in situ(DCIS)(IBC-DCIS)from simple IBC.Methods A total of 19 patients with IBC-DCIS(IBC-DCIS group)and 53 patients with simple IBC(IBC group)confirmed by surgery and postoperation pathology were retrospectively enrolled.Breast td-dMRI acquired with oscillating gradient spin-echo(OGSE)and pulsed gradient spin-echo(PGSE)sequences before operation were interpreted,and apparent diffusion coefficient(ADC)and microstructure parameters,including OGSE-ADC value,PGSE-ADC value,cellularity,cell diameter,intracellular volume fraction and extracellular diffusion coefficient were obtained and compared between groups.Receiver operating characteristic(ROC)curves of parameters being significantly different between groups were plotted,and the area under the curve(AUC)was calculated to evaluate the efficacy of these parameters for differentiating IBC-DCIS from IBC.Results Significant differences of OGSE-ADC value,PGSE-ADC value,cellularity,cell diameter,intracellular volume fraction and extracellular diffusion coefficient were found between groups(all P＜0.05).The AUC of the above parameters for differentiating IBC-DCIS from IBC was 0.81,0.79,0.78,0.68,0.77 and 0.81,respectively.Conclusion td-dMRI parameters could be used to noninvasively and effectively differentiate IBC-DCIS from simple IBC.

Objective:Using methylation characteristics of human genes to construct machine learning predictive models for screening cervical cancer and precancerous lesions.Methods:Human DNA methylation detection was performed on 224 cervical exfoliated cell specimens from the Cancer Hospital of the Chinese Academy of Medical Sciences, Tianjin Central Hospital of Gynecology Obstetrics, Xinmi Maternal and Child Health Hospital of Henan Province, West China Second Affiliated Hospital of Sichuan University, and Heping Hospital Affiliated to Changzhi Medical College collected during April 2014 and March 2015. The hypermethylated gene fragments related to cervical cancer were selected by high-density, high-association, and hypermethylated gene fragment screening and the LASSO regression algorithm. Taking cervical intraepithelial neoplasia grade 2 (CIN2) or more severe lesions as the research outcome, machine learning predictive models based on the random forest (RF), naive Bayes (NB), and support vector machine (SVM) algorithm, respectively, were constructed. A total of 144 outpatient specimens were used as the training set and 80 cervical exfoliated cell specimens from women participating in the cervical cancer screening program were used as the test set to verify the predictive models. Using histological diagnosis results as the gold standard, the detection efficacy for CIN2 or more severe lesions of the three machine learning predictive models were compared with that of the human papilloma virus (HPV) detection and cytological diagnosis.Results:In the training set of 144 cases, there were 34 cases of HPV positivity, with a positive rate of 23.61%. Cytologically, there were 37 cases diagnosed as no intraepithelial lesion or malignancy (NILM), and 107 cases diagnosed as atypical squamous cells of undetermined significance (ASC-US) or above. Histologically, there were 28 cases without cervical intraepithelial neoplasia or benign cervical lesions, 31 cases of CIN1, 18 cases of CIN2, 31 cases of CIN3, and 36 cases of squamous cell carcinoma. Seven hypermethylated gene fragments were selected from 45 genes, and three machine learning prediction models based on the RF, NB, and SVM algorithm, respectively, were constructed. In the validation set of 80 cases, there were 28 cases of HPV positivity, with a positive rate of 35.00%. Cytologically, there were 65 cases diagnosed as NILM and 15 cases as ASC-US or above. Histologically, there were 39 cases without cervical intraepithelial neoplasia or benign cervical lesions, 10 cases of CIN1, 10 cases of CIN2, 11 cases of CIN3, and 10 cases of squamous cell carcinoma. In the validation set, the area under the curve (AUC) values of the RF model, NB model, SVM model, HPV detection, and cytological diagnosis of CIN2 or above were 0.90, 0.88, 0.82, 0.68, and 0.45, respectively. The DeLong test showed that there was no statistically significant difference in the AUC values between the RF, NB, and SVM models (all P＞0.05), and the AUC values of the RF and NB models were higher than that of HPV detection (both P＜0.01), and the AUC values of the RF, NB, and SVM models were higher than that of cytological diagnosis (all P＜0.01). Compared with the NB model, the sensitivity of the RF model was similar (80.65% vs. 77.42%), but the specificity of the NB model was much higher than that of the RF model (93.88% vs. 73.47%). Conclusion:Among the machine learning prediction models for cervical cancer and precancerous lesions constructed based on human DNA methylation, the NB model has good predictive performance for CIN2 and above lesions, and may be used for screening of cervical cancer and precancerous lesions.

More than 300 million people world-wide suffer from asthma,and the incidence is in-creasing year by year.As one of the most common chronic diseases,asthma is an immune-mediated inflammatory disease with complex triggering mechanisms and strong heterogeneity.With the in-depth study of physiological and pathological mech-anisms,therapeutic small molecule and hormone drugs have been introduced to control and treat most patients,but about 5％-10％of patients still suffer from various subtypes of difficult to control and treat asthma,that is,severe asthma.In the past decade,with the rapid development of bio-pharmaceutical research,protein and antibody have become the key drugs for the treatment of se-vere asthma with high efficacy,high specificity and high safety.However,biological drugs are usually administered by injection,they cannot be noninva-sive and directly delivered into the lung to quickly absorb and take effect.Therefore,there is an ur-gent need for the introduction of inhaled biologics with quick effectiveness,convenience,economy and safety in clinical.The review summarizes the existing small molecule,hormone and biological therapy drugs,and summarizes the development of inhalable biological agents of asthma,and ana-lyzes the future prospects of the inhalable biologi-cal drugs,which is designed to deepen the percep-tion of the direction of the inhalable biological drugs research,and update the information of the field,in order to provide reference for the develop-ment of more inhalable biologics.