1.Influence of short peptide enteral nutrition preparation on nutritional status, immune function and adverse actions in children with acute lymphoblastic leukemia
Zongxuan HUANG ; Tao LI ; Fen SHENG ; Man WANG ; Tiantian HU ; Teng ZHANG ; Wenli ZHANG ; Yanling TAO
Clinical Medicine of China 2025;41(4):273-281
Objective:To investigate the effects of short-peptide-based enteral nutrition on nutritional status, immune function, and chemotherapy-related adverse reactions in children with acute lymphoblastic leukemia (ALL).Methods:A total of 106 children with ALL receiving chemotherapy at the Affiliated Hospital of Jining Medical University between January 2021 and April 2022 were enrolled. According to the principle of between-group baseline data matching, the patients were divided into observation group and control group by random number table method, with 53 cases in each group. All patients received chemotherapy according to the CCCG-ALL-2020 protocol established by the Multi-center Cooperative Group of the Chinese Society of Pediatric Oncology (2020). The control group received a regular diet, while the observation group received a regular diet supplemented with short-peptide-based enteral nutrition. The incidence rates of malnutrition, hypoproteinemia, hypoalbuminemia, abnormal immunoglobulin levels (IgG, IgM, IgA), and adverse reactions (liver injury, infection) were compared between both groups before chemotherapy and at the end of each of the following seven chemotherapy phases: Induction remission therapy (PVDL), Induction remission therapy (CAT), Early intensification therapy (CAT+), Consolidation therapy (HDMTX), Interim maintenance therapy, Reinduction therapy, and prior to the end of Maintenance therapy. Normally or approximately normally distributed measurement data were expressed as xˉ± s and compared by independent samples t-test. Counting data were expressed as n (%) and compared by χ2 test. Results:During the CAT phase, the incidence of malnutrition was significantly lower in the observation group than in the control group [20.8% (11/53) vs. 39.6% (21/53), χ2=4.48, P=0.034]. The incidence of hypoproteinemia was significantly lower in the observation group during HDMTX, Reinduction, Interim maintenance, and prior to the end of Maintenance therapy [47.2% (25/53) vs. 69.8% (37/53), χ2=5.60, P=0.018; 45.3% (24/53) vs. 67.9% (36/53), χ2=5.53, P=0.019; 41.5% (24/53) vs. 64.2% (34/53), χ2=5.45, P=0.020; 28.3% (15/53) vs. 54.7% (29/53), χ2=7.62, P=0.006, respectively]. The incidence of hypoalbuminemia was significantly lower in the observation group during CAT+, HDMTX, Reinduction, Interim maintenance, and prior to the end of Maintenance therapy [5.7% (3/53) vs. 22.6% (12/53), χ2=6.29, P=0.012; 9.4% (5/53) vs. 26.4% (14/53), χ2=5.19, P=0.023; 9.4% (5/53) vs. 28.3% (15/53), χ2=6.16, P=0.013; 7.6% (4/53) vs. 24.5% (13/53), χ2=5.68, P=0.017; 3.8% (2/53) vs. 18.9% (10/53), χ2=6.01, P=0.014, respectively]. For IgG, incidence was significantly lower in the observation group during Interim maintenance, Reinduction, and prior to the end of Maintenance therapy [7.6% (4/53) vs. 22.6% (12/53), χ2=4.71, P=0.030; 20.8% (11/53) vs. 39.6% (21/53), χ2=4.48, P=0.034; 11.3% (6/53) vs. 26.4% (14/53), χ2=3.94, P=0.047, respectively]. For IgM, incidence was significantly lower in the observation group during the CAT and CAT+ phases [45.3% (24/53) vs. 66.0% (35/53), χ2=4.63, P=0.032; 58.5% (31/53) vs. 77.4% (41/53), χ2=4.33, P=0.037, respectively]. For IgA, incidence was significantly lower in the observation group during Reinduction therapy and Interim maintenance [22.6% (12/53) vs. 45.3% (24/53), χ2=6.06, P=0.014; 9.4% (5/53) vs. 24.5% (13/53), χ2=4.28, P=0.038, respectively]. For liver injury, incidence was significantly lower in the observation group during the CAT, CAT+, and prior to the end of Maintenance phases [22.6% (12/53) vs. 43.4% (23/53), χ2=5.16, P=0.023; 26.4% (14/53) vs. 50.9% (27/53), χ2=6.72, P=0.010, 11.3% (6/53) vs. 26.4%(14/53), χ2=3.94、 P=0.047,respectively]. For infection, incidence was significantly lower in the observation group during the CAT+ and HDMTX phases [35.9% (19/53) vs. 56.6% (30/53), χ2=4.59, P=0.032; 24.5% (13/53) vs. 43.4% (23/53), χ2=4.21, P=0.040, respectively]. Conclusions:Short-peptide-based enteral nutrition demonstrates significant advantages in the treatment of pediatric ALL. It provides substantial support for patient treatment and recovery by improving nutritional status, modulating immune function, and reducing chemotherapy-related adverse reactions.
2.Synthesis of A New Naphthalenesulfonamide-based"Turn-on"Fluorescent Probe for Rapid Detection of Glyphosate
Rong-Rong ZHAO ; Hong-Lin LIU ; Ying-Ping HUANG ; Cui-Wen DENG ; Song-Yan LI ; Shui-Lian YU ; Mao-Sheng TAO ; Yi-Qun TIAN ; Xi YUAN
Chinese Journal of Analytical Chemistry 2025;53(6):903-913
Widespread utilization of glyphosate has led to environmental residues,posing potential threats to ecological systems and human health.Traditional methods for detection of glyphosate are limited by specialized equipment and operational techniques,resulting in inefficient responses.Therefore,it is urgent to develop a convenient,sensitive and accurate detection method for detection of glyphosate.Herein,a new naphthalenesulfonamide-based"Turn-on"fluorescent probe was synthesized using 2-chloroaniline and dansyl chloride as raw materials through a one-step process,which showed a good linear relationship between the glyphosate concentration in concentration range of 0.003-70 μmol/L and the fluorescence intensity(R2=0.995),with a detection limit of 2.73 nmol/L(S/N=3).Analytical techniques such as nuclear magnetic resonance(NMR)spectroscopy and high-resolution mass spectrometry(HRMS)were used to investigate the interaction mechanism between the fluorescent probe and glyphosate.The results indicated that a nucleophilic substitution reaction occurred between the probe and the secondary amine(—NH—)of glyphosate,inducing a photoinduced electron transfer(PET)effect which enhanced the fluorescence intensity by 11.2 times.The probe showed good anti-interference ability towards coexisting metal ions,anions and pesticides in water.When applied to determination of glyphosate in the samples such as tap water,river water(Xiangxi River Reservoir),soil,soybeans,and corn,the spiking recoveries ranged from 94.7%to 109.9%,demonstrating the high accuracy and broad applicability of this detection method.A portable test strip based on this fluorescent probe was developed for rapid semi-quantitative analysis of glyphosate.The developed method was rapid,sensitive,and portable,providing theoretical and technical support for on-site measurement of environmental contaminants.
3.Polarity-extended Liquid Chromatography-Mass Spectrometry System for Prostate Cancer Biomarker Screening Based on Extracellular Vesicles
Lu-Lu XIAO ; Meng-Xuan CHEN ; Shan-Shan PAN ; Yi-Chen WANG ; Tao-Hong HUANG ; Qi-Sheng ZHONG ; Yong CHEN ; Teng-Fei XU ; Jia-Hui ZHAO ; Xue-Song LIU
Chinese Journal of Analytical Chemistry 2025;53(11):1848-1859,中插4-中插29
Integrated metabolomic and lipidomic profiling,utilizing liquid chromatography coupled with high-resolution mass spectrometry(LC-HRMS),has emerged as a pivotal strategy for biomarker discovery.However,the inherent polarity disparity between metabolites and lipids complicates simultaneous analysis.To address this,a dual-stationary phase polarity-extended liquid chromatography(PELC)system was developed,which surpassed conventional one-dimensional LC(1D-LC)by enabling comprehensive coverage of both polar and non-polar compounds within a single injection.This system enhanced chromatographic resolution,peak capacity,and throughput while minimizing analytical variability.Extracellular vesicles(EVs),lipid bilayer-enclosed nanoparticles ubiquitously present in biofluids,had gained prominence as reservoirs of cancer biomarkers due to their cargo stability and pathophysiological relevance.Herein,the application of PELC-HRMS for concurrent metabolome-lipidome profiling in EVs was pioneered.A total of 193 metabolites were identified using this technique coupled with MS-DIAL software and Human Metabolome Database.Subsequently,this technique was employed to explore potential biomarkers for prostate cancer(PCa).Multivariate analysis identified 17 differentially abundant metabolites in PCa,implicating dysregulated pathways including purine metabolism,starch and sucrose metabolism,galactose metabolism,cysteine and methionine metabolism,and biosynthesis of unsaturated fatty acids.Notably,creatine(AUC=0.92)and DG 42:5(AUC=0.80)demonstrated robust diagnostic efficacy,attributable to their broad polarity ranges and EV-specific enrichment.This study established PELC as a high-fidelity platform for multi-omics integration in complex biospecimens,advancing mechanistic insights into metabolic rewiring and disease pathophysiology.
4.A prediction model for in-hospital mortality in elderly patients undergoing unsynchronous cardioversion in ICU
Dan HUANG ; Manli YUAN ; Xiaowen ZUO ; Yongjie XU ; Ye TAO ; Sheng MA ; Zhao YIN
Chinese Journal of Geriatric Heart Brain and Vessel Diseases 2025;27(9):1193-1198
Objective To construct a prediction model for in-hospital mortality in the elderly(≥65 years)patients undergoing unsynchronous cardioversion in ICU and to evaluate its effectiveness.Methods A retrospective study was conducted on 276 elderly eligible patients in the ICU of the Ninth and the First Medical Centers of Chinese PLA General Hospital between June 2022 and August 2024.According to their clinical outcomes,they were divided into a non-in-hospital dead group(111 cases)and an in-hospital dead group(165 cases).Clinical data were collected,and pre-dictive factors for in-hospital mortality were screened.And then a nomogram prediction model was developed based on the obtained predictive factors,which was evaluated with ROC curve and deci-sion curve analyses.Results When compared to the non-in-hospital dead group,the in-hospital dead group had significantly higher heart rate,ratio of hemodialysis,and levels of alanine amin-otransferase,aspartate aminotransferase,lactate dehydrogenase,alkaline phosphatase,serum cre-atinine,blood glucose,lactate,low base excess,sequential organ failure assessment(SOFA)score,model for end-stage liver disease score,and larger proportions of ventricular fibrillation/flutter and structural heart disease induced by pulseless ventricular tachycardia,and had significantly lower Glasgow Coma Scale(GCS)(P<0.05).Multivariate logistic regression analysis identified body temperature>37℃(OR=0.426,95%CI:0.198-0.915,P=0.029),chronic obstructive pulmonary disease(OR=2.333,95%CI:1.217-4.473,P=0.011),GCS score(OR=0.622,95%CI:0.410-0.944,P=0.026),hemoglobin(OR=0.817,95%CI:0.715-0.934,P=0.003),lactate(OR=1.365,95%CI:1.174-1.587,P=0.000),heart rate>100 bpm(OR=2.757,95%CI:1.397-5.441,P=0.003),and SOFA score(OR=1.112,95%CI:1.032-1.198,P=0.005)as pre-dictors of in-hospital mortality.ROC curve analysis showed an AUC value of above indicators combined together in the prediction was 0.797,with a sensitivity of 76.97%and a specificity of 65.77%.Calibration curve analysis demonstrated good consistency between predicted and observed outcomes.Decision curve analysis indicated favorable clinical utility of the model.Conclusion This study identifies independent risk factors for in-hospital mortality among elderly patients in the ICU who underwent asynchronous cardioversion.Based on these factors,a nomo-gram model is established,demonstrating good discrimination,calibration,and model fit,with high clinical applicability.
5.Influence of short peptide enteral nutrition preparation on nutritional status, immune function and adverse actions in children with acute lymphoblastic leukemia
Zongxuan HUANG ; Tao LI ; Fen SHENG ; Man WANG ; Tiantian HU ; Teng ZHANG ; Wenli ZHANG ; Yanling TAO
Clinical Medicine of China 2025;41(4):273-281
Objective:To investigate the effects of short-peptide-based enteral nutrition on nutritional status, immune function, and chemotherapy-related adverse reactions in children with acute lymphoblastic leukemia (ALL).Methods:A total of 106 children with ALL receiving chemotherapy at the Affiliated Hospital of Jining Medical University between January 2021 and April 2022 were enrolled. According to the principle of between-group baseline data matching, the patients were divided into observation group and control group by random number table method, with 53 cases in each group. All patients received chemotherapy according to the CCCG-ALL-2020 protocol established by the Multi-center Cooperative Group of the Chinese Society of Pediatric Oncology (2020). The control group received a regular diet, while the observation group received a regular diet supplemented with short-peptide-based enteral nutrition. The incidence rates of malnutrition, hypoproteinemia, hypoalbuminemia, abnormal immunoglobulin levels (IgG, IgM, IgA), and adverse reactions (liver injury, infection) were compared between both groups before chemotherapy and at the end of each of the following seven chemotherapy phases: Induction remission therapy (PVDL), Induction remission therapy (CAT), Early intensification therapy (CAT+), Consolidation therapy (HDMTX), Interim maintenance therapy, Reinduction therapy, and prior to the end of Maintenance therapy. Normally or approximately normally distributed measurement data were expressed as xˉ± s and compared by independent samples t-test. Counting data were expressed as n (%) and compared by χ2 test. Results:During the CAT phase, the incidence of malnutrition was significantly lower in the observation group than in the control group [20.8% (11/53) vs. 39.6% (21/53), χ2=4.48, P=0.034]. The incidence of hypoproteinemia was significantly lower in the observation group during HDMTX, Reinduction, Interim maintenance, and prior to the end of Maintenance therapy [47.2% (25/53) vs. 69.8% (37/53), χ2=5.60, P=0.018; 45.3% (24/53) vs. 67.9% (36/53), χ2=5.53, P=0.019; 41.5% (24/53) vs. 64.2% (34/53), χ2=5.45, P=0.020; 28.3% (15/53) vs. 54.7% (29/53), χ2=7.62, P=0.006, respectively]. The incidence of hypoalbuminemia was significantly lower in the observation group during CAT+, HDMTX, Reinduction, Interim maintenance, and prior to the end of Maintenance therapy [5.7% (3/53) vs. 22.6% (12/53), χ2=6.29, P=0.012; 9.4% (5/53) vs. 26.4% (14/53), χ2=5.19, P=0.023; 9.4% (5/53) vs. 28.3% (15/53), χ2=6.16, P=0.013; 7.6% (4/53) vs. 24.5% (13/53), χ2=5.68, P=0.017; 3.8% (2/53) vs. 18.9% (10/53), χ2=6.01, P=0.014, respectively]. For IgG, incidence was significantly lower in the observation group during Interim maintenance, Reinduction, and prior to the end of Maintenance therapy [7.6% (4/53) vs. 22.6% (12/53), χ2=4.71, P=0.030; 20.8% (11/53) vs. 39.6% (21/53), χ2=4.48, P=0.034; 11.3% (6/53) vs. 26.4% (14/53), χ2=3.94, P=0.047, respectively]. For IgM, incidence was significantly lower in the observation group during the CAT and CAT+ phases [45.3% (24/53) vs. 66.0% (35/53), χ2=4.63, P=0.032; 58.5% (31/53) vs. 77.4% (41/53), χ2=4.33, P=0.037, respectively]. For IgA, incidence was significantly lower in the observation group during Reinduction therapy and Interim maintenance [22.6% (12/53) vs. 45.3% (24/53), χ2=6.06, P=0.014; 9.4% (5/53) vs. 24.5% (13/53), χ2=4.28, P=0.038, respectively]. For liver injury, incidence was significantly lower in the observation group during the CAT, CAT+, and prior to the end of Maintenance phases [22.6% (12/53) vs. 43.4% (23/53), χ2=5.16, P=0.023; 26.4% (14/53) vs. 50.9% (27/53), χ2=6.72, P=0.010, 11.3% (6/53) vs. 26.4%(14/53), χ2=3.94、 P=0.047,respectively]. For infection, incidence was significantly lower in the observation group during the CAT+ and HDMTX phases [35.9% (19/53) vs. 56.6% (30/53), χ2=4.59, P=0.032; 24.5% (13/53) vs. 43.4% (23/53), χ2=4.21, P=0.040, respectively]. Conclusions:Short-peptide-based enteral nutrition demonstrates significant advantages in the treatment of pediatric ALL. It provides substantial support for patient treatment and recovery by improving nutritional status, modulating immune function, and reducing chemotherapy-related adverse reactions.
6.A prediction model for in-hospital mortality in elderly patients undergoing unsynchronous cardioversion in ICU
Dan HUANG ; Manli YUAN ; Xiaowen ZUO ; Yongjie XU ; Ye TAO ; Sheng MA ; Zhao YIN
Chinese Journal of Geriatric Heart Brain and Vessel Diseases 2025;27(9):1193-1198
Objective To construct a prediction model for in-hospital mortality in the elderly(≥65 years)patients undergoing unsynchronous cardioversion in ICU and to evaluate its effectiveness.Methods A retrospective study was conducted on 276 elderly eligible patients in the ICU of the Ninth and the First Medical Centers of Chinese PLA General Hospital between June 2022 and August 2024.According to their clinical outcomes,they were divided into a non-in-hospital dead group(111 cases)and an in-hospital dead group(165 cases).Clinical data were collected,and pre-dictive factors for in-hospital mortality were screened.And then a nomogram prediction model was developed based on the obtained predictive factors,which was evaluated with ROC curve and deci-sion curve analyses.Results When compared to the non-in-hospital dead group,the in-hospital dead group had significantly higher heart rate,ratio of hemodialysis,and levels of alanine amin-otransferase,aspartate aminotransferase,lactate dehydrogenase,alkaline phosphatase,serum cre-atinine,blood glucose,lactate,low base excess,sequential organ failure assessment(SOFA)score,model for end-stage liver disease score,and larger proportions of ventricular fibrillation/flutter and structural heart disease induced by pulseless ventricular tachycardia,and had significantly lower Glasgow Coma Scale(GCS)(P<0.05).Multivariate logistic regression analysis identified body temperature>37℃(OR=0.426,95%CI:0.198-0.915,P=0.029),chronic obstructive pulmonary disease(OR=2.333,95%CI:1.217-4.473,P=0.011),GCS score(OR=0.622,95%CI:0.410-0.944,P=0.026),hemoglobin(OR=0.817,95%CI:0.715-0.934,P=0.003),lactate(OR=1.365,95%CI:1.174-1.587,P=0.000),heart rate>100 bpm(OR=2.757,95%CI:1.397-5.441,P=0.003),and SOFA score(OR=1.112,95%CI:1.032-1.198,P=0.005)as pre-dictors of in-hospital mortality.ROC curve analysis showed an AUC value of above indicators combined together in the prediction was 0.797,with a sensitivity of 76.97%and a specificity of 65.77%.Calibration curve analysis demonstrated good consistency between predicted and observed outcomes.Decision curve analysis indicated favorable clinical utility of the model.Conclusion This study identifies independent risk factors for in-hospital mortality among elderly patients in the ICU who underwent asynchronous cardioversion.Based on these factors,a nomo-gram model is established,demonstrating good discrimination,calibration,and model fit,with high clinical applicability.
7.Design and application of clinical disease database system based on post-structured technology
Nan LI ; Mi-Ye WANG ; Tao ZHENG ; Yan-Sheng LI ; Da-Peng JIANG ; Yong HUANG
Chinese Medical Equipment Journal 2024;45(4):20-26
Objective To design a post-structured technology-based clinical disease database system to solve the problems of the traditional disease database system in dependence on manual judgment,lack of auxiliary annotation and poor availability of electronic medical record data.Methods An entity recognition model was constructed with I2B2 standard and bi-directional long short-term memory(BiLSTM)model to form a medical record template library,and some relational templates were generated and complex medical entities were extracted to realize post-structuring of electronic medical records.Then a clinical disease database system was established based on the post-structured electronic medical record technology,which was composed of the modules for medical record structuring,structured assessment,data annotation,routine functions and system management.Results The system developed transformed the text of electronic medical records into structured language,contributed to data element extraction and intelligent structured service and enhanced the efficiency of clinical treatment and scientific research.Conclusion The system developed improves the data availability of clinical diseases,reduces the workload of user data processing,ensures the quality of data application and lays a foundation for assisted decision making during clinical treatment and scientific research.[Chinese Medical Equipment Journal,2024,45(4):20-26]
8.Analysis on genetic etiology of disorders of sex development using whole exome sequencing
Liu-Jiao WU ; Chan-Chan JIN ; Shu ZHU ; Wen-Ming HUANG ; Jian-Hong YE ; Tao LYU ; Bao-Sheng ZHU ; Jing HE
Medical Journal of Chinese People's Liberation Army 2024;49(5):495-503
Objective To elucidate the molecular genetic etiology of patients with disorders of sex development(DSD)using whole exome sequencing(WES),thereby enhancing our understanding of the underlying mechanisms of sexual development abnormalities.Methods Retrospective analysis was conducted on clinical data of 60 DSD patients diagnosed in the First People's Hospital of Yunnan Province between March 2008 and August 2021,with an additional family study for one proband.Genomic DNA was extracted from patients for WES analysis.Single nucleotide polymorphism(SNP)and insertions/deletion(InDel)tests were identified using SAMtools software in conjunction with established SNP and InDel databases.Copy number variations(CNVs)at the exon level were detected using ExomeDepth,while the potential pathogenicity of mutations was predicted with PolyPhen-2,Mutation taster and PyMol software,with Sanger sequencing employed for confirmation.Results The study included 22 patients with 46,XX DSD and 38 with 46,XY DSD.Among the 46,XX DSD patients,the SRY gene was detected in 14 patients.In the remaining 8 patients and a proband's families,single nucleotide site variations(SNVs)of NR5A1,PROKR2 and ANOS1 genes were identified in 2 patients,and CNVs in CYP21A2 gene were found in 4 patients.The pathogenicity of CYP21A2 EX1 Dup has been previously reported,while the remaining 3 CNVs were of uncertain significance,and no DSD-related mutations were detected in 2 patients.In the WES analysis of 46,XY DSD patients,10 pathogenic or likely pathogenic SNVs across 5 genes(SRY,AR,SRD5A2,CYP17A1,and NR5A1)were identified in 14 patients.Additionally,5 likely pathogenic CNVs involving the CYP21A2,AKR1C2,CBX2,and NR5A1 genes were detected in 5 patients,comprising 3 deletions and 2 duplications.Novel SNVs in NR5A1(c.722G>T,c.48C>G)and ANOS1 c.564A>T were identified,with no prior reports in relevant databases.The pathogenicity of CYP21A2 EX1 Dup is documented in related databases,while the remaining CNVs have not been previously reported.Conclusion The utilization of WES technology has enhanced the diagnostic potential for DSD,broadened the spectrum of known DSD-related gene mutations,and deepened our comprehension of DSD pathogenesis,offering valuable support for genetic counseling.
9.Research status of quercetin-mediated MAPK signaling pathway in prevention and treatment of osteoporosis
Ke-Xin YUAN ; Xing-Wen XIE ; Ding-Peng LI ; Yi-Sheng JING ; Wei-Wei HUANG ; Xue-Tao WANG ; Hao-Dong YANG ; Wen YAN ; Yong-Wu MA
The Chinese Journal of Clinical Pharmacology 2024;40(9):1375-1379
Quercetin can mediate the activation of mitogen-activated protein kinase(MAPK)signaling pathways to prevent osteoporosis(OP).This paper comprehensively discusses the interrelationship between MAPK and osteoporosis-related cells based on the latest domestic and international research.Additionally,it elucidates the research progress of quercetin in mediating the MAPK signaling pathway for OP prevention.The aim is to provide an effective foundation for the clinical prevention and treatment of OP and the in-depth development of quercetin.
10.Artificial intelligence predicts direct-acting antivirals failure among hepatitis C virus patients: A nationwide hepatitis C virus registry program
Ming-Ying LU ; Chung-Feng HUANG ; Chao-Hung HUNG ; Chi‐Ming TAI ; Lein-Ray MO ; Hsing-Tao KUO ; Kuo-Chih TSENG ; Ching-Chu LO ; Ming-Jong BAIR ; Szu-Jen WANG ; Jee-Fu HUANG ; Ming-Lun YEH ; Chun-Ting CHEN ; Ming-Chang TSAI ; Chien-Wei HUANG ; Pei-Lun LEE ; Tzeng-Hue YANG ; Yi-Hsiang HUANG ; Lee-Won CHONG ; Chien-Lin CHEN ; Chi-Chieh YANG ; Sheng‐Shun YANG ; Pin-Nan CHENG ; Tsai-Yuan HSIEH ; Jui-Ting HU ; Wen-Chih WU ; Chien-Yu CHENG ; Guei-Ying CHEN ; Guo-Xiong ZHOU ; Wei-Lun TSAI ; Chien-Neng KAO ; Chih-Lang LIN ; Chia-Chi WANG ; Ta-Ya LIN ; Chih‐Lin LIN ; Wei-Wen SU ; Tzong-Hsi LEE ; Te-Sheng CHANG ; Chun-Jen LIU ; Chia-Yen DAI ; Jia-Horng KAO ; Han-Chieh LIN ; Wan-Long CHUANG ; Cheng-Yuan PENG ; Chun-Wei- TSAI ; Chi-Yi CHEN ; Ming-Lung YU ;
Clinical and Molecular Hepatology 2024;30(1):64-79
Background/Aims:
Despite the high efficacy of direct-acting antivirals (DAAs), approximately 1–3% of hepatitis C virus (HCV) patients fail to achieve a sustained virological response. We conducted a nationwide study to investigate risk factors associated with DAA treatment failure. Machine-learning algorithms have been applied to discriminate subjects who may fail to respond to DAA therapy.
Methods:
We analyzed the Taiwan HCV Registry Program database to explore predictors of DAA failure in HCV patients. Fifty-five host and virological features were assessed using multivariate logistic regression, decision tree, random forest, eXtreme Gradient Boosting (XGBoost), and artificial neural network. The primary outcome was undetectable HCV RNA at 12 weeks after the end of treatment.
Results:
The training (n=23,955) and validation (n=10,346) datasets had similar baseline demographics, with an overall DAA failure rate of 1.6% (n=538). Multivariate logistic regression analysis revealed that liver cirrhosis, hepatocellular carcinoma, poor DAA adherence, and higher hemoglobin A1c were significantly associated with virological failure. XGBoost outperformed the other algorithms and logistic regression models, with an area under the receiver operating characteristic curve of 1.000 in the training dataset and 0.803 in the validation dataset. The top five predictors of treatment failure were HCV RNA, body mass index, α-fetoprotein, platelets, and FIB-4 index. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the XGBoost model (cutoff value=0.5) were 99.5%, 69.7%, 99.9%, 97.4%, and 99.5%, respectively, for the entire dataset.
Conclusions
Machine learning algorithms effectively provide risk stratification for DAA failure and additional information on the factors associated with DAA failure.

Result Analysis
Print
Save
E-mail