1.Safety and effectiveness of lecanemab in Chinese patients with early Alzheimer's disease: Evidence from a multidimensional real-world study.
Wenyan KANG ; Chao GAO ; Xiaoyan LI ; Xiaoxue WANG ; Huizhu ZHONG ; Qiao WEI ; Yonghua TANG ; Peijian HUANG ; Ruinan SHEN ; Lingyun CHEN ; Jing ZHANG ; Rong FANG ; Wei WEI ; Fengjuan ZHANG ; Gaiyan ZHOU ; Weihong YUAN ; Xi CHEN ; Zhao YANG ; Ying WU ; Wenli XU ; Shuo ZHU ; Liwen ZHANG ; Naying HE ; Weihuan FANG ; Miao ZHANG ; Yu ZHANG ; Huijun JU ; Yaya BAI ; Jun LIU
Chinese Medical Journal 2025;138(22):2907-2916
INTRODUCTION:
Lecanemab has shown promise in treating early Alzheimer's disease (AD), but its safety and efficacy in Chinese populations remain unexplored. This study aimed to evaluate the safety and 6-month clinical outcomes of lecanemab in Chinese patients with mild cognitive impairment (MCI) or mild AD.
METHODS:
In this single-arm, real-world study, participants with MCI due to AD or mild AD received biweekly intravenous lecanemab (10 mg/kg). The study was conducted at Hainan Branch, Ruijin Hospital Shanghai Jiao Tong University School of Medicine. Patient enrollment and baseline assessments commenced in November 2023. Safety assessments included monitoring for amyloid-related imaging abnormalities (ARIA) and other adverse events. Clinical and biomarker changes from baseline to 6 months were evaluated using cognitive scales (mini-mental state examination [MMSE], montreal cognitive assessment [MoCA], clinical dementia rating-sum of boxes [CDR-SB]), plasma biomarker analysis, and advanced neuroimaging.
RESULTS:
A total of 64 patients were enrolled in this ongoing real-world study. Safety analysis revealed predominantly mild adverse events, with infusion-related reactions (20.3%, 13/64) being the most common. Of these, 69.2% (9/13) occurred during the initial infusion and 84.6% (11/13) did not recur. ARIA-H (microhemorrhages/superficial siderosis) and ARIA-E (edema/effusion) were observed in 9.4% (6/64) and 3.1% (2/64) of participants, respectively, with only two symptomatic cases (one ARIA-E presenting with headache and one ARIA-H with visual disturbances). After 6 months of treatment, cognitive scores remained stable compared to baseline (MMSE: 22.33 ± 5.58 vs . 21.27 ± 4.30, P = 0.733; MoCA: 16.38 ± 6.67 vs . 15.90 ± 4.78, P = 0.785; CDR-SB: 2.30 ± 1.65 vs . 3.16 ± 1.72, P = 0.357), while significantly increasing plasma amyloid-β 42 (Aβ42) (+21.42%) and Aβ40 (+23.53%) levels compared to baseline.
CONCLUSIONS:
Lecanemab demonstrated a favorable safety profile in Chinese patients with early AD. Cognitive stability and biomarker changes over 6 months suggest potential efficacy, though high dropout rates and absence of a control group warrant cautious interpretation. These findings provide preliminary real-world evidence for lecanemab's use in China, supporting further investigation in larger controlled studies.
REGISTRATION
ClinicalTrials.gov , NCT07034222.
Humans
;
Alzheimer Disease/drug therapy*
;
Male
;
Female
;
Aged
;
Middle Aged
;
Cognitive Dysfunction/drug therapy*
;
Aged, 80 and over
;
Amyloid beta-Peptides/metabolism*
;
Biomarkers
;
East Asian People
2.Circadian and non-circadian regulation of the male reproductive system and reproductive damage: advances in the role and mechanisms of clock genes.
Meng-Chao HE ; Ying-Zhong DAI ; Yi-Meng WANG ; Qin-Ru LI ; Si-Wen LUO ; Xi LING ; Tong WANG ; Jia CAO ; Qing CHEN
Acta Physiologica Sinica 2025;77(4):712-720
Recently, male reproductive health has attracted extensive attention, with the adverse effects of circadian disruption on male fertility gradually gaining recognition. However, the mechanism by which circadian disruption leads to damage to male reproductive system remains unclear. In this review, we first summarized the dual regulatory roles of circadian clock genes on the male reproductive system: (1) circadian regulation of testosterone synthesis via the hypothalamic-pituitary-testicular (HPT) and hypothalamic-pituitary-adrenal (HPA) axes; (2) non-circadian regulation of spermatogenesis. Next, we further listed the possible mechanisms by which circadian disruption impairs male fertility, including interference with the oscillatory function of the reproductive system, i.e., synchronization of the HPT axis, crosstalk between the HPT axis and the HPA axis, as well as direct damage to germ cells by disturbing the non-oscillatory function of the reproductive system. Future research using spatiotemporal omics, epigenomic assays, and neural circuit mapping in studying the male reproductive system may provide new clues to systematically unravel the mechanisms by which circadian disruption affects male reproductive system through circadian clock genes.
Male
;
Humans
;
Animals
;
Circadian Clocks/physiology*
;
Hypothalamo-Hypophyseal System/physiology*
;
Circadian Rhythm/genetics*
;
Spermatogenesis/physiology*
;
Pituitary-Adrenal System/physiology*
;
Testis/physiology*
;
Testosterone/biosynthesis*
;
CLOCK Proteins
;
Infertility, Male/physiopathology*
3.Characteristics of changes in HBsAg,HBV DNA and ALT after cessation of nucleos(t)ide analogue therapy in patients with HBeAg-negative chronic hepatitis B
Qiao-Lan LYU ; Li LIU ; Song QING ; Ying-Jie JI ; Yan LIU ; Chao ZHANG ; Fu-Sheng WANG ; Yong-Qian CHEN
Medical Journal of Chinese People's Liberation Army 2025;50(9):1076-1082
Objective To investigate the characteristics of changes in hepatitis B surface antigen(HBsAg),hepatitis B virus(HBV)deoxyribonucleic acid(DNA),and alanine aminotransferase(ALT)levels following the cessation of nucleos(t)ide analogues(NAs)therapy in hepatitis B e antigen(HBeAg)-negative chronic hepatitis B(CHB)patients with baseline HBsAg levels<1000 IU/ml.Methods This retrospective cohort study analyzed 73 HBeAg-negative CHB patients treated at the Fifth Medical Centre of Chinese PLA General Hospital from January 2020 to June 2023.Patients were divided into 3 groups according to baseline HBsAg level and discontinuation strategy:HBsAg-negative discontinuation group(n=14),HBsAg-positive discontinuation group(n=25),and HBsAg-positive continuation group(n=34).All patients were followed for 48 weeks.Baseline clinical characteristics and changes in virological and hepatic biochemical indicators during follow-up were compared among the 3 groups.Univariate logistic regression analysis was performed to assess the correlation between clinical indicators and HBV DNA reappearance in HBsAg-positive discontinuation group,and between clinical indicators and HBsAg decline>0.5 log IU/ml in this group.Results There were no significant differences in the baseline levels of gender,age,albumin,and total bilirubin among the 3 groups(P>0.05).The baseline direct bilirubin level was significantly higher in HBsAg-positive discontinuation group than that in other groups(P<0.05),while the lymphocyte counts were significantly higher in HBsAg-negative discontinuation group(P<0.05).During the 48-week follow-up period,the HBV DNA reappearance rate in HBsAg-positive discontinuation group(72.0%)was significantly higher than that in other groups(P<0.001).There was no significant difference in the incidence of ALT elevation among the three groups(P=0.260).The proportion of patients with HBsAg decline>0.5 log IU/ml in HBsAg-positive discontinuation group(24.0%)was significantly higher than that in HBsAg-positive continuation group(5.9%,P<0.05).The proportion of patients with HBsAg increase>0.5 log IU/ml in HBsAg-positive discontinuation group(12.0%)was also significantly higher than that in HBsAg-positive continuation group(0%,P<0.05).Univariate logistic regression analysis revealed no significant association between the analyzed clinical indicators and HBsAg decline(P>0.05).Conclusions Discontinuation of NAs therapy in HBsAg-negative patients demonstrates high safety,with sustained HBsAg negativity post-cessation and low risks of viral relapse and liver function abnormalities.For HBsAg-positive patients,discontinuation may promote HBsAg decline in some individuals but is associated with risks of HBV DNA reappearance and HBsAg elevation.The decision to discontinue therapy should be comprehensively evaluated based on patients'baseline HBsAg levels and clinical characteristics.
4.Machine learning-based predictive model for severe pneumonia in children
Qing DU ; Mingzhao HUANG ; Ying LI ; Kai CHEN ; Lianting HU ; Chao XIONG ; Xiaoxia LU
Chinese Journal of Preventive Medicine 2025;59(10):1716-1724
Objective:To develop and validate a clinical warning model for severe pediatric community-acquired pneumonia (CAP) using electronic health records.Methods:A retrospective cohort study was conducted, analyzing clinical data of 15 750 children hospitalized for CAP at Wuhan Children′s Hospital between January 1, 2019, and December 31, 2023. Patient data were randomly split into training and testing sets at a 7∶3 ratio. Six supervised machine learning models were constructed in the training set, optimized using five-fold cross-validation, and evaluated in the testing set. Model performance was assessed using ROC-AUC, sensitivity, specificity, positive predictive value, negative predictive value, calibration curves, and clinical decision curve analysis at optimal thresholds. The best-performing model was selected, and SHapley Additive exPlanations (SHAP) were used to interpret feature importance. A program interface was developed based on the model results, enabling integration into clinical decision support systems for automated early warning.Results:A total of 15 750 participants, ranging in age from 28 days to 18 years, were included in the study. The median age was 2 years [interquartile range (IQR): 0-4 years], with 9 555 males (60.67%) and 6 195 females (39.33%). Among them, 2 211 (14.04%) developed severe pneumonia. In the prediction models, XGB outperformed other models with an ROC-AUC of 0.884 (95% CI: 0.870-0.898), sensitivity (0.803, 95% CI: 0.772-0.832), specificity (0.828, 95% CI: 0.816-0.839). Calibration analysis showed strong agreement between predicted and observed risks (Brier score: 0.081, 95% CI: 0.075-0.086). The analysis based on the SHAP method revealed that respiratory rate, heart rate, T-lymphocyte subsets, and red blood cell volume distribution width-SD are predictive factors for severe progression of community-acquired pneumonia (CAP) in children. Conclusion:An interpretable machine learning model was developed for the early detection and personalized treatment planning of severe CAP in children, providing valuable support to clinicians.
5.Changing prevalence and antibiotic resistance profiles of carbapenem-resistant Enterobacterales in hospitals across China:data from CHINET Antimicrobial Resistance Surveillance Program,2015-2021
Wenxiang JI ; Tong JIANG ; Jilu SHEN ; Yang YANG ; Fupin HU ; Demei ZHU ; Yuanhong XU ; Ying HUANG ; Fengbo ZHANG ; Ping JI ; Yi XIE ; Mei KANG ; Chuanqing WANG ; Pan FU ; Yingchun XU ; Xiaojiang ZHANG ; Ziyong SUN ; Zhongju CHEN ; Yuxing NI ; Jingyong SUN ; Yunzhuo CHU ; Sufei TIAN ; Zhidong HU ; Jin LI ; Yunsong YU ; Jie LIN ; Bin SHAN ; Yan DU ; Sufang GUO ; Lianhua WEI ; Fengmei ZOU ; Yunjian HU ; Xiaoman AI ; Chao ZHUO ; Danhong SU ; Dawen GUO ; Jinying ZHAO ; Hua YU ; Xiangning HUANG ; Wen'en LIU ; Yanming LI ; Yan JIN ; Chunhong SHAO ; Xuesong XU ; Chao YAN ; Shanmei WANG ; Yafei CHU ; Lixia ZHANG ; Juan MA ; Shuping ZHOU ; Yan ZHOU ; Lei ZHU ; Jinhua MENG ; Fang DONG ; Zhiyong LÜ ; Fangfang HU ; Han SHEN ; Wanqing ZHOU ; Wei JIA ; Gang LI ; Jinsong WU ; Yuemei LU ; Jihong LI ; Jinju DUAN ; Jianbang KANG ; Xiaobo MA ; Yanping ZHENG ; Ruyi GUO ; Yan ZHU ; Yunsheng CHEN ; Qing MENG ; Shifu WANG ; Xuefei HU ; Hong ZHANG ; Chun WANG ; Wenhui HUANG ; Ruizhong WANG ; Hua FANG ; Bixia YU ; Yong ZHAO ; Ping GONG ; Kaizhen WENG ; Yirong ZHANG ; Jiangshan LIU ; Longfeng LIAO ; Hongqin GU ; Lin JIANG ; Wen HE ; Shunhong XUE ; Jiao FENG ; Chunlei YUE
Chinese Journal of Infection and Chemotherapy 2025;25(4):445-454
Objective To summarize the changing prevalence of carbapenem resistance in Enterobacterales based on the data of CHINET Antimicrobial Resistance Surveillance Program from 2015 to 2021 for improving antimicrobial treatment in clinical practice.Methods Antimicrobial susceptibility testing was performed using a commercial automated susceptibility testing system according to the unified CHINET protocol.The results were interpreted according to the breakpoints of the Clinical & Laboratory Standards Institute(CLSI)M100 31st ed in 2021.Results Over the seven-year period(2015-2021),the overall prevalence of carbapenem-resistant Enterobacterales(CRE)was 9.43%(62 342/661 235).The prevalence of CRE strains in Klebsiella pneumoniae,Citrobacter freundii,and Enterobacter cloacae was 22.38%,9.73%,and 8.47%,respectively.The prevalence of CRE strains in Escherichia coli was 1.99%.A few CRE strains were also identified in Salmonella and Shigella.The CRE strains were mainly isolated from respiratory specimens(44.23±2.80)%,followed by blood(20.88±3.40)%and urine(18.40±3.45)%.Intensive care units(ICUs)were the major source of the CRE strains(27.43±5.20)%.CRE strains were resistant to all the β-lactam antibiotics tested and most non-β-lactam antimicrobial agents.The CRE strains were relatively susceptible to tigecycline and polymyxins with low resistance rates.Conclusions The prevalence of CRE strains was increasing from 2015 to 2021.CRE strains were highly resistant to most of the antibacterial drugs used in clinical practice.Clinicians should prescribe antimicrobial agents rationally.Hospitals should strengthen antibiotic stewardship in key clinical settings such as ICUs,and take effective infection control measures to curb CRE outbreak and epidemic in hospitals.
6.Changing distribution and antibiotic resistance profiles of the respiratory bacterial isolates in hospitals across China:data from CHINET Antimicrobial Resistance Surveillance Program,2015-2021
Ying FU ; Yunsong YU ; Jie LIN ; Yang YANG ; Fupin HU ; Demei ZHU ; Yingchun XU ; Xiaojiang ZHANG ; Fengbo ZHANG ; Ping JI ; Yi XIE ; Mei KANG ; Chuanqing WANG ; Pan FU ; Yuanhong XU ; Ying HUANG ; Ziyong SUN ; Zhongju CHEN ; Yuxing NI ; Jingyong SUN ; Yunzhuo CHU ; Sufei TIAN ; Zhidong HU ; Jin LI ; Bin SHAN ; Yan DU ; Sufang GUO ; Lianhua WEI ; Fengmei ZOU ; Hong ZHANG ; Chun WANG ; Yunjian HU ; Xiaoman AI ; Chao ZHUO ; Danhong SU ; Dawen GUO ; Jinying ZHAO ; Hua YU ; Xiangning HUANG ; Wen'en LIU ; Yanming LI ; Yan JIN ; Chunhong SHAO ; Xuesong XU ; Chao YAN ; Shanmei WANG ; Yafei CHU ; Lixia ZHANG ; Juan MA ; Shuping ZHOU ; Yan ZHOU ; Lei ZHU ; Jinhua MENG ; Fang DONG ; Zhiyong LÜ ; Fangfang HU ; Han SHEN ; Wanqing ZHOU ; Wei JIA ; Gang LI ; Jinsong WU ; Yuemei LU ; Jihong LI ; Jinju DUAN ; Jianbang KANG ; Xiaobo MA ; Yanping ZHENG ; Ruyi GUO ; Yan ZHU ; Yunsheng CHEN ; Qing MENG ; Shifu WANG ; Xuefei HU ; Jilu SHEN ; Ruizhong WANG ; Hua FANG ; Bixia YU ; Yong ZHAO ; Ping GONG ; Kaizhen WENG ; Yirong ZHANG ; Jiangshan LIU ; Longfeng LIAO ; Hongqin GU ; Lin JIANG ; Wen HE ; Shunhong XUE ; Jiao FENG ; Chunlei YUE ; Wenhui HUANG
Chinese Journal of Infection and Chemotherapy 2025;25(4):431-444
Objective To characterize the changing species distribution and antibiotic resistance profiles of respiratory isolates in hospitals participating in the CHINET Antimicrobial Resistance Surveillance Program from 2015 to 2021.Methods Commercial automated antimicrobial susceptibility testing systems and disk diffusion method were used to test the susceptibility of respiratory bacterial isolates to antimicrobial agents following the standardized technical protocol established by the CHINET program.Results A total of 589 746 respiratory isolates were collected from 2015 to 2021.Overall,82.6%of the isolates were Gram-negative bacteria and 17.4%were Gram-positive bacteria.The bacterial isolates from outpatients and inpatients accounted for(6.0±0.9)%and(94.0±0.1)%,respectively.The top microorganisms were Klebsiella spp.,Acinetobacter spp.,Pseudomonas aeruginosa,Staphylococcus aureus,Haemophilus spp.,Stenotrophomonas maltophilia,Escherichia coli,and Streptococcus pneumoniae.Each microorganism was isolated from significantly more males than from females(P<0.05).The overall prevalence of methicillin-resistant S.aureus(MRSA)was 39.9%.The prevalence of penicillin-resistant S.pneumoniae was 1.4%.The prevalence of extended-spectrum β-lactamase(ESBL)-producing E.coli and K.pneumoniae was 67.8%and 41.3%,respectively.The overall prevalence of carbapenem-resistant E.coli,K.pneumoniae,Enterobacter cloacae,Pseudomonas aeruginosa,and Acinetobacter baumannii was 3.7%,20.8%,9.4%,29.8%,and 73.3%,respectively.The prevalence of β-lactamase was 96.1%in Moraxella catarrhalis and 60.0%in Haemophilus influenzae.The H.influenzae isolates from children(<18 years)showed significantly higher resistance rates to β-lactam antibiotics than the isolates from adults(P<0.05).Conclusions Gram-negative bacteria are still predominant in respiratory isolates associated with serious antibiotic resistance.Antimicrobial resistance surveillance should be strengthened in clinical practice to support accurate etiological diagnosis and appropriate antimicrobial therapy based on antimicrobial susceptibility testing results.
7.Novel Structural Features of Isoflavone Synthase from Medicago truncatula Shed Light on Its Unique Enzymatic Mechanism
Chao SHI ; Zhao-Yang YE ; Fei XU ; Xiang-Ning DU ; Zhang-Xin CHEN ; Ming-Yue GU ; Jie DENG ; Wei WANG ; Liang-Yu LIU ; Mei-Ying WANG ; Xiao-Dong SU ; He-Li LIU ; Ming-Ying SHANG ; Li-Xin HUANG ; Zhen-Zhan CHANG
Chinese Journal of Biochemistry and Molecular Biology 2025;41(8):1204-1213,中插1-中插6
Isoflavones which mainly distributed in leguminous plants have plenty of health benefits.Isoflavone synthase(IFS)is a membrane-associated cytochrome P450 enzyme(CYP450)which carries out the unique aryl-ring migration and hydroxylation.So far,few crystal structures of plant P450s have been obtained.We determined the crystal structure of IFS from Medicago truncatula at 1.9 ? by MAD method using a selenomethionine substituted crystal and conducted molecular docking and mutagenesis study.The structure of IFS complexed with imidazole exhibits the helix Ⅰa-loop-helix Ⅰβ motif which cor-responds to helix Ⅰ of other P450s.Compared with structures of common P450s,IFS/imidazole structure contains an extra domain,i.e.,the γ-domain.The structure reveals a homodimer in which the γ-domain of one molecule interacts with the β-domain of another.The plane of heme group makes an angle of ap-proximately 40° with the helix Ⅰa-loop-helix Ⅰβ motif.Molecular docking combined with mutagenesis study suggested that Trp-128 and Asp-300 might play important roles in substrate binding and recogni-tion.Phe-301,Ser-303 and Gly-305 from the helix Ⅰa-loop-helix Ⅰβ motif may play important roles in the aryl-ring migration.These novel structural features reveal insights into the unique reaction mechanism of IFS and provide a basis for engineering IFS in leguminous crops for health purpose.
8.Balanophora polysaccharide improves kidney injury in mice with diabetic nephropathy via regulating TLR4/MyD88/NF-κB signaling pathway
Tian-ying SONG ; Xiao-ling ZHOU ; Jian-hong GAO ; Yi-duo HE ; Chao-xi TIAN ; Xian-bing CHEN
Chinese Pharmacological Bulletin 2025;41(9):1659-1664
Aim To study the renal protective effect of balanophora polysaccharide(BPS)on diabetic nephrop-athy(DN)mice and explore the related mechanisms.Methods A DN mouse model was induced using a high-fat diet combined with intraperitoneal injection of streptozotocin(STZ),which was indicated by fasting blood glucose higher than 11.1 mmol·L-1,accompa-nied by diabetic symptoms such as polydipsia,polydia-gia,polyuria and weight loss,then BPS intervention was performed.Body weight and fasting blood glucose of each group mice were detected;automatic biochemical analyzer was used to detect blood creatinine(SCr),blood urea nitrogen(BUN),24 h urinary protein(24 h UP),triglycerides(TG),total cholesterol(TC),alanine aminotransferase(ALT)content;ELISA was applied to determine serum inflammatory factor interleukin-6(IL-6)and tumor necrosis factor-α(TNF-α)level;HE and Masson staining were employed to observe renal his-topathological morphology;Western blot was used to de-tect Toll-like receptor 4(TLR4),myeloid differentiation factor 88(MyD88),nuclear factor κB(NF-κB)for pro-tein expression.Results Compared with the model group,after BPS,body weight and fasting blood glucose decreased(P<0.01 or P<0.05);SCr,BUN,24 h UP,TC,TG and ALT significantly decreased(P<0.01 or P<0.05);the levels of the proinflammatory factors TNF-α and IL-6 were significantly reduced(P<0.01 or P<0.05);renal tissue injury and fibrosis decreased;TLR4,MyD88,NF-κB protein expression significantly decreased(P<0.01 or P<0.05).Conclusion BPS has a protective effect on the kidneys of DN mice,re-ducing the blood glucose level,improving liver and kid-ney function,alleviating renal tissue damage and renal fibrosis,and reducing inflammation response.Its mecha-nism may be related to the regulation of TLR4/MyD88/NF-κB signaling pathway.
9.Mechanism of action of D-limonene on steatosis in primary hepatocytes based on AMPK/ACC/CPT1A signaling pathway
Qian-jun REN ; Su LI ; Yu-qing CHEN ; Yin-ying LIAO ; Chun-ni LIANG ; Rui-chao FANG ; Xu-dong LIU ; Xiao-fang ZHAO
Chinese Pharmacological Bulletin 2025;41(9):1665-1672
Aim To explore the effects of D-limonene on the steatosis of primary mouse hepatocytes and its potential mechanism of action.Methods Oleic acid-induced steatosis in primary mouse hepatocytes was used as a model to observe the effects of D-limonene on cell viability,cellular lipid content,and intracellular expression of proteins such as AMP-activated protein kinase(AMPK),acetyl-coenzyme A carboxylase 1(ACC1),and carnitine palmitoyl transferase 1A(CPT1A).Results It was found that a low dose of D-limonene could effectively enhance the viability of primary mouse hepatocytes.When oleic acid at a con-centration of 300 μmol·L-1 successfully induced steatosis in primary mouse hepatocytes,D-limonene re-duced the lipid content of the cells,and D-limonene up-regulated the cellular AMPK expression level,down-regulated the cellular ACC1 and fatty acid synthetase(FAS)expression levels,which in turn promoted the overexpression of CPT1A.Conclusions D-limonene has the effect of reducing lipid deposition in primary mouse hepatocytes,and the mechanisms may be related to the activation of AMPK,the inhibitions of ACC1 and FAS,and the up-regulation of CPT1A protein expres-sion level.
10.Machine learning-based predictive model for severe pneumonia in children
Qing DU ; Mingzhao HUANG ; Ying LI ; Kai CHEN ; Lianting HU ; Chao XIONG ; Xiaoxia LU
Chinese Journal of Preventive Medicine 2025;59(10):1716-1724
Objective:To develop and validate a clinical warning model for severe pediatric community-acquired pneumonia (CAP) using electronic health records.Methods:A retrospective cohort study was conducted, analyzing clinical data of 15 750 children hospitalized for CAP at Wuhan Children′s Hospital between January 1, 2019, and December 31, 2023. Patient data were randomly split into training and testing sets at a 7∶3 ratio. Six supervised machine learning models were constructed in the training set, optimized using five-fold cross-validation, and evaluated in the testing set. Model performance was assessed using ROC-AUC, sensitivity, specificity, positive predictive value, negative predictive value, calibration curves, and clinical decision curve analysis at optimal thresholds. The best-performing model was selected, and SHapley Additive exPlanations (SHAP) were used to interpret feature importance. A program interface was developed based on the model results, enabling integration into clinical decision support systems for automated early warning.Results:A total of 15 750 participants, ranging in age from 28 days to 18 years, were included in the study. The median age was 2 years [interquartile range (IQR): 0-4 years], with 9 555 males (60.67%) and 6 195 females (39.33%). Among them, 2 211 (14.04%) developed severe pneumonia. In the prediction models, XGB outperformed other models with an ROC-AUC of 0.884 (95% CI: 0.870-0.898), sensitivity (0.803, 95% CI: 0.772-0.832), specificity (0.828, 95% CI: 0.816-0.839). Calibration analysis showed strong agreement between predicted and observed risks (Brier score: 0.081, 95% CI: 0.075-0.086). The analysis based on the SHAP method revealed that respiratory rate, heart rate, T-lymphocyte subsets, and red blood cell volume distribution width-SD are predictive factors for severe progression of community-acquired pneumonia (CAP) in children. Conclusion:An interpretable machine learning model was developed for the early detection and personalized treatment planning of severe CAP in children, providing valuable support to clinicians.

Result Analysis
Print
Save
E-mail