1.Effectiveness of "Internet plus" continuous intervention among patients with depression after hospital discharge
CHEN Chunmian ; JIN Tingting ; GAO Si ; CHEN Haoran ; LU Xiaoyuan ; ZHENG Lidan
Journal of Preventive Medicine 2025;37(10):1049-1053
Objective:
To evaluate the effectiveness of "Internet plus" continuous intervention on psychological status and nursing satisfaction of patients with depression after hospital discharge, so as to provide the reference for reducing the recurrence risk of patients with depression and improving the quality of life.
Methods:
From January to December 2024, patients with mild to moderate depression who were hospitalized in a tertiary grade-a mental health specialized hospital in Wenzhou City and met the discharge criteria were selected as the research objects. The patients were divided into the control group and the intervention group according to a ratio of 1∶1 by the random number table method. Hamilton Depression Scale, Hamilton Anxiety Scale, and Nursing Satisfaction Questionnaire were used to evaluate depressive symptoms, anxiety symptoms, and nursing satisfaction before and after intervention. Covariance analysis was used to compare the differences between the two groups before and after the intervention.
Results:
A total of 62 patients with mild to moderate depression were enrolled, with 31 patients in the intervention group and 31 patients in the control group. Before the intervention, there were no statistically significant differences in gender, age, course of disease, educational level, marital status, depression symptoms score, anxiety symptoms score, and nursing satisfaction score between the two groups (all P>0.05). After the intervention, the scores of depression and anxiety symptoms in the intervention group decreased by 8.87 and 5.01 points, respectively, compared with those before the intervention, and the scores of depression and anxiety symptoms in the control group decreased by 2.52 and 1.16 points, respectively (all P<0.05). After the intervention, the scores of depression and anxiety symptoms in the intervention group decreased more than those in the control group (both P<0.05). The nursing satisfaction score of the intervention group increased by 6.57 points on average compared with that before the intervention, and that of the control group increased by 4.23 points on average (both P<0.05). There was no statistically significant difference in the increase of nursing satisfaction scores between the two groups before and after intervention (P>0.05).
Conclusion
The "Internet plus" continuous intervention has a good effect on improving the depressive symptoms and anxiety symptoms of patients with depression after haspital discharge, which can consolidate the treatment effect and improve nursing satisfaction.
2.Programmed death-ligand 1 tumor proportion score in predicting the safety and efficacy of PD-1/PD-L1 antibody-based therapy in patients with advanced non-small cell lung cancer: A retrospective, multicenter, observational study.
Yuequan SHI ; Xiaoyan LIU ; Anwen LIU ; Jian FANG ; Qingwei MENG ; Cuimin DING ; Bin AI ; Yangchun GU ; Cuiying ZHANG ; Chengzhi ZHOU ; Yan WANG ; Yongjie SHUI ; Siyuan YU ; Dongming ZHANG ; Jia LIU ; Haoran ZHANG ; Qing ZHOU ; Xiaoxing GAO ; Minjiang CHEN ; Jing ZHAO ; Wei ZHONG ; Yan XU ; Mengzhao WANG
Chinese Medical Journal 2025;138(14):1730-1740
BACKGROUND:
This study aimed to investigate programmed death-ligand 1 tumor proportion score in predicting the safety and efficacy of PD-1/PD-L1 antibody-based therapy in treating patients with advanced non-small cell lung cancer (NSCLC) in a real-world setting.
METHODS:
This retrospective, multicenter, observational study enrolled adult patients who received PD-1/PD-L1 antibody-based therapy in China and met the following criteria: (1) had pathologically confirmed, unresectable stage III-IV NSCLC; (2) had a baseline PD-L1 tumor proportion score (TPS); and (3) had confirmed efficacy evaluation results after PD-1/PD-L1 treatment. Logistic regression, Kaplan-Meier analysis, and Cox regression were used to assess the progression-free survival (PFS), overall survival (OS), and immune-related adverse events (irAEs) as appropriate.
RESULTS:
A total of 409 patients, 65.0% ( n = 266) with a positive PD-L1 TPS (≥1%) and 32.8% ( n = 134) with PD-L1 TPS ≥50%, were included in this study. Cox regression confirmed that patients with a PD-L1 TPS ≥1% had significantly improved PFS (hazard ratio [HR] 0.747, 95% confidence interval [CI] 0.573-0.975, P = 0.032). A total of 160 (39.1%) patients experienced 206 irAEs, and 27 (6.6%) patients experienced 31 grade 3-5 irAEs. The organs most frequently associated with irAEs were the skin (52/409, 12.7%), thyroid (40/409, 9.8%), and lung (34/409, 8.3%). Multivariate logistic regression revealed that a PD-L1 TPS ≥1% (odds ratio [OR] 1.713, 95% CI 1.054-2.784, P = 0.030) was an independent risk factor for irAEs. Other risk factors for irAEs included pretreatment absolute lymphocyte count >2.5 × 10 9 /L (OR 3.772, 95% CI 1.377-10.329, P = 0.010) and pretreatment absolute eosinophil count >0.2 × 10 9 /L (OR 2.006, 95% CI 1.219-3.302, P = 0.006). Moreover, patients who developed irAEs demonstrated improved PFS (13.7 months vs. 8.4 months, P <0.001) and OS (28.0 months vs. 18.0 months, P = 0.007) compared with patients without irAEs.
CONCLUSIONS
A positive PD-L1 TPS (≥1%) was associated with improved PFS and an increased risk of irAEs in a real-world setting. The onset of irAEs was associated with improved PFS and OS in patients with advanced NSCLC receiving PD-1/PD-L1-based therapy.
Humans
;
Carcinoma, Non-Small-Cell Lung/metabolism*
;
Male
;
Female
;
Retrospective Studies
;
Middle Aged
;
Lung Neoplasms/metabolism*
;
Aged
;
B7-H1 Antigen/metabolism*
;
Programmed Cell Death 1 Receptor/metabolism*
;
Adult
;
Aged, 80 and over
;
Immune Checkpoint Inhibitors/therapeutic use*
3.Tranexamic acid-fatty alcohol polyoxyethylene ether conjugation/PVA foam for venous sclerotherapy via vascular damage and inhibiting plasmin system.
Jizhuang MA ; Keda ZHANG ; Wenhan LI ; Yu DING ; Yongfeng CHEN ; Xiaoyu HUANG ; Tong YU ; Di SONG ; Haoran NIU ; Huichao XIE ; Tianzhi YANG ; Xiaoyun ZHAO ; Xinggang YANG ; Pingtian DING
Acta Pharmaceutica Sinica B 2025;15(6):3291-3304
Venous system diseases mainly include varicose veins and venous malformations of lower limbs and the genital system. Most of them are chronic diseases that cause serious clinical symptoms to patients and affect their health and quality of life. Sclerotherapy has become the first-line therapy for venous system diseases. However, there are problems such as incomplete fibrosis and vascular recanalization after sclerotherapy, and improper operation will cause serious adverse consequences. Therefore, exploring a safe and effective sclerotherapy strategy is essential for developing clinically successful sclerotherapy. To solve the above problems, we proposed a new sclerotherapy strategy with a dual mechanism of "vascular damage and plasmin (PLA) system inhibition." We intended to construct a novel cationic surfactant (AEOx-TA) by reacting tranexamic acid (TA), a parent structure, with fatty alcohol polyoxyethylene ether (AEOx) by ester bonds. AEOx-TA could damage vascular endothelium and initiate a coagulation cascade effect to induce thrombus. Furthermore, AEOx-TA could be degraded by esterase and release the parent drug, TA, which could inhibit the PLA system to inhibit the degradation of thrombus and extracellular matrix and promote the process of vascular fibrosis. In addition, such surfactant-based sclerosants have foam-forming properties, and they can be blended with polyvinyl alcohol (PVA) to prepare a highly stable foam formulation (AEOx-TA/P), which can achieve a precise drug delivery and prolonged drug retention time, thereby improving drug efficacy and reducing the risk of ectopic embolism. Overall, the novel cationic surfactant AEOx-TA provides a new avenue to resolve the bottleneck: surfactant sclerosants' efficiency is relatively low in the current sclerotherapy.
4.Exploration of the renal protective effect and mechanism of Shenbining granule on IgA nephropathy rats based on the CXCL12/CXCR4/STAT3 signaling pathway
Xu WANG ; Chundong SONG ; Chenchen CHEN ; Haoran JIANG
China Pharmacy 2025;36(23):2912-2917
OBJECTIVE To investigate the renal protective effect and mechanism of Shenbining granule on IgA nephropathy (IgAN) rats by regulating the CXC chemokine motif ligand 12 (CXCL12)/CXC chemokine receptor 4 (CXCR4)/signal transducer and activator of transcription 3 (STAT3) signaling pathway. METHODS A total of 60 rats were randomly assigned into blank group (n=12) and modeling group (n=48). IgAN model of modeling group was induced by using bovine serum albumin, carbon tetrachloride and lipopolysaccharide, followed by model validation. Ultimately, a total of 55 rats (9 in the blank group, 46 in the modeling group) were included in the subsequent study. The rats in the modeling group were randomly divided into model group (n=10), prednisone acetate group [positive control group, 6.25 mg/(kg·d), n=12], Shenbining granule low- and high-dose groups [4.17, 8.33 g/(kg·d), n=12]. They were given relevant medicine/distilled water intragastrically, once a day, for 4 consecutive weeks. After the last medication, biochemical indicators in the urine and serum of rats were measured, and pathological morphological changes in the renal tissues of rats were observed. IgA deposition in the renal tissues, as well as the mRNA expression levels of CXCL12, CXCR4 and STAT3, and the protein expression levels of CXCL12, CXCR4, STAT3 and phosphorylated STAT3 (p-STAT3) were detected. Additionally, the level of interleukin-6 (IL-6) in the renal tissue was measured. RESULTS Compared with the model group, the low-dose and high-dose Shenbining granule groups showed significantly decreased urinary red blood cell count, 24 h total urinary protein, blood urea nitrogen, serum creatinine, and alanine amino-transferase, along with increased Alb levels (P<0.05). Pathological damage in the renal tissues was alleviated, with reduced IgA deposition in the mesangial region (P<0.05); protein and mRNA expressions of CXCL12, CXCR4 and STAT3, as well as phosphorylation level of STAT3 protein and the IL-6 level, were significantly decreased in renal tissue (P<0.05). CONCLUSIONS Shenbining granule may exert its renal protective effects in IgAN rats by inhibiting the activation of the CXCL12/ CXCR4/STAT3 signaling pathway, downregulating the expression of inflammatory factors such as IL-6, alleviating renal inflammation, and thereby improving renal pathological damage.
5.Shenqi Buzhong Formula ameliorates mitochondrial dysfunction in a rat model of chronic obstructive pulmonary disease by activating the AMPK/SIRT1/PGC-1α pathway.
Lu ZHANG ; Huanzhang DING ; Haoran XU ; Ke CHEN ; Bowen XU ; Qinjun YANG ; Di WU ; Jiabing TONG ; Zegeng LI
Journal of Southern Medical University 2025;45(5):969-976
OBJECTIVES:
To explore the mechanism of Shenqi Buzhong (SQBZ) Formula for alleviating mitochondrial dysfunction in a rat model of chronic obstructive pulmonary disease (COPD) in light of the AMPK/SIRT1/PGC-1α pathway.
METHODS:
Fifty male SD rat models of COPD, established by intratracheal lipopolysaccharide (LPS) instillation, exposure to cigarette smoke, and gavage of Senna leaf infusion, were randomized into 5 groups (n=10) for treatment with saline (model group), SQBZ Formula at low, moderate and high doses (3.08, 6.16 and 12.32 g/kg, respectively), or aminophylline (0.024 g/kg) by gavage for 4 weeks, with another 10 untreated rats as the control group. Pulmonary function of the rats were tested, and pathologies and ultrastructural changes of the lung tissues were examined using HE staining and transmission electron microscopy. The levels of SOD, ATP, MDA, and mitochondrial membrane potential in the lungs were detected using WST-1, colorimetric assay, TBA, and JC-1 methods. Flow cytometry was used to analyze ROS level in the lung tissues, and the protein expression levels of P-AMPKα, AMPKα, SIRTI, and PGC-1α were detected using Western blotting.
RESULTS:
The rat models of COPD showed significantly decreased lung function, severe histopathological injuries of the lungs, decreased pulmonary levels of SOD activity, ATP and mitochondrial membrane potential, increased levels of MDA and ROS, and decreased pulmonary expressions of P-AMPKα, SIRTI, and PGC-1α proteins. All these changes were significantly alleviated by treatment with SQBZ Formula and aminophylline, and the efficacy was comparable between high-dose SQBZ Formula group and aminophylline group.
CONCLUSIONS
SQBZ Formula ameliorates mitochondrial dysfunction in COPD rats possibly by activating the AMPK/SIRT1/PGC-1α pathway.
Animals
;
Pulmonary Disease, Chronic Obstructive/drug therapy*
;
Drugs, Chinese Herbal/therapeutic use*
;
Sirtuin 1/metabolism*
;
Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1-alpha
;
Rats, Sprague-Dawley
;
Male
;
Rats
;
AMP-Activated Protein Kinases/metabolism*
;
Mitochondria/metabolism*
;
Disease Models, Animal
;
Signal Transduction/drug effects*
6.Nucleated red blood cells ≥ 1% on the first day of intensive care unit admission is a risk factor for 28-day mortality in patients with sepsis.
Haoran CHEN ; Yao YAN ; Xinyi TANG ; Haoyue XUE ; Xiaomin LI ; Yongpeng XIE
Chinese Critical Care Medicine 2025;37(8):701-706
OBJECTIVE:
To investigate the correlation between nucleated red blood cell (NRBC) level on the first day of intensive care unit (ICU) admission and 28-day mortality in adult septic patients, and to evaluate the value of NRBC as an independent predictor of death.
METHODS:
Single-cell transcriptomic analysis was performed using the GSE167363 dataset from the Gene Expression Omnibus (including 2 healthy controls, 3 surviving septic patients, and 2 non-surviving septic patients). A retrospective clinical analysis was conducted using the America Medical Information Mart for Intensive Care-IV (MIMIC-IV) database, including adult patients (≥ 18 years) with first-time admission who met the Sepsis-3.0 criteria, excluding those without NRBC testing on the first ICU day. The demographic information, vital signs, laboratory test indicators, disease severity score and survival data on the first day of admission were collected. The restricted cubic spline (RCS) curve was used to determine the optimal cut-off value of NRBC for predicting 28-day mortality in patients. Patients were divided into low-risk and high-risk groups based on this cut-off value for intergroup comparison, with Kaplan-Meier survival curve analysis conducted. Independent risk factors for 28-day mortality were analyzed using Logistic regression and Cox regression analysis, followed by the construction of regression models.
RESULTS:
NRBC were detected in the peripheral blood of septic patients by single-cell transcriptomic. A total of 1 291 sepsis patients were included in the clinical analysis, with 576 deaths within 28 days, corresponding to a 28-day mortality of 44.6%. RCS curve analysis showed a nonlinear relationship between the first-day NRBC level and the 28-day mortality. When NRBC ≥ 1%, the 28-day mortality of patients increased significantly. Compared to the low-risk group (NRBC < 1%), the high-risk group (NRBC ≥ 1%) had significantly higher respiratory rate, heart rate, sequential organ failure assessment (SOFA), and simplified acute physiology score II (SAPSII), and significantly lower hematocrit and platelet count. The high-risk group also had a significantly higher 28-day mortality [49.8% (410/824) vs. 35.5% (166/467), P < 0.05], and shorter median survival time (days: 29.8 vs. 208.6, P < 0.05). Kaplan-Meier survival curve showed that compared with the low-risk group, the survival time of high-risk group was significantly shortened (Log-rank test: χ 2 = 25.1, P < 0.001). After adjusting for potential confounding factors including body mass, temperature, heart rate, respiratory rate, mean arterial pressure, serum creatinine, pulse oximetry saturation, hemoglobin, hematocrit, Na+, K+, platelet count, and SOFA score, multivariate regression analysis confirmed that NRBC ≥ 1% was an independent risk factor for 28-day mortality [Logistic regression: odds ratio (OR) = 1.464, 95% confidence interval (95%CI) was 1.126-1.902, P = 0.004; Cox regression: hazard ratio (HR) = 1.268, 95%CI was 1.050-1.531, P = 0.013].
CONCLUSIONS
NRBC ≥ 1% on the first day of ICU admission is an independent risk factor for 28-day mortality in septic patients and can serve as a practical indicator for early prognostic assessment.
Humans
;
Sepsis/blood*
;
Intensive Care Units
;
Risk Factors
;
Retrospective Studies
;
Prognosis
;
Male
;
Female
;
Hospital Mortality
;
Middle Aged
;
Aged
7.Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit
Chao GONG ; Na YU ; Haoran CHEN
Medical Journal of Peking Union Medical College Hospital 2025;16(3):710-721
To identify and validate the clinical phenotypes of patients with sepsis in the intensive care unit(ICU). We applied unsupervised machine learning algorithms (K-means clusteringand hierarchical clustering) to identify the phenotypes of sepsis patients in the Medical Information Mart for Intensive Care Ⅳ (MIMIC-Ⅳ) 2.2 database, based on 89 clinical features including demographic characteristics, laboratory indicators and treatment measures on the first day in ICU. Then, supervised machine learning algorithms (lightweight gradient boosting machine) were used for the prediction of the patient's phenotypes, and were further combined with SHAP (Shapely Additive eXplanations) for the identification of important features. Finally, traditional statistical methods were used to validate the differences in clinical characteristics and clinical outcomes among the phenotypes. We identified three phenotypes in 22 517 sepsis patients. The phenotype 1 patients had the highest risk of death (28-day mortality of 46.4%), dominated by abnormal renal function and elevated disease severity scores, while the phenotype 3 patients had the lowest risk of death (28-day mortality of 11.2%), and the best neurological function score. Using interpretable machine learning, we identified six features (all the worst value on the first day) that showed good performance in phenotypic identification(AUC≥0.89) and phenotypic prognostic prediction (AUC≥0.74): anion gap, blood urea nitrogen, creatinine, Glasgow Coma Scale score, prothrombin time, and Sequential Organ Failure Assessment score. The mortality risk of phenotype 3 patients was the lowest at 28 days, 60 days, 90 days, and 1 year after ICU discharge ( Using machine learning methods, we successfully identified three clinical phenotypes of sepsis patients with different clinical characteristics and prognosis and screened out six key clinical features, which are expected to play an important role in the phenotype classification and prognostic assessment of sepsis and are conducive to individualized treatment.
8.SMARCA4-deficient uterine sarcoma: a clinicopathological analysis of five cases
Cheng XU ; Gang CHEN ; Haoran SUN ; Hai LI
Chinese Journal of Pathology 2025;54(9):958-963
Objective:To investigate the clinicopathological features of SMARCA4-deficient uterine sarcoma.Methods:Five cases of SMARCA4-deficient uterine sarcoma at the Department of Pathology, the First Affiliated Hospital of Nanjing Medical University from 2018 to 2024 were collected. The morphological and immunohistochemical features were observed and analyzed. A follow-up study was also carried out.Results:Five female patients, aged 24, 54, 56, 61, and 41 years, respectively, presented with vaginal bleeding or abdominal pain. All patients had imaging findings of intracavitary lesion in the uterus, with tumor sizes ranging from 3.0 cm to 8.8 cm. The patients were followed up for 2 to 14 months. Case 1 died 9 months after surgery, whereas the remaining four patients were still alive. Histologically, the tumor cells exhibited a diffuse growth pattern, with an infiltration depth involving more than half of the myometrium. Portions of the interstitium appeared sclerosed. Benign endometrial glandular structures were observed in a leaf-like or fissured pattern, resembling those of uterine adenosarcoma. The tumor cells were large epithelioid with abundant or faintly eosinophilic cytoplasm, and the nuclei were moderately to markedly atypia with prominent nucleoli and brisk mitosis. Rhabdoid cells were seen. Some areas showed small round blue cells, with occasional spindle cells and myxoid stroma. Additionally, widespread or focal lymphovascular space invasion was observed within the myometrium. All five cases exhibited absence of SMARCA4 (BRG1) expression and retained SMARCB1 (INI1). Claudin4 expression was negative. There was no deficient expression of mismatch repair proteins MLH1, PMS2, MSH2 and MSH6. p53 showed wild-type expression. Ki-67 index ranged from 30% to 60%. CKpan, CK7, ER, PR, and PAX8 were negative.Conclusions:SMARCA4-deficient uterine sarcoma is rare, highly aggressive, and has a poor prognosis. The tumor exhibits a broad morphological spectrum, with rhabdoid cells and adenosarcoma-like structures serving as important diagnostic clues. The absence of BRG1 expression lends support to a definitive diagnosis.
9.Mechanism of the regulation of prostate cancer stem cells by CAF:Based on the Wnt/β-catenin and SDF-1/CXCR4 pathways
Haoran CHEN ; Xudong ZHU ; Jiazheng WANG ; Yafei CHEN ; Yilin WANG ; Hao LIU
National Journal of Andrology 2025;31(10):867-873
Objective To investigate the mechanism by which cancer-associated fibroblast(CAF)in the tumor microenvironment regulate key pathways in prostate cancer stem cells(PCSCs).Methods An in vitro co-culture system of CAF and PCSC was established to observe the effects of CAF on PCSC proliferation and sphere formation.Prostate cancer stem cells were treated with CAF conditioned medium pre-treated with Wnt inhibitor DKK-1 and SDF-1 neutralizing anti-body(MAB310).Western blot was used to detect the expression of β-catenin,CXCR4,CD133 and CD44 in PCSCs.And PCR was used to detect the expression of β-catenin,CXCR4,TCF,and LEF mRNA.TOPflash/FOPflash dual-luciferase reporter assays were conducted to detect the effects of SDF-1 on Wnt/β-catenin signaling activity in PCSCs.Results ELISA results showed that the secretion of Wnt3a and SDF-1 in CAF supernatant was significantly higher than that in WPMY-1 cells(P<0.05).The A value of PCSCs co-cultured with CAF at a 1∶6 ratio was significantly higher than that of the PCSC-only group(P<0.0 1),and CAF promoted sphere formation in PCSCs(P<0.05).Western blot results showed that CAF-CM significantly increased the relative expression of β-catenin,CXCR4,CD133 and CD44 in PCSCs(P<0.01).Compared to CAF-CM,CAFanti-Wnt-CM significantly reduced the expression of β-catenin,CD133 and CD44(P<0.01).CAFanti-SDF-1-CM also significantly inhibited the expression of CXCR4,β-catenin,CD133 and CD44(P<0.01).PCR results showed that CAFanti-SDF-1-CM inhibited the expression of β-catenin,CXCR4 and downstream Wnt signaling effectors TCF and LEF(P<0.01).Dual-luciferase reporter assay results showed that luciferase activity in the CAFanti-SDF-1-CM group was significantly lower than that in the CAF-CM group(P<0.05).Conclusion CAF reg-ulates the stemness of PCSCs through the Wnt/β-catenin and SDF-1/CXCR4 pathways.CXCR4 may enhance the mainte-nance of stemness by activating β-catenin.
10.Construction of a machine learning model based on the Ki67 positive index to predict the recurrence risk of hepatocellular carcinoma
Haoran LI ; Yan YU ; Fangying FAN ; Wenzhen DING ; Hui FENG ; Minghua YING ; Jiawei LI ; Qingqing SUN ; Lele BIAN ; Haokai XU ; Zhanyue CHEN ; Jie YU ; Ping LIANG
Chinese Journal of Hepatology 2025;33(9):898-909
Objective:To screen the optimal machine learning model for predicting the recurrence condition of hepatocellular carcinoma (HCC) at different time points post-surgery, based on the cutoff value of the Ki67 positive proliferation index condition calculated from recurrence-free survival and combined with various clinical features.Methods:retrospective study included initially treated patients with solitary HCC who underwent radical surgery at the Fifth Medical Center of the PLA General Hospital from January 2013 to March 2023. Data included general clinical data, preoperative laboratory parameters, and surgical pathology information about the subjects. The postoperative recurrence status was assessed by querying the medical record system or by telephone follow-up. The Ki67 positive index cutoff value was determined by the X-tile software based on the patient's recurrence-free survival status and time analysis. Survival rates were calculated using the Kaplan-Meier method, and survival curves were plotted. The study population was randomly divided into training and testing groups in a 7:3 ratio using a computer-generated random number method. The minimum redundancy maximum relevance (mRMR) method was used for feature variable selection. Predictive models for postoperative HCC recurrence conditions in patients with HCC were constructed using random forest, support vector machine, logistic regression, and gradient boosting decision tree machine learning algorithms. Inter-group comparisons for continuous data were performed using the t-test or Mann-Whitney U test. Inter-group comparisons of enumeration data were performed using the Pearson χ2 test, continuity-corrected χ2 test, or Fisher's exact test. Results:The cutoff values for the Ki67 positivity index were 0.3 and 0.5 in 510 cases, with a follow-up time ranging from 1.2 to 11.4 years (median: 6.2 years). The recurrence-free survival time was between 1 and 135 months (median: 32 months), with recurrence-free survival rates post-surgery at 1, 2, 3, and 5 years were 87.5%, 77.1%, 61.2%, and 54.5%, respectively. The top five variables predicted HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years, in accordance with information obtained by the mRMR screen out. The Ki67 positivity index screened a successfully constructed machine learning model to predict HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years. The machine learning model based on the gradient boosting decision tree algorithm had the best prediction performance among them (areas under the receiver operating characteristic curves for predicting HCC recurrence within six months in the training and validation sets were 0.996 and 0.946, and accuracies were 0.972 and 0.935, respectively).Conclusion:A machine learning model was successfully constructed using the Ki67 positivity index combined with four readily available clinical features to predict HCC recurrence. The machine learning model based on the gradient boosting decision tree algorithm demonstrated the best performance in terms of predicting HCC recurrence within six months after surgery.


Result Analysis
Print
Save
E-mail