1.Integrated Transcriptomic Landscape and Deep Learning Based Survival Prediction in Uterine Sarcomas
Yaolin SONG ; Guangqi LI ; Zhenqi ZHANG ; Yinbo LIU ; Huiqing JIA ; Chao ZHANG ; Jigang WANG ; Yanjiao HU ; Fengyun HAO ; Xianglan LIU ; Yunxia XIE ; Ding MA ; Ganghua LI ; Zaixian TAI ; Xiaoming XING
Cancer Research and Treatment 2025;57(1):250-266
Purpose:
The genomic characteristics of uterine sarcomas have not been fully elucidated. This study aimed to explore the genomic landscape of the uterine sarcomas (USs).
Materials and Methods:
Comprehensive genomic analysis through RNA-sequencing was conducted. Gene fusion, differentially expressed genes (DEGs), signaling pathway enrichment, immune cell infiltration, and prognosis were analyzed. A deep learning model was constructed to predict the survival of US patients.
Results:
A total of 71 US samples were examined, including 47 endometrial stromal sarcomas (ESS), 18 uterine leiomyosarcomas (uLMS), three adenosarcomas, two carcinosarcomas, and one uterine tumor resembling an ovarian sex-cord tumor. ESS (including high-grade ESS [HGESS] and low-grade ESS [LGESS]) and uLMS showed distinct gene fusion signatures; a novel gene fusion site, MRPS18A–PDC-AS1 could be a potential diagnostic marker for the pathology differential diagnosis of uLMS and ESS; 797 and 477 uterine sarcoma DEGs (uDEGs) were identified in the ESS vs. uLMS and HGESS vs. LGESS groups, respectively. The uDEGs were enriched in multiple pathways. Fifteen genes including LAMB4 were confirmed with prognostic value in USs; immune infiltration analysis revealed the prognositic value of myeloid dendritic cells, plasmacytoid dendritic cells, natural killer cells, macrophage M1, monocytes and hematopoietic stem cells in USs; the deep learning model named Max-Mean Non-Local multi-instance learning (MMN-MIL) showed satisfactory performance in predicting the survival of US patients, with the area under the receiver operating curve curve reached 0.909 and accuracy achieved 0.804.
Conclusion
USs harbored distinct gene fusion characteristics and gene expression features between HGESS, LGESS, and uLMS. The MMN-MIL model could effectively predict the survival of US patients.
2.Integrated Transcriptomic Landscape and Deep Learning Based Survival Prediction in Uterine Sarcomas
Yaolin SONG ; Guangqi LI ; Zhenqi ZHANG ; Yinbo LIU ; Huiqing JIA ; Chao ZHANG ; Jigang WANG ; Yanjiao HU ; Fengyun HAO ; Xianglan LIU ; Yunxia XIE ; Ding MA ; Ganghua LI ; Zaixian TAI ; Xiaoming XING
Cancer Research and Treatment 2025;57(1):250-266
Purpose:
The genomic characteristics of uterine sarcomas have not been fully elucidated. This study aimed to explore the genomic landscape of the uterine sarcomas (USs).
Materials and Methods:
Comprehensive genomic analysis through RNA-sequencing was conducted. Gene fusion, differentially expressed genes (DEGs), signaling pathway enrichment, immune cell infiltration, and prognosis were analyzed. A deep learning model was constructed to predict the survival of US patients.
Results:
A total of 71 US samples were examined, including 47 endometrial stromal sarcomas (ESS), 18 uterine leiomyosarcomas (uLMS), three adenosarcomas, two carcinosarcomas, and one uterine tumor resembling an ovarian sex-cord tumor. ESS (including high-grade ESS [HGESS] and low-grade ESS [LGESS]) and uLMS showed distinct gene fusion signatures; a novel gene fusion site, MRPS18A–PDC-AS1 could be a potential diagnostic marker for the pathology differential diagnosis of uLMS and ESS; 797 and 477 uterine sarcoma DEGs (uDEGs) were identified in the ESS vs. uLMS and HGESS vs. LGESS groups, respectively. The uDEGs were enriched in multiple pathways. Fifteen genes including LAMB4 were confirmed with prognostic value in USs; immune infiltration analysis revealed the prognositic value of myeloid dendritic cells, plasmacytoid dendritic cells, natural killer cells, macrophage M1, monocytes and hematopoietic stem cells in USs; the deep learning model named Max-Mean Non-Local multi-instance learning (MMN-MIL) showed satisfactory performance in predicting the survival of US patients, with the area under the receiver operating curve curve reached 0.909 and accuracy achieved 0.804.
Conclusion
USs harbored distinct gene fusion characteristics and gene expression features between HGESS, LGESS, and uLMS. The MMN-MIL model could effectively predict the survival of US patients.
3.Integrated Transcriptomic Landscape and Deep Learning Based Survival Prediction in Uterine Sarcomas
Yaolin SONG ; Guangqi LI ; Zhenqi ZHANG ; Yinbo LIU ; Huiqing JIA ; Chao ZHANG ; Jigang WANG ; Yanjiao HU ; Fengyun HAO ; Xianglan LIU ; Yunxia XIE ; Ding MA ; Ganghua LI ; Zaixian TAI ; Xiaoming XING
Cancer Research and Treatment 2025;57(1):250-266
Purpose:
The genomic characteristics of uterine sarcomas have not been fully elucidated. This study aimed to explore the genomic landscape of the uterine sarcomas (USs).
Materials and Methods:
Comprehensive genomic analysis through RNA-sequencing was conducted. Gene fusion, differentially expressed genes (DEGs), signaling pathway enrichment, immune cell infiltration, and prognosis were analyzed. A deep learning model was constructed to predict the survival of US patients.
Results:
A total of 71 US samples were examined, including 47 endometrial stromal sarcomas (ESS), 18 uterine leiomyosarcomas (uLMS), three adenosarcomas, two carcinosarcomas, and one uterine tumor resembling an ovarian sex-cord tumor. ESS (including high-grade ESS [HGESS] and low-grade ESS [LGESS]) and uLMS showed distinct gene fusion signatures; a novel gene fusion site, MRPS18A–PDC-AS1 could be a potential diagnostic marker for the pathology differential diagnosis of uLMS and ESS; 797 and 477 uterine sarcoma DEGs (uDEGs) were identified in the ESS vs. uLMS and HGESS vs. LGESS groups, respectively. The uDEGs were enriched in multiple pathways. Fifteen genes including LAMB4 were confirmed with prognostic value in USs; immune infiltration analysis revealed the prognositic value of myeloid dendritic cells, plasmacytoid dendritic cells, natural killer cells, macrophage M1, monocytes and hematopoietic stem cells in USs; the deep learning model named Max-Mean Non-Local multi-instance learning (MMN-MIL) showed satisfactory performance in predicting the survival of US patients, with the area under the receiver operating curve curve reached 0.909 and accuracy achieved 0.804.
Conclusion
USs harbored distinct gene fusion characteristics and gene expression features between HGESS, LGESS, and uLMS. The MMN-MIL model could effectively predict the survival of US patients.
4.Mechanism and application in tissue engineering of the active ingredient of Drynariae Rhizoma promoting bone defect repair
Zhijun DENG ; Wenlong YANG ; Zhijun YANG ; Bin ZHAO ; Dian LI ; Fengyun YANG
China Pharmacy 2024;35(8):1023-1028
Bone defect has always been a major clinical challenge because of its great difficulty and long period of treatment. Drynariae Rhizoma is a commonly used medicine in osteology and traumatology of traditional Chinese medicine, and its active ingredients(mainly flavonoids) facilitate osteoblast differentiation of bone marrow mesenchymal stem cells, osteoclast proliferation, vascular-osteogenic coupling, and inhibit osteoclast activity to promote bone mineralization, and repair and reconstruction of bone defect. As a good substitute for bone regeneration drugs, the active constituents of Drynariae Rhizoma can be loaded on scaffold materials of tissue engineering, which greatly improves the bioavailability of the drug. Meanwhile, the sustained-release microspheres also solve some problems such as sudden drug release from the scaffolds, and the composite scaffolds with active ingredient of Drynariae Rhizoma prepared by them have good ossification activity and osteoinduction, with precise bone repair effects, which meet the diverse performance requirements of bone grafts and have a promising clinical application prospect.
5.Efficacy and safety of dimethyl fumarate in the treatment of multiple sclerosis:a meta-analysis
Fengyun LI ; Rui LAN ; Duo ZHAO ; Fugui LIU ; Liangchen CHEN
China Pharmacy 2024;35(14):1776-1780
OBJECTIVE To evaluate the efficacy and safety of dimethyl fumarate (DMF) in the treatment of multiple sclerosis (MS). METHODS Retrieved from CBM, Web of Science, PubMed, the Cochrane Library, Embase, CNKI, Wanfang Data, and VIP, randomized controlled trials (RCTs) about DMF (trial group) versus other drugs or placebo (control group) were collected. After data screening and extraction, quality evaluation, meta-analysis was conducted by using RevMan 5.3 software. RESULTS A total of 6 literature were included, involving 638 patients. Results of meta-analysis showed that the proportion of patients with lesion changes after treatment in the trial group was lower than control group [MD=-0.65, 95%CI(-1.27, -0.02), P=0.04]; there was no statistical significance in recurrence rate [RR=1.06, 95%CI(0.52,2.17), P=0.88], the proportion of patients with new lesions after treatment [RR=1.05, 95%CI(0.62,1.80), P=0.85], expanded disability status scale after treatment [MD=0.02,95%CI (-0.18, 0.23), P=0.82], the incidence of adverse events [RR=1.33, 95%CI(0.97, 1.84), P=0.08] or severe adverse events [RR=0.95,95%CI(0.48,1.90),P=0.89] between 2 groups. Results of sensitivity analysis showed the study obtained unstable recurrence rate and the incidence of adverse events, while other results were robust. CONCLUSIONS DMF can control the lesion progression in MS patients to some extent and doesn’t increase the incidence of adverse events and serious adverse events, but there is no significant advantage in reducing the recurrence rate and controlling the disability progression.
6.Protective effect of compound drug Weng-Li-Tong on cisplatin-induced hepatocyte injury
Liangwen YAN ; Xinyan LI ; Jiayi XU ; Fengyun BAI ; Fenyue YUAN ; Ying SUN ; Pengfei LIU
Journal of Xi'an Jiaotong University(Medical Sciences) 2024;45(5):815-821
Objective To investigate the protective effect of the compound drug Weng-Li-Tong(WLT)against cisplatin(CDDP)-induced hepatocyte injury.Methods A cellular injury model was established by treating murine hepatocyte line BNL CL.2 with CDDP(80 μmol/L).Experimental groups were divided as follows:CDDP group(modeling only),WLT group(intervention with 1 g/L WLT),WLT+CDDP group(co-administration of CDDP and 1 g/L WLT),and a control group(normal culture).The protective effect of the compound drug WLT on CDDP-mediated hepatocyte injury was evaluated using CCK-8 assay,PI staining,crystal violet staining,Western blotting,reactive oxygen species(ROS)detection,and apoptosis analysis.Results Compared with the control group,the number of dead cells increased significantly(P<0.001)in the CDDP group,but no cytotoxicity was observed in the WLT group.The hepatocyte morphology in the WLT+CDDP group showed improvement with no obvious shrinkage compared to the CDDP group,as evidenced by the reduced proportion of PI-positive cells.Crystal violet staining results also indicated a higher cell count in the WLT+CDDP group than in the CDDP group,suggesting the protective effect of WLT against CDDP-mediated liver injury.Under CDDP intervention,the expression of the apoptosis-related protein Cleaved Caspase-3 increased.However,in the WLT+CDDP group,the expression of Cleaved Caspase-3 decreased,while the expression of the anti-apoptotic protein Bcl-2 increased.Additionally,compared to the CDDP group,the WLT+CDDP group showed a reduction in ROS production[DCFH-DA staining positive rate(%):56.20±1.65 vs.44.57±0.31]and a decrease in the proportion of apoptotic cells[proportion of early and late apoptotic cells(%):43.60±0.44 vs.19.57±0.78;33.30±1.02 vs.14.83±0.57].Conclusion The compound drug WLT exhibits a protective effect against CDDP-mediated hepatocyte injury,suggesting potential therapeutic value in acute liver injury models.
7.Analysis on TCM Syndrome and Treatment Rule of Chronic Atrophic Gastritis Based on Data Mining
Xiaoxue GUO ; Ping WANG ; Jinghua LI ; Beihua ZHANG ; Fengyun WANG ; Xudong TANG
Chinese Journal of Information on Traditional Chinese Medicine 2024;31(4):38-43
Objective To analyze the TCM syndrome and treatment rule for the treatment of chronic atrophic gastritis based on data mining;To provide reference for its clinical treatment.Methods With the help of the big data platform of evidence-based literature of TCM covering several dominant diseases in the field of digestion,literature on the experience of famous doctors,theoretical discussion and case reports in the treatment of chronic atrophic gastritis published from 1 January 2000 to 31 July 2022 was retrieved.Medical records were screened according to the inclusion and exclusion criteria.The data of TCM disease names,clinical symptoms,tongue manifestation,pulse manifestation,TCM syndrome,TCM treatment method and other terms in medical records were standardized and statistically analyzed.Results Totally 169 articles were included,including 228 medical cases,involving 228 patients,with a cumulative number of 361 visits,with an average age of(53.17±11.11)years old,with the largest number of 50-60 years old.Chronic atrophic gastritis mostly belonged to the categories of"stomach pain"and"epigastric puffiness"in TCM,and its main symptoms were stomach pain,epigastric puffiness and belching.The common tongue manifestation include dark red tongue,white moss,thin or greasy moss.The common pulse manifestation included wiry slippery pulse,wiry thready pulse and wiry pulse.The high-frequency syndromes included liver-stomach qi stagnation syndrome,spleen-stomach dampness-heat syndrome,liver-qi stagnation and spleen-qi deficiency syndrome.The disease locations were mainly in the stomach,spleen and liver,and the syndrome elements were mainly qi deficiency,qi stagnation,blood stasis,dampness and heat,etc.The complex syndrome elements were more than the single syndrome elements,and the two syndrome elements of qi deficiency + qi stagnation was the most common.The most commonly used treatment methods were supplementing,regulating qi and regulating blood.Conclusion The treatment of chronic atrophic gastritis with TCM mostly focuses on the stomach,spleen and liver.The core syndrome elements are qi deficiency,qi stagnation,and blood stasis.The methods of nourishing qi and strengthening the spleen,soothing the liver and stomach,regulating qi and blood are often used.
8.Protective effect of polydatin on beta-amyloid-induced mitochondrial oxidative stress and dysfunction in cortical neurons in vitro
Tingzhen ZANG ; Yang GUO ; Fengyun WANG ; Yanhui LI ; Yanxia SHEN ; Zhenyan ZHANG
Chinese Journal of Geriatric Heart Brain and Vessel Diseases 2024;26(5):567-572
Objective To investigate the protective effect of polydatin(PLD)on primary cultured rat cortical neurons induced by β-amyloid protein(Aβ25-35)and its possible mechanism.Methods Cortical neurons were isolated from SD rats and primarily cultured,and then divided into control group,PLD group,Aβ25-35 group and Aβ25-35+PLD group.Cell viability was detected by MTT as-say,intracellular and mitochondrial reactive oxygen species(ROS)levels were detected by dichlo-rofluorescein diacetic acid probe or MitoSOX Red staining,opening of mitochondrial membrane permeability transition pore(MPTP)was detected by MPTP kit,and contents of cytochrome C(Cyt C)and mitochondrial transcription factor A(TFAM)were detected by Western blotting.In addition,the activity of intracellular electron transport chain complexes(including complexes Ⅰ,Ⅱ,Ⅲ and Ⅳ)and the level of adenylate triphosphate(ATP)were determined.High performance liquid chromatography(HPLC)was conducted to determine the content of 8-hydroxydeox-yguanosine(8-OHdG)in the mitochondria.Results Compared with the control group,cell vitali-ty,mitochondrial fluorescence intensity,activities of mitochondrial respiratory chain enzymes(complex Ⅰ,Ⅱ,Ⅲ and Ⅳ),intracellular ATP and mitochondrial TFAM expression were signifi-cantly decreased in the cortical neuron from the Aβ25~35 group(P<0.05,P<0.01).Compared with the Aβ25-35 group,mitochondrial fluorescence intensity,mitochondrial respiratory chain enzyme activity(complex Ⅰ,Ⅱ,Ⅲ,Ⅳ),intracellular ATP and mitochondrial TFAM expression were obviously increased in the PLD group,and intracellular and mitochondrial ROS levels were nota-bly decreased after cortical neuron exposure for 3,6,12 and 24 h(P<0.05,P<0.01).The ratio of cytoplasmic/mitochondrial Cyt C and mitochondrial level of 8-OHdG were statistically lower in the Aβ25-35+PLD group than the Aβ25-35 group[3.02±0.28 vs 5.73±0.45,P<0.05;(8.07±1.45)× 106 dG vs(16.07±2.29)X 106 dG,P<0.05].Conclusion PLD can effectively protect cortical neu-rons against Aβ25-35-induced injury,which may be partially by its inhibiting mitochondrial oxida-tive stress and improving mitochondrial function.
9.Portable spirometer-based pulmonary function test willingness in China: A nationwide cross-sectional study from the "Happy Breathing Program"
Weiran QI ; Ke HUANG ; Qiushi CHEN ; Lirui JIAO ; Fengyun YU ; Yiwen YU ; Hongtao NIU ; Wei LI ; Fang FANG ; Jieping LEI ; Xu CHU ; Zilin LI ; Pascal GELDSETZER ; Till B?RNIGHAUSEN ; Simiao CHEN ; Ting YANG ; Chen WANG
Chinese Medical Journal 2024;137(14):1695-1704
Background::Understanding willingness to undergo pulmonary function tests (PFTs) and the factors associated with poor uptake of PFTs is crucial for improving early detection and treatment of chronic obstructive pulmonary disease (COPD). This study aimed to understand willingness to undergo PFTs among high-risk populations and identify any barriers that may contribute to low uptake of PFTs.Methods::We collected data from participants in the "Happy Breathing Program" in China. Participants who did not follow physicians’ recommendations to undergo PFTs were invited to complete a survey regarding their willingness to undergo PFTs and their reasons for not undergoing PFTs. We estimated the proportion of participants who were willing to undergo PFTs and examined the various reasons for participants to not undergo PFTs. We conducted univariable and multivariable logistic regressions to analyze the impact of individual-level factors on willingness to undergo PFTs.Results::A total of 8475 participants who had completed the survey on willingness to undergo PFTs were included in this study. Out of these participants, 7660 (90.4%) were willing to undergo PFTs. Among those who were willing to undergo PFTs but actually did not, the main reasons for not doing so were geographical inaccessibility ( n = 3304, 43.1%) and a lack of trust in primary healthcare institutions ( n = 2809, 36.7%). Among the 815 participants who were unwilling to undergo PFTs, over half ( n = 447, 54.8%) believed that they did not have health problems and would only consider PFTs when they felt unwell. In the multivariable regression, individuals who were ≤54 years old, residing in rural townships, with a secondary educational level, with medical reimbursement, still working, with occupational exposure to dust, and aware of the abbreviation "COPD" were more willing to undergo PFTs. Conclusions::Willingness to undergo PFTs was high among high-risk populations. Policymakers may consider implementing strategies such as providing financial incentives, promoting education, and establishing community-based programs to enhance the utilization of PFTs.
10.Association between preconception oral microbiome and fetal overgrowth
Qiuli XIAO ; Xushan CAI ; Lifeng ZHANG ; Fengyun YANG ; Xingying LI ; An CHEN ; Huajun ZHENG ; Hong JIANG
Chinese Journal of Perinatal Medicine 2024;27(6):457-467
Objective:To analyze the association between the pre-pregnancy oral microbiota of women and fetal overgrowth, and the possible mechanisms involved.Methods:A nested case-control study design based on a pre-pregnancy cohort was used to select 51 mothers who delivered macrosomia and/or large-for-gestational-age (LGA) infants from the population recruited at the Maternal and Child Health Care Hospital of Jiading District in Shanghai from October 2016 to December 2021 as the case group. A control group was formed by selecting 204 mothers who delivered infants with normal birth weight and appropriate for gestational age during the same period, in a 1:4 ratio. The LGA subgroup consisted of 48 mothers who delivered LGA infants from the total population, and a corresponding control group of 192 was randomly selected from the remaining mothers who delivered non-LGA infants in a 1∶4 ratio for the LGA subgroup analysis. The 16S rRNA gene sequencing technique was utilized to detect pre-pregnancy saliva samples to compare the characteristics of the oral microbiota, differential microorganisms, and differential functional pathways between groups. Nonparametric Wilcoxon rank-sum tests, two independent samples t-tests, or Chi-square (or Fisher's exact) tests were used for statistical analysis. Factor analysis was conducted on the pre-pregnancy diet data of women, and the primary dietary pattern of each study subject was identified based on the highest score of the dietary pattern factors. For microbiota count data, α and β diversity indices were calculated using R and QIIME2 software, and the corresponding microbiota functional count data were acquired through PICRUSt2. Results:(1) General data: There was no significant difference in the time interval from pre-pregnancy sampling to pregnancy and from sampling to delivery between the two groups. In the case group, there were three cases of macrosomia and 48 cases (94.1%) of LGA. The corresponding control group for the LGA subgroup consisted of 192 cases. There were no significant differences in dietary patterns between the case group and the control group. (2) α diversity analysis: The species richness index of the case group was lower than that of the control group [(367.27±84.57) vs. (408.71±93.08), multivariate analysis, P=0.009], while no significant differences were found between the two groups in the Shannon and Simpson indices; the species richness index of the LGA subgroup was also lower than that of the corresponding control group [(371.04±83.92) vs. (408.04±94.21), multivariate analysis, P=0.033], with no significant differences in the Shannon and Simpson indices. (3) β diversity analysis: There was a statistically significant difference in the unweighted UniFrac distance of the oral microbiota between the case group and the control group ( R2=0.006, F=1.479, P=0.048). No significant differences were found in the β diversity indices of the oral microbiota between the LGA subgroup and the corresponding control group. (4) Differential microbiota analysis: There were 14 differential microbiotas from phylum to genus between the case group and the control group. At the genus level, members of the G1 genus of the Streptococcaceae were enriched in the case group, while the Lautropia, Dialister, Leptotrichia, and Rothia were enriched in the control group. In the LGA subgroup and its corresponding control group, there were 14 differential microbiota from phylum to genus; at the genus level, Leptotrichia, Rothia, G6 genus of the Saccharibacteria, and Selenomonas were enriched in the control group (all LDA value>2, and all P<0.05). (5) Differential functional analysis: In the case group, metabolic pathways such as nicotinate degradation [log 2 fold change ( FC)=3.510, q=0.005], de novo synthesis of pyrimidine nucleotides (log 2FC=0.078, q=0.005), and L-tyrosine degradation pathway (log 2FC=0.710, q=0.034) were enriched in the oral microbiota of women. In the LGA subgroup, compared to the corresponding control group, metabolic pathways related to nicotinate degradation were enriched in the oral microbiota (log 2FC=3.660, q=0.012). Conclusions:There are differences in the structure of the pre-pregnancy oral microbiota of mothers with overgrown fetuses compared to those with normally grown fetuses, and mothers of normally grown fetuses show higher diversity in their pre-pregnancy oral microbiota. The enrichment of certain pathogenic bacteria and the reduction of symbiotic bacteria in the pre-pregnancy oral microbiota are associated with fetal overgrowth, and this association may be mediated by functional pathways such as nicotinate degradation.

Result Analysis
Print
Save
E-mail