1.Pregnancy probability prediction models based on 5 machine learning algorithms and comparison of their performance
Chao REN ; Huan YANG ; Niya ZHOU ; Qing CHEN ; Wenzheng ZHOU ; Tong WANG ; Xi LING ; Lei SUN ; Peng ZOU ; Zhuoyue LIANG ; Lin AO ; Jinyi LIU ; Jia CAO
Journal of Army Medical University 2025;47(12):1376-1387
Objective To construct 5 machine-learning models and compare their performance in predicting the associations between pre-pregnancy socio-psycho-behavioral exposures of both spouses and preconception outcomes.Methods Based on Chongqing Preconception Reproductive Health and Birth Outcome Cohort of volunteers recruited from Chongqing Health Center for Women and Children during January 2019 and March 2022,5 447 couples were recruited and surveyed through interviewer-interview for the demographic and social-psychological-behavioral data of both spouses(221 variables).According to the inclusion and exclusion criteria,4 097 couples were finally included,and randomly assigned into a training set(n=2 867 spouses)and a validation set(n=1 230 spouses)at a ratio of 7∶3.Feature analysis and collinear screening were applied to select the potential exposure factors.In consideration of difficulty to carry out semen parameters analysis in primary healthcare institutions,feature Set 1 including sperm parameters and feature Set 2 excluding semen parameters were constructed by including or excluding sperm quality simultaneously in the training set and the validation set.Five algorithms,that is,Logistic Regression,Naive Bayes,Random Forest,Gradient Boosting Machine,and Support Vector Machine,were used to construct preconception outcome prediction models,and the parameters of each model were optimized using random search combined with grid search.The predictive performance of each model was compared using precision,recall,F1 score,area under the receiver operating characteristic curve(AUC),and calibration curve.The optimal model was then selected by comparing the changes in the predictive ability of the questionnaire data for fertility outcomes with or without semen parameters.Results There were 24 variables screened out in feature Set 1,and 16 variables in feature Set 2.In feature Set 1,the gradient boosting machine performed better,with a relatively higher AUC value(0.651)and better F1 score(0.61).The logistic regression model performed stably(AUC value=0.647)and was suitable as the reference model.The random forest(AUC value=0.641),Naive Bayes(AUC value=0.641),and support vector machine(AUC value=0.634)performed second-best.By utilizing the gradient boosting machine,comparable results were found between the predictions from feature sets with or without semen parameters,as in feature Set 1,the AUC value of its validation set was 0.651(95%CI:0.629~0.681),the prediction accuracy was 0.63,the recall rate was 0.65,and the average precision value F1 was 0.61;and in feature Set 2,the AUC value of its validation set was 0.649(95%CI:0.624~0.663),and both the calibration curves were close to the ideal curve.The prediction results indicated that in feature Set 1,the features highly negatively correlated with preconception outcomes were female age,male age,and no pregnancy within 1 year without contraception,while the features highly positively correlated with preconception outcomes were female pregnancy history,total sperm vitality,and use of contraceptive measures before enrollment.Conclusion Among the 5 machine-learning algorithms performed in this cohort data,the gradient boosting machine shows slightly better performance.There are 24 factors being associated with preconception outcomes in both spouses,and the performance of the simplified model excluding semen parameters is not significantly declined.It is feasible to use machine-learning methods to predict human preconception outcomes through social-psychological-behavioral questionnaires.
2.Therapeutic effects of tofacitinib on steroid-resistant immune checkpoint inhibitor-associated myocarditis
Yuchen XU ; Jian ZHANG ; Yan WANG ; Jinyi LIN ; Yuhong ZHOU ; Leilei CHENG ; Junbo GE
China Oncology 2024;34(4):400-408
Background and purpose:Outcomes for cancer patients with steroid-resistant immune checkpoint inhibitor-associated myocarditis(srICIAM)are poor.Intensified immunosuppressive therapies,including tofacitinib,a novel Janus kinase(JAK)inhibitor,may have some therapeutic benefits.However,due to the lack of sufficient clinical data,the effectiveness of such treatments and their impact on cardiovascular outcomes remain unclear.This study aimed to investigate the therapeutic effect of tofacitinib on srICIAM.Methods:This retrospective case-control study included 36 malignant tumor patients who received immune checkpoint inhibitor treatment at Zhongshan Hospital affiliated to Fudan University from July 2019 to May 2022 and developed srICIAM.Patients receiving corticosteroids in combination with tofacitinib were assigned to the tofacitinib group(n=19),while those not treated with tofacitinib were allocated to the control group(n=17).The study compared clinical characteristics,laboratory findings,and imaging results between the two groups.Additionally,follow-up was conducted to monitor the incidence of cardiovascular endpoints in these patients.The research plan was approved by the Ethics Committee of Zhongshan Hospital Affiliated to Fudan University(Approval Number:B2021-275R).This study was conducted in accordance with the ethical guidelines of the Helsinki Declaration.Results:Compared to the control group,and with no significant difference in the cumulative dose and duration of corticosteroids(P<0.05),the tofacitinib group showed a shorter myocarditis recovery time(median recovery time:86.5 days vs 126.5 days,P=0.021).The myocarditis-related mortality rate was significantly lower in the tofacitinib group than in the control group(5%vs 35%,P=0.025).Conclusion:Tofacitinib may reduce mortality and promote cardiac recovery in srICIAM patients without impeding the anti-tumor effect.It may become one of the potential treatment strategies in the future.
3.Trends of Incidence and Age at Onset of Female Breast Cancer in Jiangsu Cancer Registration Areas from 2009 to 2019
Yingying HAN ; Bo CAI ; Ling LIN ; Jinyi ZHOU ; Ran TAO ; Weigang MIAO ; Renqiang HAN
China Cancer 2024;33(12):970-976
[Purpose]To investigate the trends of incidence and age at onset of female breast can-cer in Jiangsu cancer registration areas from 2009 to 2019.[Methods]The continuous monitoring data of female breast cancer from 2009 to 2019 were collected from 16 cancer registries in Jiangsu Province.All datasets were checked and evaluated based on data quality control criteria and were included in the analysis.Crude rate(CR),age-standardized incidence rate by Chinese standard population(ASIRC),the average annual percentage change(AAPC),crude and adjusted mean age at onset were calculated.Incidence rates stratified by regions and age groups were calculated.Age-period-cohort model was used to analyze the changes of cancer incidence and age-adjusted mean ages.[Results]The incidence CR of female breast cancer was significantly increased from 24.39/105 in 2009 to 46.72/105 in 2019 with an AAPC in CR of 6.97%(95%CI:5.26%~8.70%)and AAPC in ASIRC of 4.67%(95%CI:3.00%~6.37%).The crude mean age and adjusted mean age at onset increased from 54.10 and 52.89 years old in 2009 to 56.23 and 53.36 years old in 2019,respectively.Crude mean age at onset increased significantly over time in all registry areas(β=0.21,P<0.001),urban(β=0.1 8,P<0.001)and rural(β=0.25,P<0.001)areas,while adjusted mean age at onset remained stable in all registry areas(β=0.05,P=0.024)and rural areas(β=0.09,P=0.008).From 2009 to 2019,the incidence rate of female breast cancer of all age groups showed upward trends,with an average annual growth rate of 3.26%to 7.79%(all P<0.05),and the incidence rate in rural areas increased faster than that in urban areas.The age composition of breast cancer onset and standardized age composition of onset in women over 60 years old showed an upward trend(all P<0.05).[Conclusion]The incidence rate of female breast cancer in Jiangsu Province in-creased from 2009 to 2019,and the mean age of onset showed a backward trend.
4.Hua Xian Fang alleviates radiation-induced pulmonary fibrosis by upregulating the level of IFN-γ in blood and tissues
Junyang CHEN ; Pingjin ZOU ; Zengyi FANG ; Cuicui GONG ; Jie YIN ; Meihua CHEN ; Bing LIN ; Jinyi LANG
Chinese Journal of Radiation Oncology 2024;33(6):554-561
Objective:To investigate the therapeutic efficacy and underlying mechanisms of the traditional Chinese medicine formula "Hua Xian Fang" (HXF) in the treatment of radiation-induced pulmonary fibrosis (RIPF).Methods:In vivo experiment, 36 male specific pathogen free (SPF)-grade C57BL/6 mice aged 6-8 weeks were randomly divided into the control, irradiation (17 Gy thoracic irradiation), and irradiation+HXF groups (17 Gy thoracic irradiation+HXF). After 16 weeks, lung coefficient, HE staining and Masson staining were used to evaluate the degree of pulmonary inflammation and fibrosis. Immunohistochemistry was performed to measure the expression levels of α-smooth muscle actin (α-SMA) in lung tissues. Quantitative real-time PCR (qPCR) was performed to detect the mRNA expression levels of interferon-γ (IFN-γ). Enzyme linked immunosorbent assay (ELISA) was conducted to determine the levels of IFN-γ in serum and bronchoalveolar lavage fluid (BALF). During in vitro experiment, NIH/3T3 fibroblasts were stimulated with IFN-γ after 6 Gy irradiation, followed by 48 hours of culture. qPCR, immunofluorescence staining, and Western blot were used to assess the expression of α-SMA and collagen Ⅰ at the transcription and protein levels. One way ANOVA was used for multiple group comparisons, and Tukey test was used for inter group multiple comparisons. Results:Compared to the control group, mice in the irradiation group showed significant increases in lung coefficient, Szapiel score, Ashcroft score, and α-SMA expression in lung tissues (all P<0.001). Compared to the irradiation group, the irradiation+HXF group exhibited significant decreases in the above indicators (all P<0.001). qPCR demonstrated that the mRNA expression levels of IFN-γ were significantly higher in the irradiation+HXF group than that in the irradiation group ( P=0.001). ELISA results showed that the levels of IFN-γ in serum and BALF were significantly elevated in the irradiation+HXF group compared to those in the irradiation group ( P=0.032, 0.037). In vitro experiment revealed that after irradiation, the expression levels of α-SMA and collagen Ⅰ mRNA and protein in NIH/3T3 cells were significantly increased, while decreased after IFN-γ stimulation. Conclusion:HXF effectively alleviates RIPF, probably by the upregulation of IFN-γ in blood and tissues and inhibition of fibroblast activation.
5.Effects of chronic psychological stress on testicular cell senescence in rats and role of glucocorticoid receptor in the process
Binwei YANG ; Lei SUN ; Xi LING ; Qing CHEN ; Tong WANG ; Huan YANG ; Jinyi LIU ; Jia CAO ; Lin AO ; Peng ZOU
Journal of Army Medical University 2024;46(24):2698-2706
Objective To investigate the effect of chronic psychological stress on testicular cell senescence in rats and the role of glucocorticoid receptor(GR)in the process.Methods Thirty-two male SD rats(4~6 weeks old,weighing 120~140 g)were randomly assigned into dimethyl sulfoxide(DMSO)group,mifepristone(RU486)group,chronic unpredictable mild stress(CUMS)group,and CUMS+RU486 group.DMSO of 0.5 ml/(kg·d)was injected subcutaneously into the rats of DMSO group.After the rats of the CUMS group were exposed to CUMS,mifepristone of 1 mg/(kg·d)was given to the rats of the RU486 group and CUMS+RU486 group.All the above 4 groups were inflicted for 42 consecutive days.After modeling,the psychological stress status of rats was evaluated with animal behavioral experiments.Computer-aided sperm analysis(CASA)system was used to detect sperm density.ELISA was employed to measure the serum levels of corticosterone(CORT),sex hormones and inflammation factors.The contents of malondialdehyde(MDA),H2O2,NAD+/NADH and ATP in the testicular tissues were tested with corresponding reagent kits.Western blotting was utilized to detect protein levels of[3-gal,p16,and GR in testicular tissue,and transmission electron microscopy was used to observe mitochondrial structure of spermatogenic cells.Results Compared with the DMSO group,the sperm density,serum levels of testosterone(T),luteinizing hormone(LH)and follicle stimulating hormone(FSH)were decreased while those of CORT and estradiol(E2)were increased in the CUMS group(all P<0.001).Serum pro-inflammatory factors IL-6 and TNF-α concentrations were increased while anti-inflammatory factor IL-10 concentration was decreased in the CUMS group than the DMSO group(all P<0.05).And,ATP content and NAD+/NADH ratio were decreased while MDA and H2O2 levels were elevated in testicular tissues in the CUMS group(all P<0.05).Moreover,the protein levels of testicular cell aging markers β-gal,p16 and GR were significantly enhanced in the CUMS group(all P<0.01),accompanied by swelling and vacuolization of mitochondrial structures in spermatogenic cells.RU486 treatment effectively improved sperm density and increased CORT level and down-regulated the testicularβ-gal,p16 and GR protein level(all P<0.05).Conclusion Chronic psychological stress induces testicular damage in rats,presenting declines in sperm density and CORT level,which is possibly associated with GR-mediated cell senescence.
6.Preconception reproductive health and birth outcome cohort in Chongqing: the cohort profile
Qing CHEN ; Wenzheng ZHOU ; Niya ZHOU ; Huan YANG ; Yimeng WANG ; Haiyan ZHANG ; Qiuhong LI ; Nianrong WANG ; Hongyan CHEN ; Lin AO ; Jinyi LIU ; Ziyuan ZHOU ; Hua ZHANG ; Wei ZHOU ; Hongbo QI ; Jia CAO
Chinese Journal of Epidemiology 2022;43(7):1134-1139
Birth cohort is an important platform to study the effect of early-life exposure on health outcome, but large cohorts to investigate the effect of preconception exposure, especially paternal exposure, on reproductive health and birth outcome are limited. The Preconception Reproductive Health and Birth Outcome Cohort (PREBIC) is a prospective birth cohort study which pays equal attention to the contribution of environmental, psychological, behavioral as well as other factors to reproductive health and adverse birth outcomes in both men and women in Chongqing, China. PREBIC started in 2019 and plans to recruit 20 800 reproductive-age couples with child-bearing willingness. Followed up was conducted to understand the conception status of the women within two years. Women in pregnancy would be visited at first, second, third trimesters and after delivery. The offspring would be monitored until 2 years old to understand the incidences of preterm birth, low birth weight, birth defects, neurodevelopmental disorders and other outcomes. Related information and biospecimen collections (including semen, peripheral blood, urine, placenta, umbilical cord, cord blood and oral swab) were scheduled in each period. By January 2022, PREBIC had recruited 8 698 participants from all 38 districts in Chongqing. The goal of PREBIC is to establish one of the largest prospective preconception birth cohorts covering both men and women, which might provide a unique insight to understand the effects of the full reproductive cycle on reproductive health and adverse outcomes, with especial emphasis on preconception exposures.
7.Inhibition of MYC suppresses programmed cell death ligand-1 expression and enhances immunotherapy in triple-negative breast cancer
Xintong LI ; Lin TANG ; Qin CHEN ; Xumin CHENG ; Yiqiu LIU ; Cenzhu WANG ; Chengjun ZHU ; Kun XU ; Fangyan GAO ; Jinyi HUANG ; Runtian WANG ; Xiaoxiang GUAN
Chinese Medical Journal 2022;135(20):2436-2445
Background::Cancer immunotherapy has emerged as a promising strategy against triple-negative breast cancer (TNBC). One of the immunosuppressive pathways involves programmed cell death-1 (PD-1) and programmed cell death ligand-1 (PD-L1), but many patients derived little benefit from PD-1/PD-L1 checkpoint blockades treatment. Prior research has shown that MYC, a master transcription amplifier highly expressed in TNBC cells, can regulate the tumor immune microenvironment and constrain the efficacy of immunotherapy. This study aims to investigate the regulatory relationship between MYC and PD-L1, and whether a cyclin-dependent kinase (CDK) inhibitor that inhibits MYC expression in combination with anti-PD-L1 antibodies can enhance the response to immunotherapy. Methods::Public databases and TNBC tissue microarrays were used to study the correlation between MYC and PD-L1. The expression of MYC and PD-L1 in TNBCs was examined by quantitative real-time polymerase chain reaction and Western blotting. A patient-derived tumor xenograft (PDTX) model was used to evaluate the influence of a CDK7 inhibitor THZ1 on PD-L1 expression. Cell proliferation and migration were detected by 5-ethynyl-2′-deoxyuridine (EdU) cell proliferation and cell migration assays. Tumor xenograft models were established for in vivo verification. Results::A high MYC expression level was associated with a poor prognosis and could alter the proportion of tumor-infiltrating immune cells (TIICs). The positive correlation between MYC and PD-L1 was confirmed by immunostaining samples from 165 TNBC patients. Suppression of MYC in TNBC caused a reduction in the levels of both PD-L1 messenger RNA and protein. In addition, antitumor immune response was enhanced in the TNBC cancer xenograft mouse model with suppression of MYC by CDK7 inhibitor THZ1. Conclusions::The combined therapy of CDK7 inhibitor THZ1 and anti-PD-L1 antibody appeared to have a synergistic effect, which might offer new insight for enhancing immunotherapy in TNBC.
8. Study on the health literacy and related factors of the cancer prevention consciousness among urban residents in China from 2015 to 2017
Chengcheng LIU ; Chunlei SHI ; Jufang SHI ; Ayan MAO ; Huiyao HUANG ; Pei DONG ; Fangzhou BAI ; Yunsi CHEN ; Debin WANG ; Guoxiang LIU ; Xianzhen LIAO ; Yana BAI ; Xiaojie SUN ; Jiansong REN ; Li YANG ; Donghua WEI ; Bingbing SONG ; Haike LEI ; Yuqin LIU ; Yongzhen ZHANG ; Siying REN ; Jinyi ZHOU ; Jialin WANG ; Jiyong GONG ; Lianzheng YU ; Yunyong LIU ; Lin ZHU ; Lanwei GUO ; Youging WANG ; Yutong HE ; Peian LOU ; Bo CAI ; Xiaohua SUN ; Shouling WU ; Xiao QI ; Kai ZHANG ; Ni LI ; Wanghong XU ; Wuqi QIU ; Min DAI ; Wanqing CHEN
Chinese Journal of Preventive Medicine 2020;54(1):47-53
Objective:
To understand the health literacy and relevant factors of cancer prevention consciousness in Chinese urban residents from 2015 to 2017.
Methods:
A cross-sectional survey was conducted in 16 provinces covered by the Cancer Screening Program in Urban China from 2015 to 2017. A total of 32 257 local residents aged ≥18 years old who could understand the investigation procedure were included in the study by using the cluster sampling method and convenient sampling method. All local residents were categorized into four groups, which contained 15 524 community residents, 8 016 cancer risk assessment/screening population, 2 289 cancer patients and 6 428 occupational population, respectively. The self-designed questionnaire was used to collect the information of demographic characteristics and cancer prevention consciousness focusing on nine common risk factors, including smoking, alcohol, fiber food, food in hot temperature or pickled food, chewing betel nut, helicobacter pylori, moldy food, hepatitis B infection, estrogen, and exercise. The logistic regression model was adopted to identify the influencing factors.
Results:
The overall health literacy of the cancer prevention consciousness was 77.4% (24 980 participants), with 77.4% (12 018 participants), 79.9% (6 406 participants), 77.2% (1 766 participants) and 74.5% (4 709 participants) in each group (
9. Analysis on the consciousness of the cancer early detection and its influencing factors among urban residents in China from 2015 to 2017
Ayan MAO ; Jufang SHI ; Wuqi QIU ; Chengcheng LIU ; Pei DONG ; Huiyao HUANG ; Kun WANG ; Debin WANG ; Guoxiang LIU ; Xianzhen LIAO ; Yana BAI ; Xiaojie SUN ; Jiansong REN ; Li YANG ; Donghua WEI ; Bingbing SONG ; Haike LEI ; Yuqin LIU ; Yongzhen ZHANG ; Siying REN ; Jinyi ZHOU ; Jialin WANG ; Jiyong GONG ; Lianzheng YU ; Yunyong LIU ; Lin ZHU ; Lanwei GUO ; Youqing WANG ; Yutong HE ; Peian LOU ; Bo CAI ; Xiaohua SUN ; Shouling WU ; Xiao QI ; Kai ZHANG ; Ni LI ; Min DAI ; Wanqing CHEN
Chinese Journal of Preventive Medicine 2020;54(1):54-61
Objective:
To understand the consciousness of the cancer early detection among urban residents and identify the influencing factors from 2015 to 2017.
Methods:
A cross-sectional survey was conducted in 16 provinces covered by the Cancer Screening Program in Urban China from 2015 to 2017. A total of 32 257 local residents aged ≥18 years old who could understand the investigation procedure were included in the study by using the cluster sampling method and convenient sampling method. All local residents were categorized into four groups, which contained 15 524 community residents, 8 016 cancer risk assessment/screening population, 2 289 cancer patients and 6 428 occupational population, respectively. Self-designed questionnaires were used to collect population, socioeconomic indicators, self-cancer risk assessment, regular participation in physical examination and other information. The multivariate logistic regression model was used to identify the factors of people who had not regularly participated in the regular physical examination in the past five years.
Results:
The self-assessment results of 32 357 residents showed that there were 27.54% (8 882) of total study population with self-reported cancer risk, 45.48% (14 671) without cancer risk and 26.98% (8 704) with unclear judgement on their own cancer risk. Among population with cancer risk, 79.84% (7 091) considered physical examination accounted. In the past five years, there were 21 105 (65.43%) residents participated in regular physical examination and 11 148 (34.56%) participated in non-scheduled one, respectively. The multivariate logistic regression analysis showed that compared with unmarried and western region residents, divorced, middle and eastern region residents had a stronger consciousness to participate in the regular physical examination (
10. Analysis on the consciousness of the early cancer diagnosis and its related factors among urban residents in China from 2015 to 2017
Xuan CHENG ; Pei DONG ; Jufang SHI ; Wuqi QIU ; Chengcheng LIU ; Kun WANG ; Huiyao HUANG ; Yana BAI ; Xiaojie SUN ; Debin WANG ; Guoxiang LIU ; Xianzhen LIAO ; Li YANG ; Donghua WEI ; Bingbing SONG ; Haike LEI ; Yuqin LIU ; Yongzhen ZHANG ; Siying REN ; Jinyi ZHOU ; Jialin WANG ; Jiyong GONG ; Lianzheng YU ; Yunyong LIU ; Lin ZHU ; Lanwei GUO ; Youqing WANG ; Yutong HE ; Peian LOU ; Bo CAI ; Xiaohua SUN ; Shouling WU ; Xiao QI ; Kai ZHANG ; Ni LI ; Jiansong REN ; Wanqing CHEN ; Min DAI ; Ayan MAO
Chinese Journal of Preventive Medicine 2020;54(1):62-68
Objective:
To understand the consciousness of the cancer early diagnosis among urban residents and identify the related factors from 2015 to 2017.
Methods:
A cross-sectional survey was conducted in 16 provinces covered by the Cancer Screening Program in Urban China from 2015 to 2017. A total of 32 257 local residents aged ≥18 years old who could understand the investigation procedure were included in the study by using the cluster sampling method and convenient sampling method. All local residents were categorized into four groups, which contained 15 524 community residents, 8 016 cancer risk assessment/screening population, 2 289 cancer patients and 6 428 occupational population, respectively. The general demographic characteristics, the consciousness of the cancer early diagnosis (whether people would have a willingness or encourage their relatives/friends to confirm the abnormal results once which were detected from the physical examination) and other information were collected by using the self-designed questionnaire. The non-conditional logistic regression model was used to identify the relateol factors related to the consciousness of the cancer early diagnosis.
Results:
As for residents with abnormal result from the physical examination, 89.29% (28 802) of residents would choose to seek medical treatment for further diagnosis. If their relatives/friends had abnormal results from the physical examination, 89.55% (28 886) of residents would encourage their relatives/friends to confirm the diagnosis in time. The non-conditional logistic regression model analysis showed that compared with the public institution staff/civil servants, annual household income less than 20 000 CNY, the western region and the cancer risk assessment/screening intervention population, the company staff, annual household income about 40 000 CNY and more, and the residents from the middle and eastern region had a stronger consciousness to seek further diagnosis; while the unemployed residents and community residents were less likely to seek further diagnosis (

Result Analysis
Print
Save
E-mail