Objective:To develop a deep learning model to predict lymph node(LN)status in clinical stage IA lung adeno-carcinoma patients.Methods:This diagnostic study included 1,009 patients with pathologically confirmed clinica...Objective:To develop a deep learning model to predict lymph node(LN)status in clinical stage IA lung adeno-carcinoma patients.Methods:This diagnostic study included 1,009 patients with pathologically confirmed clinical stage T1N0M0 lung adenocarcinoma from two independent datasets(699 from Cancer Hospital of Chinese Academy of Medical Sciences and 310 from PLA General Hospital)between January 2005 and December 2019.The Cancer Hospital dataset was randomly split into a training cohort(559 patients)and a validation cohort(140 patients)to train and tune a deep learning model based on a deep residual network(ResNet).The PLA Hospital dataset was used as a testing cohort to evaluate the generalization ability of the model.Thoracic radiologists manually segmented tumors and interpreted high-resolution computed tomography(HRCT)features for the model.The predictive performance was assessed by area under the curves(AUCs),accuracy,precision,recall,and F1 score.Subgroup analysis was performed to evaluate the potential bias of the study population.Results:A total of 1,009 patients were included in this study;409(40.5%)were male and 600(59.5%)were female.The median age was 57.0 years(inter-quartile range,IQR:50.0-64.0).The deep learning model achieved AUCs of 0.906(95%CI:0.873-0.938)and 0.893(95%CI:0.857-0.930)for predicting pN0 disease in the testing cohort and a non-pure ground glass nodule(non-pGGN)testing cohort,respectively.No significant difference was detected between the testing cohort and the non-pGGN testing cohort(P=0.622).The precisions of this model for predicting pN0 disease were 0.979(95%CI:0.963-0.995)and 0.983(95%CI:0.967-0.998)in the testing cohort and the non-pGGN testing cohort,respectively.The deep learning model achieved AUCs of 0.848(95%CI:0.798-0.898)and 0.831(95%CI:0.776-0.887)for predicting pN2 disease in the testing cohort and the non-pGGN testing cohort,respectively.No significant difference was detected between the testing cohort and the non-pGGN testing cohort(P=0.657).The recalls of this model for predicting pN2 disease were 0.903(95%CI:0.870-0.936)and 0.931(95%CI:0.901-0.961)in the testing cohort and the non-pGGN testing cohort,respectively.Conclusions:The superior performance of the deep learning model will help to target the extension of lymph node dissection and reduce the ineffective lymph node dissection in early-stage lung adenocarcinoma patients.展开更多
In order to investigate the clinical significance of 99mTc-Tetrofosmin （TF） scintigraphy in the evaluation of lung cancer and mediastinal lymphoid node involvement, 33 patients with pulmo- nary neoplasmas were subje...In order to investigate the clinical significance of 99mTc-Tetrofosmin （TF） scintigraphy in the evaluation of lung cancer and mediastinal lymphoid node involvement, 33 patients with pulmo- nary neoplasmas were subjected to both 99mTc-TF scintigraphies and CT scans in one week before their operations or puncturations. All the images were judged visually and the emission images were analyzed with semi-quantitative methods in addition. The results of each group were compared. There was marked difference in target/non-target （T/N） ratio between the lung cancer group and the benign lesion group （P〈0.001）. Moreover, in the lung cancer group, T/N ratio in tomographies was signifi- cantly higher than that in planar images （P〈0.01）. The sensitivity and accuracy of semi-quantitative analysis in 99mTc-TF SPECT were significantly higher than those of CT in the diagnosis of pulmonary neoplasmas （P〈0.05 and P〈0.01 respectively）, so was the sensitivity of 99mTc-TF SPECT vs CT in the diagnosis of mediastinal lymphoid node metastasis （P〈0.05）. It was also found that epidermoid squamous cell carcinomas and adenocarcinomas had a higher T/N ratio than in small cell carcinomas （P〈0.05）, and 2 h washout rate （WR） of adenocarcinomas was higher than that of epidermoid squamous cell carcinomas （P〈0.05）. In conclusion, 99mTc-TF scintigraphy showed a favorable diag- nostic accuracy in appraising lung cancers and mediastinal lymph node metastases. Furthermore semi-quantitative technology can improve the accuracy, and is potential to offer some information about histological type of the cancer tissue. Therefore, 99mTc-TF scintigraphy will be a useful tool in the diagnosis and staging of lung cancer.展开更多
Corona virus disease 2019(COVID-19)infection has become a major public health issue affecting human health.The main goal of epidemic prevention and control at the current stage in China is to“protect people’s health...Corona virus disease 2019(COVID-19)infection has become a major public health issue affecting human health.The main goal of epidemic prevention and control at the current stage in China is to“protect people’s health and prevent severe cases”.Patients with lung cancer who receive antitumor therapy have low immunity,and the risk of severe illness and death once infected is much higher than healthy people,so they are vulnerable to COVID-19 infection.At present,less attention has been paid to the prevention and treatment of COVID-19 infection in patients with lung cancer in domestic guidelines and consensus.Based on the published data in China and abroad,we proposed recommendations and formed expert consensus on the vaccination of COVID-19,the use of neutralizing antibodies and small molecule antiviral drugs for patients with lung cancer,for physician’s reference.展开更多
Background: Resistance to cisplatin (DDP) leads to poor prognosis in patients with Lung Adenocarcinoma (LUAD) and limits its clinical application. It has been confirmed that autophagy promotes chemoresistance and, the...Background: Resistance to cisplatin (DDP) leads to poor prognosis in patients with Lung Adenocarcinoma (LUAD) and limits its clinical application. It has been confirmed that autophagy promotes chemoresistance and, therefore, novel strategies to reverse chemoresistance by regulating autophagy are desperately needed. Methods: The differentially expressed lncRNAs (DElncRNAs), miRNAs (DEmiRNAs), and mRNAs (DEmRNAs) between A549 and A549/DDP cell lines were identified using the limma package in R, after gene expression profiles were obtained from Gene Expression Omnibus (GEO) database. By combining Autophagy-Related Genes (ARGs) from Human Autophagy Database (HADb), the interactions lncRNA-miRNAs and the interactions miRNAs-mRNAs respectively predicted by miRcode and miRDB/Targetscan database, the autophagy-related ceRNA network was constructed. Then, extraction of ceRNA subnetwork and Cox regression analyses were performed. A prognosis-related ceRNA subnetwork was constructed, and the upstream Transcription Factors (TFs) regulating lncRNAs were predicted by the JASPAR database. Finally, the expression patterns of candidate genes were further verified by quantitative real-time polymerase chain reaction (qRT-PCR) experiments. Results: A total of 3179 DEmRNAs, 180 DEmiRNAs, and 160 DElncRNAs were identified, and 35 DEmRNAs were contained in the HADb. Based on the ceRNA hypothesis, we established a ceRNA network, including 10 autophagy-related DEmRNAs, 9 DEmiRNAs, and 14 DElncRNAs. Then, LINC00520, miR-181d, and BCL2 were identified to construct a risk score model, which was confirmed to be a well-predicting prognostic factor. Furthermore, 5 TF ZNF family members were predicted to regulate LINC00520, whereas the RT-PCR results showed that the 5 ZNFs were consistent with the bioinformatics analysis. Finally, a ZNF regulatory LINC00520/miR-181d/BCL2 ceRNA subnetwork was constructed. Conclusions: An ZNFs/LINC00520/miR-181d/BCL2 axis as a novel network in DDP-resistant LUAD has been constructed successfully, which may provide potential therapeutic targets for LUAD.展开更多
背景与目的囊腔型肺癌作为一种特殊类型的肺癌逐步得到人们的关注,其最常见的病理类型为腺癌。囊腔型肺腺癌的浸润性对诊疗方案的选择和预后至关重要。本研究旨在分析囊腔型肺腺癌临床多特征,探讨其浸润性的独立危险因素并建立风险预测...背景与目的囊腔型肺癌作为一种特殊类型的肺癌逐步得到人们的关注,其最常见的病理类型为腺癌。囊腔型肺腺癌的浸润性对诊疗方案的选择和预后至关重要。本研究旨在分析囊腔型肺腺癌临床多特征,探讨其浸润性的独立危险因素并建立风险预测模型。方法回顾性分析2021年1月至2022年7月于南京医科大学第一附属医院胸外科行手术治疗的129例囊腔型肺腺癌患者,根据病理结果分成浸润前组:非典型腺瘤样增生(atypical adenomatous hyperplasia,AAH)、原位腺癌(adenocarcinoma in situ,AIS)、微浸润型腺癌(minimally invasive adenocarcinoma,MIA)与浸润组:浸润性腺癌(invasive adenocarcinoma,IAC)。其中浸润前组47例,男性19例,女性28例,平均年龄(51.23±14.96)岁;浸润组82例,男性60例,女性22例,平均年龄(61.27±11.74)岁。收集两组病例多组临床特征,采用单因素分析、LASSO回归、多因素Logistic回归分析得出囊腔型肺腺癌浸润性的独立危险因素,建立浸润性风险预测模型。结果单因素分析显示年龄、性别、吸烟史、肺气肿、神经元特异性烯醇化酶(neuron-specific enolase,NSE)、囊腔数、病灶直径、囊腔直径、结节直径、实性成分直径、囊壁结节、囊壁光滑程度、囊腔形状、分叶征、短毛刺征、胸膜牵拉、血管穿行与支气管穿行在囊腔型肺腺癌浸润前组与浸润组间存在统计学差异(P<0.05)。上述变量经LASSO回归降维处理,进一步筛选出的变量包括:年龄、性别、吸烟史、NSE、囊腔数、病灶直径、囊腔直径、囊壁结节、囊壁光滑程度与分叶征,并纳入多因素Logistic回归分析,发现囊壁结节(P=0.035)与分叶征(P=0.001)是囊腔型肺腺癌浸润性的独立危险因素(P<0.05)。建立预测模型如下:P=e^x/(1+e^x),x=-7.927+1.476*囊壁结节+2.407*分叶征,曲线下面积(area under the curve,AUC)为0.950。结论囊壁结节及分叶征为囊腔型肺腺癌浸润性的独立危险因素,对囊腔型肺腺癌的浸润性预测具有一定的指导意义。展开更多
目的构建并验证一个模型以预测肺癌脑转移(lung cancer with brain metastases,LCBM)患者确诊后三个月内死亡的风险。方法本研究纳入监测,流行病学和最终结果(Surveillance,Epidemiology and End Results,SEER)数据库内2010年1月至2015...目的构建并验证一个模型以预测肺癌脑转移(lung cancer with brain metastases,LCBM)患者确诊后三个月内死亡的风险。方法本研究纳入监测,流行病学和最终结果(Surveillance,Epidemiology and End Results,SEER)数据库内2010年1月至2015年12月期间确诊为LCBM的患者。以是否发生早期死亡为研究终点将患者分为早期死亡组和非早期死亡组。以8∶2为比例将数据分为训练集和验证集。在训练集上采用最小绝对值收缩和筛选算子(least absolute shrinkage and selection operator,LASSO)回归法筛选预测因子,并使用多因素Logistic回归构建预测模型并创建列线图。使用受试者工作特征(receiver operating characteristic,ROC)曲线、校准曲线和临床决策曲线(decision curve analysis,DCA)分别在训练集和验证集上评估模型性能。结果共纳入5035例患者,早期死亡发生率28.3%。LASSO回归筛选出13个变量,Logistic回归最终保留了13个与LCBM患者早期死亡相关的危险因素,包括年龄、从诊断到开始治疗时间、肿瘤大小、肿瘤部位、肿瘤分化程度和组织学类型、T分期、N分期、手术、放疗、化疗、肝转移和骨转移。验证集的一致性指数(concordance index,C-index)为0.84,校准曲线和DCA显示模型具有较好的预测效能和临床净效益。结论基于多因素Logistic回归构建的LCBM患者发生早期死亡的预测模型的区分度较好,能够为临床决策提供一定的帮助。展开更多
基金supported by the National Key R&D Program of China(grant numbers:2020AAA0109504,2023YFC2415200)CAMS Innovation Fund for Medical Sciences(grant number:2021-I2M-C&T-B-061)+5 种基金Beijing Hope Run Special Fund of Cancer Foundation of China(grant number:LC2022A22)the National Natural Science Foundation of China(grant numbers:81971619,81971580,92259302,82372053,91959205,82361168664,82022036,81971776)Beijing Natural Sci-ence Foundation(grant number:Z20J00105)Key-Area Research and Development Program of Guangdong Province(grant number:2021B0101420005)Strategic Priority Research Program of Chinese Academy of Sciences(grant number:XDB38040200)the Youth In-novation Promotion Association CAS(grant number:Y2021049).
文摘Objective:To develop a deep learning model to predict lymph node(LN)status in clinical stage IA lung adeno-carcinoma patients.Methods:This diagnostic study included 1,009 patients with pathologically confirmed clinical stage T1N0M0 lung adenocarcinoma from two independent datasets(699 from Cancer Hospital of Chinese Academy of Medical Sciences and 310 from PLA General Hospital)between January 2005 and December 2019.The Cancer Hospital dataset was randomly split into a training cohort(559 patients)and a validation cohort(140 patients)to train and tune a deep learning model based on a deep residual network(ResNet).The PLA Hospital dataset was used as a testing cohort to evaluate the generalization ability of the model.Thoracic radiologists manually segmented tumors and interpreted high-resolution computed tomography(HRCT)features for the model.The predictive performance was assessed by area under the curves(AUCs),accuracy,precision,recall,and F1 score.Subgroup analysis was performed to evaluate the potential bias of the study population.Results:A total of 1,009 patients were included in this study;409(40.5%)were male and 600(59.5%)were female.The median age was 57.0 years(inter-quartile range,IQR:50.0-64.0).The deep learning model achieved AUCs of 0.906(95%CI:0.873-0.938)and 0.893(95%CI:0.857-0.930)for predicting pN0 disease in the testing cohort and a non-pure ground glass nodule(non-pGGN)testing cohort,respectively.No significant difference was detected between the testing cohort and the non-pGGN testing cohort(P=0.622).The precisions of this model for predicting pN0 disease were 0.979(95%CI:0.963-0.995)and 0.983(95%CI:0.967-0.998)in the testing cohort and the non-pGGN testing cohort,respectively.The deep learning model achieved AUCs of 0.848(95%CI:0.798-0.898)and 0.831(95%CI:0.776-0.887)for predicting pN2 disease in the testing cohort and the non-pGGN testing cohort,respectively.No significant difference was detected between the testing cohort and the non-pGGN testing cohort(P=0.657).The recalls of this model for predicting pN2 disease were 0.903(95%CI:0.870-0.936)and 0.931(95%CI:0.901-0.961)in the testing cohort and the non-pGGN testing cohort,respectively.Conclusions:The superior performance of the deep learning model will help to target the extension of lymph node dissection and reduce the ineffective lymph node dissection in early-stage lung adenocarcinoma patients.
文摘In order to investigate the clinical significance of 99mTc-Tetrofosmin （TF） scintigraphy in the evaluation of lung cancer and mediastinal lymphoid node involvement, 33 patients with pulmo- nary neoplasmas were subjected to both 99mTc-TF scintigraphies and CT scans in one week before their operations or puncturations. All the images were judged visually and the emission images were analyzed with semi-quantitative methods in addition. The results of each group were compared. There was marked difference in target/non-target （T/N） ratio between the lung cancer group and the benign lesion group （P〈0.001）. Moreover, in the lung cancer group, T/N ratio in tomographies was signifi- cantly higher than that in planar images （P〈0.01）. The sensitivity and accuracy of semi-quantitative analysis in 99mTc-TF SPECT were significantly higher than those of CT in the diagnosis of pulmonary neoplasmas （P〈0.05 and P〈0.01 respectively）, so was the sensitivity of 99mTc-TF SPECT vs CT in the diagnosis of mediastinal lymphoid node metastasis （P〈0.05）. It was also found that epidermoid squamous cell carcinomas and adenocarcinomas had a higher T/N ratio than in small cell carcinomas （P〈0.05）, and 2 h washout rate （WR） of adenocarcinomas was higher than that of epidermoid squamous cell carcinomas （P〈0.05）. In conclusion, 99mTc-TF scintigraphy showed a favorable diag- nostic accuracy in appraising lung cancers and mediastinal lymph node metastases. Furthermore semi-quantitative technology can improve the accuracy, and is potential to offer some information about histological type of the cancer tissue. Therefore, 99mTc-TF scintigraphy will be a useful tool in the diagnosis and staging of lung cancer.
文摘Corona virus disease 2019(COVID-19)infection has become a major public health issue affecting human health.The main goal of epidemic prevention and control at the current stage in China is to“protect people’s health and prevent severe cases”.Patients with lung cancer who receive antitumor therapy have low immunity,and the risk of severe illness and death once infected is much higher than healthy people,so they are vulnerable to COVID-19 infection.At present,less attention has been paid to the prevention and treatment of COVID-19 infection in patients with lung cancer in domestic guidelines and consensus.Based on the published data in China and abroad,we proposed recommendations and formed expert consensus on the vaccination of COVID-19,the use of neutralizing antibodies and small molecule antiviral drugs for patients with lung cancer,for physician’s reference.
文摘Background: Resistance to cisplatin (DDP) leads to poor prognosis in patients with Lung Adenocarcinoma (LUAD) and limits its clinical application. It has been confirmed that autophagy promotes chemoresistance and, therefore, novel strategies to reverse chemoresistance by regulating autophagy are desperately needed. Methods: The differentially expressed lncRNAs (DElncRNAs), miRNAs (DEmiRNAs), and mRNAs (DEmRNAs) between A549 and A549/DDP cell lines were identified using the limma package in R, after gene expression profiles were obtained from Gene Expression Omnibus (GEO) database. By combining Autophagy-Related Genes (ARGs) from Human Autophagy Database (HADb), the interactions lncRNA-miRNAs and the interactions miRNAs-mRNAs respectively predicted by miRcode and miRDB/Targetscan database, the autophagy-related ceRNA network was constructed. Then, extraction of ceRNA subnetwork and Cox regression analyses were performed. A prognosis-related ceRNA subnetwork was constructed, and the upstream Transcription Factors (TFs) regulating lncRNAs were predicted by the JASPAR database. Finally, the expression patterns of candidate genes were further verified by quantitative real-time polymerase chain reaction (qRT-PCR) experiments. Results: A total of 3179 DEmRNAs, 180 DEmiRNAs, and 160 DElncRNAs were identified, and 35 DEmRNAs were contained in the HADb. Based on the ceRNA hypothesis, we established a ceRNA network, including 10 autophagy-related DEmRNAs, 9 DEmiRNAs, and 14 DElncRNAs. Then, LINC00520, miR-181d, and BCL2 were identified to construct a risk score model, which was confirmed to be a well-predicting prognostic factor. Furthermore, 5 TF ZNF family members were predicted to regulate LINC00520, whereas the RT-PCR results showed that the 5 ZNFs were consistent with the bioinformatics analysis. Finally, a ZNF regulatory LINC00520/miR-181d/BCL2 ceRNA subnetwork was constructed. Conclusions: An ZNFs/LINC00520/miR-181d/BCL2 axis as a novel network in DDP-resistant LUAD has been constructed successfully, which may provide potential therapeutic targets for LUAD.
文摘背景与目的囊腔型肺癌作为一种特殊类型的肺癌逐步得到人们的关注,其最常见的病理类型为腺癌。囊腔型肺腺癌的浸润性对诊疗方案的选择和预后至关重要。本研究旨在分析囊腔型肺腺癌临床多特征,探讨其浸润性的独立危险因素并建立风险预测模型。方法回顾性分析2021年1月至2022年7月于南京医科大学第一附属医院胸外科行手术治疗的129例囊腔型肺腺癌患者,根据病理结果分成浸润前组:非典型腺瘤样增生(atypical adenomatous hyperplasia,AAH)、原位腺癌(adenocarcinoma in situ,AIS)、微浸润型腺癌(minimally invasive adenocarcinoma,MIA)与浸润组:浸润性腺癌(invasive adenocarcinoma,IAC)。其中浸润前组47例,男性19例,女性28例,平均年龄(51.23±14.96)岁;浸润组82例,男性60例,女性22例,平均年龄(61.27±11.74)岁。收集两组病例多组临床特征,采用单因素分析、LASSO回归、多因素Logistic回归分析得出囊腔型肺腺癌浸润性的独立危险因素,建立浸润性风险预测模型。结果单因素分析显示年龄、性别、吸烟史、肺气肿、神经元特异性烯醇化酶(neuron-specific enolase,NSE)、囊腔数、病灶直径、囊腔直径、结节直径、实性成分直径、囊壁结节、囊壁光滑程度、囊腔形状、分叶征、短毛刺征、胸膜牵拉、血管穿行与支气管穿行在囊腔型肺腺癌浸润前组与浸润组间存在统计学差异(P<0.05)。上述变量经LASSO回归降维处理,进一步筛选出的变量包括:年龄、性别、吸烟史、NSE、囊腔数、病灶直径、囊腔直径、囊壁结节、囊壁光滑程度与分叶征,并纳入多因素Logistic回归分析,发现囊壁结节(P=0.035)与分叶征(P=0.001)是囊腔型肺腺癌浸润性的独立危险因素(P<0.05)。建立预测模型如下:P=e^x/(1+e^x),x=-7.927+1.476*囊壁结节+2.407*分叶征,曲线下面积(area under the curve,AUC)为0.950。结论囊壁结节及分叶征为囊腔型肺腺癌浸润性的独立危险因素,对囊腔型肺腺癌的浸润性预测具有一定的指导意义。
文摘目的构建并验证一个模型以预测肺癌脑转移(lung cancer with brain metastases,LCBM)患者确诊后三个月内死亡的风险。方法本研究纳入监测,流行病学和最终结果(Surveillance,Epidemiology and End Results,SEER)数据库内2010年1月至2015年12月期间确诊为LCBM的患者。以是否发生早期死亡为研究终点将患者分为早期死亡组和非早期死亡组。以8∶2为比例将数据分为训练集和验证集。在训练集上采用最小绝对值收缩和筛选算子(least absolute shrinkage and selection operator,LASSO)回归法筛选预测因子,并使用多因素Logistic回归构建预测模型并创建列线图。使用受试者工作特征(receiver operating characteristic,ROC)曲线、校准曲线和临床决策曲线(decision curve analysis,DCA)分别在训练集和验证集上评估模型性能。结果共纳入5035例患者,早期死亡发生率28.3%。LASSO回归筛选出13个变量,Logistic回归最终保留了13个与LCBM患者早期死亡相关的危险因素,包括年龄、从诊断到开始治疗时间、肿瘤大小、肿瘤部位、肿瘤分化程度和组织学类型、T分期、N分期、手术、放疗、化疗、肝转移和骨转移。验证集的一致性指数(concordance index,C-index)为0.84,校准曲线和DCA显示模型具有较好的预测效能和临床净效益。结论基于多因素Logistic回归构建的LCBM患者发生早期死亡的预测模型的区分度较好,能够为临床决策提供一定的帮助。
