预测肝癌肝切除术后肝衰竭的影像组学列线图模型建立
|
摘要
目的探讨基于术前CT图像的影像组学特征对肝细胞癌(肝癌)肝切除术后肝衰竭(PHLF)的预测价值,并建立预测肝癌PHLF的影像组学列线图(nomogram)模型。方法回顾性分析2014年9月至2017年10月在中山大学附属第一医院行半肝切除术的51例肝癌患者临床资料。患者均签署知情同意书,符合医学伦理学规定。其中男44例,女7例;平均年龄(48±12)岁。患者术前均接受增强CT检查,使用A.K.软件进行肝癌影像特征提取。采用单因素Logistic回归分析筛选与PHLF发生相关的影像组学特征和临床变量;将相关的影像组学特征和临床变量纳入多因素分析,得到与PHLF发生相关的独立危险因素。根据独立危险因素,建立预测肝癌PHLF的nomogram模型。结果低密度短域补偿(LISAE)和白蛋白-胆红素(ALBI)评分是PHLF发生的独立影响因素(OR=27.93,15.53;P<0.05)。联合LISAE和ALBI评分预测PHLF的曲线下面积为0.883,明显大于单独ALBI评分的0.700(Z=-2.460,P<0.05)。肝癌预测PHLF的nomogram模型成功建立,校正的一致性系数(C-index)为0.863,标准曲线与校准预测曲线贴合良好,预测值与观察值符合度良好。结论LISAE是肝癌PHLF发生的独立影响因素,能提高临床变量ALBI评分的预测价值;预测PHLF发生的nomogram模型对肝癌发生PHLF有较好预测价值。
Objective To explore the value of preoperative CT image omics features in predicting posthepatectomy liver failure(PHLF) in patients with hepatocellular carcinoma(HCC) and to establish an image omics nomogram for predicting PHLF. Methods Clinical data of 51 HCC patients who underwent hemihepatectomy in the First Affiliated Hospital of Sun Yat-sen University from September 2014 to October2017 were retrospectively analyzed. The informed consents of all patients were obtained and the local ethical committee approval was received. Among them, 44 patients were male and 7 female, aged(48±12) years on average. All patients underwent enhanced CT scan before operation and the HCC CT-imaging features were extracted by A.K. software. The image omics features and clinical variables related to the occurrence of PHLF were screened by univariate Logistic regression, and were included in multivariate analysis to obtain the independent risk factors for the occurrence of PHLF. The nomogram for predicting PHLF in HCC patients was established according to the independent risk factors. Results Low intensity small area emphasis(LISAE) and albumin-bilirubin(ALBI) score were the independent factors affecting the occurrence of PHLF(OR=27.93, 15.53; P<0.05). The area under the curve for predicting PHLF by LISAE combined with ALBI score was 0.883, significantly larger than the 0.700 of single ALBI score(Z=-2.460, P<0.05). Nomogram for predicting PHLF in HCC patients was successfully established. The corrected consistency index(C-index) was 0.863. The standard curve was well fitted with the calibrated predictive curve, and the predicted value was highly consistent with the observed value. Conclusions LISAE is an independent factor affecting the occurrence of PHLF in HCC patients, which can improve the predictive value of clinical variable ALBI score. The nomogram for predicting PHLF yields high predictive value for PHLF in HCC patients.
Objective To explore the value of preoperative CT image omics features in predicting posthepatectomy liver failure(PHLF) in patients with hepatocellular carcinoma(HCC) and to establish an image omics nomogram for predicting PHLF. Methods Clinical data of 51 HCC patients who underwent hemihepatectomy in the First Affiliated Hospital of Sun Yat-sen University from September 2014 to October2017 were retrospectively analyzed. The informed consents of all patients were obtained and the local ethical committee approval was received. Among them, 44 patients were male and 7 female, aged(48±12) years on average. All patients underwent enhanced CT scan before operation and the HCC CT-imaging features were extracted by A.K. software. The image omics features and clinical variables related to the occurrence of PHLF were screened by univariate Logistic regression, and were included in multivariate analysis to obtain the independent risk factors for the occurrence of PHLF. The nomogram for predicting PHLF in HCC patients was established according to the independent risk factors. Results Low intensity small area emphasis(LISAE) and albumin-bilirubin(ALBI) score were the independent factors affecting the occurrence of PHLF(OR=27.93, 15.53; P<0.05). The area under the curve for predicting PHLF by LISAE combined with ALBI score was 0.883, significantly larger than the 0.700 of single ALBI score(Z=-2.460, P<0.05). Nomogram for predicting PHLF in HCC patients was successfully established. The corrected consistency index(C-index) was 0.863. The standard curve was well fitted with the calibrated predictive curve, and the predicted value was highly consistent with the observed value. Conclusions LISAE is an independent factor affecting the occurrence of PHLF in HCC patients, which can improve the predictive value of clinical variable ALBI score. The nomogram for predicting PHLF yields high predictive value for PHLF in HCC patients.
引文
[1]Siegel RL,Miller KD,Jemal A.Cancer Statistics,2017[J].CACancer J Clin,2017,67(1):7-30.
[2]El-Serag HB.Hepatocellular carcinoma[J].N Engl J Med,2011,365(12):1118-1127.
[3]van Mierlo KM,Schaap FG,Dejong CH,et al.Liver resection for cancer:new developments in prediction,prevention and management of postresectional liver failure[J].J Hepatol,2016,65(6):1217-1231.
[4]Rahbari NN,Garden OJ,Padbury R,et al.Posthepatectomy liver failure:a defnition and grading by the International Study Group of Liver Surgery(ISGLS)[J].Surgery,2011,149(5):713-724.
[5]Zou H,Wen Y,Yuan K,et al.Combining albumin-bilirubin score with future liver remnant predicts post-hepatectomy liver failure in HBV-associated HCC patients[J].Liver Int,2018,38(3):494-502.
[6]Lafaro K,Buettner S,Maqsood H,et al.Defining post hepatectomy liver insufficiency:where do we stand?[J].J Gastrointest Surg,2015,19(11):2079-2092.
[7]Reissfelder C,Rahbari NN,Koch M,et al.Postoperative course and clinical significance of biochemical blood tests following hepatic resection[J].Br J Surg,2011,98(6):836-844.
[8]Yokoyama Y,Ebata T,Igami T,et al.Predictive power of prothrombin time and serum total bilirubin for postoperative mortality after major hepatectomy with extrahepatic bile duct resection[J].Surgery,2014,155(3):504-511.
[9]Johnson PJ,Berhane S,Kagebayashi C,et al.Assessment of liver function in patients with hepatocellular carcinoma:a new evidencebased approach-the ALBI grade[J].J Clin Oncol,2015,33(6):550-558.
[10]Fan ST.Liver functional reserve estimation:state of the art and relevance for local treatments:the Eastern perspective[J].J Hepatobiliary Pancreat Sci,2010,17(4):380-384.
[11]Fan ST,Lai EC,Lo CM,et al.Hospital mortality of major hepatectomy for hepatocellular carcinoma associated with cirrhosis[J].Arch Surg,1995,130(2):198-203.
[12]Lam CM,Fan ST,Lo CM,et al.Major hepatectomy for hepatocellular carcinoma in patients with an unsatisfactory indocyanine green clearance test[J].Br J Surg,1999,86(8):1012-1017.
[13]D'Onofrio M,De Robertis R,Demozzi E,et al.Liver volumetry:is imaging reliable?personal experience and review of the literature[J].World J Radiol,2014,6(4):62-71.
[14]Yoon JH,Choi JI,Jeong YY,et al.Pre-treatment estimation of future remnant liver function using gadoxetic acid MRI in patients with HCC[J].J Hepatol,2016,65(6):1155-1162.
[15]Gillies RJ,Kinahan PE,Hricak H.Radiomics:images are more than pictures,they are data[J].Radiology,2016,278(2):563-577.
[16]Li Y,Liu X,Xu K,et al.MRI features can predict EGFR expression in lower grade gliomas:a voxel-based radiomic analysis[J].Eur Radiol,2018,28(1):356-362.
[17]Esteva A,Kuprel B,Novoa RA,et al.Dermatologist-level classification of skin cancer with deep neural networks[J].Nature,2017,542(7639):115-118.
[18]Braman NM,Etesami M,Prasanna P,et al.Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI[J].Breast Cancer Res,2017,19(1):57.
[19]Romero-Gómez M1,Gómez-González E,Madrazo A,et al.Optical analysis of computed tomography images of the liver predicts fibrosis stage and distribution in chronic hepatitis C[J].Hepatology,2008,47(3):810-816.
[20]Barry B,Buch K,Soto JA,et al.Quantifying liver fibrosis through the application of texture analysis to diffusion weighted imaging[J].Magn Reson Imaging,2014,32(1):84-90.
[21]Marrero JA,Kulik LM,Sirlin C,et al.Diagnosis,staging and management of hepatocellular carcinoma:2018 Practice Guidance by the American Association for the Study of Liver Diseases[J].Hepatology,2018,68(2):723-750.
[22]Dindo D,Demartines N,Clavien PA.Classification of surgical complications:a new proposal with evaluation in a cohort of 6 336patients and results of a survey[J].Ann Surg,2004,240(2):205-213.
[23]Cescon M,Colecchia A,Cucchetti A,et al.Value of transient elastography measured with fibroscan in predicting the outcome of hepatic resection for hepatocellular carcinoma[J].Ann Surg,2012,256(5):706-712.
[24]Pak LM,Chakraborty J,Gonen M,et al.Quantitative imaging features and postoperative hepatic insufficiency:a multi-institutional expanded cohort[J].J Am Coll Surg,2018,226(5):835-843.
[25]Toyoda H,Lai PB,O'Beirne J,et al.Long-term impact of liver function on curative therapy for hepatocellular carcinoma:application of the ALBI grade[J].Br J Cancer,2016,114(7):744-750.
[26]Okabe H,Beppu T,Chikamoto A,et al.Remnant liver volume-based predictors of postoperative liver dysfunction after hepatectomy:analysis of 625 consecutive patients from a single institution[J].Int JClin Oncol,2014,19(4):614-621.
[2]El-Serag HB.Hepatocellular carcinoma[J].N Engl J Med,2011,365(12):1118-1127.
[3]van Mierlo KM,Schaap FG,Dejong CH,et al.Liver resection for cancer:new developments in prediction,prevention and management of postresectional liver failure[J].J Hepatol,2016,65(6):1217-1231.
[4]Rahbari NN,Garden OJ,Padbury R,et al.Posthepatectomy liver failure:a defnition and grading by the International Study Group of Liver Surgery(ISGLS)[J].Surgery,2011,149(5):713-724.
[5]Zou H,Wen Y,Yuan K,et al.Combining albumin-bilirubin score with future liver remnant predicts post-hepatectomy liver failure in HBV-associated HCC patients[J].Liver Int,2018,38(3):494-502.
[6]Lafaro K,Buettner S,Maqsood H,et al.Defining post hepatectomy liver insufficiency:where do we stand?[J].J Gastrointest Surg,2015,19(11):2079-2092.
[7]Reissfelder C,Rahbari NN,Koch M,et al.Postoperative course and clinical significance of biochemical blood tests following hepatic resection[J].Br J Surg,2011,98(6):836-844.
[8]Yokoyama Y,Ebata T,Igami T,et al.Predictive power of prothrombin time and serum total bilirubin for postoperative mortality after major hepatectomy with extrahepatic bile duct resection[J].Surgery,2014,155(3):504-511.
[9]Johnson PJ,Berhane S,Kagebayashi C,et al.Assessment of liver function in patients with hepatocellular carcinoma:a new evidencebased approach-the ALBI grade[J].J Clin Oncol,2015,33(6):550-558.
[10]Fan ST.Liver functional reserve estimation:state of the art and relevance for local treatments:the Eastern perspective[J].J Hepatobiliary Pancreat Sci,2010,17(4):380-384.
[11]Fan ST,Lai EC,Lo CM,et al.Hospital mortality of major hepatectomy for hepatocellular carcinoma associated with cirrhosis[J].Arch Surg,1995,130(2):198-203.
[12]Lam CM,Fan ST,Lo CM,et al.Major hepatectomy for hepatocellular carcinoma in patients with an unsatisfactory indocyanine green clearance test[J].Br J Surg,1999,86(8):1012-1017.
[13]D'Onofrio M,De Robertis R,Demozzi E,et al.Liver volumetry:is imaging reliable?personal experience and review of the literature[J].World J Radiol,2014,6(4):62-71.
[14]Yoon JH,Choi JI,Jeong YY,et al.Pre-treatment estimation of future remnant liver function using gadoxetic acid MRI in patients with HCC[J].J Hepatol,2016,65(6):1155-1162.
[15]Gillies RJ,Kinahan PE,Hricak H.Radiomics:images are more than pictures,they are data[J].Radiology,2016,278(2):563-577.
[16]Li Y,Liu X,Xu K,et al.MRI features can predict EGFR expression in lower grade gliomas:a voxel-based radiomic analysis[J].Eur Radiol,2018,28(1):356-362.
[17]Esteva A,Kuprel B,Novoa RA,et al.Dermatologist-level classification of skin cancer with deep neural networks[J].Nature,2017,542(7639):115-118.
[18]Braman NM,Etesami M,Prasanna P,et al.Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI[J].Breast Cancer Res,2017,19(1):57.
[19]Romero-Gómez M1,Gómez-González E,Madrazo A,et al.Optical analysis of computed tomography images of the liver predicts fibrosis stage and distribution in chronic hepatitis C[J].Hepatology,2008,47(3):810-816.
[20]Barry B,Buch K,Soto JA,et al.Quantifying liver fibrosis through the application of texture analysis to diffusion weighted imaging[J].Magn Reson Imaging,2014,32(1):84-90.
[21]Marrero JA,Kulik LM,Sirlin C,et al.Diagnosis,staging and management of hepatocellular carcinoma:2018 Practice Guidance by the American Association for the Study of Liver Diseases[J].Hepatology,2018,68(2):723-750.
[22]Dindo D,Demartines N,Clavien PA.Classification of surgical complications:a new proposal with evaluation in a cohort of 6 336patients and results of a survey[J].Ann Surg,2004,240(2):205-213.
[23]Cescon M,Colecchia A,Cucchetti A,et al.Value of transient elastography measured with fibroscan in predicting the outcome of hepatic resection for hepatocellular carcinoma[J].Ann Surg,2012,256(5):706-712.
[24]Pak LM,Chakraborty J,Gonen M,et al.Quantitative imaging features and postoperative hepatic insufficiency:a multi-institutional expanded cohort[J].J Am Coll Surg,2018,226(5):835-843.
[25]Toyoda H,Lai PB,O'Beirne J,et al.Long-term impact of liver function on curative therapy for hepatocellular carcinoma:application of the ALBI grade[J].Br J Cancer,2016,114(7):744-750.
[26]Okabe H,Beppu T,Chikamoto A,et al.Remnant liver volume-based predictors of postoperative liver dysfunction after hepatectomy:analysis of 625 consecutive patients from a single institution[J].Int JClin Oncol,2014,19(4):614-621.