Wang, Jiao-Yang BM; Jiang, Yan MD; Liu, Zhi-Peng MD; Yin, Xian-Yu MD; Chen, Zhi-Yu PhD Author Information

BACKGROUND:Early recurrence is the leading cause of death for patients with perihilar cholangiocarcinoma (pCCA) after surgery. Identifying high-risk patients preoperatively is important. This study aimed to construct a preoperative prediction model for the early recurrence of patients with pCCA to facilitate planned treatment with curative resection. METHODS:This study ultimately enrolled 400 patients with pCCA after curative resection in 5 hospitals between 2013 and 2019. They were randomly divided into training (n = 300) and testing groups (n = 100) at a ratio of 3:1. Associated variables were identified via least absolute shrinkage and selection operator (LASSO) regression. Four machine learning models were constructed: support vector machine, random forest (RF), logistic regression, and K-nearest neighbors. The predictive ability of the models was evaluated via receiving operating characteristic (ROC) curves, precision-recall curve (PRC) curves, and decision curve analysis. Kaplan-Meier (K-M) survival curves were drawn for the high-/low-risk population. RESULTS:Five factors: carbohydrate antigen 19-9, tumor size, total bilirubin, hepatic artery invasion, and portal vein invasion, were selected by LASSO regression. In both the training and testing groups, the ROC curve (area under the curve: 0.983 vs 0.952) and the PRC (0.981 vs 0.939) showed that RF was the best. The cutoff value for distinguishing high- and low-risk patients was 0.51. K-M survival curves revealed that in both groups, there was a significant difference in RFS between high- and low-risk patients (P < .001). CONCLUSION:This study used preoperative variables from a large, multicenter database to construct a machine learning model that could effectively predict the early recurrence of pCCA in patients to facilitate planned treatment with curative resection and help clinicians make better treatment decisions.

Background: Cholecystectomy, hepatectomy, and lymphadenectomy are recommended as the curative treatment for resectable gallbladder cancer (GBC). Textbook outcomes in liver surgery (TOLS) is a novel composite measure that has been defined by expert consensus to represent the optimal postoperative course after hepatectomy. This study aimed to determine the incidence of TOLS and the independent predictors associated with TOLS after curative-intent resection in GBC patients. Methods: All consecutive GBC patients who underwent curative-intent resection between 2014 and 2020 were enrolled from a multicenter database from 11 hospitals as the training and the internal testing cohorts, and Southwest Hospital as the external testing cohort. TOLS was defined as no intraoperative grade greater than or equal to 2 incidents, no grade B/C postoperative bile leaks, no postoperative grade B/C liver failure, no 90-day postoperative major morbidity, no 90-day readmission, no 90-day mortality after hospital discharge, and R0 resection. Independent predictors of TOLS were identified using logistic regression and were used to construct the nomogram. The predictive performance was assessed using the area under the curve and calibration curves. Results: TOLS was achieved in 168 patients (54.4%) and 74 patients (57.8%) from the training and internal testing cohorts, and the external testing cohort, respectively. On multivariate analyses, age less than or equal to 70 years, absence of preoperative jaundice (total bilirubin≤3 mg/dl), T1 stage, N0 stage, wedge hepatectomy, and no neoadjuvant therapy were independently associated with TOLS. The nomogram that incorporated these predictors demonstrated excellent calibration and good performance in both the training and external testing cohorts (area under the curve: 0.741 and 0.726). Conclusions: TOLS was only achieved in approximately half of GBC patients treated with curative-intent resection, and the constructed nomogram predicted TOLS accurately.

目的 定义肝门部胆管癌(pCCA)根治性切除术后早期复发的时间,并开发与验证一个预测pCCA根治性切除术后早期复发的模型.方法 回顾性收集2011年1月至2020年1月间我院收治的接受根治性切除的pCCA患者318例.按照3∶1的比例,将患者分为建模组(n=238)与验证组(n=80).建模组和验证组分别用于早期复发的预测与验证.采用建模组数据,通过线性拟合法确定pCCA根治性切除术后早期复发的时间.将建模组和验证组依据早期复发时间,各自分为早期复发组和非早期复发组,Log-rank检验比较两组患者的总生存(OS)时间和无复发生存(RFS)时间.在建模组中,比较两组的基线资料,并通过Logistics回归模型确定影响早期复发发生的独立危险因素.通过R语言制作Nomogram模型实现对早期复发的预测.在建模组和验证组中,通过预测5年OS率的受试者特征曲线下面积(AUC)与校正曲线评估模型的预测性能.结果 pCCA根治性切除术后早期复发的时间为术后20个月.在建模组和验证组中,早期复发组患者的5年OS率和5年RFS率均显著低于非早期复发组(P＜0.05).影响pCCA根治性切除术后发生早期复发的独立危险因素是淋巴结阳性(N1、N2)、糖类抗原19-9＞150 U/L、年龄＞70岁、大血管侵犯、微血管侵犯、肿瘤直径＞5 cm.在建模组与验证组中,校正曲线均提示了模型具有良好的预测性能,预测5年OS的AUC分别为0.759(建模)和0.838(验证).结论 pCCA根治性切除术后患者早期复发定义为术后20个月.发生早期复发的患者其远期生存更差.本研究的模型能够准确预测pCCA根治性切除术后早期复发,为临床术后治疗提供参考.

Background & Aims:Tumor-associated chronic inflammation has been determined to play a crucial role in tumor progression, angiogenesis and immunosuppression. The objective of this study was to assess the prognostic value of the neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) in perihilar cholangiocarcinoma (pCCA) patients following curative resection.Methods:Consecutive pCCA patients following curative resection at 3 Chinese hospitals between 2014 and 2018 were included. The NLR was defined as the ratio of neutrophil count to lymphocyte count. PLR was defined as the ratio of platelet count to lymphocyte count. The optimal cutoff values of preoperative NLR and PLR were determined according to receiver operating characteristic (ROC) curves for the prediction of 1-year overall survival (OS), and all patients were divided into high- and low-risk groups. Kaplan-Meier curves and Cox regression models were used to investigate the relationship between values of NLR and PLR and values of OS and recurrence-free survival (RFS) in pCCA patients. The usefulness of NLR and PLR in predicting OS and RFS was evaluated by time-dependent ROC curves.Results:A total of 333 patients were included. According to the ROC curve for the prediction of 1-year OS, the optimal cutoff values of preoperative NLR and PLR were 1.68 and 113.1, respectively, and all patients were divided into high- and low-risk groups. The 5-year survival rates in the low-NLR (<1.68) and low-PLR groups (<113.1) were 30.1% and 29.4%, respectively, which were significantly higher than the rates of 14.9% and 3.3% in the high-NLR group (≥1.68) and high-PLR group (≥113.1), respectively. In multivariate analysis, high NLR and high PLR were independently associated with poor OS and RFS for pCCA patients. The time-dependent ROC curve revealed that both NLR and PLR were ideally useful in predicting OS and RFS for pCCA patients.Conclusions:This study found that both NLR and PLR could be used to effectively predict long-term survival in patients with pCCA who underwent curative resection.