Objective:To construct and validate a machine learning model for preoperative prediction of perineural invasion (PNI) status in intrahepatic cholangiocarcinoma (ICC).Methods:Clincial data of 329 patients, including 245 admitted to Zhengzhou University People's Hospital from January 2018 to June 2023 and 84 admitted to the Affiliated Cancer Hospital of Zhengzhou University from January 2013 to January 2020 were retrospectively analyzed. Patients were divided into a training set ( n=231) and a validation set ( n=98). Clinicopathological data including age, gender, hepatitis B virus (HBV) infection status were collected. Predictive variables were determined using least absolute shrinkage and selection operator (LASSO) regression analysis. Six machine learning algorithms including random forest (RF), logistic regression, and linear kernel-based support vector machine were selected to construct the preoperative prediction model for PNI in ICC. Performance metrics of the model were calculated using a confusion matrix, and the final model was selected. The model performance was evaluated in the validation set. Calibration curves were plotted to evaluate the final model, and a Pareto chart was used to visualize the importance of predictive variables. Results:LASSO regression identified nine predictive variables included in the prediction model, including carbohydrate antigen 19-9 (CA19-9), HBV infection status, alkaline phosphatase, alanine aminotransferase, prothrombin time, total bilirubin, albumin, neutrophil times gamma-glutamyl transferase to lymphocyte ratio, and tumor burden score. Among the trained six models, the area under the curve (AUC) of the RF model was 0.909, with a sensitivity of 0.842 and an accuracy of 0.870. Compared with the AUC of the RF model, the AUCs of the other 5 models were lower (all P<0.05). The AUC of the RF model for predicting PNI in ICC in validation set was 0.736. Calibration curves showed good fit of the RF model's prediction of PNI in ICC in both training and validation sets. The Pareto chart showed that CA19-9 was the most important predictive variable in the model, followed by HBV infection status. Conclusion:The machine learning model based on the RF algorithm has a high accuracy in preoperative prediction of PNI status in ICC.

Objective:To analyze the effect of sarcopenia on the prognosis of patients with hepatocellular carcinoma (HCC) after laparoscopic radical resection.Methods:Clinical data of 165 patients with HCC undergoing laparoscopic radical resection in Henan University People's Hospital from January 2018 to December 2021 were retrospectively analyzed, including 122 males and 43 females, aged (55.5±11.4) years. Patients were divided into sarcopenia group ( n=79) and control group (non-sarcopenia, n=86) according to the skeletal muscle index. The survivals were analyzed using the Kaplan-Meier method, and were compared by the log-rank test. Univariate and multivariate Cox regression were utilized to analyze the effect of sarcopenia on the prognosis of HCC after laparoscopic radical surgery. Results:The 1- and 3-year cumulative survival rates of control group were 96.4% and 81.2%, which were higher than those of the sarcopenia group (83.2% and 48.9%, respectively, χ2=19.67, P<0.001). The 1- and 3-year recurrence-free survival (RFS) rates of control group were 88.4% and 66.1%, which were higher than those of sarcopenia group (70.9% and 37.7%, respectively, χ2=18.80, P<0.001). Multivariate Cox regression analysis showed that the risk of recurrence ( HR=1.35, 95% CI: 1.20-1.59, P<0.001) and the risk of death ( HR=2.21, 95% CI: 1.23-3.41, P=0.001) after laparoscopic radical resection for HCC in patients with sarcopenia rises compared to non-sarcopenic patients. Conclusion:Sarcopenia is a risk factor for the survival and recurrence of HCC after laparoscopic radical surgery.

Objective:To establish and validate a preoperative differentiateon model of intrahepatic cholangiocarcinoma (ICC) and combined hepatocellular carcinoma (CHC) based on the inflammatory markers and conventional clinical indicators.Methods:The clinical data of 116 patients with ICC or CHC admitted to Henan Provincial People's Hospital from January 2018 to March 2023 were retrospectively analyzed, including 74 males and 42 females, aged (58.5±9.4) years old. The data of 83 patients were used to establish the differentiation model as the training group, including 50 cases of ICC and 33 cases of CHC. The data of 33 patients were used to validate the model as the validation group, including 20 cases of ICC and 13 cases of CHC. The clinical data including the platelet-to-lymphocyte ratio (PLR), systemic immune inflammation index (SII), prognostic inflammatory index (PII), prognostic nutritional index (PNI), neutrophil-to-lymphocyte ratio (NLR) and lymphocyte-to-monocyte ratio (LMR) were collected and analyzed. The receiver operating characteristic (ROC) curve was used to analyze the best cut-off values of PLR, SII, PII, PNI, NLR and LMR. Univariate and multivariate logistic regression analysis were used to determine the differential factors between ICC and CHC. The R software was used to draw the nomogram, calculate the area under the curve (AUC) to evaluate the model accuracy, and draw the calibration chart and the decision curve to evaluate the predictive efficacy of the model.Results:Univariate logistic regression analysis showed that liver cirrhosis, history of hepatitis, alpha fetoprotein, carbohydrate antigen 199, gamma-glutamyltransferase (GGT), PLR, PNI and inflammation score (IS) could be used to differentiate ICC from CHC (all P<0.05). The indicators identified in univariate analysis were included in multivariate logistic regression analysis. The results showed that absence of liver cirrhosis, GGT>60 U/L, PNI>49.53, and IS<2 indicated the pathology of ICC (all P<0.05). Based on the above four factors, a nomogram model was established to differentiate the ICC and CHC. The AUC of ROC curve of the nomogram model in the training and validation groups were 0.851 (95% CI: 0.769-0.933) and 0.771 (95% CI: 0.594-0.949), respectively. The sensitivities were 0.760 and 0.750, and the specificities were 0.818 and 0.769, respectively. The calibration chart showed that the predicted curve fitted well to the reference line. The decision curve showed that the model has a clear positive net benefit. Conclusion:The nomogram model based on inflammatory markers showed a good differentiation performance of ICC and CHC, which could benefits the individualized treatment.

Objective:A predictive nomogram model for the prognosis of intrahepatic cholangiocarcinoma (ICC) patients after curative resection was constructed based on the albumin-bilirubin score and tumor burden score (ATS) grade, and the predictive performance of the nomogram model was evaluated.Methods:Retrospective analysis of clinical data was made, from ICC patients who underwent curative resection at Zhengzhou University People's Hospital and Zhengzhou University Cancer Hospital from January 2016 to January 2020. A total of 258 patients were included in the study, with 140 males and 118 females, with an average age of (56.5±9.5) years. The 258 ICC patients were randomly divided into a training set ( n=174) and a testing set ( n=84) in a 7∶3 ratio. Single-factor and multi-factor Cox regression analyses were performed to identify prognostic factors for ICC patients of the training set, and then a nomogram model was constructed. The performance of the nomogram model was evaluated by using the concordance index (C-index), calibration curve, and risky decision curve analysis. Results:In the training set, univariate Cox regression analysis indicated that albumin-bilirubin (ALBI), tumor burden score (TBS), carcinoembryonic antigen (CEA), tumor differentitation, lymphvascular invasion and ATS significantly influenced overall survival after radical resection for ICC (all P<0.05). Multifactorial Cox regression analysis revealed that ATS grade, CEA, tumor differentiation, lymphovascular invasion, and AJCC N stage are independent risk factors for the prognosis of ICC patients after curative resection (all P<0.05). Assessment of the postoperative survival prediction model based on multifactorial Cox regression yielded a C-index of 0.775(95% CI: 0.747-0.841) for the training set and 0.731(95% CI: 0.668-0.828) for the testing set. The calibration curves for both the training and testing sets indicated strong predictive capability of the model. Additionally, the risk decision curve also suggested high net benefit of the model. Conclusions:The preoperative ATS grade is an independent factor affecting the survival after ICC radical resection. The nomogram model constructed based on ATS grade demonstrates excellent predictive value for postoperative prognosis in ICC patients.

Objective:To explore the relevant risk factors and prognosis of patients with intrahepatic cholangiocarcinoma (ICC) who experienced recurrence within 6 months after surgeryMethods:This retrospective study included a total of 259 patients with ICC a treated at He'nan Provincial People's Hospital and He'nan Cancer Hospital from Jan 2018 to Jan 2020. The clinical and pathological data ,differences between the group with recurrence within 6 months and the group without recurrence within 6 months were compared using the chi-square test. Logistic regression analysis was used to determine the relevant risk factors for recurrence within 6 months. Kaplan-Meier method was used to construct survival and recurrence curves, and survival rates were calculated.Results:The overall survival and recurrence-free survival of patients in the group with recurrence within 6 months were significantly shorter. CA19-9, tumor longitudinal diameter, microvascular invasion, and neural invasion were identified as independent risk factors for recurrence within 6 months after ICC surgery ( P<0.001). Conclusion:The patient population experiencing recurrence within 6 months after ICC surgery has an extremely poor prognosis and possesses a specific tumor microenvironment. CA19-9, tumor longitudinal diameter, microvascular invasion, and neural invasion were identified as independent risk factors for recurrence within 6 months after ICC surgery.

Objective:To explore the prognostic value of systemic immune-inflammatory index(SII)combined with tumor burden score (TBS) (hereinafter referred to as STS) in patients with intrahepatic cholangiocarcinoma (ICC) after radical resection, and to construct a nomogram model.Methods:The clinical data (including the degree of tumor differentiation, vascular cancer thrombus, and lymph node metastasis, etc.) of 258 ICC patients who received radical resection at People′s Hospital of Zhengzhou University (170 cases, training set) and Cancer Hospital of Zhengzhou University (88 cases, validation set) from January 1, 2016 to January 31, 2020 were retrospectively analyzed and graded by SII, TBS and STS. Multivariate Cox regression analysis were used to identify independent factors affecting the prognosis of patients with ICC. Kaplan-Meier survival curve and receiver operating characteristic curve (ROC) were drawn to evaluate the predictive efficiency of SII, TBS and STS in the overall survival of patients with ICC after radical resection. The nomogram prediction model was constructed and evaluate the performance of nomogram model using consistency index (C-index) and calibration curve.Results:Among 170 ICC patients in the training set, there were 106 cases of SII grade 1 and 64 cases of SII grade 2; 137 cases of TBS grade 1 and 33 cases of TBS grade 2; and 98 cases of STS grade 1, 47 cases of STS grade 2, and 25 cases of STS grade 3. Among 88 ICC patients in the validation set, there were 33 cases of SII grade 1 and 55 cases of SII grade 2; 66 cases of TBS grade 1 and 22 cases of TBS grade 2; and 30 case of STS grade 1, 39 cases of TBS grade 2, and 19 cases of TBS grade 3.The multivariate Cox regression analysis showed that tumor differentiation degree (highly differentiated vs. moderately differentiated HR=0.157, 95% confidence interval(95% CI) 0.045 to 0.546, highly differentiated vs. poorly differentiated HR=0.452, 95% CI 0.273 to 0.750), STS (grade 3 vs. grade 2 HR=1.966, 95% CI 1.148 to 3.469; grade 3 vs. grade 1 HR=1.405, 95% CI 0.890 to 2.216), vascular cancer thrombus ( HR=2.006, 95% CI 1.313 to 3.066), nerve invasion ( HR=1.865, 95% CI 1.221 to 2.850), and lymph node metastasis ( HR=1.802, 95% CI 1.121 to 2.896) were independent influencing factors of overall survival in ICC patients after radical resection (all P<0.05). The Kaplan-Meier survival curve showed that SII, TBS, and STS were independent influencing factors of overall survival in ICC patients (all P<0.05). The results of ROC analysis showed that the areas under the curve of SII, TBS and STS in predicting overall survival of ICC patients after radical resection were 0.566 (95% CI 0.479 to 0.652), 0.585 (95% CI 0.499 to 0.672), and 0.657 (95% CI 0.522 to 0.692), respectively. Tumor differentiation, vascular tumor thrombus, nerve invassion, lymph node metastasis, and STS were included to constract the nomogram model. The C-indexes of the training set and validation set based on the nomogram model were 0.792 (95% CI 0.699 to 0.825) and 0.776 (95% CI 0.716 to 0.833), respectively. The calibration curves of the survival rate of the training set and the validation set were close to the reference lines, and the nomogram model had better predictive ability in both the training set and the validation set. Conclusions:Preoperative STS grading is an effective and practical predictor of overall survival in ICC patients after radical section. Compared with SII and TBS alone, it has better predictive value for the prognosis of patients with ICC.