Construction And Utilization Of A Survival Prediction Model For Hepatocellular Carcinoma Patients After Liver Transplantation Based On Up-to-seven Criteria

Objective&Background:Hepatocellular Carcinoma?HCC?is one of the most common malignant tumors of digestive tract.The morbidity and mortality of HCC are ranked 5th and 3rd in all malignant tumors,respectively.Due to the lack of specific symptoms and signs in the early stage,most patients have not been diagnosed until advanced stage,thus the prognosis is very poor^[1].According to statistics,the number of patients diagnosed with HCC in China has accounted for more than half of the whole number in the world recently,so that HCC has become a sever threat to the health of Chinese people^[2,3].At present,the main therapy of HCC is still surgical resection of local lesions^[4],and hepatic arterial chemoembolization and radiotherapy are also used as adjuvant treatments preoperatively and postoperatively^[5,6].For patients who cannot be treated with surgery,simple radiotherapy or oral targeted drug sorafenib can be applied^[7].However,the therapy strategies described above cannot work very well for patients with HCC because of its high recurrence rate and metastasis rate^[8].With the development of clinical medicine,liver transplantation has been applied more and more during the treatment of HCC,and HCC has been considered as one of the main indications of liver transplantation as well^[9].After several decades of exploration,the efficacy of liver transplantation in HCC has been analyzed.Currently,there are 4 recognized criteria for liver transplantation in HCC:Up-to-seven criteria?Up7?,UCSF criteria,Milan criteria,and Shanghai Fudan criteria.The prognosis of the HCC patients who meet the criteria and underwent liver transplantation is significantly better than that of patients who do not meet the criteria.Therefore,usually the condition of patients is evaluated before surgery and priority for liver transplantation could be gave to those up to the standards^[10-13].In many previous studies,the above criteria were used to predict the prognosis of HCC patients before liver transplantation,while there are few studies about the influencing factors of the prognosis of the eligible patients after liver transplantation.The application of the Milan standard resulted in a 5-year survival rate of 70%after transplantation and a5-year recurrence rate of less than 15%^[14,15].In recent years,tumor recurrence happened in some patients who meet the above criteria,while some patients just had a small tumor without clear macrovascular invasion,which brings us some new directions for clinical research.Methods:A total of 251 patients undergoing liver transplantation in accordance with the Up7criteria were selected as study subjects.Through the analysis of basic clinical information,a long-term survival model?ATMD?of liver transplantation in HCC based on the Up7 criteria was established.At the same time,the model we built was also verified in UCSF criteria,Milan criteria,and Shanghai Fudan criteria to test its predictive efficacy.It was found that this model also has good predictive efficacy for the survival prognosis of patients meeting each criteria.Results:According to the results of multivariate COX regression analysis,preoperative AFP,preoperative total bilirubin?t-Bil?,microvascular invasion?MVI?,and tumor diameter?Diameter?were independent predictors of long-term survival in HCC patients undergoing liver transplantation.The ATMD model was established and high/low risk groups were defined.Kaplan-Meier analysis revealed that the survival conditions of the HCC patients undergoing liver transplantation meeting different standards had significant differences:Up7 standard?P<0.001?,Milan standard?P<0.001?,UCSF standard?P=0.001?and Shanghai Fudan standard?P=0.008?.The area under the ROC curve for three-year survival predicted by ATMD of HCC patients undergoing liver transplantation meeting the four different standards was 76.63%,69.41%,73.32%,and 75.87%,respectively.Conclusion:The ATMD model has good survival prediction ability for HCC patients meeting the Up7 standard,UCSF criteria,Milan criteria,and Shanghai Fudan criteria.It is of great significance to the preoperative decision-making and postoperative risk assessment of the HCC patients meeting one of the above standards,who underwent liver transplantation.Part 1 Construction of a survival prediction model for hepatocellular carcinoma patients after liver transplantation based on up-to-seven criteriaSelect the cases of patients undergoing liver transplantation in our department for nearly 10 years through the Lianzhong system of our hospital.The first entry condition of the group was the number of tumors and the maximum tumor diameter which in accordance with the Up-7 criteria.The indices of the each patient were collected,including?1?general information,:age,gender,hepatitis history,liver function grade?Child-Pugh score?,splenectomy history,upper gastrointestinal hemorrhage history,oncotherapy history?hepatic carcinectomy,,,microwave ablation of HCC,transcatheter arterial chemoembolization?,whether had antiviral treatment or not;?2?transplantation information:graft survival condition and cross-blood type transplantation condition;?3?iconography features:intraperitoneal varices,HCC feature?"fast in and out"?and tumor number;?4?intraoperative situation:tumor capsule,the amount of ascites,intraoperative hemorrhage volume,intraoperative cold ischemia time,intraoperative anhepatic phase timer;?5?the preoperative peripheral blood test indices,including the number of preoperative platelet,preoperative neutrophile granulocyte,preoperative c-reactive protein,preoperative creatinine,preoperative GGT,preoperative ALT,preoperative AFP;?6?postoperative peripheral blood test indices:the number of postoperative neutrophile granulocyte,postoperative blood platelet and postoperative white blood cell;?7?pathology,results:pathological type,Edmondson classification,MVI,tumor capsule.After integrating the data statistics,using COX regression analysis finding independent predictors,and constructing ATMD model,using software to get cut-off value of the model,dividing the cases which conformed to each criteria standard into high-risk and low-risk groups,verify if there was a difference between the survival curves under each criteria.Part 2According to the first part of the HCC survival prediction ATMD and the condition of postoperative patients taking imunodepressants and using HCC targeted drugs,choosing and collecting again the cases which postoperative pathology indicating microvascular invasion,further divided the above cases into high-risk and low-risk groups according to ATMD model.Verifying Chi-square test,comparing the first year and the second year tumor recurrence rate,which were the group taking Tac and the group taking sirolimus+sorafenib,evaluating if there was a difference between high-risk and low-risk groups under different immune cases,analyzed more appropriate immunosuppressive program for postoperative patients who were at high-risk and low-risk of recurrence.