| Background and objectiveHepatocellular carcinoma(HCC)is one of the most common malignant tumors and the third largest cause of cancer-related death in the world.Due to the concealment of HCC,most of patients were in the middle or advanced stage at the time of diagnosis,without the chance of radical surgery.Transcatheter arterial chemoembolization(TACE)is the most effective treatment for these patients.In recent years,drug-eluting beads(DEB)have been widely used in clinical practice with the advantages of loading rapidly and releasing slowly chemotherapy drugs in local tumor.However,each patient has different tumor response and survival after DEB-TACE because of the different physical condition,liver function,tumor staging and so on.Therefore,there is an urgent need for an effective tool to evaluate the prognosis of DEB-TACE based on the preoperative clinical characteristics,so as to screen HCC patients suitable for DEB-TACE.Currently,there are several staging systems widely applied to assess the outcomes of HCC patients,such as Barcelona Clinic Liver Cancer staging,Okuda staging and China liver cancer staging system.However,these staging system s have some limitations in accurately evaluating the prognosis of patients treated with DEB-TACE.At present,the value of nomogram model in the diagnosis and prognosis of various tumors has been gradually recognized.However,up to now,no such nomogram model has been established for the prognosis of HCC treated with DEB-TACE.In this study,the clinical characteristics of HCC patients undergoing DEB-TACE as initial treatment were analyzed to screen out the factors influencing postoperative tumor response and survival,to develop a nomogram model for predicting 6-month,1-year and 2-year survival rates.In addition,this study also established a prognostic risk stratification system based on the nomogram model,which can provide theoretical foundation for clinicians to accurately evaluate the prognosis of patients,individualize treatment plan and follow-up plan.Materials and methodsThe data of patients with HCC who refused or were unable to undergo surgical resection but DEB-TACE in the First Affiliated Hospital of Zhengzhou University from June 2016 to may 2020 were retrospectively analyzed.The data of preoperative baseline characteristics and postoperative adverse reactions or complications were collected.One month after operation and every three months thereafter,enhanced CT or MRI was performed.The difference between the preoperative and postoperative imaging were compared to evaluate the tumor response according to Modified Response Evaluation Criteria in Solid Tumors(mRECIST),and the survival of each patient was recorded.If the tumor response did not achieve complete release,repeated DEB-TACE was performed.Logistic regression analysis was used to analyze the independent risk factors of complete release(CR)one month after operation.Then 70%of the patients were randomly divided into the training group and the remaining 30%into the validation group.Cox regression model was used to screen the independent prognostic factors of OS in the training group,and a nomogram integrating all prognostic factors was established for the prediction of 6-month,1-year and 2-year survival rates.In the validation group,the discrimination and calibration of the nomogram model were validated,and the accuracy of prediction was compared with Barcelona Clinic Liver Cancer staging,Okuda staging and China liver cancer staging system.In addition,the prognostic risk stratification system was established based on nomogram,and the relationship between tumor response and prognostic risk stratification system was further explored.Result1.Patient characteristics A total of 302 patients with HCC were included in this study,with 212 in the training group and 90 in the validation group.Except for age and ECOG score,there was no significant difference in other factors between the training group and the validation group(P>0.05).2.Operation results and adverse reactions The success rate of DEB-TACE was 100%.The particle size of DEB was 100-300μm in 50.3%patients and 300-500μm in 49.7%patients.The main adverse reaction was postembolism syndrome and 23(7.6%)patients suffered from liver abscess,bile tumor or myelosuppression,which were improved after symptomatic treatment.There was no operation related death.3.Independent influencing factors of CR One month after DEB-TACE,73(24.1%)cases were CR,179(59.2%)were partial release(PR),27(8.9%)were stable disease(SD)and 23(7.6%)were progressive disease(PD),with objective response rate(ORR)of 83.4%and disease control rate(DCR)of 92.4%.Multivariate logistic regression analysis showed that ALP,tumor diameter and presence of tumor capsule were independent influencing factors of CR(P<0.05).4.Independent prognostic factors The median follow-up period was 18.0 months with a range of 1.1 to 46 months.The median survival was 24.9(8.5-42.4)months for all patients,22.5(8.5-42.4)months in the training group and 25.5(7.5-32.5)months in the validation group.Multivariate Cox regression revealed that AFP,AST-to-lymphocyte ratio index(ALRI),Child-Pugh grade,tumor diameter,half range,portal vein invasion and distant metastasis were independent prognostic factors for OS after DEB-TACE in HCC patients(P<0.05).5.Development and validation of nomogram model Based on independent prognostic factors,a nomogram model was established to predict the 6-month,1-year and 2-year survival rates after DEB-TACE.To facilitate the use of nomogram,this study provides a corresponding web calculator(https://jikun.shinyapps.io/deb_tace/).The calibration chart shows a high consistency between the predicted survival and the actual survival of the nomogram model.The C-index of nomogram model in training and validation groups were 0.767 and 0.755 respectively,which were higher than those of BCLC,Okuda and CNLC staging models with a range from 0.557 to 0.676.Moreover,AUC of nomogram model at all time were significantly higher than those of CNLC,Barcelona and Okuda staging models(P<0.05).6.Prognostic risk stratification Based on the total scores from nomogram,the patients were divided into three risk groups:low-risk group(0-117 points),medium-risk group(118-254 points)and high-risk group(255-415 points),with the median survival time of 37.2,15.8 and 7 months respectively.The difference of survival among three risk groups was statistically significant(P<0.001).The higher the risk is,the worse prognosis the patients have.7.Relationship between tumor response and prognosis The median survival time of CR,PR,SD and PD was 37.2,20.5,7.0 and 4.5 months,respectively.There was statistical difference between any two groups(P<0.05).In addition,significant differences existed in CR,ORR or DCR among low-,medium-and high-risk groups(P<0.05).Conclusion1.The nomogram model have superior ability of prognostic evaluation to BCLC,Okuda and CNLC staging models.2.The prognostic risk stratification system based on nomogram model provides theoretical foundation for selection of DEB-TACE indications and formulation of treatment and follow-up plans for HCC patients. |
