The Study Of A Prediction Model For Hepatocellular Carcinoma Risk In Chronic Hepatitis B Patients Treated With Antiviral Therapy In Primary Hospitals

Background Hepatocellular carcinoma(HCC)is one of the most common malignant tumors worldwide,with high morbidity and mortality.Persistent infection with hepatitis B virus(HBV)is still the main cause of liver cirrhosis and hepatocellular carcinoma in my country.At present,in order to effectively control chronic hepatitis B(CHB),the guidelines recommend that potent oral nucleotide analogues(Nucleoside analogue,NA)be the first choice,which can inhibit the replication of HBV DNA to the greatest extent and reduce the risk of liver cancer.risk.However,even under effective antiviral therapy,since HBV has not been eradicated,HCC can still occur and is a key factor affecting mortality from liver-related diseases.Therefore,for CHB patients receiving NAs treatment,it is necessary to establish a risk prediction model for HCC occurrence to identify high-risk patients,implement key monitoring,early diagnosis and early intervention,and improve their prognosis.Objective To explore the independent risk factors for the progression of CHB patients to HCC,and establish a 5-year prediction model for HCC occurrence.Provide guidance for the regular monitoring of CHB patients in primary hospitals.Methods The clinical data of CHB patients who received continuous NAs treatment from May 2017 to May 2022 in Yichang Central People’s Hospital were collected retrospectively.Using univariate and multivariate COX regression analysis to screen out the independent risk factors for the progression of CHB patients to HCC,set their weights according to the risk ratio of each independent risk factor,and establish a 5-year risk prediction model for HCC,and use this as a basis According to draw Nomogram diagram.The accuracy of the model was measured by the area under the receiver operating characteristic curve(ROC).Results Analysis of the clinical characteristics of CHB patients and HCC patients showed that PLT and PAB of HCC patients were significantly lower than those of CHB patients,and AST,ALT,GGT,AKP,AFP,PT were significantly higher than those of CHB patients.The proportion of liver cirrhosis in HCC patients was significantly higher than that in CHB patients.RBC,WBC,HB,TBIL,DBIL,ALB,TBA,CA199,CEA,APTT,age,E antigen status had no statistical difference.Univariate COX regression analysis showed that PLT,AST,ALT,PAB,GGT,AKP,AFP,CA199,PT,liver cirrhosis,age at first diagnosis,gender were significantly correlated with the 5-year HCC occurrence(P<0.05).Multivariate COX regression analysis showed that PLT,PAB,AFP,liver cirrhosis,and gender were independent risk factors for the occurrence of HCC in CHB patients within 5 years.The SPPAC model based on gender,prealbumin,platelet count,AFP,and liver cirrhosis has good predictive performance for the occurrence of HCC in 5 years.The AUC of this model is 0.84(95%CI 0.75,0.93).Conclusion 1.Male,low PAB,low PLT,high AFP,and CHB patients with cirrhosis have an increased risk of HCC.Clinicians need to focus on monitoring such patients.2.The variables included in the SPPAC model are all simple and easy-to-obtain clinical indicators,which play an important guiding role in the regular review of CHB patients in primary hospitals and HCC monitoring.3.The SPPAC model constructed based on the above five indicators has good predictive value for the occurrence of liver cancer in CHB patients.Compared with the existing models,it has certain advantages.