| Background and Objective:Primary liver cancer(PLC) is one of the world’s most malignant tumor, and hepatocellular carcinoma(HCC) is the most common subtype. In recent years, the incidence and mortality of HCC are very high, especially in Asian. Of all treatments, surgery liver resection is still the first choice. Although the survival of HCC has improved as the improvement of surgical technique and perioperative managements, the long-term prognosis of postoperative HCC patients is still unsatisfactory due to high distant metastasis and recurrence rate. So far, there are many studies reported about postoperative prognosis factors, but there is no unified preoperative evaluation standard. The aim of this study is to analyze the prognostic factors related to survival rate of primary HCC after radical resection, establish a new prognostic model to judge the prognosis of the HCC patients and to provide theory for individualized treatment.Methods 1. The clinicopathological data of 141 patients who were from the first Hospital Affiliated to Dalian Medical University from January 1, 2001 to March 1, 2010 were retrospectively collected and analyzed. All of them underwent radical hepatic resection and were confirmed to be hepatocellular carcinoma by liver pathology. They were reviewed with regard to their clinicopathological data including 27 items that may be related to the prognosis of liver cancer, i.e. age, gender, liver cirrhosis, tumor size, tumor capsule, NLR( Neutrophil: Lymphocyte Ratio),AFP serum level, liver function, liver Child-Pugh grading and tumor clinical TNM staging et al.2. Kaplan-Meier univariate analyses were applied. Survival curves were compared between groups using log-rank method, so factors that affecting overall survival were concluded(p<0.05).Postoperative independent prognostic factors were examined by the Cox multivariate analysis and then the Cox proportional hazard model was established. 3. The receiver operating characteristic(ROC) and the area under the ROC(AUC)for preliminary evaluation of the model’s ability to estimate 3-year and 5-year survival outcome were used. 4. We applied independent samples of 21 cases to validate the model’s ability to predict prognosis of HCC, evaluated the consistency between predicted results and the actual situation by calculating the accuracy and Kappa consistency coefficient.Results 1. The median survival time is 34 months, and 1-year,3-year,5-year overall survival rates were 93.2%,80% and 68.9%,respectively. 2. Univariate analysis showed that AFP serum level, tumor size, tumor capsule, liver cirrhosis, NLR, total bilirubin(TBIL) were correlated significantly with patients’ overall survival(p<0.05),while gender, age, blood style, smoking history, diabetes mellitus history, HBV infection, liver Child-Pugh grading, glutamic oxalacetic transaminase(AST), glutamic-pyruvictransaminase(ALT), albumin(ALB), γ-glutamyl-transferase(γ-GT), alkaline-phosphatase(ALP), prothrombin-time(PT), INR, CEA, platelate(PLT), tumor number, tumor location, tumor clinical TNM staging, portal vein tumor thrombosis, and blood transfusion were not relevant to overall survival of the patients after curative resection(p>0.05). 3.Cox multivariate analysis worked out that the independent prognostic factors were AFP serum level, liver cirrhosis, tumor size, tumor capsule, and NLR. Prognosis mathematical model was: Prognostic Index(PI)=1.725*liver cirrhosis+0.783*NLR +1.046*AFP+0.595*tumor size-0.811*tumor capsule. 4. According to quartiles of PI, i.e. 3.933(25%),4.716(50%),5.195(75%), all thepatients were divided into four groups: low risk group(PI﹤3.933), moderate-risk group(3.933≤PI<4.716), high-risk group(4. 761≤PI<5.195), and very high risk group(PI≥5.195). Median survival time were 60,34,32,20 months, respectively by using Kaplan-Meier, and 1, 2,3, 5- year cumulative survival rate were 100%,96 %, 96 %,86%;89%,75%,68%,68%;77%,68%,57%,44%;50%,34%,29%,29%, separately. The prognostic difference of the four groups was statistically significant(p=0.000). 5. It was regarded PI as examine variables and 3, 5 years survival outcomes(survival or death) as the state variables. The AUC were 0.726(95% confidence index: 0.637-0.816),0.770(95% confidence index: 0.672-0.868), respectively. 6. The outcomes from the prognostic model had good consistency with the actual result. The total accuracy was 80.9%,and Kappa consistency coefficient was 0.571(p = 0.009).Conclusion: The prognosis of patients with HCC after radical liver resection depends on many factors. The mathematical model of prognosis based on liver cirrhosis degree, AFP serum level, NLR, tumor size, tumor capsule(PI=1.725*liver cirrhosis +0.783* NLR +1.046*AFP+0.595*tumor size-0.811*tumor capsule). The higher PI, the lower postoperative cumulative survival rate and the worse prognosis. This model can be used as an effective method to assess the prognosis of HCC patients after resection. |
PDF Full Text Request