Background and Study Aims

The risk of hepatocarcinogenesis depends on background liver factors, of which fibrosis is a major determinant. Serum markers and scores are of increasing importance in non-invasive diagnosis of hepatic fibrosis. Our aim was to predict the occurrence of hepatocellular carcinoma (HCC) using a non-invasive fibrosis score calculated using routine patient data.

Patients and mthods

Our retrospective study included 1,291 hepatitis C related-HCC Egyptian patients (Group 1) recruited from the multidisciplinary HCC clinic, Faculty of Medicine, Cairo University in the period between February 2009 and June 2016 and 1072 chronic hepatitis C-naïve patients (Group 2) with advanced fibrosis (≥F3) and cirrhosis (F4). King score, Fibro Q score, Aspartate aminotransferase-to-platelet ratio index (APRI), AST to ALT ratio (AAR), LOK score, Göteborg University Cirrhosis Index (GUCI), Fibro-α and Biotechnology Research Center (BRC) scores were calculated for all patients. Regression analysis and receiver operating characteristics (ROC) were used to calculate the sensitivity, specificity and predictive values for significant scores with the best cut-off for predicting HCC. A regression equation was used to calculate predicted probabilities of HCC using the following variables; age, gender, haemoglobin, international normalised ratio (INR), albumin and alpha fetoprotein. The appropriate score cut-off points yielding optimal sensitivity and specificity were determined by ROC curve analysis.

Results

There was a highly significant difference between the two groups for all calculated scores (P = 0.0001). Our new score, the Hepatocellular Carcinoma Multidisciplinary Clinic-Cairo University (HMC-CU) score (Logit probability of HCC = − 2.524 + 0.152*age – 0.121*Hb − 0.696*INR – 1.059*Alb + 0.022*AFP + 0.976*Sex. Male = 1, Female = 0), with a cut-off of 0.559 was superior to other scores for predicting HCC, having a sensitivity of 90% and specificity of 80.6%.

Conclusion

The HMC-CU score is a promising, easily calculated, accurate, cost-effective score for HCC prediction in chronic HCV patients with advanced liver fibrosis.

中文翻译：

---

肝细胞癌多学科诊所-开罗大学（HMC-CU）评分：用于诊断HCC的新的简单评分。

背景和学习目标

肝癌发生的风险取决于背景肝脏因素，其中纤维化是主要的决定因素。血清标志物和评分在肝纤维化的非侵入性诊断中越来越重要。我们的目的是通过使用常规患者数据计算出的非侵入性纤维化评分来预测肝细胞癌（HCC）的发生。

患者和方法

我们的回顾性研究包括2009年2月至2016年6月期间从开罗大学医学院多学科HCC诊所招募的1,291例与丙型肝炎相关的HCC埃及患者（第1组）和1072例初治的慢性C型肝炎患者（第2组）患有晚期纤维化（≥F3）和肝硬化（F4）。King评分，Fibro Q评分，天冬氨酸转氨酶与血小板之比指数（APRI），AST与ALT之比（AAR），LOK得分，哥德堡大学肝硬化指数（GUCI），Fibro-α和生物技术研究中心（BRC）得分均为为所有患者计算。回归分析和接收者操作特征（ROC）用于计算有意义分数的敏感性，特异性和预测值，其中最佳临界值可预测HCC。使用回归方程，使用以下变量来计算HCC的预测概率；年龄，性别，血红蛋白，国际标准化比率（INR），白蛋白和甲胎蛋白。通过ROC曲线分析确定产生最佳灵敏度和特异性的合适的分数截止点。

结果

两组之间在所有计算得分上都有极显着差异（P  = 0.0001）。我们的新评分，即开罗肝细胞癌多学科诊所（HMC-CU）评分（HCC的Logit概率= − 2.524 + 0.152 *年龄– 0.121 * Hb − 0.696 * INR – 1.059 * Alb + 0.022 * AFP + 0.976 * Sex男性= 1，女性= 0），对于HCC的预测，临界值为0.559，优于其他得分，其敏感性为90％，特异性为80.6％。

结论

对于患有晚期肝纤维化的慢性HCV患者，HMC-CU评分是一项有前途，易于计算，准确，具有成本效益的HCC预测评分。