肝癌血清炎症指标及影像特征与病理分级的相关性研究

Correlation between serum inflammatory indicators, imaging features and the pathological grading of hepatocellular carcinoma

目的:探讨术前血清炎症指标及肿瘤影像特征与肝细胞癌(HCC)病理分级的相关性,分析其对HCC病理分级的诊断效能。方法:搜集150例经术后病理证实并按照Edmondson分级法进行病理分级的HCC患者,回顾性分析其临床及影像资料。记录患者的术前血清炎症指标[碱性磷酸酶/淋巴细胞计数(ALR)、中性粒细胞计数/淋巴细胞计数(NLR)、血小板计数/淋巴细胞计数(PLR)、天冬氨酸转氨酶/血小板计数(APRI)]及肿瘤影像特征(肿瘤长径、肝叶受累范围、瘤内坏死、卫星结节、数量、包膜),将150例患者分为低分化(Ⅲ-Ⅳ级)与中-高分化(Ⅰ-Ⅱ级)两组。利用单因素非参数Mann-Whitney U秩和检验、Pearson卡方检验和Fisher精确检验比较两组患者相关指标的差异,采用Spearman秩相关分析其与不同病理分级间的相关性;绘制ROC曲线,采用曲线下面积(AUC)评估单指标及多指标联合预测肝癌病理分级的效能。结果:ALR、NLR、APRI、长径、包膜、卫星结节、肝叶受累范围、肿瘤数量、瘤内坏死分别与不同病理分级呈正相关,且其诊断肿瘤病理分级为低分化的最佳临界值分别为ALR>63.6(AUC为0.672)、NLR>2.22(AUC为0.666)、APRI>0.19(AUC为0.644)、长径>5.2 cm(AUC为0.810)。多指标联合诊断的AUC为0.825,明显高于上述单一指标。结论:术前血清炎症标志物联合肿瘤影像特征作为HCC病理分级的有效预测指标具有一定可行性,且多指标联合预测病理分级的诊断效能明显优于单一指标。

Objective:This paper aimed to explore the correlation between serum inflammatory indicators combined with tumor imaging features and the pathological grading of hepatocellular carcinoma(HCC),and analyze its diagnostic efficacy for the pathological grading.Methods:Retrospective study was conducted on 150 patients with HCC which confirmed by post-operative pathology, and graded according to Edmondson pathology.The preoperative serum inflammatory indicators [alkaline phosphatase-to-lymphocyte ratio(ALR),neutrophil-to-lymphocyte ratio(NLR),platelet-to-lymphocyte ratio(PLR),and aspartate aminotransferase-to-platelet ratio index(APRI)] and tumor imaging features(tumor maximum diameter, liver lobe involvement range, intra-tumoral necrosis, satellite nodules, tumor number and capsule) of these patients were analyzed.All the patients were divided into two groups: poorly differentiated group(grade Ⅲ-Ⅳ) and moderately to well differentiated group(grade Ⅰ-Ⅱ).Statistical analysis was conducted by using single-factor non-parametric Mann-Whitney U-rank sum test, Pearson Chi-square test and Fisher’s exact test to analyze the differences between two groups.Spearman rank correlation analysis was used to analyze the correlation between these indicators and different pathological grades.Receiver operating characteristic(ROC) curves were drawn to calculate areas under the ROC curve(AUCs) for evaluating the efficacy of single-indicator and multi-indicator combination in predicting the pathological grade of HCC.Results:ALR,NLR,APRI,tumor maximum diameter, capsule, satellite nodules, liver lobe involvement range, tumor number, and intra-tumoral necrosis were positively correlated with different pathological grades.And the ROC curve obtained the best cut-off values for predicting the poorly differentiated pathological grade were >63.6,>2.22,>0.19,and >5.2 cm for ALR(AUC:0.672),NLR(AUC:0.666),APRI(AUC:0.644),and tumor maximum diameter(AUC:0.810),respectively.In addition, the AUC of combining multi-indicators was 0.825,which was significantly higher than the above single index. Conclusion:It is feasible to use preoperative serum inflammatory indicators combined with tumor imaging features as effective predictors for HCC pathological grading, and the combination of multiple indicators for predicting pathological grading is significantly better than single indicator.