Department of Surgery, Graduate School of Medicine, Kyoto University, Kyoto, Japan.
National Institute of Radiological Sciences, National Institutes of Quantum and Radiological Science and Technology, Chiba, Japan.
Ann Surg Oncol. 2018 Feb;25(2):542-549. doi: 10.1245/s10434-017-6262-z. Epub 2017 Nov 22.
Existing prognostic systems were not developed using only objective variables available preoperatively, and therefore do not provide ideal prognostication for patients undergoing hepatectomy for hepatocellular carcinoma (HCC). We aimed to develop a preoperative prognostic model using objective variables involving two parameters: F-fluorodeoxyglucose-positron emission tomography (F-FDG-PET) and the albumin-bilirubin (ALBI) grade.
This study included 207 consecutive patients with solitary HCC who underwent F-FDG-PET prior to hepatectomy. The tumor to non-tumor maximum standardized uptake value ratio (TNR) was used as an F-FDG PET imaging parameter.
The 5-year overall survival (OS) and disease-free survival (DFS) rates were 58.6% and 28.8%, respectively. Multivariate analysis of OS identified TNR ≥ 2 (hazard ratio [HR] 1.743, 95% confidence interval [CI] 1.114-2.648, p = 0.016) and ALBI grade 2 (HR 1.966, 95% CI 1.349-2.884, p < 0.001) as the only significant prognostic factors; tumor diameter and tumor markers were not significant. Patients were divided into low- (TNR < 2 and ALBI grade 1), intermediate- (TNR < 2 and ALBI grade 2, or TNR ≥ 2 and ALBI grade 1), and high-risk (TNR ≥ 2 and ALBI grade 2) groups, which differed significantly in terms of survival (5-year OS: 75.7, 49.6, and 27.3%, respectively, p < 0.001; 5-year DFS: 37.0, 24.9, and 13.6%, respectively, p < 0.001). Compared with other staging systems, our model had the best discriminatory ability (corrected Akaike information criteria 1054.8, p < 0.001) and homogeneity (likelihood ratio Chi square value 27.6, p < 0.001).
A preoperative prognostic model incorporating F-FDG-PET imaging with the ALBI grade may be useful for estimating the prognosis of selected patients with solitary HCC.
现有的预后系统并非仅使用术前可获得的客观变量进行开发，因此无法为接受肝细胞癌（HCC）肝切除术的患者提供理想的预后预测。我们旨在开发一种使用客观变量的术前预后模型，该模型涉及两个参数：氟代脱氧葡萄糖-正电子发射断层扫描（F-FDG-PET）和白蛋白-胆红素（ALBI）分级。
本研究纳入了 207 例接受肝切除术的单发 HCC 患者，这些患者在术前均接受了 F-FDG-PET 检查。肿瘤与非肿瘤最大标准化摄取值比（TNR）被用作 F-FDG PET 成像参数。
5 年总生存率（OS）和无病生存率（DFS）分别为 58.6%和 28.8%。OS 的多变量分析确定 TNR≥2（风险比[HR] 1.743，95%置信区间[CI] 1.114-2.648，p=0.016）和 ALBI 分级 2（HR 1.966，95%CI 1.349-2.884，p<0.001）为唯一显著的预后因素；肿瘤直径和肿瘤标志物无统计学意义。患者被分为低风险组（TNR<2 和 ALBI 分级 1）、中风险组（TNR<2 和 ALBI 分级 2，或 TNR≥2 和 ALBI 分级 1）和高风险组（TNR≥2 和 ALBI 分级 2），三组患者在生存方面存在显著差异（5 年 OS：分别为 75.7%、49.6%和 27.3%，p<0.001；5 年 DFS：分别为 37.0%、24.9%和 13.6%，p<0.001）。与其他分期系统相比，我们的模型具有最佳的区分能力（校正后的赤池信息量准则 1054.8，p<0.001）和同质性（似然比卡方值 27.6，p<0.001）。
纳入 F-FDG-PET 成像与 ALBI 分级的术前预后模型可能有助于估计选定的单发 HCC 患者的预后。
