【作者】 苏永杰；

【导师】 董家鸿；

【作者基本信息】 第三军医大学 ， 外科学， 2008， 博士

【摘要】 一、背景与目的对于早期HCC的治疗,目前学术界公认根治性切除是其最佳选择,而根治性切除术的术式有两种:解剖性肝切除术和非解剖性肝切除术。解剖性肝切除术是指:按照肝脏Couinaud分段进行肝切除术,包括单肝段切除及多肝段切除;解剖性肝切除术一般不需要Pringle法阻断第一肝门,关键是解剖出进入肝段的肝蒂和出肝静脉,分别予以结扎然后切除肝段。而非解剖性肝切除术定义为:完整切除肿瘤不考虑肝内解剖,一般要求切缘（RM）至少1.0cm。非解剖性肝切除术一般需要Pringle法阻断第一肝门或者半肝阻断,也可以不阻断第一肝门。目前国内外关于HCC手术方式对预后影响仍无定论。从循证医学的标准界定已经发表的国内外文献临床证据级别是B级或C级,未发现A级的临床证据。为了研究术式对HCC预后的影响,最佳的方法是应用临床随机对照研究（RCT）,比较解剖性肝切除与非解剖性肝切除哪种术式更能减少术后复发,提高HCC术后无瘤生存率和累积生存率,从而为HCC根治性切除术的术式选择提供A级的证据。大病理切片可以清楚的显示瘤体结构、交界区结构和“正常组织”区的连续过程。用于研究微转移的分布、数目、距离瘤体的距离等,较普通的病理切片有明显的优点。本研究沿在肝内门静脉血流方向的瘤体最大直径所在切面切开标本,行大病理切片检查,研究HCC微转移灶的数目、类型和转移距离及癌细胞Edmondson分级,为术式的选择提供病理学的证据。血源性转移是HCC转移复发的主要方式。目前AFP mRNA被认为是比较有价值的循环HCC细胞标志物。实时荧光定量RT-PCR技术以其高效性、高特异性、高敏感性及计量准确等优点,成为目前检测CTCs的理想工具。手术可以刺激HCC细胞入外周血循环,因此收集HCC患者手术前后外周静脉血,提取外周血单核细胞（PBMCs,包含CTCs）,采用实时荧光定量RT-PCR技术,检测外周血AFP mRNA手术前后的变化,估计两种术式外周血中CTCs数量变化是否有差异,为术式的选择提供细胞学的证据。二、方法与结果1.根据术前制定的患者纳入和排除标准,按照随机化的要求,每一个入组的HCC患者行相应的肝脏根治性切除术,收集临床、手术、病理资料,每2月随访1次,测定肝功能、血AFP和肝脏超声检查,每6个月检查胸片1次,怀疑复发时行肝脏或肺部CT检查。根据制定的判断复发标准界定有无复发,记录是否生存。结合随访信息,分析术式不同,术后复发率、术后死亡率、手术并发症的发生率是否不同;并采用Kaplan-Meier法生存分析,Log-rank法检验两组之间的术后早期复发率、术后早期无瘤生存率、术后早期累积生存率的生存曲线总体水平有无差异。本研究还分析了与预后有关的其他因素。本研究自2006年1月1日起到2007年6月30日止,共纳入就诊于重庆西南医院肝胆外科研究所符合要求病例133例,其中解剖性肝切除67例,非解剖性肝切除66例。随访截至到2008年3月31日,中位随访时间是19个月（8～27个月）,其中失访3例。两组的性别、年龄、HBV感染、Child-Pugh分级、ICGR-15、血AFP、术前肝功能、肿瘤直径、肿瘤包膜、肿瘤数目、HCC分级、肝硬化、TNM分级等一般情况和病理检查结果无统计学差异,说明两组具有可比性。术后分析的有关变量包括:两组手术切缘、手术出血量、围手术期输血量、术后肝功能、手术时间、住院时间、手术并发症、复发率、死亡率、无瘤生存和累积生存。解剖组术中出血量明显低于非解剖组,分别是（744.8±539.0）ml和（952.3±634.1）ml（p=0.044）;两组围手术期输血量的中位数分别是0（0-5100）ml和600（0-4660）ml,解剖组明显低于非解剖组（p=0.035）。术后1天和5天的血ALT和TBIL值,解剖组也明显低于非解剖组。每组各有1例患者死于手术并发症。两组术后共46（34.6%, 46/133）人发生手术并发症,共发生57人次,其中解剖组19（28.4%, 19/67）人共21人次;非解剖组27（40.9%, 27/66）人共36人次。发生人数比例经过X2检验,两组无统计学差异（X²=1.658, p=0.198）;但发生人次比例经过X2检验,两组有明显统计学差异（X²=6.045, p=0.014）,说明解剖性肝切除术更能减少手术并发症的发生。两组共有43例发生复发或转移,占32.3%（43/133）,其中肝内复发33例,占24.8%（33/133）;肺部、椎骨或后腹膜等肝外转移10例占7.5%（10/133）。解剖组复发转移15例,占本组的22.4%（15/67）;而非解剖组复发转移28例,占本组的42.4%（28/66）,经X2检验,有明显统计学差异（X2=5.219,p=0.022）。共死亡25人,占18.8%（25/133）,包括解剖组1例围手术期死于肝功能衰竭,非解剖组1例围手术期死于上消化道大出血。解剖组死亡10例,占本组的14.9%（10/67）;而非解剖组死亡15例,占本组的22.7%（15/66）,经X²检验,无统计学差异（X2=0.864, p=0.353）。Kaplan-Meier法分析解剖组和非解剖组术后1年累积复发率是分别22.9%和41.8%,Log-rank法检验两组之间生存曲线总体水平有显著性差异（X2=5.614, p=0.018）;解剖组和非解剖组术后1年无瘤生存率分别是75.5%和48.2%,总体水平有显著性差异（X2=8.601, p=0.003）;解剖组和非解剖组术后1年累积生存率分别是86.8%和78.6%,总体水平无显著性差异（X²=1.184, p=0.277）。结合单因素分析结果,COX回归模型多因素分析表明:HCC术后复发与手术方式、微转移、失血量、术前外周血AFP mRNA和肿瘤直径统计学明显相关。结合单因素分析结果,COX回归模型多因素分析表明:HCC术后早期无瘤生存率与手术方式、微转移和肿瘤边界是否清楚等3因素统计学明显相关。2.观察并记录HCC切除标本肿瘤边界是否清楚、有无肿瘤包膜、有无肉眼转移灶,测量肿瘤直径、切缘距离、标本长度,标本固定后做成大病理切片,光学显微镜观察微转移数目、类型、测定微转移的距离及肿瘤细胞分级。运用组织等比回缩规律,换算出新鲜活体标本的微转移距离。本研究微转移分为两种:微癌栓和微卫星灶。本研究133例患者都进行了大病理切片检查。54.9% （73/133）患者的标本中发现微转移灶。在73例大标本肿瘤外肝组织内共检出了136个微转移:其中13.2%（18/136）为微卫星灶,86.8%（118/136）为微癌栓。在这136个微转移中,61.0%（83/136）为门静脉癌栓,15.4%（21/136）为肝静脉癌栓,还有10.3%（14/136）的癌栓难以分辨是门静脉还是肝静脉。肿瘤扩散距离的中位数是3.6mm,范围是1.1mm～20.2mm。肝内有无微转移与肿瘤大小、包膜完整性、TNM分期、癌细胞分化程度的相关性有统计学意义;与血清AFP浓度、肿瘤数目的相关性无统计学意义。解剖组共有39例（58.2%）患者发现微转移,共发现微转移灶78个,其中门静脉癌栓51个,肝静脉癌栓9个,其他癌栓8个,卫星灶10个。非解剖组共有34（51.5%）例患者发现微转移,共发现微转移灶58个,其中门静脉癌栓32个,肝静脉癌栓12个,其他癌栓6个,卫星灶8个。两组发现微转移病例数经X2检验无统计学差异（X²=0.935,p=0.334）。两组微转移总数和门静脉癌栓数经过Mann-Whitney U检验有统计学差异（p=0.043和p=0.000）。Kaplan-Meier法分析无微转移组和有微转移组术后1年复发率是分别19.7%和43.1%,Log-rank法检验两组之间生存曲线总体水平有显著性差异（X²=8.037, p=0.005）;无微转移组和有微转移组术后1年无瘤生存率分别是76.0%和49.4%,总体水平有显著性差异（X²=10.996,p=0.001）;无微转移组和有微转移组术后1年累积生存率分别是88.5%和78.3%,总体水平有显著性差异（X²=4.579, p=0.032）。3.分别在术前和术后抽取133患者的5ml外周静脉血,肝素抗凝,收集PBMCs（包含CTCs）,提取PBMCc的总RNA,利用实时荧光定量RT-PCR技术,检测外周血AFP mRNA在手术前后的变化。我们还检测了8例肝硬化、4例肝增生、8例肝血管瘤和10例健康志愿者外周血AFP mRNA的含量作为对照。研究表明实时荧光定量RT-PCR检测外周血微转移癌细胞的敏感性可以达1/10⁶-10⁷。HCC组的外周血AFP mRNA相对表达量中位数是3.97×10^-3(9.1×10^-7～6.52×10^-1),明显高于肝硬化组、肝血管瘤组、肝脏增生组和健康对照组,差异有统计学意义（p=0.000）。HCC患者术后外周血AFP mRNA相对表达量的中位数是:1.673×10^-2(范围8.0×10^-7～5.170×10^-1),明显高于术前3.974×10^-3(范围9.0×10^-7～6.518×10^-1)的水平（p=0.000）。经过统计学分析,两组手术前外周血AFP mRNA水平无统计学差异。非解剖组术后外周血AFP mRNA水平的中位数是2.24×10^-2 (范围4.3×10^-7～5.10×10^-1),而解剖组是1.35×10^-2 (范围1.32×10^-6～4.17×10^-1),非解剖组高于解剖组,但是无统计学差异（p=0.092）。HCC组术前AFP mRNA水平相对表达量与TNM分期、微转移、肿瘤大小有明显统计学意义（p=0.003, p=0.002和p=0.000）,与肿瘤数目、包膜完整性均、癌细胞分化程度与血清AFP浓度相关性无统计学意义。外周血术前后AFP mRNA相对表达量与术后早期复发、术后无瘤生存和术后累积生存相关性有明显统计学意义。Kaplan-Meier法分析术前AFP mRNA表达低组和表达高组术后1年复发率是分别20.7%和44.7%,Log-rank法检验两组之间生存曲线总体水平有显著性差异（X2=8.868,p=0.003）;术后AFP mRNA表达低组和表达高组术后1年累积复发率是分别23.4%和43.3%,总体水平有显著性差异（X²=5.888,p=0.015）。三、结论1.与非解剖性肝切除相比较,解剖性肝切除可以降低术后早期复发率,提高术后早期无瘤生存率,其主要原因是:HCC的微转移灶以门静脉癌栓为主,HCC解剖性肝切除能清除更多的微转移灶。2.与非解剖性肝切除相比较,解剖性肝切除可以减少术中出血,降低肝功能损害,减少手术并发症的发生。3. HCC早期复发还与术中出血、微转移、术前外周血AFP mRNA和肿瘤直径密切相关。4.实时荧光定量RT-PCR法检测HCC患者外周血AFP mRNA表达量有很高的敏感性,可以反映HCC患者血中是否存在肝癌细胞。手术可刺激HCC细胞脱落入血。监测术前外周血AFP mRNA水平对预测HCC术后早期复发转移有价值更多还原

【Abstract】 Introduction and ObjectiveCurrently, curative hepatectomy remains the best option for early hepatocellular carcinoma（HCC）.However, there are two types hepatectomy for HCC, one is anatomical hepatectomy and the other is non-anatomical hepatectomy. The anatomical hepatectomy（A group） is the liver segments and lobes resection according to Couinaud’s views. The anatomic hepatectomy general isn`t need block the first hepatic hilum by Pringle method, and request dissect the hepatic-stem and the hepatic vein, then deligation and shearing them. While non- anatomical resection （B group） is the liver resection of a tumor without regard to segmental or lobar anatomy with resection margin at least 1.0cm. Non-anatomical resection usually need block hepatic blood flow by Pringle method, occlusion of hemihepatic blood flow or without occlusion the blood flow. Currently, there wasn`t final conclusion about the prognostic effect of anatomical versus non-anatomical resection for HCC. The standards of evidence based medicine of the published documents were B grade evidence or C grade evidence, without A grade evidence. To research this question, the best method is randomized controlled trial. So we designed this RCT and wanted to reveal which hepatectomy could decrease the recurrence rate, elevate the disease-free survival rate and the overall survival rate.Pathologic large slice can be observed the process of tumour cell and constitution, junctional zone cell and constitution, normal tissue. It can be used to study the distribution, number, and distance of micrometastasis. The liver specimen go along the portal venous flow and the tumour max diameter location cross section were slivered, then manufacture the pathologic large slice of operation specimen. We studied the number,type and distance of micrometastasis, wanted to offer the pathologic evidence for operation type.Hematogenous metastasis is the main metastasis mode of HCC. Currently, AFP mRNA was considered the useful marker of circulating tumor HCC cells. Real time fluorescent quantitation RT-PCR is the ideal method to detect the CTCs because of high performance, high specificity, high sensitivity. 5ml peripheral blood before and after hepatectomy for HCC were collected, then peripheral blood mononuclear cells （PBMCs, contain the CTCs） were extracted, and the AFP mRNA was detected by real-time fluorescent quantitative RT-PCR analysis. To compare the level change of AFP mRNA, we want to offer the cytological evidence for operation type.Methods and Results1. According to the including criteria, excluding criteria and randomization procedure, every HCC patient was done corresponding curative hepatectomy, collected the clinical, operation and pathology datas. All patients were followed with liver biochemistry, AFP serum samples and abdominal ultrasonography every 2 months and chest X-ray every 6 months. A CT scan was performed when a recurrence or metastasis was suspected.The patients were judged whether recurrence according to the recurrence criteria, recorded the death, then analysised whether the recurrence rate, death rate, and operation complication rate were different according to different operation type. Recurrence rates, disease-free survival rate and overall survival rates were evaluated using the Kaplan-Meier method and compared using the Log-rank test. Other factors were analysised about prognosis also.As well as this survey, an analysis of patients with HCC who underwent curative hepatic resection at Institute of Hepatobiliary Surgery, Southwest Hospital between January 2006 and June 2007 was carried out. There were 133 patients were enrolled in this study. Anatomical resection was performed in 67 patients （A group） and non-anatomical resection（ B group） in 66 patients. The follow-up by time was March 2008, and median follow-up period after the surgery was 19 months （range, 8–27 months）.The following parameters were compared in two groups: patient age, sex, Child-Pugh classification, serum HBsAg, ICGR-15, serum AFP, hepatic function, number and size of tumors, Edmondson grade, liver cirrhosis and TNM grade, and no significant differences were found between A and B groups in terms of the clinicopathologic findings. Postoperative analysis of variables including tumor-free resection margin, operation hemorrhage, perioperative blood transfusion, post-operation liver function, operation time, length of hospital stay, number and type of complications, recurrence rate, death rate, disease-free survival rates, and overall survival rates. The operation hemorrhage of A group（744.8±539.0 ml）is less than that of B group（952.3±634.1 ml）（p=0.044）, and the perioperative blood transfusion of A group is less than that of B group（p=0.035）. The ALT and TBIL level of A group are lower than those of B group at the 3th-, 5th- post operation day. A case operation-related death occurred in each group. The numbers of complication and man-time complication in two group are 46 and 57 respectively. The complication rate was higher in B group（40.9%,27/66） than in A group （28.4%,19/67）, but the difference was not significant（p=0.198）. While the man-time complication rate was higher significant in B group（54.5%,36/66） than in A group （31.3%,21/67） （p=0.014）. Altogether, 43 （32.3%,43/133） patients experienced tumor recurrence: 33（24.8%, 33/133） had an intrahepatic recurrence and 10（7.5%, 10/133） had an extrahepatic recurrence. The recurrence and metastasis rate was higher significant in B group（42.4%,28/66） than in A group （22.4%,15/67）（p=0.022）. The mortality was higher in B group （22.7%,15/66） than in A group （14.9%,10/67）, but the difference was not significant（p=0.353）.By Kaplan-Meier survival analysis, the 1-year recurrence rates were 22.9% in A group and 41.8% in B group, respectively （p=0.018）. The 1-year disease-free survival rates were 75.5% and 48.2%, respectively （p=0.003）. The corresponding 1-year overall survival rates were 86.8% and 78.6% （p=0.277）. According to univariate factor analysis, we performed the multivariate regression analysis with the Cox proportional hazard model, and multivariate analysis identified five factors （anatomic resection , micrometastasis, blood lost volume>600ml, AFP mRNA level before operation>150.0 and tumor diameter） as significantly influencing the recurrence rate, and three factors （anatomic resection , micrometastasis and tumor bouncary fuzziness ） as significantly influencing the disease-free survival rate. Anatomic resection was confirmed to be an independent favorable factor for disease-free survival.2. The boundary, tumor amicula and macroscopic metastasis of the resection specimens were observed and recorded, and tumour diameter, RM, specimens length were measured. After the specimens were fixed, they were manufactured large pathologic slices. The number, type and distance of micrometastasises and Edmondson grade were observed by light microscope. The distances of micrometastasises on unfixed liver resection specimen were calculated again by the shrinkage rate of the fixed specimen from each specimen on the pathologic section. In this study, micrometastasis was identified as two types: （1） microscopic tumor thrombus and （2） tumor satellite micronodules.In total, 133 patients were studied and their liver resection specimens were manufactured large pathologic slices. We observed 54.9% （73/133） resection specimens had micrometastases and found 136 micrometastases. There were 13.2% （18/136） tumor satellite micronodules and 86.8% （118/136） microscopic tumor thrombus. There were 61.0% （83/136） microscopic portal vein tumor thrombus, 15.4% （21/136） microscopic liver vein tumor thrombus and 10.3% （14/136） microscopic tumor thrombus difficulty identify. The distances median of micrometastasises was 3.6mm and range was （1.1mm-20.2mm）.The intrahepatic micrometastases were correlated with tumour diameter, amicula integrity, pTNM stage and Edmondson grade, not correlated with tumor numbers and serum AFP. Altogether, 58.2% （39/67） A group patients had micrometastases and 78 micrometastases were found: 51 microscopic portal vein tumor thrombus, 10 tumor satellite micronodules, 9 microscopic liver vein tumor thrombus and 8 microscopic tumor thrombus difficulty identify. While 51.5% （34/66） B group patients had micrometastases and 58 micrometastases were found: 32 microscopic portal vein tumor thrombus, 8 tumor satellite micronodules, 12 microscopic liver vein tumor thrombus and 6 microscopic tumor thrombus difficulty identify. The micrometastases rate was no significant in A group （58.2%, 39/67） and in B group （51.5%, 34/66） （p=0.334）. The numbers of micrometastases and microscopic portal vein tumor thrombus in A group were higher than those in B group with Mann-Whitney U test（p=0.043, p=0.000）. By Kaplan-Meier survival analysis, the 1-year recurrence rates were 19.7% in no micrometastases group and 43.1% in micrometastases group, respectively （p=0.005）. The 1-year disease-free survival rates were 76.0% and 49.4%, respectively （p=0.001）. The corresponding 1-year overall survival rates were 88.5% and 78.3% （P = 0.032）.3. The 5ml peripheral vein blood from 133 HCC patients were collected before and after hepatectomy into heparin tubes, then the peripheral blood mononuclear cells （PBMCs, contain the CTCs） were extracted, the all RNA was extracted and the AFP mRNA was detected by real-time fluorescent quantitative RT-PCR analysis. The peripheral vein blood from 8 patients with hepatitis B and cirrhosis, 4 patients with hyperplasia, 8 patients with hepatic hemangioma, 10 healthy volunteers were collected and AFP mRNA were detected as control groups.The sensitivity was determined theoretically at 1 AFP mRNA positive cell in approximately to 10⁶-10⁷ of mononuclear blood cells.The median data of AFP mRNA level of HCC was 3.97×10^-3(9.1×10^-7～6.52×10^-1), and higher significant than those of control groups（p=0.000）. The median data of AFP mRNA level after hepatectomy was (1.673×10^-2, range: 8.0×10^-7～5.170×10^-1), and higher significant than that of before hepatectomy (3.974×10^-3, range: 9.0×10^-7～6.518×10^-1) （p=0.000）. The level of AFP mRNA was no different between A group and B group before hepatectomy. The AFP mRNA level was higher in the B group (median, 2.24×10^-2, range: 4.3×10^-6～5.10×10^-1)than in A group (median, 1.35×10^-2, range:1.32×10^-6～4.17×10^-1), but the difference was not significant （p=0.092）. The AFP mRNA level before hepatectomy was correlated with tumour diameter, micrometastasis and pTNM stage, and not correlated with tumor numbers, serum AFP, amicula integrity, and Edmondson grade. The AFP mRNA level before and after hepatectomy were correlated with recurrence, disease-free survival and overall survival. By Kaplan-Meier survival analysis, the 1-year recurrence rates were 20.7% in the low AFP mRNA level before hepatectomy group and 44.7% in the high AFP mRNA level before hepatectomy group, respectively （p=0.003）. By Kaplan-Meier survival analysis, the 1-year recurrence rates were 23.4% in the low AFP mRNA level after hepatectomy group and 43.3% in the high AFP mRNA level after hepatectomy group, respectively （p=0.015）.Conclusions1. Compare to non-anatomical hepatectomy, anatomical hepatectomy can degrade the early recurrence rate and elevate the early disease-free survival rates. The main reason is that the micrometastasis of HCC are mostly microscopic portal vein tumor thrombus and anatomical hepatectomy can clear more micrometastasis.2. Compare to non-anatomical hepatectomy, anatomical hepatectomy can decrease operation hemorrhage, depress the liver function lesion, decrease complication.3. The other early recurrence factors of HCC are operation hemorrhage, micrometastasis, AFP mRNA level before hepatectomy and tumor diameter.4. Analysis of AFP mRNA by real-time fluorescent quantitative RT-PCR is a very sensitive method for detecting circulating HCC cells,and can indicate the presence of hematogenous metastasis in patients with HCC. HCC cells can be released into the blood circulation because of hepatectomy. Monitoring AFP mRNA in peripheral vein blood may predict recurrence and prognosis for HCC.更多还原