【作者】 何健；

【导师】 章士正；

【作者基本信息】 浙江大学 ， 影像医学与核医学， 2010， 博士

【摘要】 1肝脏原发及转移性恶性肿瘤ADC值特点及ADC值测量可重复性研究目的研究肝脏原发及各种转移性恶性肿瘤在MR弥散加权成像(diffusion weighted imaging. DWI)不同b值条件下表观弥散系数(apparent diffusion coefficient, ADC)值的特点及随b值变化的规律,评价肝脏恶性病灶ADC值测量的观察者内及观察者间一致性,病灶ADC值测量的可重复性及可重现性。材料与方法该研究方案得到医院伦理委员会批准并获得患者的知情同意书。40位肝脏原发及转移性恶性肿瘤患者(男31例,女9例,年龄32-77岁,平均年龄58.7岁)共74个靶病灶(其中肝细胞肝癌31个,结直肠癌肝转移15个,胰腺癌肝转移9个,胰腺神经内分泌癌肝转移8个,胃癌肝转移7个,肺癌肝转移4个)。行MRDWI扫描(b=0,50,150,500,800 s／mm2),获取靶病灶的ADC值,分析各种转移瘤ADC值的差异及随b值升高的变化趋势。对其中15例患者(男11例,女3例,年龄49-77岁,平均年龄62岁)的30个病灶(包括肝细胞肝癌8个,结直肠癌肝转移8个,胰腺癌肝转移6个,胃癌肝转移4个,肺癌肝转移4个)重复进行3次b=500 s／mm2 DWI扫描,并进行多次重复测量,采用配对t检验,Pearson相关分析,可靠性检验及Bland-Altman分析评估观察者内、观察者间一致性以及ADC值测量的可重复性及可重现性。结果b=500 s／mm2时,病灶周围肝脏ADC值1.84×10-3mm2／s,胰腺神经内分泌癌肝转移灶ADC值为1.90×10-3mm2／s,二者不存在显著性差异。周围肝脏及胰腺神经内分泌癌肝转移ADC值显著高于胰腺癌肝转移1.42×10-3mm2／s,胃癌肝转移1.39×10-3mm2／s,肝细胞肝癌1.36×10-3mm2／s,肺癌肝转移1.30×10-3mm2／s,结直肠癌肝转移1.23×10-3mm2／s,且后5种转移瘤ADC值之间不存在显著性差异。b=800 s／mm2时,除了胰腺癌肝转移,其他所有肿瘤ADC值与正常肝脏组织存在显著性差异(p<0.05)。所有肿瘤ADC值均随b值增大而逐步下降,但存在不同的变化趋势。肝脏恶性病灶ADC值测量的观察者内,观察者间,同次检查两个序列(b=500 s／mm2)病灶ADC值,两次检查两个序列(b=500 s／mm2)病灶ADC值之间配对t检验不存在显著性差异,Pearson相关系数分别为0.994,0.983,0.753,0.712,可靠性检验组内相关系数分别为0.997,0.991,0.857,0.827,平均变异率分别为1.75%,2.86%,9.88%和9.55%,Bland-Altman分析发现观察者内、观察者间和可重复性检验所有点均位于一致性界限之内,而可重现性分析中发现有6.7%(2／30)的点位于一致性界限之外。结论MR DWI能够反映肝脏原发及各种转移性恶性肿瘤ADC值的不同特点,肝脏恶性病灶ADC值测量的观察者内及观察者间一致性,短期可重复性较好,但变异率小于18.72%的ADC值改变可能是由于测量误差引起的,同时需要注意可重现性的误差控制。2磁共振弥散加权成像对晚期肝细胞肝癌患者分子靶向药物治疗的疗效评估目的探讨晚期肝细胞肝癌(HCC)患者接受分子靶向药物(索拉非尼)治疗后影像学(包括CT与MR检查)表现的改变,重点探讨磁共振弥散加权成像(diffusion weighted imaging, DWI)表观弥散系数(apparent diffusion coefficient,ADC)的变化规律及其临床意义,为临床提供一种早期疗效评估方法。材料与方法此研究方案经过医院伦理委员会批准并获得患者的知情同意书。2008年12月-2010年1月前瞻性纳入晚期原发性肝癌患者10例(均为男性,年龄39-77岁,平均年龄60岁),治疗前一周内行上腹部MR-DWI扫描(b=500,800 s／mm2),治疗结束后1-3周、6周及12周复查MR-DWI,测定肝内直径大于10 mm典型癌灶的ADC值,根据病灶最大径的改变或坏死、结合实验室检查及临床状况对疗效进行评估。分析病灶的CT增强特点、MR T1WI、T2WI信号改变及ADC值的变化规律。结果HCC接受分子靶向药物治疗,影像学可评估病灶31个,其中14个为有反应病灶,17个为无反应病灶。有反应病灶基线最大径及ADC值(b=500 s／mm2)显著高于无反应病灶,有反应病灶治疗后ADC值(b=800 s／mm2)变化率显著高于无反应病灶。有反应病灶经过治疗后其ADC值(b=800 s／mm2)表现为先升高,后降低,随后再次升高的独特规律。传统检查手段观察到病灶T2WI信号增高提示细胞溶解与破坏,T1WI信号增高提示病灶内部出血,CT增强显示病灶强化程度减低,坏死成分增加,据此对ADC值的变化规律作出合理解释。无反应病灶T1WI、T2WI信号及CT表现治疗后未发生显著改变,ADC值在治疗开始后6周内基本保持稳定,12周观察到病灶ADC值(b=800 s／mm2)显著减低,提示病情进展。分子靶向治疗前后有反应病灶及无反应病灶的大小未见显著差异,提示传统疗效评估标准对此种药物疗效评价的局限性。此外,治疗过程中影像学表现与肿瘤指标的变化相一致。结论DWI结合传统CT及MR成像,有助于预测并早期监测HCC患者分子靶向药物治疗的疗效,能够实时评估治疗过程中的动态变化,并早期提示肿瘤复发。3磁共振弥散加权成像对肝脏转移性肿瘤全身系统性化疗的疗效评估目的研究上腹部磁共振弥散加权成像(diffusion weighted imaging. DWI)在肝脏转移性肿瘤全身系统性化疗疗效评估中的价值。材料与方法此研究方案经过伦理委员会批准并获得病人的知情同意书。2008年12月-2010年1月前瞻性地纳入肝转移瘤患者21例(其中男12例,女9例,年龄范围33-73岁,平均年龄56.6岁),包括结直肠癌肝转移10例,胃癌肝转移3例,肺癌肝转移3例,胰腺癌肝转移2例,乳腺癌肝转移1例,未分化癌肝转移1例,胰腺神经内分泌癌肝转移1例。于全身系统化疗前一周内行DWI扫描(b=500,800 s/mm2),治疗开始后1-3周,治疗结束后1周复查DWI。选择肝内最大的3-4个病灶作为靶病灶(共80个),测量靶病灶的表观弥散系数(apparent diffusion coefficient, ADC)值。根据治疗结束后靶病灶最大径的改变分为缓解组(最大径缩小>30%,20个病灶)、进展组(最大径增大>20%,包含18个病灶)和稳定组(介于前两者之间,42个病灶),比较三组靶病灶治疗前后ADC值及ADC值变化率的差异,及这些指标与病灶最大径改变率之间的相关性。结果治疗前缓解组ADC值显著低于稳定组及进展组(p<0.015),治疗后1-3周缓解组ADC值显著升高(此时病灶最大径改变率甚微),并且ADC值改变率(32-33%)显著高于稳定组(7-8%)及进展组(ADC值下降11-13%),治疗前ADC值以及ADC值早期变化率与化疗结束后病灶最大径缩小率之间存在显著的相关性,b=800 s/mm2 Pearson相关系数、R2较b=500 s/mm2更高,对肝转移瘤化疗疗效评估更为适宜。治疗结束后缓解组ADC值显著升高,稳定组及进展组ADC值无显著性差异。结论DWI在肝转移瘤全身系统化疗的疗效预测及早期评估中具有重要价值,可望为临床提供一种新的疗效评估手段。更多还原

【Abstract】 1 Reproducibility and characteristics of the apparent diffusion coefficient (ADC) values of primary and metastatic liver cancers by MR diffusion-weighted imaging (DWI)Objective To study the apparent diffusion coefficient (ADC) values of primary and metastatic liver cancers and their evolvement characteristics by MR diffusion-weighted imaging (DWI) under the condition of different b values; and to evaluate the intra-and inter-observer consistency, short-term repeatability and reproducibility of the measurement of ADC values of hepatic malignant lesions.Materials and Methods This research program was approved by the Hospital Ethics Committee and the informed consents were obtained from the patients. Seventy four cases of 40 patients (male 31 cases, female 9 cases, age range 32-77 yrs, average age 58.7 yrs) with primary and metastatic liver cancers (including 31 hepatocellular carcinomas,15 colorectal cancer liver metastasis,9 pancreatic cancer liver metastasis. 8 pancreatic neuroendocrine carcinoma liver metastasis,7 gastric cancer liver metastasis and 4 lung cancer liver metastasis) were included. MR DWI (b= 0,50,150, 500,800 s/mm2) was performed and the ADC values of the lesions were obtained. The differences among the ADC values of different liver metastasis were analyzed, and the trend of the ADC values in accordance to the increasing of b values was evaluated: Fifteen patients (male 11 cases, female 3 cases, age range 49-77 yrs, average age 62 yrs) underwent DWI scans for three times with different intervals and repeated measurements were performed on 30 lesions (including 8 hepatocellular carcinomas,8 colorectal cancer liver metastasis,6 pancreatic cancer liver metastasis,4 gastric cancer liver metastasis and 4 lung cancer liver metastasis). The intra-and interobserver consistency, short-term repeatability and reproducibility of the measurement of ADC values of hepatic malignant lesions were evaluated by using paired t test, Pearson’s correlation analysis, reliability test and Bland-Altman analysis in SPSS 13.0 software.Results When b was set as 500 s/mm, there was no significant difference between the ADC value (×10-3 mm2/s) of the surrounding liver tissue (1.84) and the pancreatic endocrine liver metastasis (1.90). The ADC values of the surrounding liver tissue and pancreatic endocrine liver metastasis were significantly higher than those of pancreatic cancer liver metastasis (1.42), gastric cancer liver metastasis (1.39), hepatocellular carcinoma (1.36), lung cancer liver metastasis (1.30) and colorectal cancer liver metastasis (1.23). And there were no significant differences between the ADC values of the latter five kinds of liver metastasis. There were significant differences between the ADC values (b= 800 s/mm2) of the surrounding liver tissue and all the primary and metastatic liver lesions except pancreatic cancer liver metastasis. All the ADC values of the lesions decreased when the b value increased with a common and somewhat unique trend of each kind of the lesion. No significant differences of ADC values of malignant liver lesions could be found by using paired t test in the intra-and inter-observer consistency, the repeatability (two sequences both with a b of 500 s/mm2 lesion in the same session) and the reproducibility(two sequences both with a b of 500 s/mm2 in two sessions) evaluation. Pearson’s correlation coefficients were 0.994,0.983, 0.753 and 0.712, intraclass correlation coefficient of the reliability test was 0.997, 0.991,0.857 and 0.827, and the average coefficient of variation was 1.75%,2.86%, 9.88% and 9.55%, respectively. Bland-Altman analysis showed that all the data points from the intra-and inter-observer consistency, as well as repeatability tests were located within the limits of consistency, while 6.7%(2/30) points of reproducibility analysis located beyond the reference lines.Conclusion MR DWI can characterize different ADC values of primary and metastatic liver cancers. Intra-and inter-observer consistency and the short-term repeatability of the ADC values measurement are satisfied. A variation rate of the ADC value less than 18.72% may be due to measurement error and attention should be paid to control the measurement error involving reproducibility.2 Assessment of the efficacy of molecular targeted therapy in patients with advanced hepatocellular carcinoma with MR Diffusion-weighted imaging (DWI)Objective To explore the findings on CT and MR scans in patients with advanced hepatocellular carcinoma (HCC) during molecular targeted therapy (Sorafenib), with a focus on the regular pattern and clinical sense of apparent diffusion coefficient (ADC) values obtained from MR diffusion-weighted imaging (DWI), and to provide an early assessment tool for such an evolving therapy strategy. Materials andMethods This research program was approved by the Hospital Ethics Committee and informed consents were obtained from the patients. From Dec.2008 to Jan.2010,10 patients with advanced hepatocellular carcinoma (all male; age range 39-77 years old, with an average age of 60) were included. Upper abdominal MR-diffusion weighted imaging (DWI) scans (b=500,800 s/mm") were performed within 1 week prior to, 1-3 weeks.6 weeks and 12 weeks after the start of molecular targeted therapy. Apparent diffusion coefficient (ADC) values of hepatic lesions with the diameter larger than 1 cm were measured. Treatment efficacy judgment was referred to RECIST as well as clinical and laboratory observations. Evolving courses of the ADC values of the lesions during therapy were demonstrated.Results Thirty-one HCC lesions which underwent molecular targeted therapy could be assessed by imaging modality,14 of which were responsive and the other 17 lesions fell into the nonresponsive group. The largest diameters and the ADC values (b= 500 s/mm2) of the responsive lesions at the baseline were significantly higher than those of the non-responsive group. The rate of change of ADC values (b= 800 s/mm2) after the start of treatment in responsive group was significantly higher than that in non-responsive group. The ADC values (b= 800 s/mm2) of the responsive lesions after therapy showed a first rise, then a decrease, and then an increase again at last. Traditional CT and MR imaging showed increased signal intensity on T2WI, suggesting cell lysis and destruction, and increased signal intensity on T1WI indicating bleeding within the tumor. Contrast enhanced CT scans demonstrated a reduction of the degree of enhancement and a increase of the necrosis area of the lesions. Based on these findings, the mechanism of the changes of ADC values could be explored. Signal intensities on T1 and T2 WI as well as the CT findings appeared unchanged in non-responsive lesions. And the ADC values remained stable within 6 weeks after the start of treatment. A significant reduction of the ADC values (b= 800 s/mm2) of those lesions was observed at the 12-week-follow-up, suggesting progression of the disease. No significant change of the largest diameter could be observed at the end of follow-up, in both responsive and non-responsive groups, suggesting a limitation of traditional criteria in evaluation of such an evolving therapy. Additionally, the imaging findings consisted with the changes of tumor marker.Conclusion Combined with traditional CT and MR scans, MR-DWI offering ADC values of the lesions can help to predict and monitor the efficacy of molecular targeted therapy in HCC patients, give real-time evaluation of dynamic changes in the course of treatment, and prompted an early tumor recurrence.3 Assessment of treatment efficacy of systemic chemotherapy in patients with liver metastases by magnetic resonance diffusion-weighted imaging (DWI)Objective To explore the utilization of upper abdominal magnetic resonance diffusion-weighted imaging (DWI) in assessment of treatment efficacy of systemic chemotherapy in patients with liver metastases.Materials and Methods This research program was approved by the Hospital Ethics Committee and informed consents were obtained from the patients. From Dec.2008 to Jan.2010,21 patients with liver metastases (male 12 cases, female 9 cases, age range 33-73 yrs, average age 56.6 yrs) were prospectively enrolled, including 10 cases of colorectal cancer liver metastases,3 gastric cancer,3 lung cancer,2 pancreatic cancer, 1 breast cancer,1 undifferentiated carcinoma and 1 pancreatic neuroendocrine carcinoma. Upper abdominal MR-DWI scans (b=500.800 s/mm2) were performed within 1 week prior to,1-3 weeks after the start of systemic chemotherapy and 1 week after the end the therapy. Apparent diffusion coefficient (ADC) values of the target lesions (the largest 3-4 lesions in the liver of each case) were measured. Totally 80 target lesions were divided into three groups according to RECIST after completion of the therapy:remission group (maximum diameter decrease rate>30%,20 lesions), progression group (maximum diameter increase rate>20%.18 lesions) and stable group (range between remission and progress,42 lesions). ADC values before and after the therapy were compared among three groups, and the correlation between the change rates of ADC values and the maximum diameters was also evaluated. Results The ADC value of remission group was significantly lower than those of stable and progression groups (p＜0.015). The ADC values of remission group significantly increased 1-3 weeks after the start of therapy (without obvious change in maximal diameter of the lesions), and the change rate of ADC values was significantly higher than in remission group (32-33%) than in the stable group (7-8%) and progression group (decreased 11-13%). Both the ADC values pretreatment and the early change rate of ADC values (1-3 weeks after the start of therapy) correlated well with the ultimate change rate of maximum diameter of lesions (1 week after the end of therapy). The correlations were higher when the b value was settled as 800 s/mm2 rather than 500 s/mm2, which suggested that DWI with a b value of 800 s/mm2 be more favorable to predict and monitor the treatment efficiency. After therapy, ADC values increased significantly compared with those on the baseline in remission group. However, no significant differenced were observed between baseline and post-treatment ADC values in stable and progression groups.Conclusion MR-DWI has great potential in prediction, early detection and monitoring the therapeutic efficacy of systemic chemotherapy in patients with hepatic metastases.更多还原