【作者】 李颖嘉；

【导师】 张雪林；

【作者基本信息】 南方医科大学 ， 影像医学与核医学， 2007， 博士

【摘要】 研究目的1．应用能量多普勒、脉冲反相谐波实时灰阶超声造影微血管成像及时间—强度曲线定量分析技术结合免疫组化方法探讨乳腺癌发生与演进过程中血管生成作用及其变化规律，为乳腺癌高危人群的监测寻找客观简便的判断指标。2．应用脉冲反相谐波实时灰阶超声造影微血管成像、时间—强度曲线定量分析、免疫组化、透射电镜技术比较良恶性不同类型肿瘤、同种类型肿瘤不同灌注区域血管在形态结构、血流动力学功能以及相关分子表达上的差异，从而为乳腺癌的临床早期诊断及鉴别提供可靠指标，并为进一步进行以肿瘤新生血管内皮为靶标的肿瘤分子成像研究奠定实验基础。材料与方法1．研究对象正常乳腺组10例，单纯增生组20例，非典型增生组(atypical hyperplasia，AH)20例，导管内原位癌组15例，浸润性导管癌组30例，共5组95例。乳腺良恶性肿瘤分组：恶性肿瘤组30例(33个病灶)，包括浸润性导管癌23个，浸润性小叶癌7个，导管内原位癌3个；良性肿瘤组15例(17个病灶)，均为纤维腺瘤。入选标准：正常乳腺组为乳腺纤维腺瘤行区段切除术后经病理证实的结构正常的瘤旁2cm处乳腺组织。乳腺增生分类参照Rosen分类标准，乳腺肿瘤组织学分类及诊断标准参照2003年新版WHO乳腺肿瘤组织学分类标准。所有病例均经病理检查证实。2．检查仪器及方法2．1仪器检查采用飞利浦公司iU22超声检查仪，L9-3宽频线阵探头，脉冲反相谐波(Pulse Inversion Harmonic，PIH)造影条件，机械指数(Mechanical index，MI)为0.07。选用乳腺常规检测程序，彩色多普勒(Color Doppler Flow Imaging，CDFI)多切面观察后以能量多普勒(Power Doppler Imaging，PDI)对观察部位血流信号最丰富区域冻结并存储图像，之后静脉弹丸式注射南方医院研制的“全氟显”造影剂，0.02ml／kg，同步存贮造影剂注射后三分钟的连续动态图像。2．2图像分析根据观察区域PDI显示的血流形态、数目及分布特点将血流按丰富程度由低到高分为4级。0级：周边及内部未见血流信号；1级：少量血流，血流环绕病灶周围，内部无或偶见点状血流；2级：中量血流，可见1条主要血管，其长度超过病灶的半径或见几条小血管由周边向内部发出分支；3级：丰富血流，可见2条以上血管伸入病灶内部，或血管相互连通成网状。开启QLAB分析软件，启动微血管成像(microvascular imaging，MVI)程序，实时观察造影剂微泡在微血管中的显像轨迹；对观察部位血流显像最丰富区域取样3次(避开坏死区)获得感兴趣区血流灌注时间-强度曲线，由曲线得到以下参数：峰值强度(peak intensity，PI)，曲线下面积(area under the curve，AUC)、达峰时间(time to peak，TTP)，廓清时间(wash-out time，WOT)，取其均值。对良、恶性肿瘤两组病灶同上分别对病灶边缘及中心血流显像最丰富区域取样获得感兴趣区时间-强度曲线，由曲线得到不同区域灌注参数，分别取其均值计算该病灶的不同区域灌注参数，(边缘灌注参数+中心部灌注参数)÷2为病灶平均灌注参数。3．病理学检查3．1 HE染色3．2免疫组化染色小鼠抗人CD34单克隆抗体、小鼠抗人VEGF单克隆抗体、兔抗人Flk-1／KDR多克隆抗体按SP试剂盒说明书操作。结果判断：以细胞浆或细胞膜出现棕黄色颗粒为阳性细胞判定标准。微血管密度(microvessel density，MVD)测定：参照Weidner方法。VEGF和Flk-1／KDR表达测定：采用半定量积分法判定结果，阳性细胞数≤5％为0分，6％-25％为1分，26-50％为2分，51％-75％为3分，＞75％为4分；无特异性染色为0分，染色强度黄色为1分，棕黄色为2分，棕褐色为3分，两者积分相乘，0-1分为阴性(-)，2—4分为弱阳性(+)，5—8分为中度阳性(++)，9—12分为强阳性(+++)。3．3电镜观察选取手术切除乳腺癌6例、乳腺纤维腺瘤及纤维腺瘤切除后瘤旁2cm处经HE证实正常乳腺组织各4例，常规处理后在PHLIPS CM10型透射电镜下观察。4．统计学处理采用SPSS13.0软件包进行统计分析。4．1 5组的计量资料用均数±标准差表示，方差齐性则采用单因素方差分析(one-way ANOVA)，方差不齐或计数资料比较采用非参数秩和检验(Games Howell检验)，以P＜0.05为有显著性差异。4．2乳腺良恶性肿瘤两组病变，计量资料比较采用两独立样本t检验，计数资料比较分别采用x~2检验，两独立样本非参数Mann-Whitney U检验，P＜0.05为有显著性差异。结果1．第一章结果1．1病理结果1．1．1乳腺癌合并AH情况45例乳腺癌，癌旁组织合并AH64.44％(29／45)，其中导管内癌合并AH73.33％(11／15)，浸润性导管癌合并AH60％(18／30)。1．1．2 CD34、VEGF、Flk-1／KDR在各组中表达情况各组病例CD34均呈阳性表达，阳性产物定位于微血管内皮细胞胞浆或胞膜。其分布呈明显异质性，微血管丰富区多位于癌巢边缘或增生活跃组织。随病程演进各组MVD计数逐渐升高，MVD值最小为正常乳腺组(17.10±4.44)，最大为浸润性导管癌组(51.89±3.05)，5组间MVD有显著性差异(P＜0.05)。组间两两比较除正常组与单纯上皮增生组，AH组与导管内原位癌组差异不显著(P值＞0.05)，其余任意两组间均有显著性差异。VEGF表达情况：VEGF主要在乳腺癌细胞胞浆或胞膜及癌组织内新生血管内皮细胞胞浆或胞膜呈高表达，在AH的导管上皮细胞及相应间质的血管内皮细胞也有阳性表达。正常乳腺、单纯上皮增生组VEGF呈阴性或弱阳性表达，随病情进展VEGF表达渐次增高，增强，VEGF表达最低为正常乳腺组(1.10±0.78分)，最高为浸润性导管癌组(7.87±2.50分)，5组间VEGF表达有显著性差异(P＜0.05)。组间比较，除正常组与单纯上皮增生组差异不显著(P＜0.05)，其余任意两组间差异有统计意义。(P＜0.05)Flk-1／KDR表达情况：Flk-1／KDR主要表达于乳腺癌间质的血管内皮细胞胞浆或胞膜。在正常乳腺、单纯上皮增生组的血管内皮细胞Flk-1／KDR不表达，只在乳腺肌上皮细胞表达，从AH组．导管内癌组一浸润性导管癌组，Flk-1／KDR在血管内皮细胞表达渐次增高，增强，而在肌上皮细胞表达逐渐消失。Flk-1／KDR在血管内皮细胞表达最低为正常乳腺组，最高为浸润性导管癌组。1．2超声检查表现1．2．1 PDI检查表现随病程进展，PDl分级逐渐增高，5组间PDI血流分级分布有显著性差异(P＜0.05)。正常乳腺组血流分级主要为0级、1级(mean rank 29.20)，AH组2级血流明显增多(mean rank 45.33)，浸润性导管癌组血流分布最为丰富，主要为3级(mean rank 66.65)。1．2．2超声造影检查表现1．2．2．1 MVI检查表现造影后MVI对血流显示能力明显提高：正常乳腺组织血流相对不丰富，血管排列规律，走行自然；单纯上皮增生组病灶周围血管稍增多；AH组血管数目增多，由病灶周边向病灶内延伸，排列不规则，出现扩张、扭曲的血管，导管内原位癌组见更多紊乱、畸形的血管在病灶内吻合，至浸润癌组血管粗细不均、异常扩张、扭曲、囊状膨大、走行紊乱，可见杂乱分布的稠密、团状新生血管网。1．2．2．2时间-强度曲线定量分析5组间峰值强度(PI)有显著差异，最低为正常乳腺组(38.26±8.52dB)，最高为浸润性导管癌组(120.06±11.00dB)，随病情进展，PI逐渐增大。组间两两比较，除正常组与单纯上皮增生组(P值＞0.05)，AH组与导管内原位癌组差异不显著(P值＞0.05)，余两组间比较，差异均有统计学意义(P值＜0.05)。5组间达峰时间(TTP)有显著差异，TTP最长为单纯增生组(22.80±1.94s)，之后随病程演进，TTP逐渐缩短，最短为浸润性导管癌组(14.59±1.02s)。组间两两比较，正常组与单纯增生组(P＞0.05)，正常组与AH组(P＞0.05)，AH组与导管内原位癌组(P＞0.05)差异不显著，余两两之间比较，差异有统计学意义(P＜0.05)。5组间曲线下面积(AUC)有显著差异，最低为正常乳腺组(363.09±28.14dB.s)，最高为浸润性导管癌组(1300.63±124.86dB.s)。随病情发展，总趋势为AUC逐渐增大。组间两两比较，除正常组与单纯上皮增生组(P＞0.05)，AH组与导管内原位癌组(P＞0.05)差异不显著，余两组间比较，差异均有统计学意义(P＜0.05)。5组间廓清时间(WOT)有显著差异，最低为正常乳腺组(31.39±1.14s)，最高为浸润性导管癌组(74.93±3.19s)。随病程演进，总的趋势为WOT逐渐延长。组间两两比较，除正常组与单纯上皮增生组，AH组与导管内原位癌组(P＞0.05)差异不显著，其它两组比较，差异均有统计学意义(P＜0.05)。由此可见，在乳腺癌发生、发展过程中，各组PI、AUC、WOT逐渐增高，但乳腺增生初期变化不明显，上述指标变化主要始于AH阶段，乳腺AH与导管内原位癌组间差异不显著，表明AH与早期乳腺癌血流动力学改变可能具有一致性，并随乳腺癌的进一步发展而变化。TTP在单纯增生组异常延长，之后逐渐缩短。2．第二章结果2．1病理结果2．1．1 CD34、VEGF、Flk-1／KDR表达情况CD34表达情况：CD34在乳腺癌中分布呈明显异质性，微血管丰富区位于癌巢边缘，血管粗细不均，见较多裂隙状血管。CD34在乳腺纤维腺瘤中分布及血管管径较为均一。恶性肿瘤组MVD均值为(34.84±8.34)显著高于良性肿瘤组MVD均值(18.65±4.69)，良恶性肿瘤两组MVD差异有统计学意义(P＜0.05)。VEGF表达情况：VEGF在乳腺癌细胞及周围血管内皮细胞呈强阳性表达，尤其弥漫分布于癌巢边缘及坏死组织附近。纤维腺瘤中VEGF呈阴性或弱阳性表达。恶性肿瘤组VEGF表达(mean rank32.02)与良性肿瘤组(mean rank12.85)比较有统计学差异(P＜0.05)。Flk-1／KDR表达情况：Flk-1／KDR在乳腺癌血管内皮细胞呈强阳性表达，尤其弥漫分布于癌巢边缘及坏死组织附近。纤维腺瘤中Flk-1／KDR在血管内皮细胞呈阴性或弱阳性表达。恶性肿瘤组Flk-1／KDR表达(mean rank30.91)与良性肿瘤组(mean rank15.00)比较有统计学差异(P＜0.05)。2．1．2透射电镜结果乳腺癌新生血管由单层内皮细胞构成，内皮细胞间隙增宽，基底膜不连续，间质水肿。内皮细胞形态失常，胞体大，胞浆丰富，吞饮泡增多，细胞核畸形，常染色质多，异染色质少，可见核仁增大，核仁边集。狭窄、闭塞的新生血管与扩张的大血管并存。乳腺纤维腺瘤及正常乳腺问质中血管内皮细胞紧密连接存在，质膜面光滑，基底膜完整。内皮细胞呈长扁形，胞浆少，吞饮泡少，核形态正常，未见核仁。2．2超声造影表现2．2．1造影前后良恶性肿瘤血流信号检出率比较造影前良恶性病灶周边及内部血流检出率无显著性差异(P＞0.05)。造影后良恶性肿瘤病灶周边及内部血流检出率有显著性差异(P＜0.05)，以病灶内部血流检出率差异更大。2．2．2 MVI检查表现病灶内不均匀增强，充盈缺损，紊乱的血管网、血管扩张、血管迂曲为恶性肿瘤征象，良恶性肿瘤有显著性差异(P＜0.05)。2．2．3时间-强度曲线分析曲线形态：恶性肿瘤组多数(87.88％，29／33)呈速升缓降型，良性肿瘤组多数(80.00％，12／15)呈缓升速降型。恶性肿瘤组AUC(1474.62±202.13dB.s)、WOT(71.51±7.43s)大于良性肿瘤组(分别为597.11±102.58dB.s，38.54+6.55s)，两组比较有统计学差异(P＜0.05)。PI、TTP两组间无统计学差异。恶性肿瘤组病灶边缘的PI(126.09±16.69dB)、AUC(1946.76±261.86dB.s)、WOT(94.98±10.19s)大于病灶中心区域(分别为64.43±9.53 dB，998.66±143.51dB.s，48.05±4.88s)，病灶边缘的TTP(12.14±1.58s)小于中心区域(30.04±15.91s)，病灶边缘及中心各区域灌注参数比较有统计学差异(P＜0.05)。良性肿瘤组病灶边缘各灌注参数与病灶中心相比无统计学差异(P＞0.05)。结论1．乳腺癌的发生是一个多因素影响的复杂过程，正常乳腺—单纯上皮增生—AH—原位癌—浸润性癌演进过程中，血管生成起着重要作用，PDI、超声造影MVI及时间-强度曲线各主要参数与免疫组化血管生成相关指标的检测具有一致性，其总体趋势是：随病程演进，VEGF及其受体Flk-1／KDR在血管内皮细胞表达渐进增高，MVD增加，PDI血流分级增高，MVI显示血管数目增多，结构趋于紊乱，各主要灌注参数PI、AUC、WOT逐渐延长。但在病程初期各主要指标改变不明显，显著变化始于AH阶段。VEGF及其受体Flk-1／KDR的表达异常可能是乳腺癌恶性转化过程中血管生成异常的主要始动因素，PDI、超声造影MVI及各主要灌注参数能够反映乳腺癌前病变阶段组织血管生成活性的变化规律，是用于高危人群监测随访无创而实用的手段。2．实时超声造影MVI结合时间-强度曲线可以同时反映肿瘤血管的解剖和生理特征，造影后灌注模式、时间．强度曲线形态、平均灌注参数及区域灌注参数的差异是乳腺良恶性肿瘤有价值的鉴别诊断依据，而良恶性肿瘤新生微血管内皮分子表达和微血管构筑的异质性是超声造影评价乳腺肿瘤血管生成的基础。3．VEGF及Flk-1／KDR在乳腺原位癌和浸润性癌血管内皮细胞呈持续高表达，在乳腺良性病变几乎不表达，乳腺癌内新生血管内皮细胞超微结构有不同于正常血管内皮细胞的特征，表明针对VEGF及Flk-1／KDR为靶点进行乳腺癌新生血管特异性分子显像可能为解决乳腺癌的早期诊断提供新的思路。更多还原

【Abstract】 Objective1.To study the effects and mechanisms of angiogenesis in the tumorgenesis andprogression of breast precancerous lesion and carcinoma. To find a simple andobjective method to screen breast carcinoma in high risk population.2.To compare the differences in histological morphology, hemodynamics andrelated molecules between benign and malignant tumor, different regional bloodperfusion in the same type tumor. To offer reliable indicators for breast cancer earlydiagnosis and identification, and therefore to lay experimental basis for molecularimaging of neovascularization of breast carcinom.Materials and methods1.Objects95 patients were included in our study which were divided into 5 groups: normalcontrols, 10 cases,simple hyperplasia, 20 cases; atypical hyperplasia, 20 cases;intraductal carcinoma in situ, 15 cases;invasive ductal carcinoma, 30 cases.30 malignant tumor patients had 33 foci, the histological typing were as follows:infiltrating ductal carcinoma, 23 foci; infiltrating lobular carcinoma, 7 foci;intraductal carcinoma in situ, 3 foci. 15 benign tumor patients had 17 foci that wereall breast fibroadenoma.All patients were accepted histological examination.The recruitment criteria for hyperplasia patients was refered to Rosen’s Breast Pathology. The histological typing and diagnosis of breast tumor were refered to"histological typing of breast tumor" made by WHO in 2003.2. Instruments and methods2.1 InstrumentsThe ultrasound imaging was performed with iU22 ultrasound system (PhilipsTechnologies) connected to a L9-3 wide frequency linear array probe. The settings ofpulse inversion harmonic(PIH) contrast imaging were as follows: the frame rate was16 HZ, mechanical index(MI) was 0.07.The perfluoropropane-containing human albumin microphere was prepared bypharmacology experiment base of Nanfang hospital. The contrast medium (0.02ml/kg) was administered with intravenous bolus injection.2.2. Image collection and analysisImage collection:Routine procedure for breast checking was used. Color Dopplerflow imaging and power Doppler imaging (PDI) to observe the blood flow rich area andimages were frozen and stored. After contrast was injected, the continuously dynamicimaging was stored synchronously.Image analysis:The regions of interest(ROI) with PDI can be divided into 4grades according to the blood flow rich degree.Turn on QLAB software and MVI procedure to observe the imaging track ofmicrosphere in microvessels in real time. To sample twice at blood flow rich area toobtain time-intensity curve of regions of interest (ROI). The related parameters suchas peak intensity, area under the curve, time to peak enhancement and wash-out timewere obtained accordingly.To sample at the outer area of foci and blood flow rich area to obtain time-intensity curve of ROI and the perfusion parameters of different region was obtainedaccordingly. (mean perfusion parameter=marginal perfusion parameters+interiorperfusion parameter/2)3. pathology examination3.1 HE staining: The tissues were fixed in 10% formalin, embedded in paraffin,serially sectioned at 4μm, and stained with hematoxylin and eosin. 3.2 Immunohistochemistry:Immunohistochemistry was performed according to the protocol of the kit. Thepositive cell was determined by the appearances of light brown granules in cytoplasmor cell membrane.Microvessel density (MVD) was determined by Weidner’s method. Theexpression of VEGF and Flk-1/KDR was determined by semi-quantitative integralmethod: positive cells≤5%, score 0; 6%～25%, score 1; 26%～50, score 2; 51%～75%,score 3;＞75%, score 4. The score of stain: no specific staining, score 0; yellowstaining, score 1; light brown, score 2; brown, score 3. To multiply the score marks,the scores were definited as flows: 0～1, negative (-), 2～4, weakly positive (+); 5～8,middle posivive (++); and 9～12, strong positive (+++).3.3 Electron microscopy observationThe samples of breast cancers (6 cases), breast fibroadenoma (4 cases) andnormal breast tissues (4cases) was analysed routinely with transmission electronmicroscopy (CM10, PHILIPS).4 Statisticalal analysisThe measurement data of the five group was presented with mean±standarddeviation. The Kolmogorov-Smimof test was used to examine the normality of thevalues’ distribution. One-way ANOVA was used when test for homogeneity ofvariance and Games Howell test was used when test for heterogeneity of variance andenumeration data analysis.For the difference between benign and malignant tumor, measurement data wastested with independent sample t test, enumeration data was tested with x~2, twoindependent sample was performed using the non-parametric Mann-CWhitney U-test.A value of P＜0.05 was considered statisticalally significant.Results:1.Results of ChaptⅠ1.1 pathology results1.1.1 Breast carcinoma combined with atypical breast hyperplasia: in all 45Breast carcinoma patients, 64.44% patients had Atypical breast hyperplasia in peri-cancer tissue in which the incidence rate of Atypical breast hyperplasia inIntraductal carcinoma patients was 73.33% (11/15) while the rate in invasive ductalcarcinoma was 60%(18/30).1.1.2 CD34、VEGF、Flk-1/KDR expressionCD34 was expressed positively in each group. CD34 was expressed in thecytoplasm or on the cell membrane that its distribution show significant heterogeneitythat expressed mainly in the cancer nest margin or hyperplasia active tissues.The MVD increased with the progression of disease. The MVD was minimal innormal breast group (17.10±4.44) and was maximal in invasive ductal carcinoma(51.89±3.05). The MVD had significant difference among groups(P＜0.05) and hadsiginificant difference between each group except normal group and hyperplasiagroup, atypical breast hyperplasia group and intraductal carcinoma group.VEGF expressed mainly in the cytoplasm and on the cell membrane of breastcancer cells and new vessel endothelial cells, and expressed on the duct epithelial cellof atypical hyperplasia and vascular endothelial cells of interstitial tissues. VEGFhardly expressed in normal breast and simple epithelium hyperplasia while expressedmaximally in the invasive ductal carcinoma that had significant different amonggroups. Compare means between groups, VEGF expression had significant differencebetween groups except normal and simple hyperplasia group.Flk-1/KDR expressed mainly in the cytoplasm and on the cell membrane ofbreast cancer interstitial vascular endothelial cells, and expressed on the vascularendothelial cells of atypical hyperplasia. Flk-1/KDR hardly expressed in vascularendothelial cells of normal breast and simple epithelium hyperplasia group, andexpressed increaseingly with the progression of disease while expressed maximally inthe invasive ductal carcinoma.1.2 Ultrasound results1.2.1 PDI resultsWith the progression of breast disease, the blood flow detected by PDI becomericher that the normal breast group was in grade 0 or 1 while the invasive ductalcarcinoma group was the richest that most was in grade 3. 1.2.2 Contrast enhanced ultrasound imaging1.2.2.1 MVI resultsThe MVI of blood flow improved significantly after use of contrast agents. Innormal breast tissues, the arrangements of the blood vessels were regularly, the sizeswere homogeneous, the distributions were natural. The number of blood vesselincreased around the focus in simple epithelium hyperplasia tissues. We could seedilated and distorted vessels extended from the margin to the interior in atypicalhyperplasia except the increased vessel number. In intraductal carcinoma in situ, theblood vessel distortion and anastomosis was obvious. In invasive ductal carcinomagroup, we could see either the blood vessel distention, distortion, cystiform dilationand the disorderly distribution and agglomerated new born blood vessel network.1.2.2.2 Time-intensity curve analysisThe peak intensity(PI) among 5 groups had significant diffierence that thelowest was normal breast group(38.26±8.52) and the highest was invasive ductalcarcinoma group (120.06±11.00). Compare means between each group, there weresignificant difference between groups except normal breast group and hyperplasiagroup, atypical hyperplasia group and intraductal carcinoma in situ group.The time to peak enhancement(TTP) among 5 groups had significant differencethat the longest was simple hyperplasia group (22.80±1.94s) and the shortest wasinvasive ductal carcinoma group (14.59±1.02s). Compare the difference betweenevery 2 groups, there weren’t significant difference between normal breast group andhyperplasia group, normal breast group and atypical hyperplasia group, atypicalhyperplasia group and intraductal carcinoma in situ group while the others hadsignificant differences.The area under the curve(AUC) among 5 groups had significant difference thatthe lowest was normal breast group(363.09±28.14) and the highest was invasiveductal carcinoma group (1300.63±124.86). Compare means of AUC between eachgroup, there were significant difference between groups except normal breast groupand hyperplasia group, atypical hyperplasia group and intraductal carcinoma in situgroup. The wash-out time (WOT) among 5 groups had significant difference that thelowest was normal breast group(31.39±1.14) and the highest was invasive ductalcarcinoma group (74.93±3.19). The WOT showed an increasing trend with theprogression of breast disease. Compare means between each group, there weresignificant difference between groups except normal breast group and hyperplasiagroup, atypical hyperplasia group and intraductal carcinoma in situ group.From the data above, we found that the PI, AUC and WOT in each groupincreased with the progression of breast cancerous disease while the TTP showed asharp increase in simple hyperplasia group,then decreased gradually.2. Results of chaptⅡ2.1 Pathology results2.1.1 CD34、VEGF、Flk-1/KDR expressionExpression of CD34: The distribution of CD34 in breast carcinoma washeterogeneous. The micro-vessel rich regions distributed diffuse around the margin ofcancer nest. The MVD of malignant group(34.48±8.34) was significantly higherthan that of benign group(18.65±4.69).Expression of VEGF: the high VEGF expression distributed diffuse or focally atthe margin of cancer nest and necrotic tissue. The VEGF expression betweenmalignant and benign group had significant difference.Expression of CD34: Flk-1/KDR expressed diffuse or focally in breastcarcinoma, especially high at the margin of cancer nest and necrotic tissue. TheFlk-1/KDR expression between malignant and benign group had significantdifference.2.1.2 Transmission electron microscope resultsThe newborn blood vessels of breast cancer were composed of monolayerendothelial cells, the gap between endothelial cells became wide. The basementmembrane not continuously had obvious interstitial edema. The shapes of endothelialcells were abnormal, the cell body became bigger and cytoplasm became richer andabnormal cell nucleus appeared and pinocytotic vesicle increased. The endothelial cell of fibroadenoma and normal breast had tight gap, thecytoplasm was smooth, the basement membrane was integral. The cell was long-flatthat had normal cytoplasm, nuclei and the pinocytotic vesicles.2.2 Contrast enhanced ultrasound imaging2.2.1 Blood signal of malignant breast carcinoma with or without contrast: thepositive rate for blood flow had no significant difference before the use of contrastagents. After the use of contrast agents, the positive rate enhanced significantlyaround and inside the focal.2.2.2 MVI: MVI inside of the focus was increased heterogeneously that hadthe characteristic of malignant tumor such as filling defect and vessel distortion.2.2.3 Time-intensity curve analysisShape of the curve: the curve of malignant tumor group was characterized asascend rapidly and drop slowly while the benign group presented as ascend slowlyand drop rapidly.The AUC and WOT of malignant tumor group were significantly higher thanthat of benign group while the PI and time to peak had no statistical difference.In malignant tumor group, the PI, AUC and WOT of the margin of focus weresignificantly higher that of inside region of focus, while time to peak was the opposite.However, the perfusion parameter between inside and outside of focus in benigngroup had no statistical differences.Conclusions1. The tumorgenesis of breast carcinoma is complicated process that affected bymulti-factors. Angiogenesis played pivotal role in the transformation process fromprecancerous lesion to malignant breast carcinoma. The ultrasound findings such asPDI, MVI and main parameters of time-intensity curve have good accordance withthe immunohistochemical results. With the progression of breast cancer, theexpression VEGF and its receptor, Flk-1/KDR, increased; at the same time, the MVDincreased, the framework of the vessel become disorder and the perfusion parameterssuch as PI, AUC, WOT increased either. However, in the very early stages, theparameters mentioned above have no obvious changing while they changed significantly from the atypical hyperplasia stage. PDI, contrast MVI and the mainperfusion parameters could reflect the rule of angiogenesis activities changing indifferent precancerous lesions that it could be a useful, non-invasive method to screenhigh risk population.2. The perfusion pattern, shape of time-intensity curve, mean perfusionparameter and variance of regional perfusion parameters are valuable diagnostic basisin discriminating benign and malignant breast tumor. The expression of endothelialmolecules of neovasculation and the heterogeneity of mierovasculature are the basisfor the ultrasound evaluation of angiogenesis in breast carcinoma.3. The VEGF and Flk-1/KDR expression are continuously high in intraductalcarcinoma in situ and invasive ductal carcinoma patients while they were hardlyexpressed in benign breast lesions. The ultrastructures of newly mierovessel in breastcarcinoma are different from that of normal vascular endothelium. The results give usa hint that the molecular imaging that targeted to VEGF and Flk-1/KD may providean new idea to explore the early diagnosis of breast carcinoma.更多还原