首页 理论教育 膀胱癌转移

膀胱癌转移

时间:2023-04-24 理论教育 版权反馈
【摘要】:膀胱癌是累及泌尿系统最常见的恶性肿瘤。美国白种人膀胱癌的发病率高于非洲裔、亚裔、土著或拉丁裔。UC来自膀胱黏膜内层,并常常呈多中心发生,包括染色体标记、遗传多态性、遗传和表观遗传改变等多种因素都可能参与了肿瘤的发生、进展和转移。此外,有50%的肌层浸润性UC患者存在隐匿性远处转移,并且5年生存率不高。

◎Neveen Said,Dan Theodorescu

膀胱癌是累及泌尿系统最常见的恶性肿瘤。2008年,美国预计有68810例新发病例,其中男女比例为4∶1,约13750人死亡[1]。美国白种人膀胱癌的发病率高于非洲裔、亚裔、土著或拉丁裔。然而,白种人男性患者的生存时间长于其他种族的男性和所有女性。中位发病年龄为70岁[1]。在美国,最常见的膀胱癌类型是尿路上皮癌(UC),以前被称为“移行细胞癌”(TCC)。UC来自膀胱黏膜内层,并常常呈多中心发生,包括染色体标记、遗传多态性、遗传和表观遗传改变等多种因素都可能参与了肿瘤的发生、进展和转移。70%~80%的UC患者不伴有肌层侵犯(以前被称为“浅表性疾病”),20%~30%伴有肌层侵袭(图7-25)。尽管无肌肉浸润的UC患者预后良好,但常常复发,且有多达30%患者会发展为肌肉浸润。此外,有50%的肌层浸润性UC患者存在隐匿性远处转移,并且5年生存率不高。

图7-25 UC进展的不同途径(A)及膀胱癌的分期(B)

注: 目前多数UC(70%~80%)被诊断时为非肌肉浸润性(浅表性),20%~30%为肌肉浸润性。对于非肌肉浸润性UC,癌前病变从具稳定遗传改变的增生(异型增生)发展到浅表性低度或高度恶性UC。肌肉浸润性UC的癌前病变是重度非典型增生/原位癌,这种癌前病变可累积不稳定遗传改变,并发展为浸润性高级别UC(苏木精-伊红染色,放大倍数为×400;病理学图像由Joe Kronz图像修改而来,http://www.hopkinsmedicine.org.)。

7.14.1 病因和发病机制

膀胱癌通常由长期接触致癌物导致尿路上皮DNA损伤的累积所引起(图7-26)。这些致癌物分泌到尿液中,可以被水解酶激活并储存在膀胱中。因此,整个尿路上皮都存在风险,这种称为“区域性癌变”的现象可能就是该病多灶性发生的原因[2,3]。其危险因素包括吸烟、职业性接触芳香胺、摄取砷污染的水、慢性感染、放疗邻近器官、治疗性使用烷类制剂或止痛类药,如对乙酰氨基酚和非那西丁[2-6]

宿主遗传因素进一步改变UC发展的易感性,如相似的吸烟史、职业特点或环境暴露因素可在人群中导致家族聚集性[6-8]。调节参与体内致癌物活化、解毒或灭活等代谢通路的酶的遗传多态性已被确认是UC易感的风险之一(图7-26)[5]。病例-对照研究证实在吸烟者中有高比例的UC患者具有一致的基因型,它们都具有广泛的代谢活化剂作用和高水平P450细胞色素酶(CYP1A2、CTP2D6、CTP3A4)活性,且都参与N-氧化芳香胺向N-脱氧芳香胺的代谢转化,这种转化是芳香胺活化为致癌物质的初始步骤[5,9,10]

N-乙酰化是最为人知的芳香胺解毒途径之一[3,5,8-11]。N-乙酰转移酶为两种遗传性异构体NAT1和NAT2,它们可改变代谢速率,前致癌物被其中和而转变为乙酰化表型NAT1和NAT2异构体,并可改变发生膀胱癌的风险。慢型NAT2的乙酰异构体是一种比较常见的类型,可因不同种族而有所变化,最低的是亚洲人群,这不同于非洲人和高加索人[12]。膀胱癌的风险增加已证实与个体的慢型乙酰化表型相关,这种表型与环境或职业相关性芳香胺诱导的UC有直接联系[3,5,8-11]

谷胱甘肽S转移酶GSTT1和GSTM1参与体内致癌物清除过程,如烟草烟雾中的氧化应激产物和多环芳香碳氢化合物[2,5,6,12]等。GSTT1 和GSTM1 基因缺失可导致GSTT1/GSTM1失效、酶活性完全丧失、致癌物解毒能力受损和肿瘤风险增加,这些改变可能产生多灶性肿瘤,包括膀胱癌[2,5,6]

图7-26 致癌物激活和解毒作用的假说模型,以及解毒机制异常病人的细胞学结局

7.14.2 临床表现

UC患者的典型表现是无痛的、间歇性肉眼或镜下血尿[13-17]。排尿症状改变如尿频、尿急,尿痛是第二个常见症状,这些通常与弥漫性原位癌(CIS)或浸润性膀胱癌相关。梗阻症状往往由局部扩展所致,包括腰痛、下肢水肿、盆腔肿块等。晚期转移症状包括体重减轻、腹痛或骨痛[16]。肿瘤局部较广泛的患者通常有一可触及的包块。肝肿大和锁骨上淋巴结肿大可视为转移的征象。如果肿瘤阻塞了尿道口,偶尔会引起闭塞性骨盆淋巴结肿大和肾积水导致的淋巴水肿。转移可发生在骨骼、肺以及一些不常见的部位如皮肤,表现为溃疡性疼痛结节[18]

7.14.3 诊断

原发性UC可通过膀胱镜检查发现,即窥视整个膀胱尿路上皮,但确诊仍需要病变组织的活检[19]。经尿道切除膀胱肿瘤(TURBT)是常用的治疗手段。在某些情况下,终段尿或膀胱冲洗液脱落细胞学检查可协助诊断,但并不能确诊,因为尿细胞学检查对原位癌的敏感性最大(在90%左右),但对低度恶性肿瘤来说假阴性率较高[20]

已发现一组尿生物学标记作为诊断工具,包括免疫细胞化学、分子生物学和蛋白质组学分析[21-23],分子遗传学、端粒酶的表达[24],肿瘤相关细胞或细胞分泌产物、基因突变和细胞凋亡标记[9,19,25-33]。最近,多种RNA表达谱和表面增强激光解吸离子化飞行时间质谱(SELDI-TOF)也筛选出了可区分膀胱肿瘤与正常尿路上皮的不同蛋白质[34-37]

7.14.4 临床分期

增强CT检查是最常用的分期依据,可提供是否有膀胱外侵袭,盆腔或腹膜后淋巴结转移,内脏、肺或骨转移,肿瘤侵犯或阻塞上尿路等相关信息(表7-16)。增强CT的主要局限性是难以区分炎性或医源性水肿与膀胱外转移,而且对淋巴结累及的检出率相对较低[38]。其他影像学诊断如MRI和骨扫描,可能对膀胱外累及和远处转移病灶的诊断更有帮助,但这些并不是常规诊断方法[38,39]

7.14.5 侵袭转移的模式

有60%的病例发生肿瘤直接播散,其特征是癌细胞直接侵袭到初期黏膜病变以下的固有层和肌层,并有触手样侵袭(25%)或在看似正常黏膜下层的横向扩散(10%)。侵袭肌层的深度与后续发生的血行或淋巴途径转移相关[40] (图7-25)。

表7-16 AJCC膀胱癌分期系统[18,41,42]

(资料来源: AJCCCancer Staging Manual. Sixth ed. New York:Springer-Verlag,2002)

接受膀胱切除术的男性肌肉浸润性膀胱癌患者中有40%累及前列腺和前列腺尿道部。前列腺累及的患者中约40%为间质浸润,而6%间质浸润患者没有前列腺尿道部累及[40]。有间质浸润的患者即使经过根治性切除,后续远处转移的发生率也很高。淋巴结转移常发生较早,这可能独立于血道转移,因为有人观察到10%~15%有淋巴结转移的患者单用手术可治愈[41,42]

肿瘤局部范围以及淋巴结转移情况可直接影响患者手术治疗后的生存[40]。淋巴结转移最常见的部位是盆腔淋巴结,累及膀胱周围(16%)、闭孔(74%)、髂外(65%)、骶前(25%)淋巴结。约20%的患者有髂总淋巴结累及,与前面提到的邻近区域淋巴结受累共同存在[40]。常见的远处转移器官是肝(38%)、肺(36%)、骨(27%)、肾上腺(21%)和肠(13%)[43-45]。骨转移在血吸虫性膀胱癌中更常见[40]。UC引起的骨骼转移病变常表现为成骨性或成骨-溶骨混合性[43]。尽管在尿路上皮癌全身转移的治疗方面有所进展,但只有少数患者可存活5年以上。

7.14.6 浸润性膀胱癌的临床处理

非肌肉浸润性肿瘤常通过TURBT治疗,这种内镜技术可切除肿瘤并保留膀胱功能。依据多种不同的临床和病理因素,可联合辅助膀胱内治疗[46,47]。本章侧重于阐述对肌肉浸润性和转移性膀胱癌的临床治疗(表7-17)。

表7-17 膀胱癌的治疗选择[18,39]

可切除的局限性肌肉浸润性UC患者可选用根治性膀胱切除术加尿流改道术或膀胱保留术联合放疗和化疗等方案[47]。膀胱保留方案所致的复发或残留接近50%,但这一比例可以通过谨慎选择患者来降低。接受膀胱切除术的局部肌肉浸润性UC患者生存因应用甲氨蝶呤、长春新碱、多柔比星、顺铂(MVAC)新辅助治疗而有所改善[33,46,48-57]。根治性膀胱切除术在男性患者包括膀胱广泛切除和前列腺切除,在女性则是膀胱、子宫、卵巢和前阴道壁的切除[58],各大临床中心的围手术期死亡率约为1%。p T2期肿瘤患者的5年无瘤生存率约为65% ~80%,p T3期肿瘤患者则为37%~61%(表7-16),根据累及的局部淋巴结数量和程度的不同可进一步下降到5%~20%[47]。膀胱切除术后骨盆内复发率为6%~10%,这取决于原发肿瘤的分期及盆腔淋巴结累及的程度[58]

7.14.7 转移性泌尿系统上皮癌的临床处理

转移性泌尿系统上皮癌(MVAC)治疗的有效率为15%~35%[59],完全缓解率约13%,平均生存时间8~12个月[59]。然而,MVAC疗法的显著毒性作用[30,33,55-57]导致越来越多地使用吉西他滨、顺铂(GC)作为可接受的替代性姑息治疗[59]。长期存活病例仅出现在一些伴有淋巴结转移的局部进展期患者,而不是远处转移的患者[60]。联合紫杉醇、顺铂、吉西他滨与曲妥珠单抗(一种针对HER-2/neu的人化单克隆抗体)疗法的总体有效率一直稳定在70%,中位至复发时间为9.3个月,中位总生存时间为14.1个月[57]。这些临床处理都是基于对进展期UC分子通路的认识[61] (图7-25、图7-27)。

图7-27(A) UC相关的信号通路

注: 受体酪氨酸激酶(EGFR、VEGFR、FGFR)含有胞质酪氨酸激酶结构域、跨膜域和胞外结构域,可结合到同源配体,并通过多种途径同时激活下游信号通路,从而导致Ras/MAPK、AKT/PI3K的激活;还可激活多个转录因子,导致细胞增殖/生存、EMT、炎症、血管生成和淋巴管增生。ECM-整合素-ILK、TGF-β及GSKβ/β-连环蛋白的相互作用直接影响细胞黏附和(或)细胞骨架的变化,这种作用是通过对Rho GTP酶的作用实现的。这些信号通路的作用集合最终导致肿瘤细胞的侵袭和转移。

图7-27(B) GTP酶循环的调节

注: GDP结合失活的GTP酶主要在细胞质,这是由于GDIs封闭了细胞膜定位所需的C末端尾。GDIs解离后,可发生翻译后修饰,GTP酶转运到血浆膜,在那里它们可以被来源于表面配体-受体系统的GEFs激活,如黏附受体(ECM-整合素)、G蛋白偶联受体(GPCRs)、受体酪氨酸激酶(RTKs)。一旦被GEFs激活,Rho GTP酶可以结合不同的效应蛋白,GEFs可介导对这些效应蛋白的选择,导致下游信号通路的激活。GAPs使Rho GTP酶失活并关闭下游信号通路。

图7-27(C) Ras的活化循环调控

注: Ras是一种单亚基小GTP酶,可行使如二进制信号开关状态的功能。在“关闭”状态,它结合到核苷酸GDP上,而在“开”状态,Ras则结合到GTP上。Ras和其他小G蛋白的激活和失活都是受活化GTP结合和失活GDP结合形式互相转换的循环调控。这种转换过程及导致的Ras活性变化由GEFs和GAPs的活性调控。Ras具有内在GTP酶活性,但这个过程对于高效功能来说太慢了,而且被RasGAP的绑定所稳定。因此,GAPs可调节Ras的活化,GEFs促进Ras活化。在GTP结合构象中, Ras对很多效应因子都有很高的亲和力,这些因子可以使Ras发挥功能,如PI3K和MAPK。通过突变手段,结构性激活的Ras可以抑制GTP水解,从而使Ras锁定在一个永久的“开”状态,已证实在信号转导、增殖和恶性转化中发挥关键作用。

7.14.8 浸润性膀胱癌的分子机制

(1) EGFR

EGFR在UC中表达水平已证实与肿瘤的进展、高病理分级和分期[62,63]以及高复发率[64]呈正相关。肿瘤组织EGFR表达增高与患者的生存率呈负相关。然而,如只比较浸润性UC患者之间的生存,EGFR表达高低患者之间并无显著性差异,表明EGFR过表达可能与非肌肉浸润性向各种浸润性疾病的表型转换相关[65]。有趣的是,基因扩增和基因重排似乎不是UC中EGFR过表达的共同机制[66]。尽管将非肌肉浸润性UC的细胞处理后过表达的无论是突变还是正常的HRAS(RAS:大鼠肉瘤癌基因)、Harvey(HRAS的原癌基因)、Kirsten(KRAS)或neuroblastoma(NRAS),它们在m RNA和蛋白水平均引起EGFR的过表达,这提示HRAS在EGFR转录调控中具有作用,并且对EGFR的信号转导也有影响(图7-27)[67]。体外研究显示EGFR对肿瘤细胞运动和侵袭的多个步骤都有作用,这支持了EGFR过表达与肿瘤进展相关的看法,并且这不只是一个偶然现象[67-69]。在染色体17q11.2-q12上ERBB2基因扩增和蛋白过表达已作为复发进展期UC患者的预后指标[70]。重要的是,UC患者EGF (即EGFR配体)的水平比正常血中的浓度大10倍以上,这可能是EGFR过表达的后果[13,64]

(2) VEGF和VEGFR

VEGF家族包括VEGF-A、VEGF-B、VEGF-C、VEGF-D和Pl GF。在人类中,VEGF-A存在3个异构体(121-、165-、189-氨基酸),可结合并激活VEGFR-1和VEGFR-2两种受体酪氨酸激酶,也是一种强力新生血管诱导剂[71]。Pl GF和VEGF-B仅结合并激活VEGFR-1。病理条件下,Pl GF和VEGF-A的增高可通过VEGFR-1使骨髓来源的单核/巨噬细胞向肿瘤组织或炎性病变处募集,并显著提高病理性血管增生。VEGFR-3主要在淋巴管内皮细胞中表达,VEGF-C和-D对其的阻断作用可调节淋巴管的生成[71]。VEGF-C或VEGF-D高表达可促进肿瘤淋巴管增生,使肿瘤易于向淋巴结转移[71]

VEGFm RNA的高水平表达预示低级或中级T1期UC患者会复发得更早,恶化风险也增加[72]。VEGF-A免疫染色增加与分期提高相关,血清VEGF水平与分期、分级、血管侵袭和原位癌存在呈正相关。VEGF水平≥400pg/ml可高度预测转移性疾病[73-75]。T24膀胱肿瘤细胞株同时表达VEGF-A各亚型和VEGFR-2(KDR/Flk-1),并且可自分泌和旁分泌VEGF诱导有丝分裂信号环已得到确认[24]。这种有丝分裂相关的VEGF途径涉及PKC、神经鞘氨醇激酶(SPK)、RAS(h-ras和n-ras基因的激活,但没有K-Ras基因激活)、丝裂原活化蛋白激酶(ERK1/2)的激活。VEGF诱导的Ras活化是通过Ras-GAP活化物介导的,这独立于RAS-GEF的相互作用[24,76,77]

应用组织微列阵芯片(TMA)研究286例膀胱全切除的肿瘤块发现,VEGF-D的过表达与肿瘤分期及区域淋巴结转移呈正相关,而与无瘤生存率呈负相关。VEGFR-3的过表达特别存在于晚期肿瘤患者亚组,并且与短期无瘤生存率相关。多因素分析显示,VEGF-D和VEGFR-3的表达独立于肿瘤分期和淋巴结转移而成为独立的预后指标[78,79]。将转移的淋巴结与相应的原发肿瘤比较表明,VEGF-D和VEGFR-3在前者有显著高表达[78,79]。因此,VEGF-D和VEGFR-3的表达可能会成为预测肿瘤进展和转移的有力工具,有望成为干预靶点[78,79]

(3) 成纤维细胞生长因子及其受体

酸性FGF1(a FGF)和碱性FGF2(b FGF)是与膀胱癌发病机制相关的两种强有力的FGF异构体,两者均可紧密结合到细胞外基质(ECM)中的硫酸乙酰肝素,被认为是在肿瘤进展中ECM被蛋白酶降解后释放出来的[80,81]。FGF1和FGF2均与UC更高侵袭性表型相关,其免疫组化检测结果可能有助于确定肿瘤是否更易侵袭[80,81]。FGF2先于VEGF在原位癌中过表达,VEGF则是在肌肉浸润性UC晚期进展阶段出现上调[80,82-84]

FGF2的高分子量(HMW)形式含有核定位信号并被证明参与促进癌转移过程。在致癌物诱导的大鼠膀胱癌细胞株(NBT-Ⅱ)中,HMW-FGF2过表达的NBT-Ⅱ克隆表现出致瘤性和肺转移潜能均明显增加,而LMW(低分子量)-胞质FGF2过表达的NBT-Ⅱ克隆及其母系NBT-Ⅱ细胞均无肺转移潜能[85]。HMW-FGF2过表达的NBT-Ⅱ株未见FGF2特异性受体的增加,这表明核型FGF2具有新的靶点,这个靶点可诱导肺转移的生长流程[85]

FGFs及其同源受体已被证实在上皮间质化的互动过程中发挥作用[86]。此外,通常上皮特异性FGFRs的配体定位在间质组分中,而间质特异性FGFRs的配体常局限在上皮来源的细胞中。

FGFR2Ⅲb的m RNA分布于除了伞细胞的整个正常尿路上皮,低水平的FGFR2Ⅲc可在基质中检测到,而与正常尿路上皮相比,膀胱癌细胞株一般不表达FGFR2Ⅲc异构体或少量表达FGFR2Ⅲb[87]。膀胱癌中,FGFR2Ⅲb的低水平表达或完全无表达与不良预后相关[88,89]。转入FGFR2Ⅲb的人T24膀胱癌细胞株,在裸鼠皮下接种后表现出增殖和生长潜能的降低[87]

(4) 转化生长因子-β家族

TGF-β蛋白家族包括TGF-β1~β5、Müllerian抑制底物、抑制素和激活素[90]。而研究发现TGF-β可促进大鼠成纤维细胞的恶性转化[90]。在大多数情况下,它们至少部分通过激活p27和p15抑制细胞增殖,p27和p15可以通过各种细胞周期蛋白依赖性激酶去抑制视网膜母细胞瘤(Rb)蛋白的磷酸化。TGF-β1和TGF-β2的m RNA在惰性上皮肿瘤的表达显著高于侵袭性肿瘤[91]。相反,宫本和他的同事[92]报道称,UC的TGF-β1表达高于正常尿路上皮。TGF-β具有强血管生成活性,而TGF-β1在浸润性和(或)3级肿瘤患者血清中的浓度显著高于非肌肉浸润性患者[93,94]。此外, TGF-β1的过表达与血管生成和炎症标记、进展期病理分期和根治性膀胱切除术后UC患者病情恶化风险等因素相关[95]。然而,膜结合受体TGF-βRI和(或)TGF-βRⅡ的低表达与膀胱癌的分期、分级、进展和生存情况相关[32,95-97]。这些看似相互矛盾的TGF-β及其受体的表达结果可能归因于其不同的细胞来源、差异表达、下游效应因子以及在原发和继发转移部位不同的肿瘤微环境影响下的TGF-β功能变化等因素有关。

(5) p53通路的调控异常

由位于染色体17p13.1[98]位点的TP53基因编码的肿瘤抑制因子p53抑制特定细胞周期阶段(G1~S)的肿瘤进展,这是通过转录激活p21WAFJ/CIP1来起作用的[99]。绝大多数UC患者都表现为一个17p等位基因缺失,另一等位基因突变,造成TP53基因失活,导致突变型p53蛋白核内蓄积[100]。由免疫组化方法(IHC)来检测核p53蛋白的过度表达常作为检测p53突变的替代性标记。基于这种分析,p53的过度表达已被认为与进展风险增加,或非肌肉浸润和肌肉浸润性肿瘤死亡率的增加有关,它们独立于肿瘤分期、分级和淋巴结转移情况[46,98,99,101-110]。TP53基因缺失与分级和分期存在着显著正相关[111-115]。TP53基因突变形成的是截短蛋白(或无蛋白),两个等位基因的纯合性缺失或两个等位基因启动子因甲基化而导致的基因沉默无法用p53蛋白核积累方法检测[116]

p53是否可作为针对UC进展、复发、死亡率及治疗反应的独立预后指标,这点存在争议,主要因为患者的遗传和表观遗传特点、每个研究的病例数目、分析技术以及统计学方法有所不同。一个有995例非肌肉浸润性UC病例的研究证实,p53蛋白的过度表达与疾病更高分级、分期相关[117]。然而,这个指标的预后意义经包括肿瘤大小、分期、分级、数目以及患者的年龄、性别和治疗方法的多因素分析后丢失。在最近的一项研究中[21,118],应用高效电泳技术评估了TP53遗传状态对非肌肉浸润性肿瘤复发和进展的影响。与TP53野生型肿瘤患者相比,复发和恶化的肿瘤患者TP53基因突变的频率更高,TP53基因突变的肿瘤患者无瘤生存期显著缩短。然而,在复发频率和复发时间方面并无显著性统计学差异。

(6) 视网膜母细胞瘤通路的调控异常

肌肉浸润性膀胱癌中已发现存在13号染色体长臂缺失的Rb基因[119]。肌肉浸润性肿瘤患者中发现Rb表达水平异常,且与增殖指数呈正相关,与患者生存呈负相关[105,120,121]。Rb和p53蛋白正常表达的T1期患者预后良好,而其中一种或两种蛋白的异常表达均会显著增加肿瘤的进展[122]。因此,核P53和Rb蛋白的表达状态可以用于对非肌肉浸润性(Ta、T1和原位癌)膀胱癌患者进行分层[123,124]。对这两种基因蛋白表达正常的患者可以应用较保守的治疗方案,而在一个或两个基因异常的患者可能就需要更加积极的治疗方案。

对多种细胞周期调节蛋白(Rb蛋白、p53和p21WAF1/CIP1蛋白,有或没有pl6INK4a)进行分析来预测UC的预后比任何单一指标更准确,并且独立于预后相关的标准临床病理因素[125,126]。除了细胞周期的异常,Rb的失活还可通过抑制E-钙黏蛋白的表达促进肿瘤进展[127]。因此,在UC患者中,13q14杂合性缺失(LOH[81])和Rb蛋白的表达缺失常在高水平分级和分期的肿瘤中检测到[125],这种缺失可能会进一步促进E-钙黏蛋白的表达缺失。

(7) Ras和Rho家族GTP酶及其调控因子

在人类较大的Ras超家族包括100多个分子量较小、鸟嘌呤核苷酸结合蛋白相关单体,共有6个亚族:Ras、Rho、Arf、Rab、Ran和Rad。H-ras分子可激活其他下游信号通路,这些通路与细胞运动和侵袭的增强相关[128]。Rho家族的多种不同功能的基础是三磷酸鸟苷(GTP)/二磷酸鸟苷(GDP)循环(图7-27B,C)。小G蛋白在GTP结合状态和GDP结合状态之间循环。在体内,这个周期被鸟嘌呤核苷酸交换因子(GEFs)严格调控,可刺激GTP与GDP的交换,并激活GTP酶(GAPs),增加GTP水解速率[129]。Rho亚家族已经成为各种信号转导通路的交汇点,这些通路可影响细胞黏附、迁移、细胞周期进程、细胞生存、膜再循环和基因表达。另外,Ras和Rho蛋白之间的相互影响存在于多种生物学过程,包括细胞转化、细胞迁移和EMT[131,130]

Ras:20世纪80年代初,编码p21Ras(一种小GTP酶)的HRAS基因成为首批在T24/T24T尿路上皮细胞系中被发现其突变并被命名的人类原癌基因[132,133]。然而,其在UC的确切作用尚不清楚[134]。在正常尿路上皮,正常的HRAS蛋白在分化过程中消失,它在多层移行上皮细胞的基底(祖)细胞中的染色最强,而在表层(分化的)染色较弱[135]。在UC中,一些研究表明,HRAS的变异与低分级的非肌层浸润乳头状UC[136]相关。另一些研究表明,这种突变存在于UC肿瘤侵袭的多个过程中[67,137]。有报道称HRAS密码子 12 的突变发生在约 40%的膀胱肿瘤中[138,139],另有报道称HRAS蛋白在肿瘤组织中的免疫组化染色与侵袭性之间存在正相关[140]

这些数据表明,UC中过度活化的HRAS可以被突变激活,也可以通过HRAS基因的过表达和(或)上游受体酪氨酸激酶(RTKs)通路的增强来激活[132,133]。HRAS对UC的诱导作用可以进一步通过如下实验证实,即转染HRAS基因的SV40永生化人上皮细胞可以转变成侵袭性移行细胞癌细胞[141,142]。Ras可与丝氨酸/苏氨酸激酶Raf相互作用,它在含一些增强的生长信号通路的肿瘤细胞中得以激活,这些通路存在于非肌肉浸润性、肌肉浸润性、转移性膀胱癌中,随后便伴有MAPK通路的激活[143]。Ras的激活依赖于另外的一种脂质(法尼基)基团加入到羧基端。因此,法尼基转移酶抑制剂和MAPK抑制剂可能是潜在的治疗手段[143](表7-17)。

Ral GTP酶:Ras样(Ral)脒基核苷酸结合蛋白Ral A和Ral B是单体G蛋白Ras家族的两个成员,拥有85%相同氨基酸[144,145]。Ral蛋白参与内吞作用、胞吐作用、肌动蛋白细胞骨架动力学改变和转录。Ral参与以上过程一般都是通过Ral结合蛋白1(Ral BP1)、Sec5、细丝蛋白、磷脂酶D1 (PLDl)以及其他等效应分子介导的。最近的研究也表明Ral蛋白在肿瘤的形成和进展中扮演重要角色[144,145]。具有Ral蛋白质生化特异性的Ral-脒基核苷酸交换因子(Ras-GEF)被确定为致癌基因Ras的直接效应因子。在细胞培养模型系统中,功能获得和丧失研究发现Ral的激活可视为Ras表达的近期后果,这可能有助于促进Ras诱导的致癌转化过程[146,147]。利用小RNA干扰(si RNA)技术的实验结果显示,Ral A和Ral B在调控包括膀胱癌细胞系等多种人类肿瘤细胞系非锚定依赖性增殖、存活及迁移过程中发挥作用[146,148,149]

有报道,Ral激活作为由EGF刺激下人膀胱癌细胞迁移的介导因子[150]。对Ral GTP酶激活、突变状态及其在人类膀胱癌和细胞株中表达情况的研究显示,GTP结合激活的RalA和RalB存在于膀胱癌来源的细胞株中,并且这种激活状态在含有G12VHRAS原癌基因细胞系中表现得更明显[151]。Ral效应因子(如Ral BP1和转移相关蛋白CD24)在膀胱癌中的过度表达[151]支持Ral GTP酶在肿瘤进展中有作用这种观点,并且指出针对它们及其效应因子的干预可能成为合理的治疗手段。

Rho-GTP酶及其调控:Kamai等首次报道Rho /ROCK通路[131,152,153]与膀胱癌的侵袭和转移显著相关[130],并指出RhoA、RhoC和ROCK蛋白在原发瘤和转移淋巴结中的表达显著高于非肿瘤性膀胱和正常淋巴结。这种在肿瘤中的高表达与分化较差、肌肉浸润、淋巴结转移和生存预后不良相关。相反,Rho B的表达与肿瘤的分级和分期呈负相关。

Rho GDP解离抑制剂(GDIs)在细胞质中与Rho蛋白质结合,增加其水溶性并阻断其激活或发挥功能,这种作用通过抑制结合状态的核苷酸解离及其与GEFs、GAPs和效应因子的相互作用来实现。目前已经确定存在3种Rho GDIs,即GDI1、GDI2、GDI3。Rho GDIl是基于它抑制GDP从Rho A、CDC42Hs及Rac1上解离被首次发现[129,153]。目前,很少有证据显示Rho GDIl或Rho GDI3在肿瘤的发生、侵袭、迁移或转移中发挥作用[131,153]。相反,Rho GDI2(也称为D4-GDI或Ly-GDI)在生殖泌尿道上皮细胞中高表达[153],而最初认为这种与Rho GDIl享有67%氨基酸共有序列的因子只在造血细胞中才有表达[154]。重要的是,Rho GDI2的表达降低与人类膀胱癌细胞系侵袭和转移潜能的增加相关[155,156]。表达分析和后续的膀胱癌转移动物模型的机制研究确定Rho GDI2为一种转移抑制基因[157]。在人类膀胱肿瘤中, Rho GDI2表达水平与转移性疾病的发展呈负相关,并且多因素分析认为Rho GDI2是膀胱切除术后膀胱癌患者复发的独立预后指标[155]

(8) 钙黏蛋白和链蛋白

E-钙黏蛋白是一种钙依赖细胞黏附分子,可以介导同型细胞的相互作用,这种作用有助于保持上皮的紧密完整性和尿液在正常尿路上皮不渗透[80,158]。在UC中,通过遗传或表观遗传机制导致的E-钙黏蛋白功能缺失已影响到肿瘤的进展和侵袭性表型[80]。E-钙黏蛋白启动子的甲基化及其所致的基因沉默在早期和晚期UC中均已有报道,并且这种甲基化的频率与肿瘤的进展和不良预后显著相关[159,160]。在另一项研究中,E-钙黏蛋白启动子的过甲基化和16q杂合性缺失(LOH)与肿瘤的分级呈显著正相关[161]。免疫组化结果显示,E-钙黏蛋白表达与肿瘤的分级和分期[162,163]、肌肉浸润深度、淋巴结转移[164-167]、肿瘤复发[168]、5年生存率[169]呈负相关。因此,E-钙黏蛋白的表达状态可视为经膀胱切除术治疗患者疾病进展的预测指标[169,170],既可作为一种独立指标,也可联合其他黏附相关指标[171]

随着E-钙黏蛋白的表达缺失,正常尿路上皮还表达的N-钙黏蛋白出现在p T1期肿瘤病例中,并且在p T2、p T3期病例中表达增加。无瘤生存期和多因素分析显示,N-钙黏蛋白的表达是由非肌肉浸润向肌肉浸润型UC发展的独立预后指标[172]。N-钙黏蛋白在决定UC细胞的侵袭能力方面具有关键性作用,这种作用是通过激活PI3激酶/ Akt信号途径实现的[172,173]。E-钙黏蛋白、β-钙黏蛋白和pl20ctn的异常表达与UC的高分级、晚期及不良生存预后相关[36]。一个或多个上述糖蛋白的膜表达缺失已一致归因于UC的侵袭性表型[36,80]。研究发现,尿和血浆可溶性E-钙黏蛋白(s E-钙黏蛋白)在UC患者的浓度比正常人高,并且在伴有区域或远处淋巴结转移的患者中浓度更高。术前s E-钙黏蛋白与区域淋巴结转移和疾病进展呈独立相关,但与死亡率不相关[32,33]

在NBT-Ⅱ亚硝胺诱发的大鼠膀胱癌细胞系中研究发现,E-钙黏蛋白的表达减少已成为Snail活化、AKT-介导的核因子κB(NF-κB)活化[174]及NF-κB诱导的Snail表达的后果。NF-κB亚基p65的表达足以诱导EMT[158],这验证了在EMT过程中这个通路的作用。NF-κB通路活化与多种人类肿瘤的进展和转移均存在相关性。因此,这个连接EMT过程中的AKT、NF-κB、Snail和E-钙黏蛋白通路的转录网络可能成为抗肿瘤转移治疗的潜在干预靶点[158,174]

最近,已有报道称,Twist作为一种基本的螺旋-环-螺旋转录因子在UC的转移性进展中扮演了重要角色。在226例包括良性、原发癌的膀胱组织及其对应的淋巴结转移病灶的组织芯片研究得出,Twist蛋白在UC样本中的表达显著高于良性组织样本,并且与肿瘤分期、分级呈正相关。另外,Twist在转移灶的表达明显高于其对应的原发灶,更重要的是,Twist与膜E-钙黏蛋白的表达呈负相关[175]。这些研究结果进一步得到一项70例膀胱癌独立研究的支持,其中Twist表达与过往吸烟状况、肿瘤分期和分级呈正相关,而这些均与E-钙黏蛋白表达呈负相关,并且预示着不良的无瘤生存预后[176]

(9) 膀胱癌细胞外基质的转变

多项研究都已报道了UC中基膜(BM)蛋白质和受体的表达。据报道,在侵袭性/转移性膀胱癌中,Sialosyl-Lewisx (SLex)表位的异常糖基化与淋巴结和远处转移及5年生存率降低相关,这种改变可增加肿瘤细胞与内皮性E-选择素的黏附,而没有表达糖化表位的肿瘤却很少发生远处转移[177]

层粘连蛋白和IV型胶原蛋白均已作为BM的标记,然而它们的预后价值并没有被普遍接受[80]。在晚期泌尿上皮肿瘤中,IV型胶原蛋白的缺失与不良生存预后相关,而层粘连蛋白的缺失与转移密切相关[178]。UC中几种BM组分(层粘连蛋白、弹性蛋白酶、纤连蛋白)在组织匀浆中的浓度均显著高于正常尿路上皮。

血清和尿层粘连蛋白的浓度在侵袭性UC中已被证实具有较高的诊断预测价值,这些连同BM层粘连蛋白染色的断裂都表明BM的破坏和缺失[80]。层粘连蛋白5(LN5)可将上皮细胞锚定于下层BM上。对切除的膀胱肿瘤组织和脱落细胞(膀胱冲洗液和尿)样本的研究发现,由于启动子的甲基化,层粘连蛋白5被灭活。甲基化程度和频率均与几个不良预后的参数(肿瘤分期、分级、生长类型、肌肉浸润以及倍增类型)显著相关,并且可协助区分侵袭性和非侵袭性肿瘤[179]

泌尿系上皮肿瘤中的整合素:在正常尿路上皮基底层最常表达的整合素是α6β4。有报道称α6β4的表达改变是UC进展过程中的早期事件,在原位癌中由于整合素β4亚基显著减少,导致α6β1异二聚体占据主要地位等已得到证实[180]。一些免疫组化结果显示,α6β4的缺失在侵袭转移表型的获得过程中扮有重要角色[181,182]。相反,αⅤ整合素亚基在UC中显示出一种分级和分期依赖性过度表达,表明其在细胞增殖和迁移过程中的重要性[183]。最近有人提出α5β1整合素(一种纤连蛋白受体)作为卡介苗的起始细胞信号通路及UC的基因疗法[184,185]

泌尿系上皮癌中的蛋白酶:有研究显示,MMP-2和MMP-9m RNA在侵袭性UC比非肌肉浸润性UC的表达水平高,并且与生存率的降低呈正相关[186],而其在尿中的浓度与肿瘤高分级和分期、组织多肽特异性抗原(TPS)及核基质蛋白-22(NMP-22)等有关[25,187]。MMP-1的水平与疾病进展和生存率下降呈正相关[25,26,31,187-189]。TIMP-1和TIMP-2 (MMP-9和MMP-2抑制剂)等的表达减少与肿瘤高分级和分期相关[25,186,187,189-192]。在侵袭性UC中,组织型和尿激酶型纤溶酶原激活剂(t PA和u PA)、尿激酶型纤溶酶原激活剂受体(u PAR)、纤溶酶原激活物抑制物-1(PAI-1)高表达与不良预后相关[193,194]

透明质酸/透明质酸酶/透明质酸合成酶:透明质酸(HA)是一种不含硫的糖胺聚糖,它是ECM的重要构成成分[195]。在肿瘤组织中,肿瘤相关基质细胞和肿瘤细胞共同导致HA水平的升高。肿瘤细胞源性透明质酸酶(HAase)尤其是透明质酸酶1(HYAL1)可导致HA的降低[196,197]。在肿瘤异种移植模型中,HA是特异性地定位于肿瘤相关基质,而HYAL1则由肿瘤细胞表达[30]

肿瘤细胞分泌透明质酸酶已证实可诱导血管生成,这种作用是通过将HA片段转变成血管生成性透明质酸片段来完成。在2级和3级膀胱癌患者尿液中发现存在血管透明质酸片段,说明UC中HA系统处于活化状态[198]。尿液中HA和透明质酸酶水平与组织水平相关,并在UC患者的尿液中升高,它们联合起来可作为准确的诊断标记[28,29,196,197]。HA的合成发生在细胞膜上,由跨膜HA合成酶(HAS1、HAS2或HAS3)催化[199-202]。HAS1在肿瘤组织中的表达可作为UC复发和治疗失败的预测指标[37,203,204]

HA与受体(如CD44)相互作用可调节细胞黏附、迁移和增殖。肿瘤细胞产生的细胞外HA结合CD44并诱导形成脂质相关的信号复合体,这种复合体包含磷酸化ERBB2 (P-ERBB2)、PI3激酶和CD44,其中前两个已被证实在UC的进展中伴有重要角色。关于CD44,它是一种参与细胞-细胞和细胞外基质相互作用的跨膜糖蛋白,并在肿瘤转移中发挥重要作用[195]。在UC中,标准CD44和CD44v6变种的逐渐丢失与高的病理分期呈正相关,但其在侵袭性UC中作为独立预后指标的价值仍不明晰[80,205-209]

CD24:是糖基化磷脂酰基肌醇连接的表面蛋白,并且在分析Ral A/B缺失的膀胱癌细胞系表达谱时发现它是Ral通路的下游靶点[151,210]。CD24在膀胱癌中与其他肿瘤一样均高表达。CD24在肿瘤细胞株中的功能缺失与细胞增殖能力的降低、非锚定依赖性增长降低、肌动蛋白细胞骨架的变化、凋亡的诱导变化相关[210]。对膀胱癌组织芯片的免疫组化结果分析显示,CD24的表达与患者无瘤生存期的减少显著性相关[210]。Choi等[211]使用抗CD24单克隆抗体并应用免疫组化方法分析了56例p Ta、29例p T1、19例p T2,31例p T3UC患者标本CD24的表达。在正常尿路上皮,CD24定位于管腔细胞层的细胞质,且染色强度非常低。CD24在非侵袭性UC中表达上调,并且高水平的表达与肿瘤分级相关。CD24的表达随着基质/肌肉浸润、分级和分期的增加而上调。

(10) UC中的内皮素轴

内皮素(ETs)是3个含有21氨基酸肽段的家族,即ET-1、ET-2和ET-3,它们通过激活Gαq及Gαs亚型的两种G蛋白偶联受体(GPCRs),即ETA受体(ETAR)和ETB受体(ETBR)来介导其作用。内皮素轴在多种肿瘤和基质细胞的相互作用中具有相似作用,导致自分泌/旁分泌环路的激活,使细胞增殖异常、逃避细胞凋亡、形成新生血管、免疫状态改变、侵袭和转移。

ET轴在UC中的重要性因Rho GDI2调节ET-1显现出来。UC细胞株中Rho GDI2的表达缺失与ET-1的表达上调相关[128]。两个泌尿研究机构在根治性膀胱切除标本中应用反转录聚合酶链反应(RT-PCR)、免疫组化方法、启动子甲基化方法进行了有关ET-1、ETAR、ETBR表达情况的独立性研究[212,213]。发现ET轴在大多数膀胱癌样本中表现出阳性信号,而正常尿路上皮呈阴性结果。与ET-1和ETAR相反,ETBR表达常与良好预后相关[212]。ETBR在UC中表现为甲基化,而正常尿路上皮则不同。甲基化的频率与肿瘤的分级和分期呈正相关,而与肿瘤的侵袭性和疾病的不良预后呈负相关[213]

(11) 泌尿系统上皮癌进展和转移的其他介导因子

与非浸润性乳头状膀胱癌相比,IL-8在肌肉浸润性肿瘤和原位癌中表达上调[74,214]。COX-2在正常膀胱尿路上皮中不表达[215],但在侵袭性UC中存在分子水平上的过表达。COX-2表达也与疾病侵袭性、血管生成的增加、淋巴结转移、疾病复发及疾病特异性死亡率风险的增加相关[213,215-217]。血小板反应蛋白-1(TSP-1)是ECM的一种内源性成分,这种成分通过抑制肿瘤新生血管形成达到肿瘤抑制的功能[75,218]。有关UC细胞株的一项研究显示,TSP可抑制血管生成[75]。UC组织中TSP的免疫组化染色强度与肿瘤分期[75]、平均血管密度[219]、5年生存率呈负相关,提示TSP可以作为预测预后的指标[220]

7.14.9 膀胱癌的动物模型

(1) 致癌物诱发的膀胱癌

N-丁基-N-(4-羟基丁基)亚硝胺(BBN)[221]是N-二甲基亚硝胺的下游代谢产物,这种化合物存在烟草烟雾、某些食物及工业品中,它是小鼠和大鼠的致癌物质。BBN诱导的小鼠肿瘤表现出尿路上皮和鳞状上皮双重特征,往往伴有肌肉浸润和转移,而在大鼠中,BBN诱导的乳头状瘤几乎完全不发生肌层浸润[222,223]。同样,大部分BBN诱导的小鼠肿瘤中包含p53 基因突变,而大部分的大鼠肿瘤则不包含[221,223]

与人膀胱癌相似,在两种啮齿类动物模型中的H-Ras基因突变都处于低频率[221,223]。与野生型相比,BBN诱导的癌变更易发生在转入H-Ras基因的转基因小鼠和p53杂合子小鼠中,其中50%肿瘤发生在p53杂合子小鼠又丢失了另一野生型等位基因的情况下[115,222,224]。BBN诱导的肿瘤也出现EGFR和COX-2的水平升高[223,225],从而使它们成为研究表皮生长因子受体抑制剂和非甾体类抗炎药(NSAIDs)对肿瘤生长和转移影响的有价值的模型系统。小鼠4号染色体上的等位基因缺失(对应人类染色体9p21-22)也很常见,正好对应人类肿瘤发生的9p21-22缺失[226]

比较人类、小鼠、大鼠UC相关的基因表达谱[227]发现,许多人类基因与致癌物诱导的啮齿动物肿瘤的差异表达基因同源,这些基因在人类疾病中刚好也存在表达差异,并且优先地与非肌肉浸润性转向肌肉浸润性疾病的过程相关。由此证明啮齿类动物肿瘤的全基因表达谱与人类侵袭性肿瘤全基因表达谱的关联更加紧密。

在烟草烟雾和某些工业品中发现的化合物4-氨基联苯(4-ABP)也与人类UC相关[222-224]。4-ABP可形成DNA加合物并诱导小鼠膀胱肿瘤生成[228]。有报道显示,4-ABP口饲约4周后,出现了明确的时间和剂量依赖性的DNA加合物和膀胱肿瘤。但是,因为这种模型的p53突变特点与人类肿瘤存在差异,所以这种模型不能完全替代人类疾病[11,229]

(2) 转基因小鼠模型

转基因/基因敲除小鼠使得模型可以通过传代而复制,这种模型可以用来研究UC的发生和肿瘤进展[230]。此外,特定的致癌基因和抑癌基因之间的共同作用可以通过制作双向甚至三相转基因动物模型来研究,这种动物含有两个或两个以上不同的基因异常,并且遗传背景清晰,可避免人类巨大的遗传背景异质性[231,232]。此外,转基因/基因敲除小鼠已被证实是绝佳的用于评估新的诊断性、预防性及治疗性方案的临床前模型[124,233,234]

启动某些特定基因在尿路上皮特异性地表达依赖尿路上皮特异性基因启动子uroplakin(UPK),它在哺乳动物物种中是保守的,并以尿路上皮特异性方式表达[235~246]。UPKⅡ启动子已被用来启动尿路上皮特异性、浓度依赖性尿路上皮基因的表达,包括癌基因、突变的肿瘤抑制基因及生长因子受体基因,这些改变涉及膀胱癌变过程中的多个时期中特定基因改变所扮演的体内角色,例如结构性活化及突变H-Ras基因的上皮性表达[246-248],p53和Rb通路在尿路上皮的失活。它们是通过SV40的大T抗原和表皮生长因子受体的过度表达来实现的[249,250]

(3) 人类肿瘤异种移植模型

人膀胱细胞株在免疫缺陷小鼠上形成的移植瘤有如下优势:可以利用建立好的并有合适特性的人类膀胱癌细胞系,可以进行比致癌物诱发或UPKⅡ作用的自发成瘤模型更快速、成本更低的实验。原位模型概括了膀胱癌进展的两种主要途径,即非肌肉浸润性和肌肉浸润性,对临床前的治疗研究来说非常有价值。另外,转移模型已发展到对过程的理解,以及可进行治疗性方案的测试。

(4) 原位移植瘤

将人类肿瘤细胞直接注入裸鼠的膀胱内,在对正常膀胱上皮细胞层无任何处理的条件下进行植入。肿瘤负荷可以通过非侵入性成像技术得到量化(增强MRI、18F-FDG-PET、各种CT),或将荧光素酶或绿色荧光蛋白(GFP)转染给肿瘤细胞,然后测量荧光素酶活性或GFP荧光而进行活体成像,或通过监测尿中的GFP荧光量来估算肿瘤负担[251,252]。这种模型已被更广泛地用于研究通过膀胱内注射病毒基因的新型疗法[124,134,233,234,251-257]

连续原位接种已用来分离具有侵袭和转移潜能的人类253JTCC细胞系变种[258]。253JB-V和253J肺-IV是其中的两个变种,它们可在膀胱中侵袭性生长或转移至肺,在上述方法下被分别分离出来。这种原位接种手段可比异位(皮下)手段更好地重演人类原发肿瘤的侵袭性行为,从而将细胞系从这些原发瘤中分离出来。例如,原位接种RT4膀胱癌细胞系可保留它们的非肌肉浸润性特征,而在小鼠接种EJ则导致浸润性表型[259,260]。虽然这些行为可反映这些原始的原发瘤特征,细胞株即来源于这些原发瘤,而皮下接种这两种细胞均不表现它们之间的任何差别。

7.14.10 转移的分析

(1) 自发性转移分析

自发转移发生于尿路上皮的“原位”种植的肿瘤细胞中[258],或发生在致癌物诱发的UC中,或发生在尿路上皮特异性(UPKⅡ驱动)的带有癌基因的SV40大T抗原过度表达模型中[246-250]。原位注射移植肿瘤模型可以重演人类肿瘤的生物学行为,包括肿瘤组织学行为、血管形成、基因表达和转移生物学行为[261]。然而,使用这种方法前也应考虑到植入过程中靶组织的机械性破坏,可导致肿瘤细胞进入血液循环,播种到距实验部位较远的地方。移植瘤也可能通过扩散性的因子来抑制继发性肿瘤的生长[262],从而抑制继发性病变的形成,继而掩盖对早期转移阶段的分析[261,262]。此外,种植过程还依赖于培养细胞的一系列体外和体内通道,有了这些通道,转移性变种细胞才能在人工的或异质的环境中选择出来[261]

不幸的是,除了分析的简便和有效外,大部分自发肿瘤模型(致癌物质诱发或遗传工程模型如UPKⅡ-SV40-T)潜伏期较长,且转移发生率不高[261]

(2) 实验性转移分析

实验性转移包括将肿瘤细胞直接注射到静脉循环,不需要使细胞自发地脱离原发肿瘤而进入血液循环[263,264]。此方法可研究转移级联终端阶段,使得较短时间内即出现继发灶的细胞克隆增生[263,264]。T24T细胞株是T24细胞株[150,157,266]的高度致瘤性和转移性变种,将其进行反复尾静脉注射[265]可产生转移性更强的细胞株FL(从肺中来)系列FL1、FL2、FL3[267]。类似的方法也用来使转染荧光素酶的UMUC3细胞(UMUC3-Luc)产生肺转移株UMUC3-Lul1-3系列(Theodorescu,未发表数据)。

将基因表达谱与临床数据经基因表达分析联系起来,发现Rho GDI2可作为转移抑制因子[266]。Rho GDI2下游的两种候选基因也是经类似的方法鉴定出来的。这两个基因,即内皮素-1(ET-1)[128]和神经介素U(NMU)[268],可能称为治疗转移性膀胱癌的绝佳靶点。膀胱骨转移的T24/TSU-Prl细胞株及两个亚株,即TSU-Pr1-B1和TSU-Pr1-B2,它们都是连续地经心腔注射循环而形成的骨转移株[269]。这种模型使得在不同继发灶转移的分子机制得以渐渐显露,并使第一个模型在 UC 患者中观察到成骨-溶骨性骨表型[270-272]

7.14.11 总结及展望

转移性UC的主要问题是其不良的预后和生存。本章介绍了促进侵袭和转移过程蛋白质分子的认识,这些成果为治疗提供了合理的靶标。对原发瘤、尿液或血清,以及UC细胞株的遗传学和表观遗传学的不断研究,将进一步识别有利于预测肿瘤的分级、分期和生存的生物标记。这些生物标记可用于与传统的临床和病理指标一起组成分类图,以进一步提高风险分层及识别高危患者的水平,并确定适当的治疗方案——无论是膀胱内治疗、膀胱切除术后的选择性辅助化疗,还是新辅助化疗。此外,新概念已应用到膀胱癌的遗传和转录数据中,使得个体化预测治疗和药物反应成为现实[273]。这将是革命性的变化,即膀胱癌和其他肿瘤患者都会得到个体化医疗服务和药物治疗。

(张晓飞 译,钦伦秀 审校)

参考文献

[1]Jemal A,et al. Cancer statistics,2008. CA Cancer J Clin,2008,58: 71-96.

[2]McGrath M,et al. Polymorphisms in GSTT1,GSTM1,NAT1 andNAT2 genes and bladder cancer risk in men and women. BMCCancer,2006,6: 239.

[3]Sanderson S,et al. Joint effects of the N-acetyltransferase 1 and 2( NAT1 and NAT2) genes and smoking on bladder carcinogenesis:a literature-based systematic HuGE review and evidence synthesis.Am J Epidemiol,2007,166: 741-751.

[4]Johansson SL,et al. Epidemiology and etiology of bladder cancer.Semin Surg Oncol,1997,13( 5) : 291-298.

[5]Hung RJ,et al. GST,NAT,SULT1A1,CYP1B1 geneticpolymorphisms,interactions with environmental exposures andbladder cancer risk in a high-risk population. Int J Cancer,2004,110( 4) : 598-604.

[6]Vineis P,et al. Genetic susceptibility according to three metabolicpathways in cancers of the lung and bladder and in myeloidleukemias in nonsmokers 10. 1093 /annonc /mdm 109. Ann Oncol,2007,18( 7) : 1230-1242.

[7]Jhamb M,et al. Urinary tract diseases and bladder cancer risk: acase-control study. Cancer Causes Control,2007,18 ( 8 ) :839-845.

[8]Vineis P,et al. Current smoking,occupation,N-acetyltransferase-2 and bladder cancer: a pooled analysis of genotype-based studies.Cancer Epidemiol Biomarkers Prev,2001,10( 12) : 1249-1252.

[9]Fanlo A,et al. Urinary mutagenicity,CYP1A2 and NAT2 activityin textile industry workers. J Occup Health,2004,46 ( 6 ) :440-447.

[10]Vaziri SA,et al. Variation in enzymes of arylamine procarcinogenbiotransformation among bladder cancer patients and controlsubjects. Pharmacogenetics,2001,11( 1) : 7-20.

[11] Poirier MC,et al. Aromatic amine DNA adduct formation inchronically-exposed mice: considerations for human comparison.Mutat Res,1997,376( 1-2) : 177-184.

[12] Dong LM,et al. Genetic susceptibility to cancer: the role ofpolymorphisms in candidate genes. JAMA,2008,299 ( 20 ) :2423-2436.

[13]Messing EM,et al. Normal and malignant human urothelium: invitro effects of epidermal growth factor. Cancer Res,1987,47( 9) : 2230-2235.

[14]Varkarakis MJ,et al. Superficial bladder tumor. Aspects ofclinical progression. J Urol,1974,4( 4) : 414-420.

[15] Messing EM,et al. Hematuria screening for bladder cancer. JOccup Med,1990,32( 9) . 838-845.

[16] Foresman WH,et al. Bladder cancer: natural history,tumormarkers,and early detection strategies. Semin Surg Oncol,1997,13( 5) : 299-306.

[17]Messing EM,et al. Home screening for hematuria: results of amulticlinic study. J Urol,1992,148( 2 Pt 1) : 289-292.

[18]Konety BR. Urothelial carcinoma: cancers of the bladder,ureter,and renal pelvis. In: Tanagho EA,McAninch JW,eds. Smith'sGeneral Urology. New York: McGraw Hill,2007: 308-327.

[19] Konety BR. Molecular markers in bladder cancer: a criticalappraisal. Urol Oncol,2006,24( 4) : 326-337.

[20] van Rhijn BW,van der Poel HG,van der Kwast TH. Urinemarkers for bladder cancer surveillance: a systematic review. EurUrol,2005,47( 6) : 736-748.

[21]Hutterer GC,et al. Urinary cytology and nuclear matrix protein 22in the detection of bladder cancer recurrence other than transitionalcell carcinoma. BJU Int,2008,101( 5) : 561-565.

[22]Raitanen MP. The role of BTA stat test in follow-up of patientswith bladder cancer: results from Finn Bladder studies. World JUrol,2008,26( 1) : 45-50.

[23]Nguyen CT,et al. Defining the role of NMP22 in bladder cancersurveillance. World J Urol,2008,26( 1) : 51-58.

[24]Wu W,et al. VEGF receptor expression and signaling in humanbladder tumors. Oncogene,2003,22( 22) : 3361-3370.

[25]di Carlo A, et al. Urinary gelatinase activities ( matrixmetalloproteinases 2 and 9) in human bladder tumors. Oncol Rep,2006,15( 5) : 1321-1326.

[26] Eissa S,et al. Noninvasive diagnosis of bladder cancer bydetection of matrix metalloproteinases ( MMP-2 and MMP-9) andtheir inhibitor ( TIMP-2 ) in urine. Eur Urol,2007,52 ( 5 ) :1388-1396.

[27]Hautmann S,et al. Immunocyt and the HA-HAase urine tests forthe detection of bladder cancer: a side-by-side comparison,EurUrol,2004,46( 4) : 466-471.

[28] Lokeshwar VB,et al. HA-HAase urine test. A sensitive andspecific method for detecting bladder cancer and evaluating itsgrade. Urol Clin North Am,2000,27( 1) : 53-61.

[29]Lokeshwar VB,et al. Urinary bladder tumor markers. Urol Oncol,2006,24( 6) : 528-537.

[30]Lokeshwar VB, et al. HYAL1 hyaluronidase: a moleculardeterminant of bladder tumor growth and invasion. Cancer Res,2005,65( 6) : 2243-2250.

[31]Monier F,et al. Urinary release of 72 and 92 kDa gelatinases,TIMPs,N-GAL and conventional prognostic factors in urothelialcarcinomas. Eur Urol,2002,42( 4) : 356-363.

[32]Matsumoto K,et al. Preoperative plasma soluble E-cadherinpredicts metastases to lymph nodes and prognosis in patientsundergoing radical cystectomy. J Urol,2003,170( 6 Pt 1) : 2248-2252.

[33] Shariat SF,et al. Urinary levels of soluble E-cadherin in thedetection of transitional cell carcinoma of the urinary bladder. EurUrol,2005,48( 1) : 69-76.

[34]Holyoake A,et al. Development of a multiplex RNA urines test for the detection and stratification of transitional cell carcinoma of thebladder. Clin Cancer Res,2008,14( 3) : 742-749.

[35]Xie XY,et al. Analysis of hTERT expression in exfoliated cellsfrom patients with bladder transitional cell carcinomas using SYBRgreen real-time fluorescence quantitative PCR. Ann Clin Biochem,2007,44( Pt 6) : 523-528.

[36]Clairotte A,et al. Expression of E-cadherin and alpha-,beta-,gamma-catenins in patients with bladder cancer,identification ofgamma-catenin as a new prognostic marker of neoplasticprogression in T1 superficial urothelial tumors. Am J Clin Pathol,2006,125( 1) : 119-126.

[37] Kuncova J,et al. Expression of CD44v6 correlates with cellproliferation and cellular atypia in urothelial carcinoma cell lines5637 and HT1197. Folia Biol ( Praha) ,2005,51( 1) : 3-11.

[38]Schrier B,et al. Imaging in the assessment of urinary bladdercancer. In: Lerner S,et al,eds. Textbook of Bladder Cancer.Abingdon: Taylor and Francis,2006: 191-205.

[39]Badalament R,et al. Imaging for transitional cell carcinomas. In:Vogelzang N, et al, eds. Comprehensive Textbook ofGenitourinary Oncology. Baltimore: Williams and Wilkins,2000: 357.

[40]Messing EM. Urothelial tumors of the bladder,in: Wein A,et al,eds. Campbell-Walsh Urology. Vol 3. Philadelphia: SaundersElsevier,2007: 75.

[41]Bochner BH. Intravesical bacillus Calmette-Guerin combined withelectromotive mitomycin for high-risk superficial bladder cancer.Nat Clin Pract Oncol,2006,3( 9) : 474-475.

[42]Bochner BH. Gene therapy in bladder cancer. Curr Opin Urol,2008,18( 5) : 519-523.

[43]Goldman SM,et al. Metastatic transitional cell carcinoma from thebladder: radiographic manifestions. Am J Roentgenol,1979,132( 3) : 419-425.

[44]Babaian RJ,et al. Metastases from transitional cell carcinoma ofurinary bladder. Urology,1980,16( 2) : 142-144.

[45] Sengelov L, et al. Pattern of metastases in relation tocharacteristics of primary tumor and treatment in patients withdisseminated urothelial carcinoma. J Urol,1996,155 ( 1 ) :111-114.

[46]Moonen PM,et al. Risk stratification of Ta,Tis,T1 cancer. In:Lerner S,et al,eds. Textbook of Bladder Cancer. Oxford: Taylorand Francis,2006: 281-286.

[47]Herr HW. Surgical factors in the treatment of superficial andinvasive bladder cancer. Urol Clin North Am,2005,32( 2) : 157-164.

[48]Grossman HB. Immunotherapy for bladder cancer. Is the black boxbecoming grayer? J Urol,2003,169( 5) : 1709.

[49]Grossman HB,Dinney CP. If cystectomy is insufficient,what is anurologist to do? Urol Oncol,2003,21( 6) : 475-478.

[50]Grossman HB,et al. Neoadjuvant chemotherapy plus cystectomycompared with cystectomy alone for locally advanced bladdercancer. N Engl J Med,2003,349( 9) : 859-866.

[51]Black PC,et al. Neoadjuvant chemotherapy for bladder cancer.World J Urol,2006,24( 5) : 531-542.

[52]Sawhney R,et al. Neoadjuvant chemotherapy for muscle-invasivebladder cancer: a look ahead. Ann Oncol,2006,17 ( 9 ) :1360-1369.

[53]McLaren DB. Neoadjuvant chemotherapy in transitional-cellcarcinoma of the bladder. Clin Oncol,2005,7( 7) : 503-507.

[54]Walz J,et al. Adjuvant chemotherapy for bladder cancer does notalter cancer-specific survival after cystectomy in a matched casecontrolstudy. BJU Int,2008,101( 11) : 1356-1361.

[55]Muramaki M,et al. Prognostic significance of adjuvant cisplatinbasedcombination chemotherapy following radical cystectomy inpatients with invasive bladder cancer. Int J Urol,2008,15( 4) :314-318.

[56]Lokeshwar VB,et al. Urinary uronate and sulfated glycosaminoglycanlevels: markers for interstitial cystitis severity. J Urol,2005,174( 1) :344-349.

[57]Ok JH,et al. Is the use of anything but MVAC justified in theevidence-based medicine era? Curt Opin Urol,2005,15 ( 5 ) :312-314.

[58]Stein JP,et al. Radical cystectomy in the treatment of invasivebladder cancer: long-term results in 1,054 patients. J Clin Oncol,2001,19( 3) : 666-675.

[59]von der Maase H,et al. Long-term survival results of a randomizedtrial comparing gemcitabine plus cis-platin,with methotrexate,vinblastine,doxorubicin,plus cisplatin in patients with bladdercancer. J Clin Oncol,2005,23( 21) : 4602-4608.

[60]Lehmann J,et al. Gemcitabine /cisplatin vs MVAC,5 yearsurvival outcome of the phase Ⅲ study of chemotherapy ofadvanced urothelial carcinoma in Germany. Urologe A,2003,42( 8) : 1074-1086.

[61]Kim WJ,et al. Molecular biomarkers in urothelial bladder cancer.Cancer Sci,2008,99( 4) : 646-652.

[62]Lipponen P,et al. Expression of epidermal growth factor receptorin bladder cancer as related to established prognostic factors,oncoprotein ( c-erbB-2,p53) expression and long-term prognosis.Br J Cancer,1994,69( 6) : 1120-1125.

[63]Gorgoulis VG, et al. Molecular and immunohisto-chemicalevaluation of epidermal growth factor receptor and c-erb-B-2 geneproduct in transitional cell carcinomas of the urinary bladder: astudy in Greek patients. Mod Pathcl,1995,8( 7) : 758-764.

[64]Chow NH,et al. Significance of urinary epidermal growth factorand its receptor expression in human bladder cancer. AnticancerRes,1997,17( 2B) : 1293-1296.

[65]Nguyen PL,et al. Expression of epidermal growth factor receptorin invasive transitional cell carcinoma of the urinary bladder. Amultivariate survival analysis. Am J Clin Pathol,1994,101( 2) :166-176.

[66] Sauter G,et al. Epidermal-growth-factor-receptor expression isassociated with rapid tumor proliferation in bladder cancer. Int JCancer,1994,57( 4) : 508-514.

[67]Theodorescu D,et al. H-ras induction of the invasive phenotyperesults in up-regulation of epidermal growth factor receptors andaltered responsiveness to epidermal growth factor in humanpapillary transitional cell carcinoma cells. Cancer Res,1991,51( 16) : 4486-4491.

[68] Theodorescu D,et al. Inhibition of human bladder cancer cellmotility by genistein is dependent on epidermal growth factorreceptor but not p21ras gene expression. Int J Cancer,1998,78( 6) : 775-782.

[69]Theodorescu D,Laderoute KR,Guiding KM. Epidermal growthfactor receptor-regulated human bladder cancer motility is in part aphosphatidylinositol 3-kinase-mediated process. Cell GrowthDiffer,1998,9( 11) : 919-928.

[70]Ravery V,et al. Evaluation of epidermal growth factor receptor,transforming growth factor alpha,epidermal growth factor andc-erbB-2 in the progression of invasive bladder cancer. Urol Res,1997,25( 1) : 9-17.

[71] Shibuya M. Vascular endothelial growth factor dependent andindependent regulation of angiogenesis. BMB Rep,2008,41( 4) :278-286.

[72]Crew JP,et al. Vascular endothelial growth factor is a predictor ofrelapse and stage progression in superficial bladder cancer. CancerRes,1997,57( 23) : 5281-5285.

[73]Bouck N,et al. Anti-cancer dividends from captopril and otherinhibitors of angiogenesis. J Nephrol,1998,11( 1) : 3-4.

[74]Campbell CL,et al. Expression of multiple angiogenic cytokines incultured normal human prostate epithelial cells: predominance ofvascular endothelial growth factor. Int J Cancer,1999,80 ( 6) :868-874.

[75]Campbell SC,et al. Molecular mediators of angiogenesis inbladder cancer. Cancer Res,1998,58( 6) : 1298-1304.

[76]Shu X,et al. Sphingosine kinase mediates vascular endothelialgrowth factor-induced activation of ras and mitogen-activatedprotein kinases. Mol Cell Biol,2002,22( 22) : 7758-7768.

[77] Wu XX,et al. Telomerase activity in urine after transurethralresection is not a predictive marker for recurrence of superficialbladder cancer. Int J Urol,2003,10( 2) : 117-118.

[78]Herrmann E,et al. VEGF-C,VEGF-D and Flt-4 in transitionalbladder cancer: relationships to clinicopathological parameters andlong-term survival. Anticancer Res,2007,27( 5A) : 3127-3133.

[79]Herrmann E,et al. New markers for pharmacological targeting inbladder cancer with lymph node metastasis. Aktuelle Urol,2007,38( 5) : 392-397.

[80]Gontero P,et al. Metastasis markers in bladder cancer: a review ofthe literature and clinical considerations. Eur Urol,2004,46( 3) :296-311.

[81]Knowles MA. Molecular subtypes of bladder cancer: Jekyll andHyde or chalk and cheese? Carcinogenesis,2006,27 ( 3 ) :361-373.

[82]Miyake H,et al. Expression of basic fibroblast growth factor isassociated with resistance to cisplatin in a human bladder cancercell line. Cancer Lett,1998,123( 2) : 121-126.

[83]Miyake H,et al. Basic fibroblast growth factor regulates matrixmetalloproteinases production and in vitro invasiveness in humanbladder cancer cell lines. J Urol,1997,157( 6) : 2351-2355.

[84]Munro NP,et al. Fibroblast growth factors and their receptors intransitional cell carcinoma. J Urol,2003,169( 2) : 675-682.

[85]Thomas-Mudge RJ,et al. Nuclear FGF-2 facilitates cell survival invitro and during establishment of metastases. Oncogene,2004,23( 27) : 4771-4779.

[86]Ornitz DM,et al. Fibroblast growth factors. Genome Biol,2001,2( 3) : 1-12.

[87]Ricol D,et al. Tumour suppressive properties of fibroblast growthfactor receptor 2-Ⅲb in human bladder cancer. Oncogene,1999,18( 51) : 7234-7243.

[88]Bernard-Pierrot I,et al. Inhibition of human bladder tumour cellgrowth by fibroblast growth factor receptor 2b is independent of itskinase activity. Involvement of the carboxy-terminal region of thereceptor. Oncogene,2004,23( 57) : 9201-9211.

[89]Bernard-Pierrot I,et al. Oncogenic properties of the mutated formsof fibroblast growth factor receptor 3b. Carcinogenesis,2006,27( 4) : 740-747.

[90]Sporn MB,Roberts AB. Transforming growth factor-beta. Multipleactions and potential clinical applications. JAMA,1989,262( 7) : 938-941.

[91]Coombs LM,et al. Reduced expression of TGF beta is associatedwith advanced disease in transitional cell carcinoma. Br J Cancer,1993,67( 3) : 578-584.

[92]Miyamoto H,et al. Expression of transforming growth factor-beta 1in human bladder cancer. Cancer,1995,75( 10) : 2565-2570.

[93]Eder IE,et al. Expression of transforming growth factors beta-L,beta 2 and beta 3 in human bladder carcinomas. Br J Cancer,1997,75( 12) : 1753-1760.

[94]Eder IE,et al. Transforming growth factors-beta 1 and beta 2 inserum and urine from patients with bladder carcinoma. J Urol,1996,156( 3) : 953-957.

[95] Kim JH,et al. Predictive value of expression of transforminggrowth factor-beta 1 and its receptors in transitional cell carcinomaof the urinary bladder. Cancer,2001,92( 6) : 1475-1483.

[96]Shariat SF, et al. The addition of urinary urokinasetypeplasminogen activator to urinary nuclear matrix protein 22 andcytology improves the detection of bladder cancer. J Urol,2003,170( 6 Pt 1) : 2244-2247.

[97]Shariat SF,et al. Correlation of cyclooxygenase-2 expression withmolecular markers,pathological features and clinical outcome oftransitional cell carcinoma of the bladder. J Urol,2003,170( 3) :985-989.

[98]Mitra AP,et al. Molecular pathways in invasive bladder cancer:new insights into mechanisms, progression, and targetidentification. J Clin Oncol,2006,24( 35) : 5552-5564.

[99] Mitra AP,et al. p53 and retinoblastoma pathways in bladdercancer. World J Urol,2007,25( 6) : 563-571.

[100]Iggo R,et al. Increased expression of mutant forms of p53oncogene in primary lung cancer. Lancet,1990,335 ( 8691) :675-679.

[101]Esrig D,et al. p53 nuclear protein accumulation correlates withmutations in the p53 gene,tumor grade,and stage in bladdercancer. Am J Pathol,1993,143( 5) : 1389-1397.

[102]Spruck CH 3rd,et al. Two molecular pathways to transitional cellcarcinoma of the bladder. Cancer Res,1994,54( 3) : 784-788.

[103]Esrig D, et al. Accumulation of nuclear p53 and tumorprogression in bladder cancer. N Engl J Med,1994,331( 19) :1259-1264.

[104] Stavropoulos NE,et al. CD44 standard form expression as apredictor of progression in high-risk superficial bladder tumors.Int Urol Nephrol,2001,33( 3) : 479-483.

[105]Ioachim E, et al. Immunohistochemical expression ofretinoblastoma gene product ( Rb) ,p53 protein,MDM2,c-erbB-2,HLA-DR and proliferation indices in human urinary bladdercarcinoma. Histol Histopathol,2000,15( 3) : 721-727.

[106]Lianes P,et al. Altered patterns of MDM2 and TP53 expressionin human bladder cancer. J Natl Cancer Inst,1994,86 ( 17) :1325-1330.

[107] Sarkis AS,et al. Nuclear overexpression of p53 protein intransitional cell bladder carcinoma: a marker for diseaseprogression. J Natl Cancer Inst,1993,85( 1) : 53-59.

[108]Sarkis AS,et al. Prognostic value of p53 nuclear overexpressionin patients with invasive bladder cancer treated with neoadjuvantMVAC. J Clin Oneal,1995,13( 6) : 1384-1390.

[109]Stavropoulos NE,et al. Prognostic significance of p53,bcl-2 andKi-67 in high-risk superficial bladder cancer. Anticancer Res,2002,22( 6B) : 3759-3764.

[110]Peyromaure M,et al. Prognostic value of p53 overexpression inbladder tumors treated with Bacillus Calmette-Guerin. Expert RevAnticancer Ther,2002,2( 6) : 667-670.

[111]Fujimoto K,et al. Frequent association of p53 gene mutation ininvasive bladder cancer. Cancer Res, 1992, 52 ( 6 ) :1393-1398.

[112] Soini Y,et al. p53 immunohistochemistry in transitional cellcarcinoma and dysplasia of the urinary bladder correlates withdisease progression. Br J Cancer,1993,68( 5) : 1029-1035.

[113]Moch H,et al. p53 and erbB-2 protein overexpression areassociated with early invasion and metastasis in bladder cancer.Virchows Arch A Pathol Anat Histopathol,1993,423 ( 5 ) :329-334.

[114]Matsuyama H,et al. p53 deletion as a genetic marker inurothelial tumor by fluorescence in situ hybridization. CancerRes,1994,54( 23) : 6057-6060.

[115] Yamamoto S,et al. Frequent mutations of the p53 gene andinfrequent H-and K-ras mutations in urinary bladder carcinomasof NON/Shi mice treated with N-butyl-N-( 4-hydroxybutyl )nitrosamine. Carcinogenesis,1995,16( 10) : 2363-2368.

[116] Thomas CY,et al. Molecular markers of prognosis and noveltherapeutic strategies for urothelial cell carcinomas. World JUrol,2006,24( 5) : 565-578.

[117]Malats N,et al. p53 as a prognostic marker for bladder cancer: ameta-analysis and review. Lancet Oneol,2005,6( 9) : 678-686.

[118]Ecke TH,et al. TP53 gene mutations as an independent markerfor urinary bladder cancer progression. Int J Mol Med,2008,21( 5) : 655-661.

[119]Cairns P,et al. Loss of heterozygosity at the RB locus is frequentand correlates with muscle invasion in bladder carcinoma.Oncogene,1991,6( 12) : 2305-2309.

[120]Cordon-Cardo C,et al. Genetic studies and molecular markers ofbladder cancer. Semin Surg Oneal,1997,13( 5) : 319-327.

[121]Cordon-Cardo C,et al. Altered expression of the retinoblastomagene product: prognostic indicator in bladder cancer. J NatlCancer Inst,1992,84( 16) : 1251-1256.

[122]Grossman HB,et al. p53 and RB expression predict progressionin Tl bladder cancer. Clin Cancer Res,1998,4( 4) : 829-834.

[123]Cordon-Cardo C. p53 and RB: simple interesting correlates ortumor markers of critical predictive nature? J Clin Oncol,2004,22( 6) : 975-977.

[124]Kikuchi E,et al. Inhibition of orthotopic human bladder tumorgrowth by lentiviral gene transfer of endo-statin. Clin Cancer Res,2004,10( 5) : 1835-1842.

[125]Shariat SF,et al. p53,p21,pRB,and p16 expression predictclinical outcome in cystectomy with bladder cancer. J Clin Oncol,2004,22( 6) : 1014-1024.

[126]Shariat SF,et al. Predictive value of cell cycle biomarkers innonmuscle invasive bladder transitional cell carcinoma. J Urol,2007,177( 2) : 481-487.

[127] Arima Y,et al. Rb depletion results in deregulation of Ecadherinand induction of cellular phenotypic changes that arecharacteristic of the epithelial-to-mesenchymal transition. CancerRes,2008,68( 13) : 5104-5112.

[128]Titus B,et al. Endothelin axis is a target of the lung metastasissuppressor gene RhoGDI2. Cancer Res,2005,65 ( 16 ) :7320-7327.

[129] Nitz MD,et al. Invasion and metastasis models for studyingRhoGDI2 in bladder cancer. Methods Enzymol,2008,439: 219-233.

[130]Kamai T,et al. Significant association of Rho /Rock pathway withinvasion and metastasis of bladder cancer. Clin Cancer Res,2003,9( 7) : 2632-2641.

[131] Ellenbroek SI,et al. Rho GTPases: functions and associationwith cancer. Clin Exp Metastasis,2007,24( 8) : 657-672.

[132]Taparowsky E,et al. Activation of the T24 bladder carcinomatransforming gene is linked to a single amino acid change.Nature,1982,300( 5894) : 762-765.

[133] Parada LF,et al. Human EJ bladder carcinoma oncogene ishomologue of Harvey sarcoma virus ras gene. Nature,1982,297( 5866) : 474-478.

[134]Kawano H,et al. New potential therapy for orthotopic bladder carcinoma by combining HVJ envelope with doxorubicin. CancerChemother Pharmacal,2008,61( 6) : 973-978.

[135]Theodorescu D. Molecular pathogenesis of urothelial bladdercancer. Histol Histopathol,2003,18( 1) : 259-274.

[136]Fitzgerald JM,et al. Identification of H-ras mutations in urinesediments complements cytology in the detection of bladdertumors. J Natl Cancer Inst,1995,87( 2) : 129-133.

[137]Theodorescu D,et al. Overexpression of normal and mutatedforms of HRAS induces orthotopic bladder invasion in a humantransitional cell carcinoma. Proc Natl Acad Sci USA,1990,87( 22) : 9047-9051.

[138]Czerniak B,et al. Concurrent mutations of coding and regulatorysequences of the Ha-ras gene in urinary bladder carcinomas. HumPathal,1992,23( 11) : 1199-1204.

[139] Czerniak B,et al. Ha-ras gene codon 12 mutation and DNAploidy in urinary bladder carcinoma. Br J Cancer,1990,62( 5) :762-763.

[140]Fontana D,et al. Evaluation of c-ras oncogene product ( p21) insuperficial bladder cancer. Eur Urol,1996,29( 4) : 470-476.

[141]Christian BJ,et al. EJ /ras neoplastic transformation of simianvirus 40-immortalized human uroepithelial cells: a rare event.Cancer Res,1990,50( 15) : 4779-4786.

[142]Pratt CI,et al. Neoplastic progression by EJ /ras at different stepsof transformation in vitro of human uroepithelial cells. CancerRes,1992,52( 3) : 688-695.

[143]Wallerand H,et al. Molecular targeting in the treatment of eitheradvanced or metastatic bladder cancer or both according to thesignalling pathways. Curr Opin Urol,2008,18( 5) : 524-532.

[144]Bos JL. Ras-like GTPases. Biochim Biophys Acta,1997,1333( 2) : MI9-31.

[145]Feig LA. Ral-GTPases: approaching their 15 minutes of fame.Trends Cell Bioi,2003,13( 8) : 419-425.

[146]Bodemann BO,et al. Ral GTPases and cancer: linchpin supportof the tumorigenic platform. Nat Rev Cancer,2008,8( 2) : 133-140.

[147]Hamad NM,et al. Distinct requirements for Ras oncogenesis inhuman versus mouse cells. Genes Dev,2002,16 ( 16 ) :2045-2057.

[148]Chien Y,et al. RAL GTPases are linchpin modulators of humantumour-cell proliferation and survival. EMBO Rep,2003,4( 8) :800-806.

[149]Oxford G,et al. RalA and RalB: antagonistic relatives in cancercell migration. Cancer Res,2005,65( 16) : 7111-7120.

[150]Gildea JJ,et al. The role of RalA in epidermal growth factorreceptor-regulated cell motility. Cancer Res,2002,62( 4) : 982-985.

[151]Smith SC,et al. Expression of Ral GTPases,their effectors,andactivators in human bladder cancer. Clin Cancer Res,2007,13( 13) : 3803-3813.

[152]Oxford G,et al. The role of Ras superfamily proteins in bladdercancer progression. J Urol,2003,170( 5) : 1987-1993.

[153] Theodorescu D. Molecular biology of invasive and metastaticurothelial cancer. In: Lerner S,Schoenberg M,Sternberg C,eds. Textbook of Bladder Cancer. Oxford: Taylor and Francis,2006: 147-156.

[154]Scherle P,et al. Ly-GDI,a GDP-dissociation inhibitor of theRhoA GTP-binding protein, is expressed preferentially inlymphocytes. Proc Natl Acad Sci USA,1993,90 ( 16 ) :7568-7572.

[155]Theodorescu D, et al. Reduced expression of metastasissuppressor RhoGDI2 is associated with decreased survival forpatients with bladder cancer. Clin Cancer Res,2004,10( 11) :3800-3806.

[156]Seraj MJ,et al. The relationship of BRMS1 and RhoGDI2 geneexpression to metastatic potential in lineage related human bladdercancer cell lines. Clin Exp Metastasis,2000,18( 6) : 519-525.

[157]Gildea JJ,et al. RhoGDI2 is an invasion and metastasissuppressor gene in human cancer. Cancer Res,2002,62( 22) :6418-6423.

[158]Thiery JP. Epithelial-mesenchymal transitions in tumourprogression. Nat Rev Cancer,2002,2( 6) : 442-454.

[159] Bornman DM,et al. Methylation of the E-cadherin gene inbladder neoplasia and in normal urothelial epithe-lium fromelderly individuals. Am J Pathol,2001,159( 3) : 831-835.

[160] Dhawan D,et al. Evidence for the early onset of aberrantpromoter methylation in urothelial carcinoma. J Pathol,2006,209( 3) : 336-343.

[161]Horikawa Y,et al. Hypermethylation of an E-cadherin ( CDH1)promoter region in high-grade transitional cell carcinoma of thebladder comprising carcinoma in situ. J Urol,2003,169 ( 4) :1541-1545.

[162]Syrigos KN,et al. E-cadherin expression in bladder cancer usingformalin-fixed, paraffin-embedded tissues: correlation withhistopathological grade,tumour stage and survival. Int J Cancer,1995,64( 6) : 367-370.

[163]Imao T,et al. Dominant role of E-cadherin in the progression ofbladder cancer. J Urol,1999,161( 2) : 692-698.

[164]Sanchez-Carbayo M,et al. Molecular profiling of bladder cancerusing cDNA microarrays: defining histogenesis and biologicalphenotypes. Cancer Res,2002,62( 23) : 6973-6980.

[165]Sun W,et al. E-cadherin expression in urothelial carcinoma insitu, superficial papillary transitional cell carcinoma, andinvasive transitional cell carcinoma. Hum Pathol,2002,33( 10) : 996-1000.

[166] Nakopoulou L,et al. Prognostic value of E-cadherin,betacatenin,P120ctn in patients with transitional cell bladder cancer.Anticancer Res,2000,20( 6B) : 4571-4578.

[167] Sun W,et al. E-cadherin expression in invasive urothelialcarcinoma. Ann Diagn Pathol,2004,8( 1) : 17-22.

[168]Mahnken A,et al. E-cadherin immunoreactivity correlates withrecurrence and progression of minimally invasive transitional cellcarcinomas of the urinary bladder. Oneal Rep,2005,14 ( 4) :1065-1070.

[169]Shariat SF,et al. E-cadherin expression predicts clinical outcomein carcinoma in situ of the urinary bladder. Urology,2001,57( 1) : 60-65.

[170]Byrne RR, et al. E-cadherin immunostaining of bladdertransitional cell carcinoma,carcinoma in situ and lymph nodemetastases with long-term follow up. J Urol,2001,165 ( 5 ) :1473-1479.

[171] Rao J,et al. Tissue microarray analysis of cytoskeletal actinassociatedbiomarkers gelsolin and E-cadherin in urothelialcarcinoma. Cancer,2002,95( 6) : 1247-1257.

[172]Rieger-Christ KM,et al. Novel expression of N-cadherin elicits invitro bladder cell invasion via the Akt signaling pathway.Oncogene,2004,23( 27) : 4745-4753.

[173]Lascombe I,et al. N-cadherin as a novel prognostic marker ofprogression in superficial urothelial tumors. Clin Cancer Res,2006,12( 9) : 2780-2787.

[174]Julien S,et al. Activation of NF-kappaB by Akt upregulates Snailexpression and induces epithelium mesenchyme transition.Oncogene,2007,26( 53) : 7445-7456.

[175] Zhang Z,et al. Significance of TWIST expression and itsassociation with E-cadherin in bladder cancer. Hum Pathol,2007,38( 4) : 598-606.

[176]Fondrevelle ME,et al. The expression of Twist has an impact onsurvival in human bladder cancer and is influenced by thesmoking status. Ural Oneal,2009,27( 3) : 268-276.

[177]Numahata K,et al. Sialosyl-Le( x) expression defines invasiveand metastatic properties of bladder carcinoma. Cancer,2002,94( 3) : 673-685.

[178] Al-Sukhun S,et al. Molecular biology of transitional cellcarcinoma. Crit Rev Oneal Hematol,2003,47( 2) : 181-193.

[179]Sathyanarayana UG,et al. Molecular detection of non-invasiveand invasive bladder tumor tissues and exfoliated cells by aberrantpromoter methylation of laminin-5 encoding genes. Cancer Res,2004,64( 4) : 1425-1430.

[180]Harabayashi T,et al. Reduction of integrin beta4 and enhancedmigration on laminin in association with intraepithelial spreadingof urinary bladder carcinomas. J Urol,1999,161 ( 4 ) :1364-1371.

[181]Grossman HB,et al. Expression of the alpha6beta4 integrinprovides prognostic information in bladder cancer. Oneal Rep,2000,7( 1) : 13-16.

[182]Liebert M,et al. Urothelial differentiation and bladder cancer.Adv Exp Med Bioi,1999,462: 437-448.

[183] Kausch I,et al. Molecular aspects of bladder cancer Ⅲ.Prognostic markers of bladder cancer. Br J Urol,2002,41( 1) :15-29.

[184] Kausch I,et al. Immune gene therapy in urology. Curr UrolRep,2002,3( 1) : 82-89.

[185] Chen F,et al. Bacillus Calmette-Guerin initiates intracellularsignaling in a transitional carcinoma cell line by cross-linkingalpha5 beta1 integrin. J Urol,2003,170( 2 Pt 1) : 605-610.

[186] Kanayama H. Matrix metalloproteinases and bladder cancer. JMed Invest,2001,48( 1-2) : 31-43.

[187]Papathoma AS, et al. Prognostic significance of matrixmetalloproteinases 2 and 9 in bladder cancer. Anticancer Res,2000,20( 3B) : 2009-2013.

[188]Monier F,et al. Gelatinase isoforms in urine from bladder cancerpatients. Clin Chim Acta,2000,299( 1-2) : 11-23.

[189]Grignon DJ, et al. High levels of tissue inhibitor ofmetalloproteinase-2 ( TIMP-2) expression are associated with pooroutcome in invasive bladder cancer. Cancer Res,1996,56( 7) :1654-1659.

[190]Wallard MJ,et al. Comprehensive profiling and localisation of thematrix metalloproteinases in urothelial carcinoma. Br J Cancer,2006,94( 4) : 569-577.

[191]Gakiopoulou H,et al. Tissue inhibitor of metalloproteinase-2 as amultifunctional molecule of which the expression is associatedwith adverse prognosis of patients with urothelial bladdercarcinomas. Clin Cancer Res,2003,9( 15) : 5573-5581.

[192] Furukawa A,et al. Role of the matrix metalloproteinase andtissue inhibitors of metalloproteinase families in noninvasive andinvasive tumors transplanted in mice with severe combinedimmunodeficiency. Urology,1998,51( 5) : 849-853.

[193]Seddigh M,et al. Expression of UPA and UPAR is associatedwith the clinical course of urinary bladder neoplasms. Int JCancer,2002,99( 5) : 721-726.

[194]Span PN,et al. Components of the plasminogen activator systemand their complexes in renal cell and bladder cancer: comparisonbetween normal and matched cancerous tissues. BJU Int,2008,102( 2) : 177-182.

[195]Naor D,et al. CD44: structure,function,and association withthe malignant process. Adv Cancer Res,1997,71: 241-319.

[196]Lokeshwar VB,et al. Urinary hyaluronic acid and hyaluronidase:markers for bladder cancer detection and evaluation of grade. JUrol,2000,163( 1) : 348-356.

[197] Lokeshwar VB,et al. Differences in hyaluronic acid-mediatedfunctions and signaling in arterial,microvessel,and vein-derivedhuman endothelial cells. J Biol Chem,2000,275 ( 36 ) :27641-27649.

[198]Lokeshwar VB,et al. Tumor-associated hyaluronic acid: a newsensitive and specific urine marker for bladder cancer. CancerRes,1997,57( 4) : 773-777.

[199]Hautmann SH,et al. Hyaluronic acid and hyaluronidase 2 newbladder carcinoma markers. Urology,2001,40( 2) : 121-126.

[200] Hautmann SH,et al. Elevated tissue expression of hyaluronicacid and hyaluronidase,validates the HA-HAase urine test forbladder cancer. J Urol,2001,165( 6 Pt 1) : 2068-2074.

[201] Lokeshwar VB,et al. Regulation of hyaluronidase activity byalternative mRNA splicing. J Biol Chem,2002,277 ( 37 ) :33654-33663.

[202] Golshani R,et al. Hyaluronic acid synthase-1 expression regulates bladder cancer growth,invasion,and angiogenesisthrough CD44. Cancer Res,2008,68( 2) : 483-491.

[203]Kuncova J, et al. Expression of CD44s and CD44v6 intransitional cell carcinomas of the urinary bladder: comparisonwith tumour grade,proliferative activity and p53 immunoreactivityof tumour cells. APMIS,2007,115( 11) : 1194-1205.

[204] Muramaki M,et al. Overexpression of CD44V8-10 in humanbladder cancer cells decreases their interaction with hyaluronicacid and potentiates their malignant progression. J Urol,2004,171( 1) : 426-430.

[205] Sugino T,et al. Progressive loss of CD44 gene expression ininvasive bladder cancer. Am J Pathol,1996,149( 3) : 873-882.

[206]Garcia del Muro X,et al. Prognostic value of the expression ofE-cadherin and beta-catenin in bladder cancer. Eur J Cancer,2000,36( 3) : 357-362.

[207]Lipponen P,et al. Expression of CD44 standard and variant-v6proteins in transitional cell bladder tumours and their relation toprognosis during a long-term follow-up. J Pathol,1998,186( 2) : 157-164.

[208]Hong RL,et al. Correlation of expression of CD44 isoforms andE-cadherin with differentiation in human urothelial cell lines andtransitional cell carcinoma. Cancer Lett,1995,89( 1) : 81-87.

[209]Hong RL,et al. Expressions of E-cadherin and exon v6-containing isoforms of CD44 and their prognostic values in humantransitional cell carcinoma. J Urol,1995,153( 6) : 2025-2028.

[210]Smith SC, et al. The metastasis-associated gene CD24 isregulated by Ral GTPase and is a mediator of cell proliferationand survival in human cancer. Cancer Res,2006,66( 4) : 1917-1922.

[211]Choi YL,et al. Overexpression of CD24: association withinvasiveness in urothelial carcinoma of the bladder. Arch PatholLab Med,2007,131( 2) : 275-281.

[212]Wulfing C,et al. Expression of endothelin-I and endothelin-Aand -B receptors in invasive bladder cancer. Oncol Rep,2005,13( 2) : 223-228.

[213]Yates DR,et al. Promoter hypermethylation identifies progressionrisk in bladder cancer. Clin. Cancer Res,2007,13 ( 7 ) :2046-2053.

[214] Black PC,et al. Bladder cancer angiogenesis and metastasistranslationfrom murine model to clinical trial. Cancer MetastasisRev,2007,26( 3-4) : 623-634.

[215] Margulis V,et al. Expression of cyclooxygenase-2 in normalurothelium, and superficial and advanced transitional cellcarcinoma of bladder. J Urol,2007,177( 3) : 1163-1168.

[216]Hammam OA,et al. Possible role of cyclooxygenase-2 inschistosomal and non-schistosomal-associated bladder cancer.Medscape J Med,2008,10( 3) : 60.

[217]Zu X,et al. Vascular endothelial growth factor C expression inbladder transitional cell cancer and its relationship to lymph nodemetastasis. BJU Int,2006,98( 5) : 1090-1093.

[218] Ioachim E,et al. Thrombospondin-1 expression in urothelialcarcinoma: prognostic significance and association with p53alterations, tumour angiogenesis and extracellular matrixcomponents. BMC Cancer,2006,6: 140.

[219] Goddard JC,et al. Reduced thrombospondin-1 at presentationpredicts disease progression in superficial bladder cancer. EurUrol,2002,42( 5) : 464-468.

[220] Grossfeld GD,et al. Thrombospondin-1 expression in bladdercancer: association with p53 alterations,tumor angiogenesis,andtumor progression. J Natl Cancer Inst,1997,89( 3) : 219-227.

[221] Cohen SM. Urinary bladder carcinogenesis. Toxicol Pathol,1998,26( 1) : 121-127.

[222]Enomoto T,et al. H-ras activation and ras p21 expression inbladder tumors induced in F344 /NCr rats by N-butyl-N-( 4-hydro-xybutyl) nitrosamine. Carcinogenesis,1990,11 ( 12 ) :2233-2238.

[223]Grubbs CJ, et al. Celecoxib inhibits N-butyl-N-( 4-hydroxybutyl ) -nitrosamine-induced urinary bladder cancers inmale B6D2F1 mice and female Fischer-344 rats. Cancer Res,2000,60( 20) : 5599-5602.

[224]Fujita J,et al. Activation of H-ras oncogene in rat bladder tumorsinduced by N-butyl-N-( 4-hydro-xybutyl ) nitrosamine. J NatlCancer Inst,1988,80( 1) : 37-43.

[225] Elmarjou A,et al. Involvement of epidermal growth factorreceptor in chemically induced mouse bladder tumourprogression. Carcinogenesis,2000,21( 12) : 2211-2218.

[226]Miyao N,et al. Role of chromosome 9 in human bladder cancer.Cancer Res,1993,53( 17) : 4066-4070.

[227] Williams PD,et al. Molecular credentialing of rodent bladdercarcinogenesis models. Neoplasia,2008,10( 8) : 838-846.

[228]Besaratinia A,et al. Mutational signature of the proximate bladdercarcinogen N-hydroxy-4-acetylaminobiphenyl: inconsistency withthe p53 mutational spectrum in bladder cancer. Cancer Res,2002,62( 15) : 4331-4338.

[229]Flammang TJ,et al. DNA adduct levels in congenic rapid andslow acetylator mouse strains following chronic administration of4-aminobiphenyl. Carcinogenesis,1992,13( 10) : 1887-1891.

[230]Wu X,et al. Mouse models for multistep tumorigenesis. TrendsCell Biol,2001,11( 11) : S2-9.

[231]Carver BS,et al. Mouse modeling in oncologic preclinical andtranslational research. Clin Cancer Res,2006,12 ( 18 ) :5305-5311.

[232]McCormick F. Cancer gene therapy: fringe or cutting edge? NatRev Cancer,2001,1( 2) : 130-141.

[233] Kikuchi E,et al. Highly efficient gene delivery for bladdercancers by intravesically administered replication-competentretroviral vectors. Clin Cancer Res,2007,13 ( 15 Pt 1 ) :4511-4518.

[234] Kikuchi E,et al. Detection and quantitative analysis of earlystage orthotopic murine bladder tumor using in vivo magneticresonance imaging. J Urol,2003,170( 4 Pt 1) : 1375-1378.

[235]Finch JL,et al. Cloning of the human uroplakin 1B cDNA and analysis of its expression in urothelial-tumor cell lines andbladder-carcinoma tissue. Int J Cancer,1999,80( 4) : 533-538.

[236]Lin JH,et al. Precursor sequence,processing,and urotheliumspecificexpression of a major 15-kDa protein subunit ofasymmetric unit membrane. J Biol Chem,1994,269( 3) : 1775-1784.

[237]Lin JH,et al. A tissue-specific promoter that can drive a foreigngene to express in the suprabasal urothelial cells of transgenicmice. Proc Natl Acad Sci USA,1995,92( 3) : 679-683.

[238]Moll R,et al. Uroplakin Ⅲ,a specific membrane protein ofurothelial umbrella cells,as a histological markers for metastatictransitional cell carcinomas. Verh Dtsch Ges Pathol,1993,77:260-265.

[239] Ogawa K,et al. Immunohistochemical analysis of uroplakins,urothelial specific proteins,in ovarian Brenner tumors,normaltissues,and benign and neoplastic lesions of the female genitaltract. Am J Pathol,1999,155( 4) : 1047-1050.

[240] Ogawa K,et al. Comparison of uroplakin expression duringurothelial carcinogenesis induced by N-butyl-N-( 4-hydroxybutyl)nitrosamine in rats and mice. Toxicol Pathol,1999,27 ( 6 ) :645-651.

[241]Olsburgh J,et al. Human uroplakin lb gene structure andpromoter analysis. Biochim Biophys Acta,2002,1576 ( 1-2 ) :163-170.

[242]Wu RL,et al. Uroplakin Ⅱ gene is expressed in transitional cellcarcinoma but not in bilharzial bladder squamous cell carcinoma:alternative pathways of bladder epithelial differentiation and tumorformation. Cancer Res,1998,58( 6) : 1291-1297.

[243]Wu XR,et al. Mammalian uroplakins. A group of highlyconserved urothelial differentiation-related membrane proteins. JBioI Chem,1994,269( 18) : 13716-13724.

[244]Xu X,et al. Uroplakin as a marker for typing metastatictransitional cell carcinoma on fine-needle aspiration specimens.Cancer,2001,93( 3) : 216-221.

[245]Yu J,et al. Uroplakin Ⅰ: a 27-kD protein associated with theasymmetric unit membrane of mammalian urothelium. J Cell Biol,1990,111( 3) : 1207-1216.

[246]Zhang ZT,et al. Urothelium-specific expression of an oncogenein transgenic mice induced the formation of carcinoma in situ andinvasive transitional cell carcinoma. Cancer Res,1999,59( 14) : 3512-3517.

[247]Cheng J,et al. Overexpression of epidermal growth factor receptorin urothelium elicits urothelial hyperplasia and promotes bladdertumor growth. Cancer Res,2002,62( 14) : 4157-4163.

[248]Zhang ZT,et al. Role of Ha-ras activation in superficial papillarypathway of urothelial tumor formation. Oncogene,2001,20( 16) : 1973-1980.

[249]Gao J,et al. p53 deficiency provokes urothelial proliferation andsynergizes with activated Ha-ras in promoting urothelialtumorigenesis. Oncogene,2004,23( 3) : 687-696.

[250] Saban MR,et al. Lymphatic vessel density and function inexperimental bladder cancer. BMC Cancer,2007,7: 219.

[251]Watanabe T,et al. An improved intravesical model using humanbladder cancer cell lines to optimize gene and other therapies.Cancer Gene Ther,2000,7( 12) : 1575-1580.

[252]Zhou JH,et al. Visualizing superficial human bladder cancer cellgrowth in vivo by green fluorescent protein expression. CancerGene Ther,2002,9( 8) : 681-686.

[253] Hadaschik BA,et al. Intravesical chemotherapy of high-gradebladder cancer with HTI-286,a synthetic analogue of the marinesponge product hemiasterlin. Clin Cancer Res,2008,14 ( 5) :1510-1518.

[254] Hadaschik BA,et al. Intravesically administered antisenseoligonucleotides targeting heat-shock protein-27 inhibit the growthof non-muscle-invasive bladder cancer. BJU Int,2008,102( 5) :610-616.

[255] Hadaschik BA,et al. Paclitaxel and cisplatin as intravesicalagents against non muscle invasive bladder cancer. BJU Int,2008,101( 11) : 1347-1355.

[256]Hadaschik BA,et al. A validated mouse model for orthotopicbladder cancer using transurethral tumour inoculation andbioluminescence imaging. BJU Int,2007,100( 6) : 1377-1384.

[257]Ioachim E,et al. A clinicopathological study of the expression ofextracellular matrix components in urothelial carcinoma. BJU Int,2005,95( 4) : 655-659.

[258] Dinney CP,et al. Isolation and characterization of metastaticvariants from human transitional cell carcinoma passaged byorthotopic implantation in athymic nude mice. J Urol,1995,154( 4) : 1532-1538.

[259]Ahlering TE,et al. A new in vivo model to study invasion andmetastasis of human bladder carcinoma. Cancer Res,1987,47( 24 Pt 1) : 6660-6665.

[260]Slaton JW,et al. Correlation of metastasis-related gene expressionand relapse-free survival in patients with locally advanced bladdercancer treated with cystectomy and chemotherapy. J Urol,2004,171( 2 Pt 1) : 570-574.

[261]Khanna C,et al. Modeling metastasis in vivo. Carcinogenesis,2005,26( 3) : 513-523.

[262]O'Reilly MS,et al. Angiostatin: a novel angiogenesis inhibitorthat mediates the suppression of metastases by a Lewis lungcarcinoma. Cell,1994,79( 2) : 315-328.

[263]Welch DR. Technical considerations for studying cancermetastasis in vivo. Clin Exp Metastasis,1997,15 ( 3 ) :272-306.

[264] Steeg PS. Tumor metastasis: mechanistic insights and clinicalchallenges. Nat Med,2006,12( 8) : 895-904.

[265]Fidler IJ. Selection of successive tumour lines for metastasis. NatNew Biol,1973,242( 118) : 148-149.

[266]Harding MA,et al. RhoGDI2: a new metastasis suppressor gene:discovery and clinical translation. Urol Oncol,2007,25 ( 5 ) :401-406.

[267]Nicholson BE,et al. Profiling the evolution of human metastatic bladder cancer. Cancer Res,2004,64( 21) : 7813-7821.

[268]Wu Y,et al. Neuromedin U is regulated by the metastasissuppressor RhoGDI2 and is a novel promoter of tumor formation,lung metastasis and cancer cachexia. Oncogene,2007,26( 5) :765-773.

[269]Chaffer CL,et al. Upregulated MT1-MMP /TIMP-2 axis in theTSU-Pr1-B1 /B2 model of metastatic progression in transitionalcell carcinoma of the bladder. Clin Exp Metastasis,2005,22( 2) : 115-125.

[270] Chaffer CL,et al. Aberrant fibroblast growth factor receptorsignaling in bladder and other cancers. Differentiation,2007,75( 9) : 831-842.

[271]Chaffer CL,et al. Mesenchymal-to-epithelial transition facilitatesbladder cancer metastasis: role of fibroblast growth factorreceptor-2. Cancer Res,2006,66( 23) : 11271-11278.

[272]Chaffer CL,et al. Mesenchymal to epithelial transition indevelopment and disease. Cells Tissues Organs,2007,185( 13) : 7-19.

[273]Lee JK,et al. A strategy for predicting the chemosensitivity ofhuman cancers and its application to drug discovery. Proc NatlAcad Sci USA,2007,104( 32) : 13086-13091.

免责声明:以上内容源自网络,版权归原作者所有,如有侵犯您的原创版权请告知,我们将尽快删除相关内容。

我要反馈