叶盛威医生的科普号

一位来自湖北鄂州农村的男性二年前得了结肠癌行手术治疗后，粘液血便3月并排便困难2周入院，直肠肿块距离肛门5cm，病灶很大，侵犯肠壁整圈，活动性较差，而且右侧腹部原切口处出现包块，当地医院认为可能术后复发并转移，建议病人行“改道”手术，病人不愿失去肛门，到湖北省肿瘤医院看病，肿瘤医院腹部外科的专家在做了肛门检查和腹部、直肠磁共振检查以后，经过多学科会诊后，认为病人为异时性多原发癌（结肠和直肠），右下腹包块可能为癌切口种植，但不排除炎性包块，给了病人如下治疗意见：一是马上手术切除右下腹包块并作预防性回肠造口；二是再接受放化疗，一月后再接受直肠手术，有保肛可能，因为肿瘤在放化疗后可能有较大退缩，可能可以增加保肛的机会。病人听从了医生的建议，在肿瘤医院外科切除右下腹包块表明为炎性包块，使病人心理包袱明显减轻，说明以前的肿瘤没有复发，再在放疗科接受放化疗，治疗结束后，复查磁共振，病灶明显退缩，之后手术成功地为病人保住了肛门，手术后继续接受化疗。目前术后已经半年，病人没有出现肿瘤复发或转移。大肠癌是一种常见的恶性肿瘤，在西方发达国家其发病率居恶性肿瘤谱的第二位。随着人民生活水平不断提高，饮食习惯的改变，中国大肠癌的发病率逐年上升，并且上升的速度很快，就城市而言，大肠癌从20世纪80年代的第四位常见恶性肿瘤上升到90年代的第三位。目前认为，大肠癌发病率上升和现代人高脂肪、高蛋白、低纤维素的饮食习惯以及缺少体力活动有密切关系。人们总认为得了癌症等于宣判死刑。其实，60%~70%的大肠癌病人是可以完全治愈的，影响疗效的关键在于病人是不是接受了合理、规范、最佳的多学科综合治疗。多学科综合治疗已经成为国际上肿瘤诊治的“金标准”。为了应对日益增多的大肠癌患者对医疗机构带来的挑战，湖北省肿瘤医院成立了大肠癌诊治中心(本人为中心专家兼秘书)，由外科、内科、放疗、影像科和病理科等的本专业顶尖专家组成，将使大肠癌的诊疗水平得到极大的提升。以下为大肠癌疑难病例会诊的流程。大肠癌疑难病例会诊流程一、时间：每周一下午，2点30分开始。二、地点：湖北省肿瘤医院住院大楼胃肠外科教室（武汉市洪山区卓刀泉南路116号）。三、参加人员：大肠癌指导小组成员，每专业至少一名，秘书提前落实；胃肠外科所有成员；指导小组成员如有特殊情况需要提前2-3天向指导小组秘书请假，组长同意后备案；四、准备会诊操作流程：（一）、住院患者：各科室协调人员于周五前将患者资料提交指导小组秘书，统计人数；同时协调人员需要进一步了解病史，而且尽可能收集齐全的辅助检查资料（CT、MRI等），于讨论当日进行病史汇报（电子PPT文档）并将病历带至讨论地点；统一收集PPT文档，安排讨论顺序；每周一上午会议秘书将需讨论的病人姓名、住院号等告知病理科及放射科，作讨论准备。（二）、门诊患者1. 指导小组成员（主治医师以上职称）预约患者流程；周五之前到指导小组秘书处预约登记（患者姓名、疾病诊断、联系电话）；根据讨论病例数目，给予确定的讨论日期通知，同时于周一下午1点至胃肠外科教室预约登记确认后，家属到门诊挂号登记；会议秘书采集患者病史、辅助检查信息，完成门诊病史书写和讨论内容的初步登记；病人携带的辅助检查资料如CT等留存，待讨论。2.其他人员预约患者流程首先需要由申请医生联系本指导小组成员（主治医师以上职称），了解病史并确认有无讨论价值；如有讨论价值，由指导小组成员到指导小组秘书处登记，其它流程同前。五、会诊讨论过程（一）、病史汇报人员：住院病例：由各科室主管医师汇报。门诊病例：由会议秘书统一汇报；（二）、汇报内容：1.基本病史，体检以及阳性检查结果；2.影像学和病理学的检查报告等，对相关辅助资料如影像学和病理学的检查图片进行现场阅片，此两项分别由放射科和病理科医生完成；3.诊治的疑难点，讨论目的。（三）、讨论规则：1. 设主席或主持人职位，掌握讨论主题，分配发言机会；2. 一般指导小组组长综合会诊意见；出现意见不一致时，根据表决结果决定会诊意见；七、会议记录讨论意见由会议秘书统一记录，作为大肠癌指导小组的工作内容。八、结果的登记和书写：应用单病种疑难病例讨论记录单，需要三名主要人员（内科、外科、放疗科）签字，载入病历。所有结果应注明属大肠癌指导小组会诊意见。十一、会议秘书的确定： 原则上由各临床科室协调员轮流担任，也可由各科室青年医生担任，由指导小组秘书安排确认。大肠癌疑难病例会诊流程图主治医师准备病人临床资料（电子PPT文档）每周五报大肠癌协作组秘书告知放射科和病理科病人信息每周一下午会诊（指导小组每专业至少一名成员参加）根据讨论结果书写会诊意见

结直肠癌的发病率逐渐升高，大约20-30%的病人在诊断时即出现肝转移，而且60-70%的病人出现异时肝转移，其中约40%的病人的转移仅局限于肝脏[1]。目前，手术切除是临床上治疗结直肠癌肝转移的有效手段，5年生存率为30-40%，但仅有6.1-20%的肝转移的病人可以获得治愈性切除[1，2]。随着化疗药物和新的靶向药物的逐渐应用，一些治疗方案可以使肿瘤的缓解率可以高达50%，不仅使病人的无进展生存期（PFS）延长，而且使一些起初不可切除的结直肠癌肝转移降期并获得治愈性切除[2]。 本研究通过应用CAPEOX方案作为一线化疗方案，治疗仅局限于肝脏的不可切除性结直肠癌肝转移的病人，评估治疗效果包括缓解率、可切除率，并分析影响预后的因素。1. 资料与方法病人选择：2005年7月至2007年8月共收治的转移局限于肝脏的结直肠癌病人，年龄小于70岁，组织学证实为结直肠癌，转移灶局限于肝脏，经外科医生评估不可切除，以前未接受过含奥沙利铂和卡培他滨方案的化疗，卡氏评分大于70分，肝肾功能显示胆红素正常，AST＜120U/L而且肌酐小于120mmol/L。排除标准：合并肝硬变或妊娠的病例，或6月内有心肌梗塞病史者，或拒绝化疗者。其中不可切除的肝转移标准为：（1）、转移病灶累及所有三支肝静脉主干、肝后下腔静脉和门静脉的分叉处；（2）、转移病灶累及一侧门静脉分支和对侧的肝静脉属支；（3）、转移灶超过6个并分布于两叶；（4）、转移灶过大需切除大于75%的肝脏导致术后出现肝功能衰竭。治疗方法：应用CapeOX方案化疗，即奥沙利铂按135mg/m2静脉滴注，第一天，卡培他滨1000 mg/m2口服，2次/日，第1-14天，每3周为一周期。化疗一直持续至病情进展，或毒副作用不可忍受，或病人拒绝继续治疗。每2周期评估一次，如可行治愈性切除则考虑手术。如四周期后评估为肿瘤缩小大于25%但小于50%，则行经皮行肝动脉栓塞化疗一次，药物以奥沙利铂和氟尿嘧啶为主。手术中行腹腔广泛探查，结合术中超声，力争达到治愈性切除，即切除所有肿瘤而且保证无瘤大于1cm。一般手术日期与末次化疗时间大于一月。另外，术中可以应用射频消融治疗手术难以切除的3个以下或直径小于3cm的转移灶。术后一般继续CapeOX方案化疗，最低2周期，一般要求一共8周期（包括术前化疗周期数）。手术后每3月评估，包括肿瘤标记物，胸腹CT等，如复发病灶可再切除则再次手术。所有病例如病情进展，可以考虑介入治疗或其他化疗方案，如CPT-11或CPT-11加爱必妥。统计学分析：疾病生存期定义为自化疗开始至任何原因死亡。应用Kaplan-Meier方法计算生存寿命，Log-Rank方法比较不同治疗方法的预后。应用卡方检验或Fisher’s检验分析影响化疗后治愈性切除的因素。2. 结果病人特征见表1。起初符合局限于肝脏但不可切除的肝转移病人一共53例，有2例因体力状况差未接受化疗，其余51例纳入本研究。这些病例接受1至8周期（中位数4）的化疗，4例出现3度以上的化疗副反应，其中2例出现3度手足综合征，呕吐及白细胞减少各1例，无化疗相关的死亡。化疗后肿瘤稳定10例，部分缓解22例，缓解率为43.1%（22/51）。目前所有临床因素均不能预测化疗效果与切除率。无淋巴结转移的病例治愈性切除率更高，但统计学无意义（Fisher’s exact test, p=0.276）。所有化疗有缓解的22例病人中，术前评估18例可以行治愈性手术切除，这些病人一共行63周期化疗，平均化疗4周期，化疗至手术时间约4月（范围2-6月），其中有2例病人接受了介入治疗。18例中实际仅15例行治愈性切除，其中包括肝右叶切除4例，左叶切除5例，联合肝段切除6例，其中有3例手术中结合射频消融治疗一些存在于对侧肝叶的小病灶，根据术后CT评估无活性，也判定为治愈性切除；另外3例中，1例拒绝手术，1例行超声引导下射频消融治疗，另1例均为肿瘤数目过多（超过20个病灶切且累及肝脏超过80%），虽然经化疗8周期后CT评估显示仅可见5个转移灶，而且PET/CT检查也未发现更多的代谢活跃灶，但剖腹探查后显示在肝脏表面可见约8个小结节，将其中2个病灶切除送冰冻病理检查显示为腺癌，故仅行原发灶切除（右半结肠切除）加姑息性肝转移灶射频消融术。16例接受手术的病人共出现了5例并发症，1例术后出血，再手术行出血点缝扎止血；1例术后第二天出现右下肺动脉栓塞，经抗凝等治疗治愈；2例出现右侧胸腔积液，包括上述姑息性手术病例；1例切口感染，均痊愈出院，无手术死亡病例。随访36月，51例病例中有4例失访，而15例治愈性手术切除病例中均全部随访，其中有6例复发，1例行肝左叶切除后13月肝内复发，仅局限于右后叶单个肿瘤，再次行肝右后叶切除，随访33月，病人仍存活，但肝右前叶出现复发病灶，未再到我院就诊；另1例术后14月肝右叶及尾状叶复发，目前正在住院期间，其他4例死亡，包括肝及肺多发转移1例，肝内复发并盆腔转移1例，肝内多发转移2例，分别于12月、13月、17月和21月死亡；而在化疗后肿瘤稳定及缓解但未能行肝切除的17病例中有9例死亡，中位生存期为19月（与治愈性切除组相比，Log-Rank值4.02，p=0.045），而那些对CapeOX方案化疗无效的19例病人中已有12例死亡，中位生存期仅12月（与治愈性切除组相比，Log-Rank值12.28，p=0.045）。生存曲线见图1。3. 讨论目前大量的回顾性分析表明，肝切除术是影响结直肠癌长期生存的独立预后因素，手术死亡率为4.3%，5年生存率为32.8%，远高于未行切除的病人，遗憾的是仅约10-20%的病人才能获得这些机会[1]。随着一些新的化疗药物和靶向治疗药物的出现，肿瘤的缓解率明显提高，一些起初不能切除的肝转移病灶在化疗后获得切除机会，如Adam等应用FOLFOX方案化疗后，1104例起初不能切除的结直肠癌肝转移病人中约12.5%的病人获得治愈性切除，5年总生存期和无病生存期分别达到了33%和22%[3]。而Alberts等应用FOLFOX4方案达到了40%的切除率，其中完全切除率为32%[4] 。CapeOX方案中卡培他滨取代了FOLFOX方案中的5-Fu，具有胃肠道反应轻等优点，而且NCCN2008年第一版已经将CapeOX加贝伐单抗方案已经作为进展性或转移性结直肠癌的一线治疗方案之一，也可作为其他一线方案无效的二线治疗方案，说明这一方案的疗效已经得到了广泛认可。本研究应用此方案一线治疗获得了43%的肿瘤缓解率，约29%的病人获得了完全切除，而且有较长的生存期。Santini等表明尽管药物与CapeOX相同，应用时辰治疗方案可以达到58.6%的缓解率和34.9%的稳定率[5]，也许将来我们应用时辰治疗方案可以达到更高的切除率。本研究中发现1例病人肿瘤数目超过20而且累及肝脏超过80%的病人，虽然化疗效果好，但剖腹探查后不能切除，而且姑息切除后出现并发症，增加病人痛苦，生活质量差且生存期短。Benoist等研究表明，66处肝转移灶在化疗后CT表现为肿瘤完全缓解，但手术中肉眼可以观察到20个转移，而不能肉眼观察到的15个病灶中切除后12个存在有活性的肿瘤细胞，而剩余不能发现的31个病灶术后有23个早期复发。这说明，尽管CT甚至PET上表明一些肝转移灶完全消失，但是实际并不能得到病理学的缓解[6]。我们的经验也印证了上述观点，遗憾的是在本组病例中未采用腹腔镜探查，提示在以后的临床工作中，在拟行肝切除前，可以应用诊断性腹腔镜技术，后者可以使10%的肝切除病人因不能获得潜在的治愈性切除而终止手术，这无疑可以提高病人的生活质量。当然并不是所有的病人都需要腹腔镜探查，Grobmyer等应用以下指标评估临床风险评分：原发灶淋巴结阳性、无病间隔小于12月、肝转移灶多于1个、肿瘤大于5CM、CEA水平大于200ug/L，每种指标为1分，病人所有指标相加即为临床风险评分，当临床风险评分越高，腹腔镜的价值越大，当评分达到4或5时，可使24%的病人免于开腹手术[7]。 目前大宗病例的研究结果表明，即使没有应用化疗，大于65岁的病例接受肝切除的并发症发生率高达31.9%，本研究中应用CapeOX方案化疗可以使部分病人肿瘤缓解后手术，16例肝切除病例中出现5例并发症，与文献报道接近，但无手术死亡，可能与本组病例少且年龄较轻有关。随着靶向治疗药物的广泛应用和给药方法的改进，不可切除性结直肠癌肝转移的病人将获得更多的肝切除机会，并延长生存期。参考文献1. cummings L C, Payes J D, Cooper G S. Survival after hepatic resection in metastatic colorectal cancer. Cancer. 2007; 109: 718-26.2. Kemeny N. Presurgical chemotherapy in patients being considered for liver resection. The Oncolologist. 2007; 12: 825-39.3. Adam R, Delvart V, Pascal G et al. Rescue surgery for unresectable colorectal liver metastases downstged by chemotherapy: a model to predict long term survival. Ann Surg. 2004; 240: 644-57.4. Alberts S R, Horvath W L,Sernfeld W C et al. Oxaliplatin, fluorouracil, and leucovorin for patients with unresectable liver-only metastases from colorectal cancer: a north central cancer treatment group phase Ⅱ study. J Clin Oncol. 2005; 23: 9243-9.5. Santini D, Vincenzi B, Schiavon G et al. Chronomodulated administration of oxaliplatin plus capetabine (XELOX) as first line chemotherapy in advanced colorectal cancer patients: phase Ⅱ study. Cancer Chemother Pharmacol. 2007; 59: 613-206. Benoist S, Brouquet A, Penna C et al. Complete response of colorectal liver metastases after chemotherapy: dose it mean cure? J Clin Oncol. 2006; 24: 3939-457. Grobmyer S R, Fong Y, D’Angelica R et al. diagnostic laparoscopy prior to planned hepatic resection for colorectal liver metastases. Arch Surg. 2004; 139:1326-30图1 结直肠癌肝转移生存曲线图表一 病人资料（起初不可切除病人，n=51）化疗后进展病人（n=19）化疗后稳定及缓解但未切除病人（n=17）化疗后完全切除病人（n=15）年龄（范围）63（40-70）60（49-69）59（28-70）性别比（男/女）10/98/98/7原发部位结肠643直肠131312T分期T1-2013T3-4433淋巴结转移无446有15139转移灶数目＜67410＞612135转移类型同期181514异时121肝受累范围＜2535525-50676＞501074基线CEA＜200ug/L998＞200ug/L1187

因为近期有病人咨询我腹膜后淋巴结转移的问题，目前我对这种疾病的外科经验仅仅10例左右，但是也有病例获得了长期生存的效果。因此，将我四年前发表在专业杂志方面的文章贴一下。初诊时结直肠癌腹膜后淋巴结转移的病例相对较少，但结直肠癌行治愈性手术切除后，约 21% 的患者会出现局部/区域复发，其中约 10% 会出现腹膜后淋巴结转移，但是仅 1% ～ 2% 为腹膜后的寡转移，而其他病例常合并其他部位的远处转移［1］。AJCC2010 版分期［2］认为仅髂内血管旁淋巴结属于直肠癌的区域淋巴结，而髂外、腹膜后及闭孔淋巴结均属于远处转移，NCCN 指南对可切除的肝肺转移主张积极手术切除，而腹膜后淋巴结转移未单独列出，则按其他远处转移处理，仅考虑化疗及靶向治疗。结直肠癌多学科联合诊疗提高晚期患者的生存期，其中手术切除起不可替代的作用，合并肝或肺转移灶的 Ｒ0 切 除 可 以 使 5 年 生 存 率 达 到 30% ～50% ，肿瘤减灭达到 CC0 的腹膜转移患者 5 年生存率为 11% ～ 32%［3-4］; 一些研究认为腹膜后淋巴结转移患者预后明显较合并区域淋巴结转移差，但是通过积极的局部治疗如手术治疗或放化疗，仍然可以获得治愈的机会，与可切除性肝转移相同［1］。当腹膜后淋巴结侵犯大血管时，完整的肿瘤切除存在困难，有时达到 Ｒ0 切除常常需要行血管切除和重建，存在较大的手术风险，这也是目前手术治疗存在争议的原因之一。那么腹膜后淋巴结转移的手术切除是否能够获得长期生存? 本文就腹膜后淋巴结转移的定义、手术治疗以及其他治疗进行综述。1 定 义结直肠癌的腹膜后淋巴结转移，上界位于胰腺后方腹腔干下方，下界位于腹主动脉分叉处，两侧界为两侧输尿管［5］。随着影像学技术的发展，腹膜后淋巴结短轴直径 ＞ 2 mm 就可以被发现，初诊患者术前检查时发现腹膜后淋巴结如何确定为转移? Lu 等［6］研究认为，当腹膜后淋巴结直径≥10 mm、原发灶存在脉管癌栓，区域淋巴结转移、癌胚抗原( carcino-embryonicantigen，CEA) ＞ 10 ng /L 是无其他远处转移的结肠癌患者术后生存的独立预后因素，因此建议将直径≥10 mm 作为腹膜后淋巴结转移的判断值; 另外将这 4 项临床病理因素各评分 1 分，其中 0 ～ 1 分为低风险组，其预后介于Ⅰ期与Ⅱ期患者之间，2 分和3 ～ 4分分别为中、高风险组，其预后分别与Ⅲ期和Ⅳa期相近，应该给予积极治疗。但是除腹膜后淋巴结肿大和 CEA 外，其他因素需原发灶切除后病理检查才能确定，所以临床实践中如何处理同时性腹膜后淋巴结转移仍然没有一致意见。腹膜后淋巴结复发即异时性转移，经常在术后常规复查时发现，患者症状不明显或有腰痛等症状，影像学检查显示，中央腹膜后区的复发( central retroperitoneal recurrences，CＲＲs) ，包括腹膜后的淋巴结转移( lymph node recurrence，NＲ) 和局部复发( local recurrence，LＲ) 。局部复发的定义为出现在瘤床或附近的肿瘤再生长，同时病理检查时没有淋巴结受累的证据。研究［5］显示，31 例患者的 CＲＲs 行手术切除后，NＲ 患者的预后明显较 LＲ 患者好，其 3年的无瘤生存率( disease free survival，DFS) 和总生存率( overall survival，OS) 分别为 27% 、0% 和 86% 、26% ; 而 NＲ 和 LＲ 的中位生存时间分别为 53 月和18 月。研究认为，患者原发灶为 T4、系膜切缘为阳性、影像学检查显示浸润性生长，更有可能为 LＲ，这类患者预后不佳，最好不要进行手术切除。首次术后出现腹膜后淋巴结转移复发的预测因素研究很少，目前仅认为结肠癌较直肠癌更多出现腹膜后淋巴结转移。Eveno 等［7］在进行细胞减量手术( cytoreductive surgery，CＲS) 的结直肠癌腹膜转移的患者中，105 例患者在中位 5 年的随访中，82%出现复发，18% 出现腹膜后淋巴结转移，多因素分析显示，卵巢转移是腹膜后淋巴结转移的预测因素。2 化疗及靶向治疗腹膜后淋巴结转移的概率较低，很少有文献对这一类患者的内科治疗作系统总结。目前，对于晚期患者的化疗，如果奥沙利铂、伊立替康和 5-氟尿嘧啶类这 3 种化疗药物都应用，中位生存期约 20月; 而靶向治疗的引入可以使中位生存期延长至 24－ 28 月左右，如果患者为 Ｒas 野生型，加入表皮生长因子受体的单抗可以使患者的中位生存期达到30 月左右。一些个案报道显示，化疗取得较好的效果，患者获得长期生存［8］。本院收治乙状结肠癌腹膜后淋巴结转移病例 1 例，应用奥沙利铂、卡培他滨加贝伐单抗方案治疗获得病理完全缓解，但患者 2年后死于腹腔转移和肝转移。3 手术治疗腹膜后淋巴结转移属于远处转移，仍然有很多外科医师尝试行腹膜后淋巴结切除。这包括 2 种情况，同时性和异时性腹膜后淋巴结转移。后者为复发时行腹膜后淋巴结切除，也称为挽救性手术［5］。3． 1 同时性腹膜后淋巴结转移的手术治疗Tentes 等［9］报道单中心的临床试验，124 例左侧结肠癌患者随机分为传统手术组和腹膜后淋巴结切除组，传统手术组采取肠系膜下动脉和静脉根部结扎，区域淋巴结清扫，腹膜后淋巴结切除组则在以上基础上进行腹膜后淋巴结切除; 两组病例数相等，各种基线条件无差异，研究终点为总生存率。结果显示，腹膜后淋巴结切除组 5 年生存率较传统手术组略高( 75% vs 67% ) ，但差异无统计学意义( P ＞0． 05) ，进一步亚组分析发现，Ⅲ期患者中腹膜后淋巴结切除组( 27 例) 5 年生存率较传统组( 23 例) 明显提高( 70% vs 19% ) 。研究表明，腹膜后淋巴结切除没有增加手术并发症和病死率，而Ⅲ期患者中腹膜后淋巴结切除组 OS 延长，值得进一步探讨; 存在以下问题: ( 1) 所作的腹膜后淋巴结切除没有单独计数并作病理学检查，难以明确患者腹膜后淋巴结转移的概率; ( 2) Ⅲ期患者中腹膜后淋巴结切除组有更高的生存率，但是不能明确这组患者是否存在腹膜后淋巴结转移，而对照的传统手术组预后远低于常规文献报道的结果; ( 3) 亚组分析并非研究的终点指标，所以结论难以令人信服。Bae 等［10］的回顾性研究说明了Ⅲ期患者中腹膜后淋巴结转移状况以及治疗效果。研究中共有1 082例Ⅲ期结肠癌患者纳入研究，术前通过 CTPET /CT 检查，129 例患者怀疑腹膜后淋巴结转移并进行腹膜后淋巴结切除，平均每例腹膜后淋巴结为5． 4 ± 5． 1 枚; 其中 49 例存在淋巴结转移，每例淋巴结为 4． 4 ± 4． 7 枚。研究显示，Ⅲ期患者中腹膜后淋巴结转移的比例为 4． 5% ，腹膜后淋巴结切除不增加手术并发症和病死率; 937 例患者接受辅助化疗，随访显示合并腹膜后淋巴结转移的患者 5 年总生存率为 33． 9% ，虽然远低于无腹膜后淋巴结转移的患者( 75． 1% ) ，但是与同时进行的同期肝脏转移切除病例比较显示，腹膜后淋巴结转移组( 49 例) 和肝转移组( 91 例) 的 5 年 OS 和 DFS 都无差异，分别为33. 9% vs 38． 7%和 26． 5% vs 27． 6% ; 腹膜后淋巴结组的复发模式与肝转移组不同，主要表现在前者复发病例的 25% 为腹膜后复发，而肝转移组的腹膜后复发仅 9% 。没有区域淋巴结转移的患者是否会有腹膜后淋巴结转移? Quadros 等［11］研究 100 例低位直肠癌患者的区域外淋巴结( 包括腹膜后和侧方淋巴结) 转移状况，所有入选的都是术前分期为Ⅱ、Ⅲ期的患者，除按直肠全系膜切除术( total mesorectal excision，TME) 原则进行手术外，所有患者均接受腹膜后及侧方淋巴结清扫，其中 44 例接受术前放化疗，19 例接受术后放疗。术后病理检查显示，区域淋巴结和区域外淋巴结获取数的中位数都是 12 枚; 17例存在区域外淋巴结转移，其中 59 例无区域淋巴结转移的患者中有 7 例存在区域外淋巴结转移。所有存在区域外淋巴结转移的患者 50 月的总生存率为28. 6% ，远低于无区域外转移患者的 80． 0% 。研究认为，CEA 升高、肿瘤较大和 T3 或 T4 肿瘤是直肠癌出现区域外淋巴结转移的预示因素。该研究存在一定缺陷，没有明确直肠癌的侧方和腹膜后淋巴结具体的转移概率，另外不能直接证实，是否行区域外淋巴结切除是治疗腹膜后淋巴结的有效方法。而Liu 等［12］的研究与上述研究相似，但结果发现，腹膜后淋巴结转移的比例高达 12% ，且非浸润型低位直肠癌行腹主动脉旁及侧方淋巴结清扫能够改善DFS。腹腔镜外科技术的进展显示，对初诊怀疑结直肠癌腹膜后淋巴结转移的患者同时行腹膜后淋巴结切除也很安全。Song 等［13］的研究包括 40 例患者，入选标准为: ( 1) 结直肠原发灶病理诊断为腺癌;( 2) CT 影像显示，腹膜后淋巴结短轴的直径 ＞ 8 mm且轮廓不规则或中心坏死，或 PET /CT 显示为高代谢灶; ( 3) 淋巴结位于左肾静脉水平以下，但不位于血管后方，无其他远处转移。手术安全性好，平均手术时间 192． 3 ± 68． 8 分钟( 100 － 400 分钟) 、失血( 65． 6 ± 52． 6) mL( 20 ～ 210 mL) 。无中转开腹，术后并发症 15% ，无 ＞ 3 级并发症。术后病理证实，16例合并腹膜后淋巴结转移，这些转移病例的 3 年 OS和 PFS 分别为 65． 7% 和 40． 2% 。本研究说明，影像学诊断为转移的病例仅 40% 为病理学转移，腹腔镜切除也能获得较佳的治疗效果。3． 2 挽救性手术治疗3． 2． 1 个案报道及病例回顾 目前有个案报道行腹膜后淋巴结切除合并血管切除和重建［14］。此外Ho、Mixter、Hashimoto 和 Tsarkov 等［1，15-17］ 也报道 4例，详见表 1。表 2 中 Abdelsattar 等［18］也有 3 例行腹主动脉或下腔静脉切除及重建，而 Gagnière 等［19］报告 25 例腹膜后淋巴结转移，其中 3 例侵犯大血管行切除和重建，故目前共 11 例行腹膜后淋巴结切除及血管重建，大于 3 级的并发症较少，获得较好的治疗效果。而病例总结报告 11 篇，一些研究中包括初诊及复发时出现的腹膜后淋巴结转移病例行手术切除，如 Gagnière 等［19］的 25 例研究中 6 例为复发病例，而 Arimoto 等［20］的 14 例研究中 6 例为复发病例，Abdelsattar 等［18］的研究为结直肠癌术后复发合并血管切除 14 例患者，其中 3 例合并腹主动脉或下腔静脉切除，根据文献描述，其中 3 例为腹膜后淋巴结转移; Min 等［21］报道 38 例腹膜后淋巴结转移仅 6例行手术切除，但是效果较放化疗好; 最近 Ｒazik等［22］报道了迄今为止最多的 48 例腹膜后复发病例行手术治疗的结果，文章中没有对腹膜后淋巴结转移和局部复发进行区分，其中有 8 例还合并其他远处器官转移，但是其远期疗效是目前所有报道的研究中最好的之一，其 5 年 OS 高达 70% ，说明选择合适的患者进行手术切除是合理的; 所有研究累计病例数 176 例，其临床病理及预后见表 2。部分患者除腹膜后淋巴结转移外，其他脏器的转移同时行 Ｒ0切除，如 Dumont 等［5］的研究中 23 例腹膜后淋巴结转移患者中有 8 例肝转移和 1 例腹股沟淋巴结转移; 而 Gagnière 等［19］报告 25 例腹膜后淋巴结转移，合并 6 例肝转移和 1 例腹膜转移，也同时行肝转移灶切除和细胞减量手术( cytoreductive surgery，CＲS)和腹 腔 热 灌 注 化 疗 ( hyperthermia intraperitonealchemotherapy，HIPEC) ; Ｒazik 等［22］的研究中合并 8例其他器官转移，5 例也同时( 4 例) 或异时( 1 例)切除了转移灶。说明腹膜后淋巴结转移合并其他器官转移通过手术切除也能获得较好的 DFS; Gagnière等［19］重点强调系统性腹膜后淋巴结切除的技术，即切除包括腹主动脉和下腔静脉后方的淋巴结，其疗效较以前的报道结果更好，但是手术风险增加，术后需要低脂饮食 1 月。除此之外，围手术期治疗在近年应用的比例更高，如果术前明确存在腹膜后淋巴结转移，几乎所有病例都考虑应用新辅助化疗，部分病例联合靶向治疗如西妥昔单抗或贝伐单抗等治疗，Dumont 等［5］ 和 Gagnière 等［19］ 的研究中均有 2例接受新辅助化疗加靶向治疗后肿瘤病理学完全缓解的病例，获得完全缓解的患者是否有必要行手术切除值得探讨。3． 2． 2 挽救性手术术后复发模式 所有研究显示挽救性手术术后有很高的复发转移率( 55% ～ 83% ) 。Arimoto 等［20］研究显示，复发的首要部位为肺，其次为淋巴结; 其他研究显示，最常见的复发部位为腹膜后淋巴结，如 Gagnière 等［19］报道的复发病例中 52%经历淋巴结复发，中位复发时间为 14 月，剩余 12 例在中位随访的 85 月( 8 － 116 月) 仍然无局部复发;这期间也有 60% 的患者出现远处转移，依次为肺( 52% ) 、肝( 47% ) 、纵隔( 33% ) 、腹膜、脑、卵巢等;Dumont 等［5］研究显示，83% 病例复发转移，中位复发时间为 15 月，48% 经历淋巴结局部复发，而 71%出现远处转移，依次为肺、肝和腹膜。这种复发模式与同时性腹膜后淋巴结转移行手术切除后复发模式相似。这些结果提示，除手术切除外，还需要其他手段降低局部复发率。3． 2． 3 影响挽救性手术预后的临床病理学因素Taylor 等［25］认为，达到 Ｒ0 切除是获得长期生存、降低局部复发的重要因素，因此很多文献对肿瘤累及邻近器官时常联合脏器切除，如腹主动脉的切除和重建。Ｒ0 切除的定义为病理学切缘阴性，但是腹膜后存在重要的血管，如果肿瘤并未侵犯血管，而仅仅为包绕，这种情况下如果未合并血管切除，行淋巴结清扫时常会导致淋巴结破碎，肉眼显示，手术野无肿瘤残留，实际上很难达到病理学切缘阴性。Gagnière等［19］的报道显示，25 例患者中 15 例淋巴结存在破碎，但是无论 OS 或 PFS 都与淋巴结破碎无相关性;另外 Min 等［21］认为，腹膜后淋巴结转移位于肾血管水平以下预后好，但其他文献未得到证实; Shibata等［24］认为，转移的淋巴结直径 ＜ 5 cm 预后好，Choi等［27］认为 ＜ 2 枚转移淋巴结预后好，但是无论大小还是淋巴结转移数与预后的关系在 Gagnière 等［19］的研究中都未得到证实; Bowne 等［26］认为患者首诊时分期对复发后预后有影响，Gagnière 等［19］的研究包括 19 例初诊病例，其中 7 例合并其他器官转移同时切除，结果显示，合并其他脏器转移明显影响 OS，但是 PFS 无影响; Ｒazik 等结果显示年轻和肿瘤巨大是 OS 预后不良因素，而年轻和肿瘤残留也影响DFS。目前的所有结果都是小样本研究，进行预后分析时存在Ⅱ型错误，因此哪些因素对预后有影响也很难确定。除此之外，越来越多的研究中除手术切除淋巴结外，近年应用围手术期治疗包括化疗、靶向治疗和放疗的比例也越来越高，同时生存率也较高，因此患者的预后可能与综合治疗的进步有关。4 放 疗因为结直肠腹膜后淋巴结转移部位邻近小肠、脊髓等，传统的放疗为避免这些器官的损伤，很难给予较大的放射剂量，只能起姑息治疗的作用。近年来放射技术的发展，应用放疗治疗腹膜后淋巴结转移的研究逐渐增多。Kim 等［28］应用立体定向放射治疗技术治疗 7例腹膜后孤立淋巴结转移的患者，淋巴结最大直径＜ 8 cm，或为融合淋巴结，或 2 ～ 3 个分散但相邻的淋巴结; 患者在放疗前接受亚叶酸钙( folinic acid ，CF) 加 5-FU 化疗后病情进展; 放射剂量为 33 ～ 51Gy，均在 3 次完成。其中 3 例获得完全缓解，4 例获得部分缓解。1 例于放疗后 4 月因空肠放射性溃疡和狭窄导致肠梗阻行手术治疗好转。中位随访 26月( 21 － 70 月) ，6 例复发，1 例 26 月仍无瘤生存，所有病例中位生存期 37 月。研究者认为腹膜后淋巴结转移可以应用立体定向放疗进行挽救治疗Yeo 等［29］报道接受放疗的 22 例腹膜后淋巴结转移患者，淋巴结转移≤3 个脊椎的高度，所采用的技术有 2 种: 20 例采用三维适形放疗，总剂量为 63Gy /35 F 或 58． 8 Gy /31 F; 2 例采用螺旋断层放疗，临床靶区( clinial target volume，CTV) -2 剂量为 60Gy /20 F，生物学剂量相当于常规分割的 66 Gy。所有病例同时接受化疗，术后 16 例接受辅助化疗，方案分别为氟尿嘧啶与奥沙利铂或伊立替康的单药或联合化疗。疗效如下: 13 例完全缓解( complete response，CＲ) ，6 例部分缓解( partial response，PＲ) ，3例稳定( stable disease，SD) 。无严重并发症，中位生存期 41 月，3 年和 5 年的 OS 为 64． 7% 和 36． 4% ，3年和 5 年的无复发生存 ( recurrence free survival，ＲFS) 分别为 34． 1% 和 25． 6% 。15 例患者出现复发，放疗野内复发 4 例，另外 11 例为放疗野外的转移( 包括 8 例远处转移) 。说明放疗局部控制率高，预后分析显示，放化疗前淋巴结位于肾静脉水平以下、CEA 正常是 ＲFS 有利因素; 而放化疗后缓解和辅助化疗是 OS 的独立预后因素。该研究首次说明放化疗在腹膜后淋巴结转移中的决定性作用。Lee 等［30］对结直肠癌腹膜后淋巴结转移的放疗时机进行研究，52 例患者都接受放化疗加或不加靶向治疗，其中 25 例先接受放疗，而另外 27 例患者先接受内科治疗，局部获得控制( 17 例) 或仅出现局部进展( 10 例) 的患者接受放疗; 是否先接受放疗由多学 科 协 作 小 组 ( multidisciplinary team /treatment，MDT) 讨论决定，但是先接受内科治疗的患者大体肿瘤体积( gross tumor volume，GTV) 更大。放疗采用的技术为三维适形放疗或螺旋断层放疗，中位生物学剂量相当于常规分割的 54 Gy( 31 ～ 88 Gy) ，其中 24 例应用三维适形放疗的中位剂量为 52 Gy，28例应用螺旋断层放疗的为 62 Gy。随访显示，31%患者获得完全缓解，62% 获得部分缓解，所有患者的中位生存期为 41 月，81% 患者出现复发，2 年 PFS为 37． 5% ，2 年 OS 为 69． 6% 。复发模式表现为局部区域复发 62% ，远处转移 60% ，远处转移的器官依次为肺( 48． 4% ) 、锁骨上或纵隔淋巴结( 38． 7% )等。尽管直接放疗组的预后较其他组好，但是 Cox分析显示，无瘤间期( disease free interval，DFI) ＜ 12月、肿瘤位于左肾静脉水平以上、GTV ＞ 30 mL、女性是影响总生存率的独立预后因素，除以上参数外，放射生物学剂量 ＞ 54 Gy 也是影响复发的独立预后因素。本研究并非随机对照研究，因此仍然不能肯定放疗时机对预后的影响。总之，随着放疗技术的发展，结合化疗及靶向治疗，结直肠癌腹膜后淋巴结转移也获得较长的生存期，较好的局部控制率，但是放疗选择的病例与手术切除的病例不同，后者都有病理学诊断的证实，而前者主要依赖 CEA 水平和影像学检查，而 Bae 等［10］以及 Song 等［13］的回顾性研究显示，＜ 50% 影像学怀疑转移的病例被病理学证实为腹膜后淋巴结转移，因此也不能排除放疗病例中包含非转移的病例。5 小结及展望结直肠癌合并孤立的腹膜后淋巴结转移率约1% ～ 2% ，其中存在区域淋巴结转移的患者腹膜后淋巴结转移率约为 4． 5% 。初诊时影像学检查考虑腹膜后淋巴结肿大≥10 mm 时考虑存在转移，CEA＞ 10 ng /L、淋巴结位于左肾静脉水平以上、淋巴结累计直径 ＞ 5 cm 或者淋巴结数 ＞ 2 枚预后差，建议给予更积极的治疗。手术切除需要较高的技巧才能达到肉眼的无瘤切除，但是如果不合并血管切除和重建，也会导致淋巴结破碎; 对侵犯血管的病例行血管切除和重建也是安全的; 最近腹腔镜技术的进展显示，经过选择的病例也可安全施行。总之，同时性腹膜后淋巴结转移行切除可达到与可切除肝转移相近的疗效; 而且复发时手术切除也可以使一些病例获得长期生存，多个研究显示 Ｒ0 切除可以提高 DFS，同时近年来 5年总生存率的提高与应用化疗和靶向治疗有关。放疗技术的进步可以使腹膜后放疗的精确性提高，增加放疗剂量，能够有较高的局部控制率，结合化疗和靶向治疗也能获得与手术切除相近的生存率。但是值得注意的是，放疗的病例可能部分为假阳性，因此其结果仍然不能与手术相提并论。无论手术切除还是放疗，即使结合目前的化疗和靶向治疗，其首要的复发仍然为局部复发，较全身转移更早出现，这也对如何完善治疗策略方面提出挑战。是否手术切除结合放疗能够改善局部复发率? 是否化疗和靶向治疗优先能够更好地筛选手术切除获益的患者? 这些均需要前瞻性多中心的临床研究来回答。

胃癌术后复发有以下几种类型：局部复发、腹腔播散、远处转移如肝肺转移等。也可以联合出现。如果肿瘤仅局部复发，仍有治愈可能。治疗包括围手术期化疗和手术。以下是本院一例病人资料。男，57岁，胃癌术后2年，上腹部疼痛并食欲减退2月。体检：消瘦，全身浅表淋巴结不大，心肺正常，腹平坦，上腹部饱满，可触及肿块，约直径5cm大小，质硬，较固定，边界欠清，压痛，无反跳痛，肝门指检：直肠正常，Douglas窝未触及结节。辅助检查：胃镜示残胃吻合口小弯侧溃疡，病检示低分化腺癌。CEA34umol/ml。肝肾功能示ALT及AST稍升高但小于参考值上限的2倍。胸部及盆腔CT正常，腹部CT示肝硬变、残胃壁增厚，与周围器官组织分界欠清，详见下图。诊断：1、胃癌术后局部复发（rT4N+M0，Ⅳ期）；2、肝硬变与病人及其家属交流，认为病人属胃癌复发，局部晚期，考虑行围手术期化疗及手术。病人认为自己合并肝硬变，坚决不同意手术前化疗，故作术前准备后行剖腹探查术。手术中情况：术中见残胃肿瘤，侵透浆膜，与邻近器官与横结肠胰腺粘连紧密，部分被横结肠包裹，后者表面未见到肿瘤侵犯，腹腔无播散，考虑肿瘤复发侵及胰腺和和横结肠，肝硬变为轻度，考虑行左上腹脏器切除。手术过程中见肝总动脉和胃左动脉旁均有淋巴结肿大，质地较硬，第一次手术后胃左动脉仍保留约3cm长。手术中切除脏器包括残胃、横结肠、胰腺体尾部及脾脏，重建为食管空肠Roux-en-Y吻合，结肠吻合，并行空肠营养管置入术，瘤床应用氟尿嘧啶缓释剂（中人氟安）0.6g，手术历时3小时40分钟，出血约800ml，术中输血800ml。术后予营养支持、抗感染等处理，顺利恢复出院。术后病检：未分化癌，侵及胃壁全层至横结肠肌层及胰腺被膜组织，可见脉管癌栓，肝总动脉旁淋巴结0/3转移，胃左动脉旁淋巴结3/3转移，而且淋巴结旁脂肪组织中可见癌浸润，腹腔干淋巴结0/5转移。图片1 CT示残胃肿瘤侵犯横结肠及胰腺湖北省肿瘤医院图片图片2 术中见胃左动脉仍残留，周围淋巴结肿大湖北省肿瘤医院图片图片3 左上腹脏器切除后腹腔干，胃左动脉根部、脾动脉根部、肝总动脉图像肝总动脉胃左动脉根部脾动脉根部图片4 切除标本体会：本例病人第一次手术由本地大学附属医院教授操刀手术，但仍然在二次手术中可以见到胃左动脉未在根部结扎，说明第一次手术不是D2手术，尽管术后局部复发的因素与病人的病期、肿瘤的生物学行为等有关，但也不能排除与手术有关。其实这不是我第一次接诊这类病人，而且一些胃癌病人的第一次手术均在负有盛名的大医院所施行。我曾经接诊一例胃癌病人，在外院手术治疗后约两个月到本院检查，MRI显示胃左动脉根部淋巴结肿大，而且胃左动脉未在根部结扎，再次手术行淋巴结清扫示淋巴结转移，无疑这例病人第一次手术并非D2根治术。其实8年前在中国肿瘤界已经达成共识，对于胃癌病人，除少数Ⅰa期病人外，均应施行D2手术。而在我省的现状表明，一些胃癌病人的第一次手术并没有采取合理的手术方式。对病人而言，肿瘤的首诊非常重要，请一定慎重。

ABSTRACTPrevious studies have indicated that the infiltration of CD8+ T cells in colorectal cancer is an independent predictor of increased survival but clinical observations have suggested that the cytotoxic function of CD8+ T cells infiltrating colorectal cancer may often be limited. In this study, we have assessed the phenotype of colorectal cancer CD8+ tumor-infiltrating lymphocytes (TILs) isolated ex-vivo from tumor tissue, and assessed the perforin content of TIL with respect to their location using immunohistochemistry. We found that CD8+ T cells TILs isolated from colorectal cancer are mainly composed of antigen-experienced cells of effector memory type (CD45RA-CCR7-, CD27+/CD28- or CD27-/CD28-) and contain only minor proportions of terminally differentiated CD8+ T cells. The perforin content of these TILs, however, is significantly lower than that of antigen-experienced T cells in PBMCs due to the much lower levels of perforin found in the CD27-CD28- subset in TILs compared with CD8+ T cells of similar phenotype in PBMCs. Colorectal carcinomas are solid epithelial tumors that have long been considered to be poorly immunogenic. However, along with the discovery of human T cell defined tumor associated- and specific-antigens (TAA/TSA), tumor antigens were also identified in colorectal cancers and both autologous cellular and/or humoral immune responses to specific tumor antigens, such as mutated TGFb receptor II (1,2), SART-1 (3), Ep-CAM (4) and NY-ESO-1 (5,6), have been reported in some colorectal cancer patients. A large body of evidence has proven that the tumor-reactive CD8+ lymphocytes play a central role in cancer immunity. Recently, Naito and colleagues (7) have classified CD8+ tumor infiltrating T lymphocytes (TILs) in primary colorectal cancer into three groups: (1) peri-tumoral, consisting of cells distributed along the invasive margin of the tumor; (2) stromal, consisting of cells infiltrating the tumor stroma; and (3) intraepithelial, corresponding to cells that infiltrates cancer cell nests and are in direct contact with tumor cells. These authors reported that the existence of CD8+ TILs in the third group was a strong independent predictor of survival. In contrast, the presence of stromal TILs did not correlate with a better prognosis. Further study of the function of CD8+ TILs collected from the tumor sites of colorectal carcinomas is needed to unveil the clinical implications of these observations.Upon antigen stimulation, naive CD8+ T cells proliferate and go through several differentiation stages that can be distinguished by the expression of surface markers, such as CD45RA/CCR7 (8) and/or CD27/CD28 (9). Their progeny include effector T cells and effector memory cells. Effector cells migrate to peripheral tissues and display immediately restrain invasion by pathogens or cancer cells, while memory cells travel through secondary lymphoid organs and can generate a new wave of effector cells upon a re-encounter with the stimulating antigen (8). In cancer tissues, TILs can be attracted to tumor environment by cancer cells themselves, the inflammatory signals caused by cancer, or naturally homing factors. Under any of the above circumstances, the differential phenotype and cytotoxic function of CD8+ T cells influence the clinical out come dramatically by (1) direct killing of cancer cells, (2) angiostasis through acting on stromal cells, or (3) maintenance a cancer tolerant environment through regulatory T cells. In this study, we examined the phenotype and cytotoxic status of CD8+ T lymphocytes from colorectal carcinomas. Generally, CD8+ T cells can be classified into two distinct effector cell types depending on their cytokine-secretion profiles following an antigen encounter (ref.13, 14). Type 1 CD8+ T cells predominantly secrete IL-2 and IFN-g, while type 2 CD8+ T cells produce IL-4, IL-5, and IL-10. Although the patterns of cytokine production are different, both populations of CD8+ T cells exert a cytotoxic function via the perforin pathway and have been demonstrated to contribute to the mediation of antitumor effects (14-16). Therefore, perforin was adopted here as the common cytotoxic indicator in the functional analysis of CD8+ TILs. We compared the differentiated phenotype and the perforin expression level in TILs and peripheral blood mononuclear cells (PBMCs) of human colorectal cancer patients. The results showed that tumor tissues of two-third of colorectal cancer patients were infiltrated by T lymphocytes. Among them, the majority of CD8+ infiltrating T lymphocytes located in tumor lesions were antigen-experienced effector memory cells instead of terminally differentiated effector cells, and that they express much less perforin than CD8+ T cells with similar phenotype in PBMCs. Furthermore, the function of the CD27-CD28- subset of CD8+ T cells in TILs was not represented by their phenotype as they were in PBMCs, indicating an antigen-experienced effector-memory phenotype and a functional tolenrant state of of low perforin expression of the CD8+ T cells that infiltrate colorectal cancer tissues. The location and expression of the CD8+ TILs were also determined by immunohistochemical staining.Materials and MethodsSample collection and preparationAfter obtaining informed consent and the approval of the Hospital Ethic Review Committee, 40 preoperative peripheral blood samples and their corresponding postoperative tumor tissues were collected from patients at Beijing Cancer Hospital, School of Oncology, Peking University. All tissues were pathologically confirmed as primary colorectal carcinomas. None of the patients enrolled in this study had been subjected to chemotherapy or any other therapy with immunosuppressive effects before lymphocyte isolation. The patients’ clinical characteristics are summarized in Table 1. Tumor infiltrating lymphocytes were obtained via mechanical dispersion of tumor tissues into a single cell suspension in calcium-magnesium free HBSS (containing EDTA) within an hour of surgery, followed by Ficoll-Hypaque gradient centrifugation as described by Watanabe et al. (17). Peripheral blood mononuclear cells (PBMCs) were isolated by density centrifugation using Ficoll-Hypaque as previously described, and the number of TILs in colorectal cancer tissues was counted. Flow cytometric analysis (FACS) was carried out when TILs numbered more than 0.2×106 cells. The number of TILs was scored as follows: 0, no TILs; +, 0.2-2×106 TILs; ++, more than 2×106 TILs. Paraffin embedded tumor tissue samples were prepared for immunohistochemical staining.Flow cytometric detectionAntibodies and associated agentsThe following mAbs, the corresponding isotype control antibody and other agents were purchased from BD PharMingen (San Diego, USA): anti-CD27-fluorescin isothiocyanate (FITC) and phycoerythrin (PE), anti-CD28-FITC and –PE-CY5, anti-CD3-PE, anti-CD8-PE-CY5, anti-CD107a-FITC, anti-perforin-FITC, Cytofix/CytopermTM, Perm/WashTM buffer. Other sources of mAbs were: Beckman Coulter, Fullerton, USA (anti-CD8-ECD), Immunotech, Marseile, France (7-AAD, anti-CD45RA-ECD and isotype antibody), and eBioscience, San Diego, USA (anti-TCRαβ-PE and anti-TCR-γδ-FITC), and RD SYSTEMS, Tustin, USA (anti-CCR7-FITC). Flow cytometric analysis After TILs or PBMCs were isolated, the cells were aliquoted and stained with anti-CD8 along with anti-CD27 and anti-CD28, anti-CD45RA and anti-CCR7 or isotype controls at 4oC for 30 minutes. The cells were then washed twice with phosphate buffered saline (PBS)/2% FCS. Intracellular levels of perforin were measured in freshly isolated CD8+ T lymphocytes without previous stimulation. In brief, after staining with the appropriate mAbs, cells were fixed with Cytofix/CytopermTM for 20 min at 4oC. After permeabilization, the cells were washed with Perm/WashTM buffer and labeled with anti-perforin mAb. Labeled cells were collected and analyzed by a COULTER EPICS XLTM with EXPO32 ADC software (Beckman Coulter, Fullerton, CA). Approximately 5,000-10,000 CD8+ gated events were acquired for analysis. Intracellular perforin expression is represented as the normalized mean fluorescence intensity (NMFI), which is the mean fluorescence intensity (MFI) of cells stained with anti-perforin-FITC mAb divided by the MFI of cells stained with mouse IgG2b-FITC mAb (control). To exclude natural killer (NK) cells, FACS analysis and sorting was performed on gated CD8bright T cells. The cells were also stained with both anti-TCRαβ-PE and anti-TCRγδ-FITC to determine which subsets of T cells were present.ImmunohistochemistryAnti-CD3 (murine, clone PS1, 1:100), anti-CD8 (murine, clone 1A5, 1:40), and anti-perforin (murine, clone 5B10, 1:20) monoclonal antibodies were obtained from Novocastra Laboratories LTD, Newcastle upon Tyne, United Kingdom. Following heat-based antigen retrieval in EDTA buffer (1 mmol/L, pH8.0), paraffin-embedded specimens were incubated overnight with mAbs at 4°C. Detection of the primary antibody was performed with PowerVisionTM two-step histostaining reagent (Zhongshan Biotechnology, Beijing, China). Normal lymphocyte nodes and T lymphoma were used as negative and positive controls.The counting of intratumor intraepithelial CD3+, CD8+ and perforin+ cells was performed independently by two pathologists using a video-assisted measuring system (MOTICAM 2000, MOTIC IMAGES ADVANCED 3.2 software; Shamen, China). Nine areas of 670 x 500 μm were chosen. The total numbers of positive cells were divided by 3 (1 mm3). Fields were chosen to contain the maximal amount of neoplastic cells with minimal necrotic debris. The cell numbers of intraepithelial or stromal CD3+ and CD8+ cells were graded as nil (no staining in 1 mm2), mild(1-19 positive cells/1 mm2), moderate(20-49 positive cells/1 mm2) or severe(>50 positive cells/1 mm2), according to the criterion proposed by Naito et al. (7). The number of intraepithelial or stromal perforin positive cells were counted and directly compared.Statistical analysisSince most of the frequency data of PBMC and TIL subsets were normally distributed according to Kolmogorov-Smirnov statistics, the SPSS program (SSPS Inc., Chicago, USA) was adopted in analyzing the data using the paired-samples T test. Data sets that did not exhibit a normal distribution were analyzed with the same software using the Wilcoxon test. Spearman’s non-paramatric correlation coefficient (rho) was used to measure the degree of association between the two methods of evaluating TILs (FCM and immunohischemical staining) and the number of CD3+ and CD8+ cells. Statistical significance was accepted at p < 0.05 (two tailed). RESULTSAssessment and isolation of CD8+ T lymphocytes infiltrating colorectal carcinomas.Lymphocytes infiltrating colorectal carcinoma tissues were isolated from fresh tumor tissue as detailed in the methods section. From 12 of 40 tumor samples, we obtained only low numbers of TILs (< 0.2 x 106) that were insufficient for analysis. The number of TILs obtained in the remaining 28 cases varied from 0.2 x 106 to 7 x 106 (median, 0.9 x106) per gram of tumor tissue (Table 1). All TILs were stained with 7-AAD and showed a content in viable cells > 80% (not shown). The number of CD3+ T cells varied from 0 to 5 x 105 (median, 2.1 x 105) per gram of tumor tissue, which was in agreement with previous data by Watanabe et al. (17). CD8+ T cells represented between 12 and 58% of total T cells in these samples. Of those, >90% were CD8bright. The large majority (91-97%) of CD8bright cells in TILs were TCR ab positive, whereas significant proportions of TCRgd T cells were only detected in four samples, at frequencies lower than 5%. The presence and localization of CD3+ and CD8+ T lymphocytes in the tumors of 32 colorectal patients was assessed by immunohistochemistry. Both CD3+ and CD8+ infiltrating T lymphocytes were present in the majority of these samples either in the stroma and/or intraepithelial. Depending of the location of the infiltrating T lymphocytes, samples were classified in three groups: (1) those showing both stromal and intraepithelial lymphocyte infiltration (12 cases); (2) those showing lymphocyte infiltration only in the stromal tissue (17 cases); and (3) those where infiltrating lymphocytes were undetectable (3 cases) (Figure 1). In general, most CD3+ and CD8+ T cells were detected in the stroma. For the 12 samples containing stromal and intraepithelial CD3+ T lymphocytes, the number of positive cells at both locations were counted and compared. The ratio of the CD3+ T cells in intraepithelial tissue versus stroma varied from 1/60 - 2/5 (median of 1/30). The CD8 expression pattern was similar to that of CD3 (Table 1). As expected, the number of CD8+ T cells in the stroma was lower than that of CD3+ T cells. However, we found a direct correlation between the numbers of CD3+ and CD8+ T cells (Spearman's rho coefficiency = 0.78, p = 0.001). We then compared the numbers of intraepithelial and stromal CD3+ or CD8+ TILs in paraffin embedded tumor tissues with the number of TILs isolated from fresh tumor tissues. We found a positive correlation between the number of isolated TILs and the number of stromal CD3+ T cells (correlation coefficiency = 0.514, p = 0.003) or CD8+ T cells (correlation coefficiency = 0.481, p = 0.006) found by immunohistochemical analysis. Thus, considering the distribution patterns of infiltrating lymphocytes in colon carcinomas, most CD8+ T infiltrating lymphocytes isolated from these tissues are likely to correspond to lymphocytes present in the tumor stroma.Phenotypic analysis of CD8+ T lymphocytes in ex-vivo isolated TILs from colorectal carcinoma samplesIn order to assess the differentiation stage of CD8+ T lymphocytes infiltrating colorectal carcinomas we stained TILs isolated ex-vivo with antibodies specific for CD8 and either CD45RA and CCR7, or CD27 and CD28 as described previously (8,9). PBMC and TIL samples from the same patients were stained at the same time to allow an internal comparison between similar subsets in PBMC and TILS (Fig. 2, Table 2). As expected, co-staining of CD8+ T lymphocytes in patients’ PMBCs with anti CD45RA and anti CCR7 antibodies identified different subsets that included, according to the classification proposed by Sallusto et al (8) a nave (CD45RA+CCR7+, 10.2 ± 6.6%) and an antigen-experienced population (CD45RA-) comprising both effectors (CD45RA+CCR7-, 46.4 ± 16.2%) and effector-memory cells (CD45RA-CCR7-, cells 39.4 ± 14.1%) in roughly similar proportions. In contrast, no significant proportions of nave CD8+ T cells were detectable in TILs from colorectal carcinoma samples, that contained predominantly antigen-experienced cells exhibiting for the largest part an effector-memory phenotype (CD45RA-CCR7- 87.5 ± 5%) and to a lesser extent an effector phenotype (). Staining of CD8+ T cells in PBMCs with antibodies against CD27 and CD28 identified four major subsets that, according to what previously described by Appay and colleagues, represents sequential differentiation stages: nave and early differentiated (CD27+CD28+, 27.2±17.4%), intermediate (CD27+CD28-, 20.5±14%), and terminally differentiated (CD27-CD28-, 55.0 ± 15.8%). In contrast in TILs, only two of these subsets were found, present at roughly similar proportions, namely CD27+CD28- (44.2 ± 21.5%) and CD27-CD28- (47.6 ± 20.1%). Assessment of perforin expression in colorectal carcinoma TILs with respect to their phenotype identifies a major effector-memory subset with low perforin content.For 10 of these samples where the number of TILs was sufficient for further analysis, we examined the presence of the cytotoxic molecule perforin by intracellular staining using specific antibodies. Approximately 30-50% (34.6 ± 2.1%) of total CD8+ T cells from PBMCs contained perforin, whereas only 0-13% (7.4 ± 1.4.1%) of CD8+ T cells from TILs contained this molecule. In addition the intensity of perforin staining was significantly lower in CD8+ T cells in TILs (TILs:1.4 ± 0.4) than in PBMCs (3.8 ± 0.9; p = 0.001, Fig. 3). It is been previously reported that the expression level of perforin increases along with the stage of differentiation of CD8+ T cells (8, 18), and negatively correlates with the expression of CD27(18). Therefore, we analyzed perforin expression in TILs with respect to their differentiation stage by co-staining with perforin, CD27 and CD28 specific antibodies. As expected, low levels of perforin were found in the CD27+CD28- subset from both PBMCs (NMFI: 2.1 ± 1.5) and TILs (NMFI: 1.2 ± 0.3). However, perforin levels in the CD27-CD28- subset in TILs (MFI: 1.5 ± 0.8) was found to be similar to the level of CD27+CD28- subsets (p = 0.586), and significantly lower than that of the corresponding subset in PBMCs (p = 0.002) (Figure 3). The low intracellular level of perforin in CD8+ T cells at the tumor site could in principle result from the degranulation occurring after interaction of cytotoxic T cells with tumor cells. To address this possibility, we assessed the expression of CD107a, a lysosomal integral membrane protein in cytotoxic granules, that is a marker of degranulation after stimulation (19). No CD107a was detected on the surface of cells with low intracellular perforin levels (Fig. 3), indicating that low perforin expression in CD8+ TILs was most likely not due to degranulation. To further exclude the possibility that the procedure of lymphocyte isolation might have resulted in partial degranulation of CD8+ T cells or in the degradation of perforin, perforin expression was further assessed by in situ immunohitochemistry Analysis of six samples containing perforin expressing CD8+ cells both stromal and intraepithelial revealed that perforin expression was less frequent in the stromal as compared to the intraepithelial lymphocytes (stroma: 16±6%; intraepithelial: 59 ± 22%, t = -4.274, p = 0.01, Figure 4). In addition, analysis of an additional 7 cases where lymphocytes were only present in the stroma revealed that less than 10% of total CD8+ T cells present at this location express perforin. DISCUSSIONAlthough some studies have suggested that the number of CD8+ tumor infiltrating lymphocytes is a good prognostic marker in colorectal cancer patients (7, 20), many patients with obviously infiltrating lymphocytes nevertheless have a short survival time. Other reports do not support a protective role for infiltrating lymphocytes (21). In several clinical trials, no survival benefit was found after adoptive therapy of TILs to treat colorectal cancer patients (22-24, ref.25). Thus, elucidating the functional state of TILs in colorectal cancer in vivo is critical in interpreting the conflicting data. To our knowledge, this study is the first attempt to detect the differentiated phenotype and cytotoxic function (represented by perforin) of TILs ex vivo of colorectal cancer. Fresh TILs were mechanically isolated from colorectal cancer tissues without the processes of enzyme digestion (26) or cryopreservation (27), which can change TIL phenotypes. Hence, the results reflect the actual phenotype and function status of the TILs in vivo.γδ T lymphocytes comprise 25-37% of intraepithelial lymphocytes residing in normal intestinal epithelium. However, the tumor infiltrating lymphocytes in colorectal cancer tumor sites were mainly αβ T lymphocytes (17). Our results also showed that more than 90% of CD8+ TILs were αβ T lymphocytes. This phenomenon is consistent with previous reports (7,17), and indicates that during tumor development local inflammatory responses and/or tumor antigens themselves attract antigen-specific T cells to tumor sites .Although several studies have confirmed that tumor-specific cellular responses do exist in colorectal cancer patients, the colorectal carcinoma-associated/specific antigens are not well defined. Therefore, it is difficult to obtain epitope-specific T lymphocytes to analyze their phenotype and function in the TILs of the majority of colorectal cancer patients. On the other hand, TILs isolated from tumor sites consist of multiple antigen-specific T lymphocytes responding to a cluster of various antigenic challenges. Either they act on cancer cells directly, or on cancer stromal cells, cytotoxic TILs will impact on cancer growth. In this situation, the TILs analyzed by this study were more representative of an evaluation of local immune status of the tumors.According to Sallusto et al. (8), antigen-experienced CD8+ T cells in PBMCs can be classified into three subsets representing the differentiation stages: nave (CD45RA+CCR7+), effector (CD45RA+CCR7-), and effector memory (CD45RA-CCR7-) cells,. In our study, CD8+ T lymphocytes isolated from tumor sites were mostly antigen-experienced T cells with an effector memory phenotype (88%), while in PBMCs the majority of antigen-experienced CD8+ T cells were composed of both effector (46%) and effector memory (39%) cells. It was reported that although both types of antigen-experienced cells are functional in target cell killing, effector memory cells are less effective (28). These results indicated that CD8+ T lymphocytes in colorectal cancer sites were phenotypically cytotoxic, though their efficacy might not be as strong as that of effector cells. On the other hand, when CD8+ TILs were analyzed according to the T-cell differentiation model defined by Appay et al. [9], two major differentiated phenotypes, CD27+CD28- and CD27-CD28-, were displayed that were different from CD8+ T cells in PBMCs. These results confirmed that the TILs we isolated were not cells from blood vessels but were tumor site infiltrators. The differentiation phenotypes were also distinct from those of metastatic lymph nodes, where the two major subsets of CD27+CD28+ and CD27-CD28-, were detected (11). We first revealed that the phenotype of CD8+ T cells from colorectal cancer tumor sites was a pre-terminally differentiated stage (CD45RA-CCR7-CD27+/-CD28-). Despite the fact that the T cells were infiltrating into a tumor site, they lacked the terminally differentiated subset of CD45RA+CCR-CD27-CD28-. In studies of chronic virus infections, it was found that distinct CD8+ T cell populations representing early to terminally differentiated stages were established corresponding to the immunogenicity of the viruses (9,28). We postulate that the pre-terminally differentiated phenotype of CD8+ T cells in TILs from colorectal cancer could be the result of low immunogenicity of the tumor. As a result, immunosuppressive factors such as IL-10 (ref.10), TGF-b (ref.10), and regulatory T cells (29) in the local microenvironment of the tumor may influence the differentiation and function of T cells.Our results further showed that CD8+ T cells from tumor sites expressed less perforin than those from PBMCs. Interestingly, we found that the low expression of perforin in these cells is the result of low or no expression by their CD27-CD28- subsets. The low level of perforin was not caused by the release of a cytotoxic molecule, since expression of the CD107a marker of degranulation was not detected (19). A previous study reported that the quantity of CD27 expression in CD8+ T cells was reversely correlated with perforin expression, as well as with cytotoxic functions (18). Our results in CD8+ T cells from PBMC are in agreement with this opinion, but perforin expression in CD8+ T cells from tumor sites did not increase while the quantity of CD27 decreased. These results suggest that the function of the CD27-CD28- subset of CD8+ T cells in TILs is not represented by their phenotype as is the case with PBMCs, and indicates a deviation in the status of phenotype and function in these cells --- they were antigen-experienced effector-memory CD8+ T cells which were functionally tolerant. Further study will be needed to clarify the function of these T cells, including if they are regulatory T cells, or they are in a state of anergy only; and if the tolerant state can be reversed by immune interference in vitro or in vivo. Immunohistochemistry labeling showed that the number of intraepithelial lymphocytes was much lower there than in the stroma, indicating that the majority of TILs isolated from fresh tumor tissues were stromal TILs. While stained by CD45RA antibody, CD45RA was not detected in the intraepithelial lymphocytes, suggesting that those cells were pre-terminally differentiated effector-memory cells (data not shown). However, perforin expression was found to be relatively higher in intraepithelial tumors than in stromal tissues. Recently, numorous evidences have been shown that cancer stromal cell killing is critical in cancer tissue eradication. The tolerant state of colorectal cancer stroma CD8+ TILs may explain the irrelative of these cells with clinical prognosis of colorectal cancer. By typing the differentiated phenotypes and functional molecules of infiltrating CD8+ T cells in colorectal cancer, we found that although the CD8+ T cell immune response did exist in tumor sites, most of the cells lacked cytotoxic molecule expression despite being in the intermediate differentiated stage. The deviation of molecular function from the differentiation phenotype could explain the observed clinical phenomenon in which infiltrating lymphocytes do not inhibit cancer growth. Further studies are needed on the mechanism and immunological intervention of TIL function.REFERENCES1 Saeterdal I, Bjorheim J, Lislerud K, et al. Frameshift-mutation-derived peptides as tumor-specific antigens in inherited and spontaneous colorectal cancer. Proc Natl Acad Sci USA.2001; 98: 13255-60.2 Linnebacher M, Gebert J, Rudy W, et al. Frameshift peptide-derived T-cell epitopes: a source of novel tumor-specific antigens. Int J Cancer 2001; 93: 6-11.3 Ito M, Shichijo S, Miyagi Y, et al. Identification of SART3-derived peptides capable of inducing HLA-A2-restricted and tumor-specific CTLs in cancer patients with different HLA-A2 subtypes. Int J Cancer 2000; 88: 633-9.4 Nagorsen D, Keilholz U, Rivoltini L, et al. Natural T-cell response against MHC class I epitopes of epithelial cell adhesion molecule, her-2/neu, and carcinoembryonic antigen in patients with colorectal cancer. Cancer Res 2000; 60: 4850-4.5 Scanlan MJ, Welt S, Gordon CM, et al. Cancer-related serological recognition of human colon cancer: identification of potential diagnostic and immunotherapeutic targets. Cancer Res 2002; 62: 4041–7. 6 Li M, Yuan Y H, Han Y, Liu Y X, Yan L, Wang Y, and Gu J. Expression profile of cancer-testis genes in 121 human colorectal cancer tissue and adjacent normal tissue. Clin Cancer Res. 2005; 11: 1809-147 Naito Y, Saito K, Shiiba K, et al. CD8+ T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res 1998; 58: 3491-4.8 Sallusto F, Lenig D, forster R, Lipp M and lanzavecchia A. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. 1999. Nature. 401: 708-712 9 Appay V, Dunbar PR, Callan M, et al. Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections. Nature Medicine. 2002; 8: 379-38510 Dunn GP, Bruce AT, Ikeda H, Old LJ, Schreiber RD. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol. 2002; 3:991-8. 11 Mortarini R, Piris A, Maurichi A, et al. Lack of terminally differentiated tumor-specific CD8+ T cells at tumor site in spite of antitumor immunity to self-antigens in human metastatic melanoma. Cancer Res. 2003; 63: 2535-4512 Zippelius A, Bartad P, Rubio-Godoy V, et al. Effector function of human tumor-specific CD8 T cells in melanoma lesions: a state of local functional tolerance. Cancer Res. 2004; 64: 2865-73 13 Sprent J, Surh CD. T cell memory. 2002; 20: 551-7914 Hishii M, Kurnick JT, Ramirez-Montagut T and Pandolfi F. Studies of the mechanism of cytolysis by tumour-infiltrating lymphocytes. Clin Exp Immunol. 1999; 116:388–39415 Smyth MJ, Thia KY, Street SE, MacGregor D, Godfrey DI, and Trapani JA. et al. Perforin-mediated cytotoxicity is critical for surveillance of spontaneous lymphoma. J Exp Med. 2000; 192: 755-60 16 Smyth MJ, Street SE, Trapani JA. Cutting edge: granzymes A and B are not essential for perforin-mediated tumor rejection. J Immunology. 2003; 171: 515–817 Watanabe N, Hizuta A, Tanaka N, Orita K. localization of T cell receptor(TCR)- γδ+ T cells into human colorectal cancer: flow cytometric analysis of TCR-γδ expression in tumor-infiltrating lymphocytes. Clin Exp Immunol. 1995; 102: 161-7318 Tomiyama H, Takata H, Matsuda T, Takiguchi M. Phenotypic classification of human CD8+ T cells reflecting their function: inverse correlation between quantitative expression of CD27 and cytotoxic effector function. Eur J Immunol. 2004; 34: 999-101019 Rubio V, Stuge T B, Singh N, et al. Ex vivo identification and anlysis of tumor-cytolytic T cells. Nature Medicine. 2003; 9(11): 1377-8220 Guidoboni M, Gafa R, Viel A, et al. Microsatellite instability and high content of activated cytotoxic lymphocytes identify colon cancer patients with a favorable prognosis. Am J Pathol. 2001;159:297-304 21 Neilsen HJ, Hansen U, Christensen IJ, Reimert CM, Brunner N, Moesggaard F. Independent prognotic value of eosiniphil and mast cell infiltration in colorectal cancer. J Pathol. 1999; 189: 487-9522 Yannelli JR, Hyatt C, McConnell S, et al. Growth of tumor-infiltrating lymphocytes from human solid cancers: summary of a 5-year experience. Int J Cancer. 1996; 65: 413-42123 Hawkins MJ, Atkins MB, Dutcher JP, et al. A phase II clinical trial of interleukin-2 and lymphokine-activated killer cells in advanced colorectal carcinoma. J Immunother 1994; 15: 74-8.24 Fabbri M, Ridolfi R, Maltoni R, et al. Tumor infiltrating lymphocytes and continuous infusion interleukin-2 after metastasectomy in 61 patients with melanoma, colorectal and renal carcinoma. Tumori. 2000; 86: 46-52.25 Marincola F M, Wang E, Herlyn M, Seliger B and Ferrone S. Tumor as elusive targets of T-cell-based active immunotherapy. Trends Immunol. 2003; 24(6): 334-4126 Diederichsen ACP, Zeuthen J, Christensen PB, Kristensen T. Characterisation of tumor infiltrating lymphocytes and correlation with immunological surface molecules in colorectal caner. Eur J Cancer. 1999; 35(5): 721-2627 Costantinia A, Mancinia S, Giuliodoroa S, Butinia L, Regnerya CM, Silvestrib G, Montronia M. Effects of cryopreservation on lymphocyte immunophenotype and function. J Immunol Methods. 2003; 278: 145-5528 Champagne P, Ogg FS, King AS, et al. Skewed maturation of memory HIV-specific CD8 T lymphocyte. Nature. 2001; 410: 106-1129 Unitt E, Rushbrook SM, marshall A, et al. Compromised Lymphocytes Infiltrate Hepatocellular Carcinoma: The Role of T-Regulatory Cells. Hepatol. 2005; 41: 722-730

叶盛威医师：首先介绍一下病人的情况。男性患者，52岁，因“阴茎龟头处肿块一年加重并出血一月”入院。病人因阴茎处肿块并性交痛致近一年无性生活，但排尿正常，一月前肿块增大呈菜花状并出血，在当地医院行包皮环切术无好转，遂入本院，入院后体检：Bp180/100mmHg,阴茎龟头处见约3×3cm菜花样肿块，部分坏死出血，尿道口被肿瘤覆盖，双侧腹股沟未触及肿大淋巴结。实验室检查：血常规示PLT51G/L，血红蛋白、红细胞及白细胞均在正常范围；生化全套示：血磷2.14mmol/L,血清总胆固醇6.32mmol/L，甘油三酯18.87mmol/L(参考值＜1.71mmol/L）；凝血功能检测因高血脂而不能检测；B超示：双侧腹股沟淋巴结肿大，最大为右侧，约1.8×1.07cm；CT示盆腔淋巴结无肿大。胸片及B超示无肝肺转移。初步诊断：1、阴茎癌？；2、高血压病；3、高脂血症；4、血小板减少待查。请各位医师谈谈病人的诊断及治疗方法。熊治国医师：根据病人症状体征，临床考虑为阴茎癌，因病人合并高血压病，需控制血压血脂，目前凝血功能无法检测与高血脂有关，是否血小板减少也与此有关？病人目前疼痛和出血症状明显，为提高病人生活质量，减轻疼痛，可以在病人血压稳定后考虑手术，手术中行快速病理切片，如系恶性行阴茎部分切除术，鉴于病人B超示双侧腹股沟淋巴结肿大，不排除转移可能，可考虑同时行双侧腹股沟淋巴结清扫术。秦敬医师：根据病人情况，诊断无异议，是否应行阴茎全切术？目前病人血小板减少，原因不明，因肿瘤考虑恶性，是否存在肿瘤骨髓转移可能？如行血液学全套检查则时间较长，而目前病人局部疼痛难以忍受。与病人充分沟通后，可以先行手术治疗解决疼痛问题，鉴于病人希望以后保留性功能，也可以行阴茎部分切除术，但应与病人说明可能复发，至于双侧腹股沟淋巴结清扫，需与病人沟通，因为病人术后约50%可能出现并发症如阴囊及下肢水肿，伤口不愈合等。熊治国医师：通过与病人交流并知晓病情后，对可能为恶性肿瘤非常担心，希望尽快手术，并希望保留部分阴茎，遂在血压控制平稳后，在连硬麻下行阴茎部分切除术，术中快速病理检查示恶性，考虑阴茎癌，遂行双侧腹股沟淋巴结清扫术。术后留置导尿管，予抗生素等处理，术后第三天开始阴囊水肿逐渐加重，术后病理检查：阴茎鳞癌（肉瘤样癌），侵及尿道海绵体，断端切缘未见癌，右侧腹股沟淋巴结1/5见癌转移，左侧腹股沟无淋巴结转移。叶盛威医师：我介绍一下术后的治疗经过。术后十二天时拔出导尿管，三天后因排尿困难，再次行导尿，见阴茎断端组织愈合不良，术后半月因阴茎处疼痛，阴囊高度肿胀，见阴茎断端肉芽组织较硬，呈腐肉状，遂取材活检，病理报告示：复发性鳞癌，，免疫组织化学检查支持肉瘤样癌。因病人疼痛症状明显，而且阴囊肿胀持续存在，且病人合并发热，血常规示白细胞升高，考虑阴茎癌残留并阴囊感染，遂于术后20天遂行膀胱耻骨上造瘘，阴囊处引流，右侧腹股沟肉芽组织活检，术后病理报告为肿瘤复发；术后一月出现呼吸困难，胸片示双肺多发结节，考虑阴茎癌肺转移（见图）。鉴于病人病情发展快，请各位医师谈谈诊治教训。夏和顺医师：结合病人临床表现及病理切片结果（图），病理诊断为阴茎癌，因本病例细胞形态不典型，进一步免疫组化示：PCK(＋),Vim(＋)，而CD68（─）,CD34（─），Melan A（─），排除淋巴瘤及恶性黑色素瘤，考虑分型为肉瘤样鳞癌，很少见，预后差。张克亮医师：病人诊断为阴茎癌，且病理类型为肉瘤样鳞癌，恶性程度高，而且肿瘤侵犯龟头超过1/2，侵犯深度至尿道海绵体，属于T2，因此病人最好行阴茎全切除术，术后十二天病人行阴茎部分切除术复发后，可以行全切除术，但病人1、阴囊严重水肿，不排除肿瘤已经侵犯阴囊及尿道，无法分离尿道，2、血常规示血小板为50万，且后来复查进行性下降，不排除病人本身存在阴茎癌骨髓转移可能，因此没有必要扩大手术。但初次手术时不行双侧腹股沟淋巴结清扫术可以提高病人的生活质量。魏少忠医师：病人入院时血小板减少，未进一步查明原因，回头分析，不能排除骨髓转移可能，因此本病人入院时可能为阴茎癌晚期，而且肿瘤病理类型为肉瘤样鳞癌，预后极差，手术前应向病人说明，采取姑息性外科治疗如阴茎部分切除术也符合治疗原则。叶章群医师：病人诊断是明确的，在分期上入院时没有明显远处转移的征象，但一月后胸片显示存在肺转移，说明原来可能存在潜在转移灶未被发现。治疗上可以采用化疗，但目前的化疗方案疗效并不确切，采取姑息性手术减轻病人的疼痛也可以，鉴于病人病期偏晚，最好行阴茎全切除术，最大程度减少病人局部复发可能，而使病人疼痛减轻。至于是否做腹股沟淋巴结清扫，目前观点认为，术前未触及淋巴结肿大可以不行淋巴结清扫，即使病人存在淋巴结肿大，也可以二期手术，因为一部分病人淋巴结肿大可能为炎症导致，原发灶手术后这些淋巴结可以消退，而没有消退的病人可能是真正的淋巴结转移，在原发灶手术后1-2月再行腹股沟淋巴结清扫，术中快速病理检查，如果转移淋巴结超过2个，再行同侧盆腔淋巴结清扫术。具体到这个病人，第一次手术时不行双侧腹股沟淋巴结清扫，术后不会出现阴囊水肿，胀痛进而影响病人的生活质量。目前，阴茎癌是一种少见的疾病，而肉瘤样癌仅占阴茎癌的1-2%，恶性程度高，而本病人病程已经有一年，病期晚，因此预后肯定不佳。此病人给我们的教训是：1、应该在术前明确病人肿瘤的病理类型及较准确的分期，与病理科医生沟通至关重要，这样让病人可以充分知晓病情；2、手术前也可以考虑诱导化疗，手术的目的是解决病人的痛苦，姑息性为主，尽可能选用创伤小，并发症少的术式，如不行腹股沟淋巴结清扫。附图