Rivera J A, Graeme-Cook F, Werner J, Z'graggen K, Rustgi A K, Rattner D W, Warshaw A L, Fernández-del Castillo C
Department of Surgery, Massachusetts General Hospital, Harvard Medical School, Boston 02114, USA.
Surgery. 1997 Jul;122(1):82-90. doi: 10.1016/s0039-6060(97)90268-3.
Current experimental models of pancreatic cancer either fail to reproduce the ductal phenotype or cause simultaneous cancers in other organs also. To develop an animal of pancreatic cancer that accurately mimics the human condition, we restricted carcinogenic exposure to the pancreas and specifically targeted ductal epithelial cells. Three different carcinogens were either implanted directly into the pancreas or infused into the pancreatic duct, with or without near-total pancreatectomy (as a means of inducing pancreatic ductal cell proliferation).
Groups of male Sprague-Dawley rats were exposed to varying doses of dimethylbenzanthracine (DMBA), methynitronitrosoguanidine, or ethylnitronitrosoguanidine either through direct implantation into the pancreas or infusion into the pancreatic duct. Near-total pancreatectomy was added in all groups except two DMBA implantation groups. Surviving rats were killed at 3, 6, 9, or 12 months, and the pancreata were evaluated histologically.
All three carcinogens caused pancreatic inflammation, ductal hyperplasia, atypia, and dysplasia beginning by 3 months and becoming more prominent at later time points. Only DMBA caused frequent invasive pancreatic ductal adenocarcinoma, which was first evident by 6 months. The prevalence of pancreatic cancer among DMBA-treated rats evaluated after 10 months was 39% (19 of 49). The addition of pancreatic resection did not enhance pancreatic cancer development.
Of the strategies tested, only direct implantation of DMBA into the rat pancreas frequently produces pancreatic cancer histologically similar to human ductal adenocarcinoma. The development of hyperplastic, atypical, and dysplastic changes preceding and accompanying carcinomas suggests that these lesions are preneoplastic. This model recapitulates the progression from normal to neoplastic epithelium and is likely to be useful for the study of morphologic and molecular mechanisms underlying the early stages of pancreatic carcinogenesis and for the investigation of novel diagnostic and therapeutic techniques.
目前的胰腺癌实验模型要么无法重现导管表型，要么还会在其他器官同时引发癌症。为了培育出能准确模拟人类病情的胰腺癌动物模型，我们将致癌物质的暴露限制在胰腺，并专门针对导管上皮细胞。三种不同的致癌物质通过直接植入胰腺或注入胰管的方式给药，同时或不进行近全胰腺切除术（作为诱导胰腺导管细胞增殖的一种手段）。
将雄性斯普拉格-道利大鼠分组，通过直接植入胰腺或注入胰管的方式，使其接触不同剂量的二甲基苯并蒽（DMBA）、甲基硝基亚硝基胍或乙基硝基亚硝基胍。除两个DMBA植入组外，所有组均进行近全胰腺切除术。存活的大鼠在3、6、9或12个月时处死，对胰腺进行组织学评估。
所有三种致癌物质在3个月时均引发胰腺炎症、导管增生、异型性和发育异常，且在后期时间点更为明显。只有DMBA频繁引发浸润性胰腺导管腺癌，6个月时首次明显出现。10个月后评估，接受DMBA治疗的大鼠中胰腺癌的患病率为39%（49只中的19只）。胰腺切除术的加入并未促进胰腺癌的发展。
在测试的策略中，只有将DMBA直接植入大鼠胰腺才能频繁产生组织学上与人类导管腺癌相似的胰腺癌。癌前和伴随癌出现的增生性、异型性和发育异常变化表明这些病变是癌前病变。该模型概括了从正常上皮到肿瘤上皮的进展过程，可能有助于研究胰腺癌发生早期阶段的形态学和分子机制，以及用于研究新型诊断和治疗技术。
