1752 K-ras Point Mutations in Cancerous and Noncancerous Biliary Epithelium in Patients with Pancreaticobiliary Maljunction Toshiki Matsubara, M.D. Yoichi Sakurai, M.D., m.0. Yoshinori Sasayama, M.D. Haruna Hori, M.D. Masahiro Ochiai, M.D. Takahiko Funabiki, M.D. Kazuyuki Matsumoto, M.D. lwao Hirono, M.D. Departments of Surgery and Pathology, Fujita Health University School of Medicine, Toyoake, Aichi. Japan. Presented at the Seventh Annual Meeting of the Japanese Research Society for Gastroenterological Carcinogenesis, Oita, Japan, September 8-9. 1995. This study was supported by a grant from Fujita Health University. Address for reprints: Takahiko Funabiki, M.D., Department of Surgery, Fujita Health University School of Medicine, 1-98 Dengakugakubo Kutsukake-cho. Toyoake Aichi 470-1 1, Japan. Received December 4, 1995; revision received January 26, 1996; accepted January 26, 1996. 0 1996 American Cancer Society BACKGROUND. Pancreaticobiliary maljunction (PBM), an anomalous union of the pancreatic duct with the common bile duct, has frequently been shown to be associated with biliary carcinoma. However, the mechanism of carcinogenesis is unknown. METHODS. Mutations of the K-ras oncogene were examined in cancerous and noncancerous biliary tract epithelium of 20 patients with PBM by an extraction of DNA from surgically resected histologic specimens. DNA was analyzed by a polymerase chain reaction single strand conformation polymorphism (PCR-SSCP) method and direct sequencing. RESULTS. An abnormally mobilized DNA band was detected not only in cancerous epithelium but also in hyperplastic, metaplastic, and inflammatory epithelium of the gallbladder andlor common bile duct in patients with PBM. Among the biliary epithelium of patients with PBM, point mutation of K-rus oncogenes were detected in 4 of 5 (80%) cancerous epithelium, 7 of 12 (58%) hyperplastic and metaplastic epithelium, and 8 of 18 (44%) inflammatory epithelium, whereas no point mutation of the K-rus oncogene was detected in the gallbladder epithelium in 3 control patients without PBM. Direct sequence analysis of the K-rus oncogene revealed the mutation at codon 12 substituting the wild-type glycine (GGT) for aspartic acid (GAT) in all cancerous lesions of patients with PBM. Simultaneous two-point mutations from the wild-type glycine (GGC) to arginine (CGC) at codon 13 associated with the mutation at codon 12 were also found in one case of gallbladder carcinoma and one case of bile duct carcinoma. CONCLUSIONS. K-ras gene mutation is involved in the carcinogenesis of biliary tract epithelium in patients with PBM, and appears to be a high risk factor for carcinogenesis of the biliary tract. Cancer 1996; 77:1752-7. 8 1996 American Cancer Society. KEYWORDS: pancreaticobiliary maljunction, K-ras, oncogene, polymerase chain reaction, single strand conformation polymorphism. S ince Irwin and Morison' reported the first case of a carcinoma arising in a congenital cyst of the common bile duct in 1944, there has been a notion that congenital biliary dilatation may be an important disease entity that causes carcinoma of the common bile duct. In 1969 Babbit et a1,' and then others, advocated that congenital biliary cyst is frequently accompanied by pancreaticobiliary maljunction (PBM). Furthermore, it has also been reported that patients with PBM whose bile duct is not dilated often demonstrate gallbladder carcinoma. Thus, PBM itself could in part be a cause of carcinoma arising in the gallbladder and common bile duct. Biliary carcinoma occurring in patients with PBM may be distinct, because this carcinoma frequently occurs in a relatively younger K-ras Mutation in Pancreaticobiliary Maljunction/Matsubara et al. age group than that occurring in patients who do not have PBM.'-5 Conversely, there have been clinical observations that retlux of pancreatic juice into the biliary tract occurs in patients with PBM and that regurgitated pancreatic juice stagnates in the dilated common bile duct or in the gallbladder. This phenomena may cause chemical irritation, various histologic changes, and, ultimately, carcinogenesis in biliary epithelium. Clinical findings suggest that patients without biliary dilatation often develop gallbladder carcinoma, whereas patients with biliary dilatation develop carcinoma of the common bile duct. This proposed mechanism of carcinogenesis in the bile duct epithelium has also been supported by our previous in vitro and in vivo studies demonstrating that the mixture of bile with pancreatic juice increased mutagenicity."-' Based on these findings, we recommended surgical resections of the extrahepatic bile duct and gallbladder combined with so-called "diversion surgery" that separates bile flow from pancreatic juice, regardless of the association of cystic dilatation of the common bile duct with PBM.' This surgical procedure was used to obtain the bile duct and/or gallbladder specimens in patients with PBM. Recent studies have shown that K-rus oncogene activation occurs in gastrointestinal carcinomas, including pancreatic c a r ~ i n o m a , ' ~colorectal -'~ c a ~ c i n o r n a , 'and ~-~~ biliary c a r c i n o ~ n a s . ~ ~ Greater -~' than 90% of pancreatic adenocarcinomas have been reported to contain a K-rus oncogene activated by a mutation at codon 12.23-25 Thus, K-rusoncogene mutation may play an important role in carcinogenesis. In the present study, we attempted to determine whether genetic alterations in the K-rm oncogene that could explain the initial event of carcinogenesis occurring at the genetic level are involved in the different types of lesions of the biliary tract epithelium in patients with PBM. The DNA extraction of surgically resected histologic specimens combined with a polymerase chain reaction single strand conformation polymorphism (PCR-SSCP)'6-'8 was used to examine the alterations of the K-rusoncogene. MATERIALS AND METHODS Tissue Specimens Surgically resected biliary duct epithelial tissues of 20 patients with PBM and 3 patients with benign biliary diseases not associated with PBM were obtained from the Department of Surgery, Fujita Health University Hospital. Among the 20 patients with PBM, 5 demonstrated carcinomas and 15 did not. Five specimens of gallbladder and biliary carcinomas, 8 specimens and 4 specimens of hyperplastic andlor metaplastic gallbladder and bile duct epithelium respectively, 6 specimens of inflammatory gallbladder epithelium, and 12 specimens of inflammatory bile duct epithelium were obtained from the patients 1753 TABLE 1 Synthetic OligonucleotidesUsed for Flanking Primers of Polymerase Chain Reaction and Sequencing Primer K-rasat codon 12/13 Flanking primer Sequencing primer "CTGAGTATAAACTTGTGGTA3' "TATCGATTAAGTCTTAGTGA3' "TGCTATGTCGATTAAGTCTT3' (Sense) (Antisense) (Antisense) with PBM. Gallbladder epithelia obtained from the patients without PBM were used for control bile duct epithelium. Renal cortical tissue and gallbladder epithelium not showing cholesterol polyps were used as negative controls. Tissues had been fixed in 10%formalin and embedded in paraffin for routine surgical and pathologic diagnoses. Extraction of DNA Three 5- to 7-pm tissue sections were cut from the tissue blocks. One of each section was stained with hematoxylin and eosin to confirm the histologic types of biliary tissues and the localization of the target lesion in which the K-ras oncogene point mutation was examined in the histologic section. For tissue sections from one tissue block that contained different types of histology, particular portions of tissue in the histologic sections targeted were identified and selectively scraped out and used for further DNA extraction. This procedure might eliminate the possibility that target tissue was contaminated with surrounding cells. The DNA of each tissue section was extracted using a Sepa-gene kit (Sanko Junyaku Co., Tokyo, Japan).29 Polymerase Chain Reaction Oligonucleotide primers for K-rus exon 1 were artificially synthesized using a DNA synthesizer (Model 391, Applied Biosystems, Foster City, CA) and then purified by high performance liquid chromatography. The oligonucleotide flanking primer and the sequencing primer used for the polymerase chain reaction (PCR) are shown in Table 1. The PCR analysis was performed using 1 yg of DNA, 0.7 unit of Taq polymerase (Takara Biochemical Co. Ltd., Kyoto, Japan) and 15 pmol of primers in which both terminals were labeled. Thirty cycles of PCR were sequentially performed at 94 "C for 30 seconds, at 55 "C for 30 seconds, and then at 72 "C for 1 minute. These reaction processes were carried out in a programmable thermo-controller (MJ Research, Inc., Tokyo, Japan). Single Strand Conformation Polymorphism Five microliters of PCR reaction mixture were diluted with 45 pL of 95% formamide, 20 mM EDTA, 0.05% bromopheno1 blue, and 0.05% xylene cyanol and heated at 80 "C 1754 CANCER Supplement April 15, 1996 I Volume 77 I Number 8 for 2 minutes. Two microliters of samples were applied to 6% polyacrylamide gel containing 90 mM Tris borate (pH 8.3) and 4 mM EDTA. Electrophoresis was performed at 40 W for 2-2.5 hours. The gel was then dried on filter paper and the exact location of the labeled bands was identified using autoradiography. PCR-amplified K-rus exon 1 fragments generated 128 base pair DNA bands, which were confirmed in the SSCP gel. Direct DNA Sequencing To determine the point mutations, direct sequencing was performed using an abnormally mobilized DNA band found in the SSCP gel. The abnormally mobilized DNA fragments in the SSCP gel were selectively isolated from the template, and then dissolved in 50 yL of distilled water for 30 minutes at 80 "C. DNA samples of 15 pL of aliquots were amplified by an asymmetric PCR with 2.5 units of Taq DNA polymerase using both 50 and 5 pmol of paired primers. Fifty cycles of PCR were sequentially performed at 94 "C for 30 seconds, at 55 "C for 1 minute, and then at 72 "C for 2 minutes. After phenollchloroform extraction and ethanol precipitation, unincorporated nucleotide and primers were eliminated from amplified DNA by filtering through a polysulfone filter that removed molecules a molecular weight of greater than 10,000. Samples were then subjected to sequence reaction using Sequennase Version 2.0 (United States Biochemical Corp., Cleveland, OH) with [a3'P]dCTP according to the manufacturer's instruction. RESULTS All histologic specimens of noncancerous biliary epithelium in patients with PBM had inflammatory changes and 10 of 20 cases (50%) also showed hyperplastic andlor metaplastic lesions (Fig. 1). DNA samples extracted from 35 lesions in 20 cases were used for further analysis. SSCP analysis of K-rus exon 1 is shown in Figure 2. Although the activated K-rus oncogene was not detected in any of the gallbladder epithelium of the patients without PBM and renal tissues, abnormal bands that corresponded to the mutant allele were detected not only in the cancerous epithelium but also in the hyperplastic, metaplastic, and inflammatory epithelium of the gallbladder andlor common bile duct in the patients with PBM. Point mutations in the K-rus oncogene were found at codon 12 substituting glycine (GGT) for aspartic acid (GAT) and at codon 13 substituting glycine (GGC) for arginine (CGC) (Fig. 3). Two point mutations were detected on exon 1 in the DNA extracted from the same specimen in one case of gallbladder carcinoma and one case of bile duct carcinoma (Patients 2 and 5). K-rus point mutation at codon 12 was found in 4 of 5 (80%) biliary carcinomas, 7 of 12 (58%) hyperplastic and/or metaplas- FIGURE 1. Representative histologic sections of cancerous (C; top), hyperplastic and/or metaplasia (H/M; middle) and inflammatory biliary epithelium (I; bottom) stained with hematoxylin and eosin. Portions of the tissues targeted were selectively scraped out and the DNA samples were extracted from these tissues after the histologic types of lesions were confirmed (H & E, original magnification x40). tic lesions, and 8 of 18 (44%) inflammatory epithelium, whereas no point mutation of the K-rus oncogene was detected in the gallbladder epithelium of patients without PBM (Table 2). DISCUSSION The primary aim of the present study was to test whether K-ras oncogene activation occurs in gallbladder and/or common bile duct epithelium during the process of carcinogenesis in patients with PBM. The present study demonstrated that regardless of the histologic types of biliary tract epithelial tissues in patients with PBM, the K-rus K-ras Mutation in Pancreaticobiliary Maljunction/Matsubara et al. FIGURE 2. Detection of the K-ras oncogene mutated at codon 12 in single strand conformation polymorphism analysis in cancerous, hyperplastic and/or metaplastic, and inflammatory biliary epithelium in patients with pancreaticobiliary maljunction and in patients without it. Arrows indicate the positions of the mutant (m) and wild-type (w) bands. (A part of the experimental results were presented.) FIGURE 3. Direct sequencing of the K-ras gene at codons 12 and 13. (1) A mutation at codon 12 substituting glycine (GGT) to aspartic acid (GAT) found in a patient with gallbladder carcinoma. (2) A mutation at codon 13 substituting glycine (GGC) to arginine (CGC) found in a patient with bile duct carcinoma. (3) A mutation at codon 12 substituting glycine (GGT) to aspartic acid (GAT) found in a patient with hyperplasia and/or metaplasia. (4) No mutations of codons 12 and 13 were found in the nonneoplastic gallbladder epithelium of a patient without pancreaticobiliary maljunction. oncogene that was mutated at either codon 12 or 13 was frequently detected in cancerous, hyperplastic andlor metaplastic, and even inflammatory biliary tract epithelium, whereas no mutated K-ras oncogene was detected in the gallbladder epithelium of patients without PBM. These results clearly suggest for the first time that chronic exposure of the biliary epithelium to the regurgitated mixture of bile and pancreatic juice could stimulate the genetic alterations that ultimately may result in carcinogenesis in the biliary tract epithelium in patients with PBM. The reflux of pancreatic juice has been shown to occur in patients with PBM and is considered to be due to the anatomic finding that in patients with PBM, the pancreatic 1755 duct is connected with the common bile duct outside of the duodenal wall where no sphincter of Oddi exists."" The reflux of pancreatic juice is known to occur as well in patients with biliary cysts and this phenomena has also been reported to possibly contribute to the chronic inflammatory changes usually observed in the biliary epithelium of these patients. The reflux of pancreatic juice back into the common bile duct that stagnates in the biliary tract may cause a variety of histologic changes in the biliary epithelium, leading to the conclusion that biliary tract epithelium in patients with PBM has a high malignant potential. This has been supported by our in vitro studies using human gallbladder bile and pancreatic juice from patients with PBM that have shown sipficant mutagenicity. Although the carcinogenic process in biliary epithelium into which pancreatic juice is regurgitated to the common bile duct has been explained by a variety of m e c h a n i ~ m s , ~no ~ " genetic ~ ~ ~ alteration in biliary tract tissues in patients with PBM has been reported in the past. Alterations in the ras oncogene have been reported in various human tumors, and the incidence rates of K-rasmutations have generally been reported to be 0-93%.'" Pancreatic adenocarcinomas in particular have a high incidence rate, between 75-93%.'3s'3-25Reports regarding the incidence rates of K-ruspoint mutation in gallbladder carcinomas show them to be relatively lower than those in pancreatic carcinomas, despite the fact that they share a common developmental origin in the embryonic foregut.'0~'8~'" Although previous reports regarding point mutations detected in human biliary tract tissues have used the PCR-SSCP technique to enhance the K-ras oncogene, the detailed methods of PCR used are different from study to study. Thus it is possible that the incidence of K-raspoint mutation detected is dependent on the different methodologies used in each study. Tada et a1.22reported that only 9 of 18 tumors (56%)of cholangiocellular carcinomas contained the ras gene mutation when PCR and direct sequencing were used. Their technique has been reported to be able to detect the mutant allele only when at least 20% of the sample contains the ras mutation. Conversely, Almoguera et al.'" have found no mutations in five gallbladder carcinomas using the RNAse A mismatch cleavage method. In the present study, we utilized the SSCP method, in which the synthesized primer was used during the process of PCR. When the extraction of DNA in cancerous tissue was performed, we selectively scraped out the cancer cells in the histologic section, identifying them under the light microscope to eliminate the surrounding inflammatory cells. In addition, we have demonstrated in the present study that no K-ras point mutation was detected in the gallbladder epithelium of patients who did not have PBM. In this study, a high incidence rate of point mutations of the K-rasoncogene in the biliary tract epithelium of patients with PBM is 1756 CANCER Supplement April 15,1996 I Volume 77 / Number 8 TABLE 2 K-rasOncogene at Codon 12 and 13 in Pancreaticobiliary Maljunction Patient Age Sex 1 2 3 4 5 6 7 8 62 57 57 F F F F F Gallbladder C 9 10 11 12 13 14 15 16 17 18 19 20 71 35 34 47 41 36 63 26 25 23 47 60 M 49 M 26 34 F M M hl 42 36 HIM Common bile duct I C HIM I 0 I 0 @ @ 0 F F F F F F F F F @ @ @ 0 @ 0 r, @ 0 0 0 0 0 0 0 Ci @ C 0 @ @ 0 0 0 C: carcinoma; HIM: hyperpiasla and/or metaplasia: I: Inflammation. 0: K-rasoncogene mutated at codon I? substituting the wild-tpe glycine IGGT) for aspartic acid (GAT); I: two simultaneous point mutations, in which the K-rasoncogene mutated both at codon 13 from glycine IGGC) or arginine (CGCI and at codon I? from glycine (GGTI to aspartic acid (GAT); 0:no mutation. surprising and would lead us to confirm the general hypothesis that the mixture of bile and pancreatic juice map cause genetic alterations in the bile duct epithelium. Our findings that the high incidence rates of K-ras point mutations in patients with PBM are detected in cancerous, hyperplastic andlor metaplastic, and inflammatory lesions of the biliary epithelium are fundamentally similar to the findings of carcinogenesisin colonic epithelial cells, in which the incidence of alteration of the gene increases as the histologic neoplastic grade increases.16 Our findings that 50% of nonneoplastic biliary epithelium in patients with PBM showed alteration of the K-ras gene is striking, because no point mutation of the K-ras gene was detected in the gaUbladder epithelium in patients without PBM. Thus, it is possible that bile duct epithelium in patients with PBM has already undergone neoplastic changes at the genetic level before these changes are histologically detected. However, these findings are not consistent with the previous report suggesting the stepwise progression of carcinogenic process regarding the adenorna-carcinoma sequence demonstrated in colonic epithelial Carcinogenic processes in biliary carcinoma, such as initiation and progression, are complex; it is not possible to explain the overall events occurring during the process of carcinogenesis by a single factor. Nonetheless, we demonstrated for the first time that K-rm oncogene mutations are found in noncancerous biliary epithelium, including hyperplasia, metaplasia, and even inflammation before the neoplastic changes are histologically detected. Further investigation is required to definitely determine whether these genetic alterations are destined to occur in patients with PBM or if they are caused by environmental factors such as chemical irritation from the reflux of pancreatic juice into the biliary tract. REFERENCES 1. 2. 3. 4. 5. Irwin ST, Morison JE. Congenital cyst of the common bile duct containing stones and undergoing cancerous change. B r ] SUrg 1944;32:319-21. Babbit DP. Congenital choledochal cysts: new etiological concept based on anomalous relationships of the common bile duct and pancreatic bulb. Ann Radio/ (Paris) 1969;12: 231-40. Voyles CR, Smadja C, Shands WC, Blumgart LH. Carcinoma in choledochal cysts: age-related incidence. Arch Surg 1983;118:986-8. Konii N, Kunimoto K, Tamiira T, Udaka H, Miyoshi Y, Takehara H. Nationwide survey of cases of choledochal cyst. Analysis of coexistent anomalies, complications and surgical treatment in 645 cases. Surg Gastroenterol 1984;3:69-73. Kimura K. Ohto M, Ono T, Tsuchiya Y, Saisho H, Kawamura K, et al. Congenital cystic dilatation of the common bile duct: relationship to anomalous pancreaticobiliary ductal union. Am 1 Roentogenol 1977;128:571-7. K-ras Mutation in Pancreaticobiliary Maljunction/Matsubara et al. ti. 7. 8. 9. 10 11. 12. 13. 14. 15. 1ti. 17. 18. 19. Matsubara T, Sasayama Y, Yamaguchi H, Futawatari H, Imazu €3, Arai K, et al. f3iochemical and histological study on a long term survived canine model of the pancreaticobiliary maljunction. J Jpn Res Soc Gastroenterol Carcinog 1994;ti: 149-53. Matsubara T, Tsuji T, Miyama A, Yamaguchi H, Funabilu T. Mutagenicity of bile and pancreatic juice from patients with pancreatico-biliary maljunction. Hepatogustroenterology 1995;42:113-6. Funabiki T, Sugiue K, Matsubara T, h n a n o H, Ochiai M. Bile acids and biliary carcinoma in pancreaticobiliary nialjunction. Keio J Med 1990;40:118-22. Funabiki T, Matsubara T, Sugiue K, Kamei K, Amano H, Ochiai M. Surgery for pancreaticobiliary maljunction. In: Takada T, Vaniyapongs T, Nimsakul N, Shaipanich T, editors. New frontiers in hepatobiliary-pancreatic surgery. Bangkok Bangkok Medical Publisher, 1993:109- 12. Alnioguera C, Shibata 11, Forrester K, Martin K, Arnheim J, Perucho M. Most human carcinomas of the exocrine pancreas contain mutant c-K-rus genes. Cell 1988;53:549-54. Mariyania M, Kishi K, Nakamura K, Obata H, Nishimura S. Frequency and types of point mutation at the 12th codon of the c-K-ras gene found in pancreatic cancers from Japanese patients. Jpn J Cancer Res 1989;80:622-6. Tada M, Omata M, Ohto M. Ras gene mutations in intraductal papillary neoplasms of the pancreas: analysis in five cascbs. Cancer 1991;67:634-7. Moiojima K, Tsunoda T, Kanematsu T, Nagata Y, Urano T, Shiku H. Distinguishing pancreatic carcinoma from other periampullaiy carcinomas by analysis of mutations in the Kirsten-ras oncogene. Ann Surg 1991;214:657-61. Forrester I<, Almoguera C, Han K, Grizzle WE, Perucho M. Detection of high incidence of K-ras oncogenes during human colon tumorigenesis. Nature 1987;327:298-303. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990;61:759-67. Vogelstein B, Fearon ER, Hamilton SR. Genomic alterations during colorectal-tumor development. N Engl J Med 1988; 319 525-32. Levi S, Urbario-Ispizua Gill R, Thomas DM, Gilbertson J , Foster C. Marshall CJ. Multiple K-ras codon 12 mutation in cholangiocarcinomas demonstrated with sensitive polymerase chain reaction technique. Cancer Res 1991;51:3497-502. Tada M, Yokosuka 0, Omata M, Ohto M, Isono K. Analysis of rus gene mutations in biliary and pancreatic tumors by polymerase chain reaction and direct sequencing. Cancer 1990;66:930-5. Imai M, Hoshi T, Ogawa K. K-ras codon 12 mutations in 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 1757 biliary tract tumors detected by polymerase chain reaction denaturing gradient gel electrophoresis. Cancer 1994;73: 2727-33. Watanabe M. Asaka M, Tanaka 1, Kurosawa M, Kasai M, Miyazaki T. Point mutation of K-rus gene codon 12 in biliary tract tumors, Gastroenterology 1994;107:1147-53. Tsuda H, Satarug S, Bhudhisawasdi V, Kihana T, Sugimura T, Hirohashi S. Cholangiocarcinomas in Japanese and Thai patients. Mol Carcinog 1992;f3266-9. Tada M, Omata M, Ohto M. High incidence of rus gene mutation in intrahepatic cholangiocarcinoma. Cancer 1992; 69:1115-8. Smit VTHBM, Boot AIM, Smit AMM, Fleuren GJ, Cornelisse CJ, Bos JL. K-ras codon 12 mutations occur very frequently in pancreatic adenocarcinomas. Nucleic Acids Res 1988;16: 7773-82. Nagata Y, Motoshima K, Nakayama E, Shiku H. Frequent glycine-aspartic acid mutations at codon 12 of c-k-ras gene in human pancreatic cancer in Japanese. Jpn J Cancer Res 1990;81:135-40. Gmnwald K, Lyons J, Frohlich A, Feichtinger H, Weger RA, Schwab G, et al. High frequency of Kwas codon 12 mutations in pancreatic carcinomas. Int J Cancer 1989;43:1037-41. Suzuki Y, Sekiya T, Hayashi K. Allele-specific polymerase chain reaction: a method for amplification and sequence determination of a single component among a mixture of sequence variants. Anal Biochem 1991;19232-4. Orita M, Iwahana H, Kanazawa H, Hayashi K, Sekiya T. Detection of polymorphism of human DNA by gel electrophoresis as single-stranded conformation polymorphism. Proc Nut1 Acad Sci USA 1989;86:2766-70. Orita M, Suzuki Y, Sekiya T, Hayashi K. Rapid and sensitive detection of point mutations and DNA polymorphisms using polymerase chain reaction. Genomics 1989;5:874-9. Funato T, Hoshino A, Kawamura T. Evaluation of a commercial extraction kit (Sepa Gene) for DNA diagnosis. Igaku to Yakugaku 1993;29:1401-4. Todani T, Arima E, Eto T, Funabiki T, Kakita A, Kimura K, et al. Diagnostic criteria of pancreaticobiliary maljunction. J Hepatobiliary Pancreatic Surg 1994;1:219-21. Hizawa K, Fukuda F, Akagi G, Kitamuro T, Tao S. Carcinoma associated with idiopathic choledochus cyst. Acta PatholJpn 1967;17:lOl-ti. Neiderhiser DH, Morningstar WA, Roth HP. Absorption of lecithin and lysolecithin by the gallbladder. J Lab Clin Med 1973;821891-7. Bos JL. ras oncogenes in human cancer: a review. Cancer Res 1989;49:4682-9.