K-ras mutations in stools and tissue samples from patients with malignant and nonmalignant pancreatic diseases

¹ Institute of Laboratory Medicine and Pathobiochemistry and
² Clinic of Surgery, Medical Faculty Charité, Humboldt-University, Schumannstrasse 20/21, D-10117 Berlin, Germany.

³ InViTek GmbH, Berlin Buch, 10362 Berlin, Germany.
^a Address correspondence to this author at: Universitätsklinikum Charité, Campus Charité-Mitte, Institut für Laboratoriumsmedizin und Pathobiochemie, Schumannstrasse 20/21, 10098 Berlin, Germany. Fax 049-30-2802-1400; e-mail christoph.berndt{at}charite.de.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Mutant-enriched PCR and reverse dot blot hybridization in microplates were applied for examining K-ras status in stools and tissue samples from patients with pancreatic tumors and chronic pancreatitis. In tissue samples, K-ras mutations were found in 32 of 35 cases of ductal adenocarcinoma, in 5 of 7 periampullary cancers, in 1 cystadenocarcinoma, and in 3 of 5 patients with chronic pancreatitis. In stools, mutated K-ras was seen in 10 of 25 cases of ductal adenocarcinoma, in 1 case of cystadenocarcinoma, and in 2 of 6 cases of chronic pancreatitis. These data indicate that the K-ras status of stool samples may help identify pancreatic carcinoma and persons at risk for cancer development; however, it does not allow discrimination of malignant from nonmalignant diseases.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
With a 5-year survival rate on the average <5%, pancreatic carcinomas are among the neoplasias with the worst prognoses (1). Because the disease is characterized by a long-term interval of nonspecific symptoms associated with an early appearance of mainly lymphogenic metastases, only a minority of patients can be treated by curative surgery at the time of diagnosis. However, the 5-year survival rates may exceed 20% if resections are performed at early tumor stages when the patients are free from infiltrated lymph nodes and perineural or duodenal invasion (2). Thus, earlier tumor detection is the major means of improving the prognosis for pancreatic cancers unless novel treatment approaches, e.g., gene or immune therapy, are broadly applicable.

Conventional tumor markers, such as CA 19-9, have a diagnostic sensitivity of >70%; however, they lack specificity and are poorly suited to reveal the initial stages of cancer progression (3). Because its potential to detect single neoplastic cells, PCR could be valuable in identifying curable pancreatic tumors provided appropriate genomic markers are available. A potential molecular marker for pancreatic cancers is K-ras. The K-ras protooncogene codes for a 21-kDa protein that belongs to a highly conserved family of GTPases involved in signal transduction (4)(5). When growth factor binds to receptor tyrosine kinases, Ras proteins become activated by an exchange of the associated guanine nucleotide GDP by GTP. Engaging several effectors, the best known of which is c-Raf, GTP-Ras initiates downstream cascades, eventually leading to the phosphorylation of key transcription factors. Oncogenic forms of K-Ras are permanently activated proteins that result from point mutations of the codons 12, 13, or 61. More than 80% of all pancreatic cancers harbor activated K-ras, with mutations being virtually restricted to codon 12 (6). This unique constellation simplifies analytical approaches and allows for mutant enrichment techniques (7). Furthermore, oncogenic K-ras occurs at early stages of tumor progression (8) and can be determined in a noninvasive manner by analyzing stool (9). Thus, the K-ras status in stool is a candidate test for screening persons at increased risk for pancreatic tumors, for example, elderly persons, [~80% of pancreatic cancers occur between the ages of 60 and 80 (10)], patients with chronic pancreatitis (11), or those with predisposing hereditary disorders, e.g., BRAC2 mutation carriers (12). To gain additional information on its diagnostic value, we have compared the K-ras status of stools with that of tissue samples and serum tumor markers CA 19-9 and CEA from patients suffering from pancreatic tumors and chronic pancreatitis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
patients and diagnosis
Stools and tissue samples were obtained with informed consent of the patients. Histological diagnosis and tumor grading and staging were done according to the Union Internationale Contre le Cancer criteria (13). Pancreatic tumors are classified as follows: pT1, tumors restricted to the pancreas; pT2, tumor invasion into duodenum, bile duct, and/or peripancreatic tissue; or pT3, tumor invasion into stomach, spleen, colon, and/or major vessels. Involvement of regional lymph nodes is indicated by pN0 or pN1 and distant metastases by pM0 or pM1.

dna preparation
Resected pancreatic tissue, transferred into lysis solution, and stool samples were stored at -80 °C. DNA from tissue samples was prepared using the InViSorb Genomic Kit II (InViTek) according to the manufacturer's protocol. Stool DNA was prepared by strictly following the protocol of Sidransky et al. (14) with the only modification that a Cleanmix kit (Fröbel) was used instead of glass powder in the last purification step.

mutant-enriched pcr
Sample DNA was amplified by mutant-enriched PCR. The principle of this method is to introduce a BstNI restriction site into the amplicons of wild-type (wt)-coding K-ras by means of primers containing a mismatch to the wt K-ras sequence (7). An ensuing incubation with BstNI cleaves wt amplicons but leaves mutant PCR products intact. This way, the ratio of mutant to wt amplicons can be increased by two orders of magnitude. A second amplification step is added to produce the amounts of PCR products required for further analysis. Mutant enrichment was carried out as described (15) except that unlabeled forward primers (5'AAC-TTG-TGG-TAG-TTG-GAG-CT 3') and 5' biotin-labeled reverse primers (5' GTT-GGA-TCA-TAT-TCG-TCC-AC 3'; Biotez) were used in the second amplification step. Each PCR run included blanks and a positive control containing 0.1% of K-ras-mutated SW 480 cells among wt-coding peripheral blood cells.

genotyping
K-ras genotyping was performed by a reverse dot blot test essentially as described (15). Amplicons were diluted 1/100 in binding buffer (2 mol/L NaCl, 10 mmol/L Tris, 1 mmol/L EDTA, pH 7.5), and 50 µL was added to each well of a strip of eight streptavidin-coated microwells (Boehringer Mannheim). The samples were immobilized by incubation for 15 min at room temperature, the solution was then aspirated, and 50 µL of 0.2 mol/L NaOH was added for strand separation. After 15 min at room temperature, the wells were washed three times with phosphate-buffered saline containing 1 mL/L Tween 20. Single-stranded amplicons were hybridized with 3 pmol of 5' digoxigenin-labeled probes representing wt K-ras coding for glycine (GGT) at codon 12 or mutated K-ras coding for alanine (GCT), arginine (CGT), aspartate (GAT), cysteine (TGT), serine (AGT), or valine (GTT) and measured luminometrically as described (15).

tumor markers
The tumor markers CA 19-9 and CEA were determined by enzyme immunoassays using the Axsym system (Abbott). Cutoff values were 37 kilounits/L for CA 19-9 and 5 µg/L for CEA.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
The total frequency of K-ras mutations in tumor samples was 86% (38 of 44 cases; Table 1 ). Of the 44 tumors analyzed, 35 were ductal adenocarcinomas, 24 were from cancers of the head or head/body, 8 were from the body or body/tail, and 3 were from the tail. Thirty-two of the thirty-five (91%) ductal adenocarcinomas harbored mutated K-ras. Among the non-ductal cancers, five of seven periampullary carcinomas and one cystadenocarcinoma exhibited mutated K-ras. The following mutations were identified: 18 (47%) were GGTGAT, 14 (37%) were GGTGTT, 4 (11%) were GGTCGT, 1 was GGTGCT/GTT, and 1 was GGTGAT/GTT. Three of five tissues samples from patients with chronic pancreatitis carried K-ras mutations (2 were GGTGAT, and 1 was GGTGTT).

Stools from 40 patients were analyzed. Mutated K-ras was seen in 10 of 25 (40%) of the patients with ductal adenocarcinomas, in 1 patient with cystadenocarcinoma, and in 2 of 6 patients with chronic pancreatitis. Nine of the detected mutations were GGTGAT transitions, four were GGTGTT transversions. No mutations were found in the stools from four patients with periampullary cancers, from three patients with endocrine tumors, and from one patient with a cystadenoma. Stools from six control persons without known pancreatic disease had wt K-ras (not shown).

Serum tumor markers CA 19-9 and CEA were increased in 31 of 40 (78%) and 14 of 36 (39%) of the patients with ductal adenocarcinomas and 5 of 7 and 2 of 8 of the patients with chronic pancreatitis, respectively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
K-ras MUTATIONS IN MALIGNANT AND PREMALIGNANT PANCREATIC LESIONS
Genotyping of tissue samples revealed K-ras mutations in >90% of the ductal adenocarcinomas, a figure in the range reported repeatedly. Because ductal cancer is by far the most abundant pancreatic tumor, this finding illustrates an outstanding diagnostic sensitivity of K-ras, which is further underlined by the similar high frequency of K-ras mutations found in papillary cancers. The latter result is in accord with previous data from Stork et al. (16) but differs from other reports showing that <50% of these tumors harbor oncogenic K-ras (17)(18)(19). This discrepancy may be partly attributable to the analytical procedures applied. Tada et al. (17) and Scarpa et al. (18) looked for K-ras mutations in PCR products that had not been enriched for mutations. Such an approach requires at least several percent of the altered alleles to be detected; thus the detection limit of this method is at least one order of magnitude above the detection limit of allele-specific amplification or mutant-enriched PCR (7)(16). Furthermore, in ampullary cancers the frequency of oncogenic K-ras might be correlated to the tumor size, (19) and all but one of the periampullary tumors analyzed here were at stage T2 or T3. Generally, the low incidence of non-ductal pancreatic tumors compromises a statistically accurate estimation of any oncogene involvement. This caveat includes endocrine tumors of the pancreas where controversial findings regarding K-ras have also been reported (20)(21).

Confirming previous evidence (9)(22)(23)(24)(25)(26), we saw K-ras mutations in the tissue samples of three of five patients with chronic pancreatitis, which points to the limitations of the diagnostic value of K-ras for malignant pancreatic diseases despite its sensitivity. These data have been extended by recent studies demonstrating that oncogenic K-ras can occur even in the pancreatic ducts of individuals without definite pancreatic disorder (23)(25)(27). Importantly, however, in all these cases mutated K-ras was not found in nondiseased ductal epithelia but always associated with hyperplasia or atypia, apparently reflecting an involvement of the K-ras oncogene in the expression of precursor lesions of pancreatic cancers (28). The prevalence of such premalignant ductal abnormalities seems rather high; these abnormalities are estimated to afflict nearly 50% of the healthy elderly population (29), and according to the published data (22)(26)(27)(28), 20–60% of those lesions may harbor mutated K-ras. Obviously, these observations pose a diagnostic dilemma: K-ras mutations may help identify curable cancers only if detectable at early tumor stages, but the earlier the gene is involved in tumor progression, the lower the probability that it is associated with malignancy. The combination of K-ras and serum tumor markers probably will not resolve this dilemma, as illustrated by our finding that five of seven patients with chronic pancreatitis showed increased CA 19-9. A multiplex molecular tumor marker analysis seems the more promising way to improve the discrimination of malignant from premalignant lesions. The tumor suppressors p16, p53, and dpc4 are frequently inactivated in pancreatic carcinomas (29). Recently, these markers were analyzed in addition to K-ras, and it was seen that in 37 of the 39 pancreatic cancers more than one of the four genes was damaged (30). It would be interesting to check a similar pattern of genes in premalignant pancreatic ducts.

K-ras TESTS IN STOOL DNA
To the best of our knowledge, a paper by Caldas et al. (9) describes the only previous study in which K-ras status was analyzed in the stool of patients suffering from pancreatic diseases. Using a colony hybridization assay and probing with radiolabeled oligonucleotides, these authors found mutated K-ras in 5 of 11 adenocarcinomas, 2 of 3 periampullary cancers, and 1 of 3 patients with chronic pancreatitis. Overall, our results fit the data of Caldas et al. (9) well. On the basis of these results, the diagnostic sensitivity of K-ras in the stool for pancreatic cancers may be preliminarily estimated as 40%. This figure is equivalent to that of serum CEA but is considerably less than the sensitivity of serum CA 19-9 and less than one-half of that of K-ras in tumor tissue. It implies that a negative K-ras test in stool is not sufficient to exclude pancreatic cancer in situations of uncertain diagnosis, e.g., in problematic cases of chronic pancreatitis. The positive predictive value is compromised because K-ras mutations may originate from colorectal tumors (13) and, as discussed, from premalignant pancreatic lesions, as exemplified here by the two patients with chronic pancreatitis who carry oncogenic K-ras in the stool. One of these cases (patient 62) was striking because analysis of the patient's tissue sample missed mutated K-ras, whereas an endoscopic secretion revealed the GGTGAT transition seen in the stool. Comparable observations were made by Caldas et al. (9), who reported discrepancies of the K-ras status between microdissected ductal lesions and tumor tissues in all five cases where parallel determinations had been done. In one of their cases, K-ras mutations were present in stool and the dissected ductal lesion but not in the tumor sample. Constellations such as those resemble the detection of dysplastic cells harboring mutated K-ras in pancreatic secretions of tumor-free patients (25). The frequent presence of oncogenic K-ras in premalignant lesions suggests that even apparently healthy persons might have mutated K-ras in the stool. Thus far we have seen K-ras mutations only in the fecal samples from patients with carcinoma or chronic pancreatitis; however, Villa et al. (31), screening for patients at risk for colorectal carcinoma, identified 2 K-ras-positive cases in the stools from 15 persons with no apparent organic disease.

Given the lack of specificity and the limited diagnostic sensitivity, is it worthwhile to analyze K-ras in the stool from subjects who are suspect for pancreatic tumors? At the moment we would not definitely give a negative answer because of the following consideration: K-ras genotyping in the stool is a noninvasive test with no risk of complications. With regard to the extremely poor prognosis of pancreatic cancers, a test contributing to an even modestly increased number of patients who can be treated by curative surgery would mark a valuable improvement. Because of its early role in carcinogenesis and its applicability to mutation enrichment, K-ras may identify curable tumors or, in the case of premalignant lesions, be used as be a novel criterion defining a patient's risk for cancer development on the molecular level. However, because the clinically important differentiation between premalignant and malignant lesions cannot be made based solely on this gene, the application of K-ras tests in stools for tumor screening seems problematic at present. It is important to establish a marker combination of K-ras, p16, and p53 suited for the analysis of stool samples.

	Acknowledgments

 
We greatly appreciate the technical assistance of Marion Bebenroth, Kathrin Lange, and Kerstin Oehlschlegel. Special thanks to Wolfgang Henke for critical reading of the manuscript. The study was supported by grant 573/97 from the Bundesministerium für Wirtschaft.

	Footnotes

 
¹ Present address: Königin-Elisabeth-Krankenhaus, Berlin Lichtenberg, 13125 Berlin, Germany.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Eckhauser FE, Colletti LM, Knol JA. Pancreatic carcinoma. Curr Opin Gastroenterol 1995;1:414-422.
2. Nitecki SS, Sarr MG, Colby TV, van Heerden JA. Long-term survival after resection for ductal adenocarcinoma of the pancreas. Is it really improving?. Ann Surg 1995;221:59-66. [ISI][Medline] [Order article via Infotrieve]
3. Plebani M, Basso D, Panozzo MP, Fogar P, Del Favero G, Naccarato R. Tumor markers in the diagnosis, monitoring and therapy of pancreatic cancer: state of the art. Int J Biol Markers 1995;10:189-199. [ISI][Medline] [Order article via Infotrieve]
4. Kronteris TG. Oncogenes. N Engl J Med 1995;333:303-306. [Free Full Text]
5. Bos JL. p21ras. An oncoprotein functioning in growth factor-induced signal transduction. Eur J Cancer 1995;31A:1051-1054.
6. Hruban RH, van Mansfeld ADM, Offerhaus GJA, van Weering DJH, Allison DC, Goodman SN, et al. K-ras oncogene activation in adenocarcinoma of the human pancreas. Am J Pathol 1993;143:545-554. [Abstract]
7. Levi S, Urbano-Ispizua A, Gill R, Thomas DM, Gilbertson J, Foster C, Marshall CJ. Multiple K-ras codon 12 mutations in cholangiocarcinomas demonstrated with a sensitive polymerase chain reaction technique. Cancer Res 1991;51:3497-3502. [Abstract/Free Full Text]
8. Motojima K, Urano T, Nagata Y, Shiku H, Tsunoda T, Kanematsu T. Mutations in the Kirsten-ras oncogene are common but lack correlation with prognosis and tumor stage in human pancreatic carcinoma. Am J Gastroenterol 1991;86:1784-1788. [ISI][Medline] [Order article via Infotrieve]
9. Caldas C, Hahn SA, Hruban RH, Redston MS, Yeo CJ, Kern SE. Detection of K-ras mutations in the stool of patients with pancreatic adenocarcinoma and pancreatic ductal hyperplasia. Cancer Res 1994;54:3568-3573. [Abstract/Free Full Text]
10. Gordis L, Gold EB. Epidemiology of pancreatic cancer. World J Surg 1984;8:808-821. [ISI][Medline] [Order article via Infotrieve]
11. Fernandez E, La Vecchia C, Porta M, Negri E, d'Avanzo B, Boyle P. Pancreatitis and the risk of pancreatic cancer. Pancreas 1995;11:185-189. [ISI][Medline] [Order article via Infotrieve]
12. Flanders TY, Foulkes WD. Pancreatic adenocarcinoma: epidemiology and genetics. J Med Genet 1996;33:889-898. [Abstract]
13. Union Internationale Contre le Cancer. TNM atlas. Illustrated guide to the TNM/pTNM-classification of malignant tumours, 3rd ed. Berlin: Springer, 1989:119–39..
14. Sidransky D, Tokino T, Hamilton SR, Kinzler KW, Levin B, Frost P, Vogelstein B. Identification of ras oncogene mutations in the stool of patients with curable colorectal tumors. Science 1992;256:102-105. [Abstract/Free Full Text]
15. Berndt C, Wolf G, Schröder G, Bebenroth M, Oehlschlegel K, Hillebrand T, Zanow J. A microplate assay for K-ras genotyping. Eur J Clin Chem Clin Biochem 1996;34:837-840. [ISI][Medline] [Order article via Infotrieve]
16. Stork P, Loda M, Bosari S, Wiley B, Poppenhusen K, Wolfe H. Detection of K-ras mutations in pancreatic and hepatic neoplasm by non-isotopic mismatched polymerase chain reaction. Oncogene 1991;6:857-862. [ISI][Medline] [Order article via Infotrieve]
17. Tada M, Yokosuka O, Omata M, Ohto M, Isono K. Analysis of ras gene mutations in biliary and pancreatic tumors by polymerase chain reaction and direct sequencing. Cancer 1990;66:930-935. [ISI][Medline] [Order article via Infotrieve]
18. Scarpa A, Zamboni G, Achile A, Capelli P, Bogina G, Iacono C, et al. Ras-family gene mutations in neoplasia of the ampulla of Vater. Int J Cancer 1994;59:39-42. [ISI][Medline] [Order article via Infotrieve]
19. Chung CH, Wilentz RE, Polak MM, Ramsoekh TB, Noorduyn LA, Gouma DJ, et al. Clinical significance of K-ras oncogene activation in ampullary neoplasms. J Clin Pathol 1996;49:460-464. [Abstract/Free Full Text]
20. Pellagata NS, Sessa F, Renault B, Bonato M, Leone BE, Solcia E, Ranzani GN. K-ras and p53 gene mutations in pancreatic cancer: ductal and nonductal tumors progress through different genetic lesions. Cancer Res 1994;54:1556-1560. [Abstract/Free Full Text]
21. Pavelic K, Hrascan R, Kapitanovic S, Vranes Z, Cabrijan T, Spaventi S, et al. Molecular genetics of malignant insulinoma. Anticancer Res 1996;16:1707-1717. [ISI][Medline] [Order article via Infotrieve]
22. Yanasigawa A, Ohtake K, Ohashi K, Hori M, Kitagawa T, Sugano H, Kato Y. Frequent c-Ki-ras oncogene activation in mucous cell hyperplasias of pancreas suffering from chronic inflammation. Cancer Res 1993;53:953-956. [Abstract/Free Full Text]
23. Suzuki H, Yoshida S, Ichikawa Y, Yokota H, Mutoh H, Koyama A, et al. Ki-ras mutations in pancreatic secretions and aspirates from two patients without pancreatic cancer. J Natl Cancer Inst 1994;86:1547-1549. [Free Full Text]
24. Brentnall TA, Chen R, Lee JG, Mimey MB, Bronner MP, Haggitt RC, et al. Microsatellite instability and K-ras mutations associated with pancreatic adenocarcinoma and pancreatitis. Cancer Res 1995;55:4624-4627.
25. Uehara H, Nakaizumi A, Baba M, Iishi H, Tsatuta M, Kitamura T, et al. Diagnosis of pancreatic cancer by K-ras point mutation and cytology of pancreatic juice. Am J Gastroenterol 1996;91:1616-1621. [ISI][Medline] [Order article via Infotrieve]
26. Rivera JA, Rall CJ, Graeme-Cook F, Fernandez del Castillo C, Shu P, Lakey N, et al. Analysis of K-ras oncogene mutations in chronic pancreatitis with ductal hyperplasia. Surgery 1997;121:42-49. [ISI][Medline] [Order article via Infotrieve]
27. Tada M, Ohashi M, Shiratori Y, Okudaira T, Komatsu Y, Kawabe T, et al. Analysis of K-ras gene mutation in hyperplastic duct cells of the pancreas without pancreatic disease. Gastroenterology 1996;110:227-231. [ISI][Medline] [Order article via Infotrieve]
28. Moskaluk C, Hruban RH, Kern SE. p16 and K-ras gene mutations in the intraductal precursors of human pancreatic adenocarcinoma. Cancer Res 1997;57:22140-22143.
29. Kern SE. Advances from genetic clues in pancreatic cancer. Cur Opin Oncol 1998;10:74-80. [Medline] [Order article via Infotrieve]
30. Rozenblum E, Schutte E, Goggins M, Hahn SA, Panzer S, Zahurak M, et al. Tumor-suppressor pathways in pancreatic adenocarcinoma. Cancer Res 1997;57:1731-1734. [Abstract/Free Full Text]
31. Villa E, Dugani A, Rebecchi AM, Vignoli A, Grottola A, Buttafoco P, et al. Identification of subjects at risk for colorectal carcinoma through a test based on K-ras determination in the stool. Gastroenterology 1996;110:1346-1353. [ISI][Medline] [Order article via Infotrieve]