The point mutation of c-Ki- ras at codon 12 in carcinoma of the pancreatic head region and in intraductal mucin-hypersecreting neoplasma of the pancreas

Summary In order to clarify whether the detection of a point mutation in the c-Ki-ras gene at codon 12 in tumor tissues can assist in predicting the tumor’s biological grade of malignancy, two types of tumors were investigated; one was called “carcinoma in the pancreatic head region,” and the other was intraductal mucin-hypersecreting neoplasm of the pancreas (IMHN). Dot hybridization and a modified PCR technique developed by Haliassos et al. were employed. Among 16 cases of tumors in the pancreatic head region, the point mutation was detected with a high frequency only in pancreatic ductal cell carcinomas (five out of six cases, 83.3%), but was not detected in extrahepatic bile duct carcinomas (0/5) or in ampullary carcinomas (0/5). In pancreatic ductal cell carcinomas, no relation was found between the occurrence of the point mutation and the histological type of the tumor. Among 20 cases of IMHNs, the point mutation was found in 11 cases (55%). No relation was found between the occurrence of the mutation and the size of IMHNs. However, as the grade of cell atypia increased, the frequency of the mutation tended to become higher. These results suggest that detection of this point mutation might be useful for distinguishing pancreatic ductal cell carcinoma from those of other origins in the pancreatic head region, and for the determination of the histopathological grade of malignancy in IMHNs.     

Significance of K-ras mutation and CEA level in pancreatic juice in the diagnosis of pancreatic cancer

The early diagnosis of pancreatic carcinoma is essential for increasing patient survival rates. In this study, 52 patients with suspected pancreatic diseases were examined to investigate the value of K‐ras codon 12 point mutation, levels of carcinoembryonic antigen (CEA) and carbohydrate antigen (CA19‐9), and cytology of pancreatic juice in the diagnosis of pancreatic carcinoma. Pancreatic juice was taken without secretin stimulation. K‐ras mutation was detected by enriched polymerase chain reaction (PCR) restriction fragment length polymorphism (RFLP). K‐ras mutation in pancreatic juice was more frequent in carcinoma than in benign diseases (P = 0.0448). The positive predictive value of K‐ras mutation for the diagnosis of neoplastic disease was 83%. The CEA level in pancreatic juice in carcinoma was significantly greater than that in benign disease (P < 0.0001). When the cutoff level of CEA was set at 50 ng/ml, its accuracy for the diagnosis of carcinoma was 85%. A multivariate analysis showed that K‐ras mutation and CEA level in pancreatic juice, as well as serum CA19‐9 level and age of the patient were independent variables for the diagnosis of carcinoma, and the accuracy of diagnosis by this analysis was increased to 90%. In conclusion, both K‐ras mutation and CEA level in pancreatic juice may be valuable for the diagnosis of carcinoma. Better discrimination was possible with a multivariate analysis.     

Detection of c-Ki-ras oncogene mutation in gastric adenomas with formalin-fixed, paraffin-embedded biopsy materials

Point mutations at codon 12 of the c-Ki-ras gene were investigated in 103 biopsy tissues from 38 patients with gastric adenoma. DNA from minute specimens of formalin-fixed paraffin-embedded tissue was amplified by nested polymerase chain reaction (PCR) and the restriction fragment length polymorphism (RFLP) method was used to detect mutations of the c-Ki-ras gene. Mutations were detected in 4 of the 38 gastric adenomas (10.5%), but were not detected in intestinal metaplasias and gastric adenocarcinomas. Of the 4 mutation-positive gastric adenomas, the GGT sequence in codon 12 changed to GAT in 2 cases, to GTT in 1 case and to TGT in 1 case. We classified gastric adenomas into three degrees of atypia; mutations were detected only in biopsy tissues with moderate and severe atypia but were not detected in biopsy tissues with mild atypia. When different biopsy tissues from a mutation-positive gastric adenoma were examined, the mutation was not always detected in every tissue sample examined. This suggests that mutations are not homogeneously distributed, even in the same adenoma. It is possible to follow up the mutation of biopsy materials of gastric adenoma by this method.     

Detection of point mutations in K- ras gene at codon 12 in bile from percutaneous transhepatic choledochal drainage tubes for diagnosis of biliary strictures

Summary Detection of K-ras mutations at codon 12 constitutes one modality for diagnosis of pancreatic tumors. We attempted to detect K-ras mutations in DNA from bile collected through percutaneous transhepatic choledochal drainage (PTCD) tubes as a diagnostic approach to biliary strictures. Since bile salts induce cell damage, we first investigated the degeneration of cells according to bile exposure time using cell lines. High-mol-wt DNA could be extracted from cells exposed to bile for 6 h, but not from those exposed for 12 h. However, DNA exposed to bile for up 12 h could be amplified by the polymerase chain reaction (PCR) method. Therefore, K-ras mutations in fresh bile specimens collected from 15 patients through PTCD tubes were examined using PCR with restriction enzyme digestion. K-ras mutations were found in five out of five (100%) pancreatic cancers, all of which were negative according to cytodiagnosis of the same bile. On the other hand, K-ras mutations were not detected in bile from biliary tract cancers or metastatic neoplasms, except for one bile duct carcinoma and one metastatic case. Thus, although K-ras mutation alone is not an absolute marker for cancer, detection of K-ras mutations in fresh bile from PTCD tubes is a useful adjunct for diagnosis of pancreatic carcinomas in cases of biliary tract strictures.     

Detection of K-<Emphasis Type="Italic">ras</Emphasis> mutations in the plasma DNA of pancreatic cancer patients

Background In pancreatic cancers, K-ras mutations have been found frequently (80%–100%), and they could be a good marker to detect tumor DNA in the plasma. Several studies have indicated that polymerase chain reaction/restriction fragment length polymorphism (PCR/RFLP) analysis of K-ras mutation was a useful method for the detection of hepatic and lymph node metastasis of pancreatic cancer. However, this method sometimes exhibited false-positive results, and the rate of K-ras mutation might thus be overestimated in these tissues. To diagnose pancreatic cancer correctly at an early stage, we attempted to detect tumor DNA in the plasma of pancreatic cancer patients using a more sensitive and specific method.Methods We examined 28 pancreatic cancer patients using a sensitive mutation-specific mismatch ligation assay for K-ras gene mutations in primary tumors and paired plasma samples.Results K-ras gene mutations were detected in 26 of the 28 (93%) pancreatic cancers. We also found the same mutations in 9 of these 26 (35%) patients in their plasma DNA. This mutation was found even in the plasma of patients with TNM stage II cancer.Conclusions Genetic alterations present in the tumors of pancreatic cancer patients can be detected in their plasma, and this approach is potentially applicable for cancer screening and the monitoring of this deadly disease.     

CYP1B1 Polymorphisms and K-ras Mutations in Patients with Pancreatic Ductal Adenocarcinoma

The frequency of CYP1B1 polymorphisms in pancreatic cancer has never been reported. There is also no evidence on the relationship between CYP1B1 variants and mutations in ras genes (K-, H- or N-ras) in any human neoplasm. We analyzed the following CYP1B1 polymorphisms in 129 incident cases of pancreatic ductal adenocarcinoma (PDA): the m1 allele (Val to Leu at codon 432) and the m2 allele (Asn to Ser at codon 453). The calculated frequencies for the m1 Val and m2 Asn alleles were 0.45 and 0.68, respectively. CYP1B1 genotypes were out of Hardy–Weinberg equilibrium; this was largely due to K-ras mutated PDA cases. The Val/Val genotype was over five times more frequent in PDA cases with a K-ras mutation than in wild-type cases (OR = 5.25; P = 0.121). In PDA, polymorphisms in CYP1B1 might be related with K-ras activation pathways. PANKRAS II study group—Members of the multicenter prospective study on the role of K-ras and other genetic alterations in the diagnosis, prognosis and etiology of pancreatic and biliary diseases (PANKRAS II) study group are mentioned in previous publications.     

Frequent multiple c‐ki‐ras oncogene activation in pancreatic juice from patients with benign pancreatic cysts

Background: We detected benign pancreatic cysts in more than half of 12 cases of in situ pancreatic cancer. A few investigators, having detected K‐ras mutation in pancreatic juice before the clinical diagnosis of pancreatic cancer, have suggested that this mutation might be an early event in pancreatic oncogenesis. Therefore, in the present study, we evaluated whether benign pancreatic cysts are a precancerous condition as reflected by K‐ras mutation in pancreatic juice. Methods: Pancreatic juice was collected through endoscopic cannulation of the pancreatic duct. Analysis of the mutations was performed using the polymerase chain reaction–preferential homoduplex formation assay. Results: The frequencies and types of K‐ras point mutations and the rate of multiplicity of K‐ras mutations in patients with benign pancreatic cyst were the same as those with pancreatic cancer. Conclusions: Multiple K‐ras codon 12 mutations in the pancreatic juice of patients with benign pancreatic cysts are present as frequently as those with pancreatic cancer. These results indicate that patients with benign cysts may be at a high‐risk for the development of pancreatic cancer.     

Mixed pleomorphic-osteoclast-like tumor of the pancreas

Summary The morphological, immunohistochemical, and molecular biological features of a case of giant cell tumor of the pancreas are described. This neoplasm showed mononuclear and multinucleated tumor giant cells as well as numerous osteoclast-like cells with multiple foci of osteoid-osseous metaplasia. The pleomorphic and osteoclastic giant cells displayed extensive homologies in their immunohistochemical profiles. Neither the pleomorphic nor osteoclast-like portion of the tumor showed neither c-Ki-ras nor p53 mutation and did not express the mutated p53 protein. The results suggest that the pleomorphic and osteoclast-like components are histogenetically related and that this rare neoplasm originates from a precursor cell capable of differentiating along divergent cell type.     

Analysis ofras mutations in human melanocytic lesions: activation of theras gene seems to be associated with the nodular type of human malignant melanoma

We have analyzed the Ha-ras, Ki-ras and N-ras gene for point mutations at codons 12, 13 and 61 via restriction fragment length polymorphism/polymerase chain reaction analysis and subsequent direct sequencing in non-cultured fresh-frozen tissues of 16 superficial spreading melanomas (SSM), 13 nodular malignant melanomas (NMM), 2 lentigo malignant melanomas (LMM), 1 dysplastic nevus, 1 congenital nevus and 5 normal nevi from 38 patients. Mutations were found in 4 melanoma samples, all belonging to the nodular malignant type. Three of them were mutated in N-ras and one in the Ha-ras gene. Mutation in N-ras was also detected in the congenital nevus. All mutations were exclusively located at the first two base pairs of codon 61. No Ki-ras mutation was detected in any lesion. No mutation could be found in SSM and LMM in addition to dysplastic and normal nevi. The frequency ofras mutation in NMM was 31%, whereas in SSM it was 0%. Our study suggests (a) an association betweenras mutations (mainly N-ras) and the NMM as a subgroup of human melanoma; (b) that activation of Ki-ras is not involved in the pathogenesis of melanoma. The role of UV radiation in point mutations ofras genes in human melanoma is discussed. This investigation was supported by grant EV5V-CT92-0096     

p53 and K-ras mutations in pancreatic juice samples from patients with chronic pancreatitis

Background: Mutations in p53 and ras genes are frequent in pancreatic carcinoma. Several ras mutations are consistently detected in the pancreatic juice from patients with chronic pancreatitis. The p53 gene mutations have been detected occasionally in chronic pancreatitis tissue. It was the aim of this study to evaluate the presence and clinical significance of p53 and ras mutations in clinical pancreatic juice samples from patients with chronic pancreatitis. Methods: Pancreatic juice was obtained from 66 patients with chronic pancreatitis and no evidence of pancreatic carcinoma (51 men, 15 women; age 17-86 years [mean 49.6 ± 12.9]). Patients were followed prospectively for 26 ± 3 (4-54) months. Detection of p53 gene mutations was by temperature gradient gel electrophoresis (TGGE) and single strand conformation polymorphism (SSCP) for exons 5-8. Analysis of ras mutations was performed by SSCP/polymerase chain reaction, restriction fragment length polymorphism/polymerase chain reaction. All mutations were confirmed by sequencing. Results: Five of 66 (7.5%) pancreatic juice samples contained p53 mutations, and ras mutations were detected in 6 cases (9%). Cytology was negative in all cases. No pancreatic carcinoma developed during follow-up and neither cancer cells nor preneoplastic lesions could be detected histologically in resected specimens. Although no correlation between p53 mutations and duration of pancreatitis or drinking habits was found, K-ras mutations correlated with both heavy smoking and severity of the disease. Conclusion: p53 and ras mutations can be detected in a minority of pancreatic juice samples from patients with chronic pancreatitis in the absence of malignancy. (Gastrointest Endosc 2001;53:734-43.)     

Adenosquamous carcinoma of the pancreas: preoperative diagnosis and molecular alterations

Adenosquamous carcinoma of the pancreas is a rare tumor which has a less favorable prognosis than common ductal cell carcinoma of the pancreas, and a definite preoperative diagnosis of this tumor is quite difficult. We herein report two cases of this rare variant. The patients were a 41-year-old man (patient 1) and a 67-year-old woman (patient 2). Patient 1 had a hypoechoic mass measuring 3cm in the uncus of the pancreas, while patient 2 had a huge mass, measuring 8cm, in the tail of the pancreas. Patient 2 was successfully diagnosed preoperatively as having an adenosquamous carcinoma, by cytological examination of the pure pancreatic juice obtained by endoscopic retrograde pancreatic juice aspiration. A pylorus-preserving pancreatoduodenectomy was performed for patient 1, and a distal pancreatectomy with resection of the spleen and the left kidney was performed for patient 2. Subsequent pathological findings of these two tumors revealed adenosquamous carcinoma of the pancreas. K-ras point mutation, p53 overexpression, and telomerase activity in both tumor specimens were detected by the mutant allele specific amplification method, immunohistochemical staining, and telomeric repeat amplification protocol assay, respectively. The two patients died of recurrent disease 5 and 4 months, respectively, after surgery. Cytological examination of pure pancreatic juice is a useful modality for the preoperative diagnosis of this tumor, and frequent molecular alterations may be associated with the poor prognosis of adenosquamous carcinoma of the pancreas.     

Clinicopathological significance of Ki- ras point mutation and p21 expression in benign and malignant exocrine tumors of the human pancreas

Summary Conclusion The present study suggests that Ki-ras point mutations may play an important role in the early stages of tumorigenesis and that a double mutation has a stronger detrimental effect than a single mutation on the survival after pancreatectomy. Background Previous studies have suggested the important role of Ki-ras point mutations inras gene codon 12 in the tumorigenesis of pancreatic cancer, but their clinicopathological significance is still unclear. The present study was designed to assess the clinicopathological significance of Ki-ras point mutations, and p21 expression in malignant and benign diseases of the pancreas. Methods Oligonucleotide dot-blot hybridization for Ki-ras point mutations in codon 12 and immunohistochemical staining for p21 expression were applied. Cases included 44 primary and 15 metastatic lesions of pancreatic cancer, and 17 benign pancreatic diseases. Results Ki-ras point mutations and p21 expression were detected in 43 and 19 primary lesions, 9 and 6 metastatic lesions, and four and five benign diseases, respectively. The patients with a single mutation had a better survival after pancreatectomy than those with a double mutation. The patients with a p21(+) GAT mutation showed the worst survival after pancreatectomy compared with other categories of patients.     

Genetic diagnosis of pancreatic cancer

Genetic analysis of pancreatic juice is a promising aid for the accurate and early diagnosis of pancreatic cancer. K‐ras mutation is frequently observed in pancreatic cancer; however, it is not specific for carcinoma because pancreatic adenoma and pancreatitis also show this mutation. Overexpression of p53 protein is solely detected in pancreatic juice from pancreatic cancer patients, but the positivity rate differs among various reports. Telomerase activity in pancreatic juice was detected in 20 of 24 (83.3%) pancreatic cancer patients and in only 1 of 23 (4.3%) pancreatic adenoma patients, while none of 23 (0%) pancreatitis patients showed evident telomerase activity. The relative value for telomerase activity was significantly higher in parcreatic cancer than in adenoma and pancreatitis. Centrosome abnormalities are very frequently seen in pancreas cancer tissues and the detection of these abnormalities is expected to be a potent new diagnostic tool for the genetic analysis of pancreatic juice. Genetic analysis of pancreatic juice will improve the sensitivity and specificity of pancreatic cancer diagnosis.     

K- ras mutation and pancreatic adenocarcinoma

Conclusion Pancreatic adenocarcinoma, the human tumor with the highest incidence of K-ras mutations, is one of the leading causes of cancer death and has a very poor prognosis. Over the past few years, a molecular genetic profile of pancreatic cancer has started to emerge: K-ras mutations in at least 90% of cases, MTS1 alterations (somatic mutations and homozygous deletions) in 80%, p53 mutations in about 70%, as well as multiple sites of allelic loss in cancer cells. The detection of K-ras mutations is done by using one of several PCR-based methods. A modified allele-specific ligation assay recently described appears particularly simple and reproducible. The possibility of pancreatic ductal lesions being the early precursors of pancreatic cancer has gained support from the finding of frequent K-ras mutations in these lesions. Although K-ras mutations are a defining event in pancreatic ductal carcinogenesis, it is unclear in which genetic context they occur. K-ras mutations as markers of cancer cells have been detected in clinical specimens from patients, including stool and blood, raising the possibility of early diagnosis. In addition K-ras can be a molecular target for therapeutic intervention, as illustrated with the use of farnesylation inhibitors.     

Evaluation of the diagnostic value of serum tumor markers,and fecal k‐ras and p53 gene mutations for pancreatic cancer

OBJECTIVE: To evaluate the diagnostic value for pancreatic cancer of four serum tumor markers, carbohydrate antigen (CA) 199, CA242, CA50 and carcino‐embryonic antigen (CEA), and fecal k‐ras and p53 gene mutations. METHODS: From February 2002 to March 2004, 136 patients were consecutively diagnosed with pancreatic cancer in the three participating medical centers. The diagnosis was confirmed by pathology in 53 patients, of whom five were excluded because they did not have measurement of serum tumor marker. The remaining 48 patients comprised the case group in the study. Ninety‐six patients with benign digestive diseases diagnosed during the same period were recruited as control subjects. They were matched by sex and age. In both groups, serum CA199, CA242, CA50 and CEA were measured by ELISA, and fecal k‐ras and p53 gene mutations were measured by PCR‐restriction fragment length polymorphism and PCR‐single strand conformational polymorphism, respectively. The receiver operating characteristic (ROC) curve and area under the curve (AUC) were used to compare their diagnostic value, as well as the sensitivity, specificity and likelihood ratio. Moreover, independent and sensitive tests from these non‐invasive approaches were selected to form a parallel test that may have further improved sensitivity for diagnosis of pancreatic cancer. RESULTS: The AUC of serum CA199 and CA242 were 0.821 (95%CI 0.725–0.917) and 0.821 (95%CI 0.723–0.919), respectively. The optimal diagnostic value of serum CA199 for pancreatic cancer was 93 U/mL, with a sensitivity of 73.7% and specificity of 91.4%. The positive likelihood ratio of CA199 was 8.57, and the negative likelihood ratio was 0.29. The optimal diagnostic value of serum CA242 was 25 U/mL, with a sensitivity of 71.1% and specificity of 93.5%. The positive likelihood ratio of CA242 was 10.94, and the negative likelihood ratio was 0.31. The sensitivity of fecal k‐ras gene mutation for diagnosis of pancreatic cancer was 77.4%, and the specificity was 81.2%. The positive and negative likelihood ratios of fecal k‐ras gene mutation were 4.12 and 0.28, respectively. The sensitivity and specificity of fecal p53 gene mutation were 25.8% and 95.3%, respectively, and its positive and negative likelihood ratios were 5.49 and 0.78. The rate of fecal k‐ras mutation was higher in patients with benign pancreatic diseases (57.14%) than that of controls with non‐pancreatic disorders. The values of serum tumor markers and fecal k‐ras and p53 gene mutation rates were not significantly different in subgroups according to site or stage of pancreatic cancer. The sensitivity and specificity of the parallel test of serum CA199 and fecal k‐ras gene mutation were 94.06% and 74.22%, respectively, while the sensitivity and specificity of the parallel test of serum CA242 and fecal k‐ras were 93.47% and 75.92%, respectively. CONCLUSIONS: Serum CA199 and CA242 are valuable diagnostic tools for pancreatic cancer. The diagnostic value is further improved when they are combined with fecal k‐ras gene mutation measurement.     

Ki -ras point mutation in different types of colorectal carcinomas in early stages

PURPOSE: The aim of this study was to elucidate pathways of carcinogenesis in the colon and rectum by investigating Ki-ras point mutation in different types of colorectal carcinomas in the early stage. METHODS: We analyzed rates of Ki-ras codon 12 mutations in 34 small, polypoid-type carcinomas (Tis or T1), 21 superficial-type carcinomas (Tis or T1), and 42 advanced carcinomas (T2, T3, and T4). RESULTS: Frequency of Ki-ras mutations in superficial-type carcinomas was 14.3 percent (3/21), which was significantly lower than 50 percent (17/34) in small polypoid carcinomas and 40.5 percent (17/42) in advanced carcinomas. These data suggest that another pathway of colorectal carcinogenesis that does not involve Ki-ras point mutation might exist. Among the 17 small polypoid carcinomas with Ki-ras point mutation in which both adenomatous and carcinomatous tissue were examined, 12 showed a mutation of the same type in both carcinomatous and adenomatous tissues. In two cases, mutation was present only in carcinomatous tissue and not in adenomatous tissue; in the other three cases, Ki-ras point mutation was present only in adenomatous tissue but not in carcinomatous tissue. CONCLUSIONS: These data suggest that carcinoma in a small polypoid lesion does not always develop from pre-existing adenoma with Ki-ras point mutation; in a small number of the polypoid-type early carcinomas, polyclonal composition concerning the Ki-ras gene may exist.Poster presentation at the meeting of The American Society of Colon and Rectal Surgeons, Seattle, Washington, June 9 to 14, 1996.     

H-ras oncogene point mutations in arthritic synovium

Objective. To examine mutational activation of ras proto-oncogenes in synovial tissue from patients with rheumatoid arthritis (RA) compared with synovial specimens from patients with osteoarthritis (OA) or other arthropathies. Synovial samples from cadavers, without any signs of joint disease, were used as control material. Methods. Using a combination of polymerase chain reaction (PCR) and automated sequencing of the amplified PCR product, regions around codons 12, 13, and 61 of the H-, K-, and N-ras proto-oncogenes were analyzed. Confirmation of mutations was based on restriction fragment length polymorphism analysis and/or oligonucleotide hybridization. Results. Four (6%) of 72 patients with RA, 2 (13%) of 16 with OA, and 1 (8%) of 12 with other arthropathies harbored mutant H-ras proto-oncogenes, and were heterozygous at codon 13 for the GGT→GAT (Gly→Asp) change. An unexpected mutation was found in the H-ras gene, in which a heterozygous GTG→ATG (Val→Met) mutation was observed over codon 14. The incidence for this mutation was 39% (28 of 72) in RA patients, 94% (15 of 16) in OA patients, and 42% (5 of 12) in patients with other arthropathies. All samples carrying the codon 13 mutation of H-ras were also codon 14-mutated, i.e., double mutations existed. Identical point mutations were also detected in a few synovial specimens obtained from cadavers (n = 8), including a single case of double mutation. All specimens showed normal K- and N-ras loci. Conclusion. Activation of proto-oncogene H-ras by point mutation in codons 13 and 14 occurred in the synovial tissue of patients with RA, OA, or other arthropathies, as well as, to some extent, in the control synovia, indicating that the phenomenon is not specific for RA. In codon 14, incidence of the H-ras point mutation was highest in OA tissue. The possible significance of this codon 14-mutated H-ras gene needs to be clarified.     

Cyclooxygenase-2 Overexpression Is Related to Polypoid Growth and K-<Emphasis Type="Italic">ras</Emphasis> Gene Mutation in T1 Colorectal Carcinomas

PURPOSE Cyclooxygenase-2 is thought to play a role in the development of intestinal tumors and levels are elevated in approximately 80 to 90 percent of human colorectal carcinomas. To clarify the role that cyclooxygenase-2 plays in the development of colorectal carcinoma, we studied the relationship between cyclooxygenase-2 expression and tumor morphology and that between cyclooxygenase-2 expression and K-ras mutation.METHODS We classified 48 T1 colorectal carcinomas as polypoid or nonpolypoid and analyzed the clinicopathologic features. The expression of cyclooxygenase-2 was determined immunohistochemically, and nested polymerase chain reaction-restriction fragment length polymorphism detected a K-ras codon 12 mutation.RESULTS Cyclooxygenase-2 expression was higher in polypoid carcinomas than in nonpolypoid carcinomas (P < 0.001). The K-ras codon 12 mutation was associated with higher levels of cyclooxygenase-2 expression compared with carcinomas without this mutation (P = 0.028).CONCLUSIONS Polypoid growth and K-ras gene mutation are both associated with increased levels of cyclooxygenase-2 expression in T1 tumors.Supported in part by grants from the Ministry of Education, Science, Sports, and Culture of Japan.     

Enrichment polymerase chain reaction for the detection of Ki-ras mutations: relevance of Taq polymerase error rate, initial DNA copy number, and reaction conditions on the emergence of false-positive mutant bands

Screening for oncogene mutations as a marker for malignancy can be a powerful tool for the early diagnosis of cancer. The enrichment polymerase chain reaction (PCR) is a sensitive method for the detection of low-frequency mutations in small samples. However, false-positive results, caused by methodological errors, may have severe clinical implications. When applied to the detection of Ki-ras mutations in pancreatic secretions, the assay sensitivity is limited to approximately 1:1400. Our investigation of Ki-ras mutations in blood samples from patients with pancreatic carcinoma revealed PCR bands presumably derived from mutant Ki-ras in samples from healthy volunteers, while all blood samples of the patients with pancreatic carcinomas showed a wild-type band pattern. Mathematical modeling of the PCR reaction reveals that the rate of false positive PCR results depends on the initial amount of DNA, the Taq polymerase error rate, the number of PCR reaction cycles, reaction efficiency and the restriction endonuclease chosen. The overall error rate of false positive results of the enrichment PCR can be reduced to the square of the rate of a single-step analysis if repeated amplifications of the same DNA specimen show an identical result. Received: 11 December 1998 / Accepted: 7 January 1999