Prognostic value of tumor progression-related gene expression in colorectal cancer patients.

PURPOSE: Colorectal cancer (CRC) is driven by genetic alterations causing its progression. Besides accepted tumor suppressor- and onco- genes, a series of genes have been identified, which contribute to transformation into a more malignant stage. We investigated whether the expression level of such genes, alone or in combination, could add to predict the prognosis of CRC patients. METHODS: Tumor samples from 118 CRC patients were screened in a retrospective analysis by qRT-PCR for expression of the four tumor progression-associated genes osteopontin (Opn), transforming growth factor ß (Tgf-ß), matrix metalloproteinase-2 (Mmp-2) and cyclooxigenase-2 (Cox-2). The resulting qRT-PCR values were related to those of housekeeping genes. All patients were clustered for similar expression levels between the four genes with R statistical software using the package pvclust, which provides bootstrap agglomerative hierarchical clustering. Clusters with similar expression of the four genes were analyzed for correlation with UICC stages and survival time. RESULTS: Expression of the four genes varied considerably within the cohort of patients. Cluster analysis of patients revealed a subgroup (n = 33) who in comparison with the other patients showed tenfold higher expression levels of all four genes (p < 0.001, respectively). However, there was no correlation between patients expressing high or low levels of these four genes and known parameters of clinical prognosis (UICC stages, survival time). CONCLUSIONS: In conclusion, tenfold increased expression levels of Opn, Tgf-ß, Mmp-2 and Cox-2 in a subset of CRC patients did not predict for a clinical outcome that is different from that of the remaining patients.

The insulin-like growth factor binding proteins 3 and 7 are associated with colorectal cancer and liver metastasis.

Over the last few decades, a great deal of attention has been directed to the IGF system for its vital role in regulating cell and tissue survival, growth and differentiation. The insulin-like growth factor binding proteins (IGFBPs), a main constituent of this system, have been implicated in the tumorigenesis of colorectal cancer (CRC). In this study, we intended to shed more light on two essential members; IGFBP3 as representative for the six main IGFBPs and IGFBP7 to represent their related proteins (IGFBP-rps). Our experiments on silencing IGFBP3 or IGFBP7 in the two human CRC cell lines SW480, Caco2, and in the rat CRC cell line CC531 show reduced proliferation, colony formation, and for IGFBP3, also reduced migration. The expression of both genes in 68 human CRC samples was higher in UICC stages II and III than in stages I and IV. Additionally, IGFBP3 was negatively correlated with age (p = 0.05) and positively related to IGFBP7 expression (p = 0.0001). Further, in a liver metastasis experiment, the expression of both genes was drastically increased in response to early metastatic growth in vivo. Since these high levels returned gradually to normal thereafter, it could be assumed that the up-regulation of IGFBPs is vital during the process of homing into the liver and early metastatic dissemination. Our results indicate that IGFBP3 and 7 cannot be simply considered as tumor suppressors but have additional properties, which become evident only during cancer progression and metastasis formation.

Disseminated colorectal tumor cells in organs prone to metastasis detected by new double enriched nested-PCR in comparison with recognized assays.

Hypothetically, K-ras mutations can be used as a marker of disseminated tumor cells (DTCs) in patients with K-ras mutated primary carcinoma. This study focused on the development of a useful assay for detecting low numbers of DTCs in potential target tissues of metastatic K-ras codon 12 mutated colorectal cancer. Tumor, liver, lymph node and bone marrow tissues from 46 colorectal carcinoma patients were examined for K-ras codon 12 mutations with a new double enriched nested (DEN)-PCR and the incidence of mutations was compared to those obtained from three established assays. DEN-PCR followed by sequencing found one mutated cell within 107 K-ras codon 12 wild-type cells and was more sensitive than other methods (1:106-1:102). Colon carcinomas (26/46) and adenomas (1/3) harbored mutations in K-ras codon 12. Sixteen of these 27 mutated tumors were found with all assays, two with three methods and one with the two most sensitive assays. In 8 cases, only DEN-PCR identified the K-ras mutation, and thus prevailed over the other methods used (p<0.002). Eight of 26 patients with K-ras mutated colorectal carcinoma also harbored K-ras mutated DTCs in liver and lymph nodes, respectively, and 4 in bone marrow. For liver and lymph node samples, DTC-mutations were identical to those in the primary carcinoma but those in bone marrow differed from the respective mutation in the primary carcinoma. In conclusion, DEN-PCR is a highly sensitive method for detecting K-ras mutations as marker of early and late tumor cell dissemination in tissues potentially harboring colorectal carcinoma metastases.

The balanced induction of K-ras codon 12 and 13 mutations in mucosa differs from their ratio in neoplastic tissues.

The aim of this study was to compare the ratio of K-ras codon 12 and 13 mutations in various tissues of colorectal cancer patients. Multiple samples of inconspicuous mucosa and a sample of carcinoma tissue were taken from 36 colorectal cancer patients (group I) and these results were compared with those from polyp and carcinoma tissues of another 48 colorectal cancer patients (group II). A polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) assay was used to detect the respective point mutations. The results of this assay were complemented by sequencing the K-ras mutations. In mucosa tissue, the ratio of codon 12 and 13 mutations was nearly equal (0.9:1) whereas the respective ratio in tumour tissue showed a strong preponderance of K-ras codon 12 mutations (14:1, p=0.004). In polyp tissue of patients from group II, the ratio was 2.7:1 and that in carcinomas was 19:1 (p=0.053). The prevalence of both types of mutation was 14.6% in all mucosa samples, corresponding to 30.6% of group I patients. The K-ras mutation rate in carcinoma tissue of the same patients was 38.9%. Similarly, 33.4% of all polyp and 41.7% of all carcinoma samples from group II harboured K-ras codon 12 and/or 13 mutations. Sequencing confirmed 59 of 60 K-ras codon 12 mutations, but due to the detection limit for sequencing (1:10(4)) only 10 of 20 K-ras codon 13 mutations were confirmed. It is concluded that after balanced induction K-ras codon 12 mutations increase in frequency relative to K-ras codon 13 mutations during tumour progression.

Molecular detection of clinical colorectal cancer metastasis: how should multiple markers be put to use?

BACKGROUND AND AIMS: Up to 45% of colorectal cancer (CRC) patients will develop local recurrence or metastasis following curative resection. The latter is due to cells shed from the primary carcinoma prior to or during surgery. The aim of this study was to contribute toward a "rational"-approach for detecting these disseminated tumor cells (DTC) using a combination of independent markers and detection methods. PATIENTS/METHODS: Liver, lymph node, and bone marrow samples from 246 CRC patients were screened for DTC using three markers: mutated K-ras was detected by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP), and cytokeratin 20 (CK20) and guanylylcyclase C (GCC), indicating circulating epithelial cells, were tracked by nested reverse-transcription (RT) PCR. RESULTS: The rate of positive findings of the individual markers (CK20: 88%; GCC: 88%; K-ras: 67%) and their combinations (88-50%) was significantly higher in biopsies from liver metastases than in liver samples from patients without evident distant metastasis (M0; p<0.03). The detection rate of individual markers (except GCC) was also significantly elevated in inconspicuous liver tissue adjacent to metastasis compared with specimens from M0 patients. When using the concomitant detection of all three markers as criterion for DTC in the liver of M0 patients, however, no patient was DTC-positive. Therefore, the concomitant presence of the two CEC markers (CK20 plus GCC) and/or the presence of mutated K-ras were preferred for a combined evaluation, which resulted in a 24% detection rate for biopsies from both liver lobes. This translates into 39% of M0 patients with at least one positive liver biopsy. CONCLUSION: Our results suggest that the concomitant detection of CK20 plus GCC and/or the presence of mutated K-ras are a rational approach for tracking CEC/DTC in CRC patients.

K-ras codon 12 and 13 mutations are correlated with differential patterns of tumor cell dissemination in colorectal cancer patients.

The aim of this prospective study was to relate the incidences of cytokeratin 20 (CK20) and guanylylcyclase C (GCC) in lymph node, liver, and bone marrow specimens of 245 colorectal cancer (CRC) patients with the K-ras oncogene status of the corresponding primary tumor. Qualitative RT-PCR detection of CK20 and GCC mRNA was used as marker of circulating epithelial cells (CEC). Samples were considered positive for CEC only when both markers were detected concomitantly. For the detection of K-ras mutations, a PCR-RFLP assay was used. In the group with K-ras mutated primary carcinomas (n=92), CEC were detected in 62% of lymph node-, 43% of liver-, and 2% of bone marrow samples. No statistical significance was found when comparing these results with those from patients with K-ras wild-type carcinoma (59%, 46%, and 0%, respectively). In contrast to this combined evaluation, separate analysis of K-ras codons 12 (n=75, 82%) and 13 (n=17, 18%) revealed significantly differing CEC incidences. Lymph node specimens from corresponding K-ras codon 13 mutated carcinomas showed a significantly higher CEC incidence (82%) than the groups with codon 12 mutation (57%, p<0.05) or K-ras wild-type sequence (59%, p<0.05). Unlike these findings in lymph nodes, liver biopsies from corresponding carcinomas with K-ras codon 12 mutation or wild-type sequence were significantly more often positive for CEC (31% and 29%) than specimens from K-ras codon 13 mutated primary CRC (12%, p<0.04, respectively). In conclusion, colorectal carcinomas with K-ras codon 12 mutation showed the same pattern of tumor cell dissemination as their K-ras wild-type counterparts. Since K-ras codon 12 mutations prevailed 4-fold over codon 13 mutations, combined analysis of the two codons showed the same result. However, sub-analysis of patients with K-ras codon 13 mutation revealed that the respective CEC incidence was significantly increased in lymph nodes, but decreased in liver biopsies.

Detection of isolated tumor cells by polymerase chain reaction-restriction fragment length polymorphism for K-ras mutations in tissue samples of 199 colorectal cancer patients.

The aim of this study was to identify K-ras mutations as marker for isolated tumor cells in liver, lymph node, and bone marrow specimens of colorectal cancer patients. To detect these, a PCR-RFLP assay was used with a sensitivity exceeding that of routine histopathology by at least 1 order of magnitude. In addition, the ratio of mutated versus wild-type alleles was determined by an internal standard. Of 199 patients, 74 (37.5%) were found to bear a K-ras-positive tumor. Of these, 60 (81%) were mutated in codon 12 and 14 (19%) in codon 13 (P < 0.001). In addition, 14 organs were found K-ras positive, 13 of which were from 12 patients with a K-ras-positive tumor (16%) and 1 from a patient with a K-ras-negative tumor (0.8%). Eight patients exhibited liver involvement and 6 showed lymph node involvement. Remarkably, no bone marrow specimen was found K-ras positive (P < 0.017 versus liver involvement). Sequence analysis of tumor DNA revealed that GGT (Gly) was replaced by GAT (Asp; 35%), GTT (Val; 32%), AGT (Ser; 13%), GCT (Ala; 10%), TGT (Cys; 8%), and CGT (Arg; 2%) for codon 12, and by GAC (Asp) as the only type of mutation for codon 13. In colorectal carcinomas the ratio of K-ras mutated versus wild-type alleles ranged over 4 orders of magnitude (10(0)-10(-4), median: 10(-2)) and was correlated with both, residual tumor load (R1/2; P = 0.028) and distant metastasis (M1; P = 0.057). These results show that detection of K-ras mutated alleles by PCR-RFLP in patients with colorectal carcinoma may aid in the identification of isolated tumor cells. High ratios of K-ras alleles were correlated with certain negative prognostic parameters (R,M). In accord with its function as a primary filter for colorectal carcinoma cells, the liver was more often contaminated with K-ras-positive cells than bone marrow.

Prognostic value of disseminated colorectal tumor cells in the liver: results of follow-up examinations.

BACKGROUND AND AIMS: Dissemination of tumor cells is an initial step in metastatic disease. Detection of disseminated tumor cells (DTC) in blood, bone marrow and lymph nodes has been associated with reduced disease-free survival, but to date there are no data for hepatic DTC. We investigated the prognostic relevance of hepatic DTC that are present in patients with colorectal cancer (CRC) at the time of surgery. PATIENTS AND METHODS: In 121 patients with CRC clinically diagnosed for liver metastasis by ultrasound, CT, and exploration during surgery DNA from liver biopsy specimens obtained during surgery was examined by a PCR-RFLP assay for K-ras mutations as a marker for DTC. At the time of surgery 54 of the 121 were mutated in K-ras codons 12 or 13. After a median follow-up of 405 days all survivors were reevaluated by ultrasound/CT. RESULTS: Patients with a K-ras mutation in their primary tumor had a significantly lower probability of survival and higher risk of harboring a synchronous second colorectal carcinoma than patients with a K-ras wild-type tumor. Based on specimens taken intraoperatively DTC were found in the liver of 14 of 54 patients (26%). At follow-up only 10 of 40 patients (25%) with DTC-free liver had died of their disease but 9 of the 14 patients with hepatic DTC. Among the 14 patients with hepatic DTC 10 (71%) had developed new liver metastasis, compared to 12 of 40 (30%) in those without hepatic DTC. CONCLUSION: Hepatic DTC in colorectal cancer patients is associated with reduced overall survival and increased risk of hepatic metastasis development. Further studies are necessary to corroborate our results since the number of patients studied is still limited.

Cytokeratin 20 and guanylyl cyclase C mRNA is largely present in lymph node and liver specimens of colorectal cancer patients.

The aim of our prospective study was to detect circulating epithelial cells (CEC) indicating the presence of disseminated tumor cells (DTC) in tissues affected by lymphatic and hematogenic colorectal cancer metastasis. DTC were tracked in lymph node, liver or bone marrow samples of 245 colorectal cancer patients using 2 independent RT-PCR assays for cytokeratin 20 (CK20) and guanylylcyclase C (GCC) that demonstrated a sensitivity of 1 colorectal cancer cell in 10(6) nucleated hematopoietic cells. CK20 mRNA was detected in 79% of lymph nodes, 35% of both liver lobes and 11% of bone marrow samples. GCC mRNA was found in 68% of lymph nodes, 60% of both liver lobes and 6% of bone marrow specimens. Both markers were recorded in 63% of lymph nodes, 45% of at least 1 liver lobe and 1% of bone marrow samples. There was no significant difference when comparing lymph node samples tested positive for both markers in patients with (N1/2; 65%) and without (N0; 56%) nodal involvement. The same was true when comparing the percentages of patients with and without clinically overt distant metastasis who were positive for both markers in at least 1 liver lobe (62% vs. 41%) or in bone marrow (4% vs. 0%). A score denoting the cumulative sum of tests indicating presence of CK20 and GCC mRNA in the liver was significantly related with UICC classification (p = 0.039). However, addition of lymph node results to this score decreased the correlation. The high incidence of clinically inconspicuous lymph node and liver samples tested positive for both markers emphasizes the function of these organs as primary filters for epithelial cells possibly shed from colorectal carcinomas. The potential prognostic significance of these findings warrants verification, especially regarding the importance of CEC or DTC resident in the liver of colorectal cancer patients.

Superiority of combined chemo-embolization and portal infusion with 5-fluorouracil over locoregional infusion concepts in Novikoff hepatoma-bearing rats.

The aim of this study was to investigate experimentally whether there is a superior effect of the combination of hepatic artery chemo-embolization with portal vein infusion over either of the two treatment modalities alone. Novikoff hepatoma cells transplanted under the liver capsule of Sprague Dawley rats were used as a model. Tumor growth was assessed at 7 and 21 days after tumor inoculation. The prolamine solution Ethibloc was employed for embolization, and 5-fluorouracil was used as a chemotherapeutic agent for both infusion and chemo-embolization. All arterial treatment modalities were administered in a super-selective manner. There was no intolerable toxicity after dosages of 55 to 125 mg 5-fluorouracil/kg body weight. With regard to therapeutic efficacy the results show that embolization is an effective therapeutic means for inducing tumor necrosis in selected liver areas. As a consequence, the ranking of all treatment modalities was based on the combined evaluation of tumor size and extent of tumor necrosis. According to this evaluation, hepatic artery chemo-embolization was superior to the respective type of infusion (P<0.01). In addition, the combination of both modalities in the form of hepatic artery chemo-embolization and portal vein infusion was effective in destroying more than 97% of vital tumor tissue (P<0.01). These results suggest the need for a comparative clinical study.

Hepatic disseminated tumor cells in colorectal cancer UICC stage 4 patients: prognostic implications.

Approximately 30-50% of all colorectal cancer patients with a resectable primary tumor will subsequently develop metastatic disease due to tumor cell dissemination. In the case of limited solid hepatic metastasis, resection of the primary tumor and the respective hepatic metastasis can be curative. Our aim was to evaluate the incidence of hepatic DTC in patients with solid liver metastasis and to describe their prognostic impact. Therefore, we applied a sensitive PCR-RFLP assay detecting one K-ras mutant among one million wild-type cells. Tumor tissue and liver biopsies from 32 colorectal cancer patients staged UICC 4 undergoing palliative surgery could be obtained intra-operatively and were thereupon screened for the presence of hepatic DTC. The primary tumor of 16 patients (50%) harbored a K-ras mutation and survival of K-ras positive patients was reduced as compared to K-ras negatives (P=0.039). In 8 of the patients (50%) with a K-ras mutated primary, no hepatic DTC were detected, whereas the liver of the remaining 8 patients (50%), showed a coincidence of solid liver metastasis and hepatic DTC. Median survival of patients with solid liver metastasis and hepatic DTC was decreased as compared with patients without any hepatic DTC burden (median survival 165 vs. 240 days; log-rank P=0.951). A subgroup analysis revealed that survival was significantly decreased in the case of bilobal DTC affection as compared with monolobal hepatic DTC affection (median survival 68 vs. 355 days; log-rank P=0.002). We conclude that the detection of hepatic DTC is a powerful prognostic factor. Furthermore, bilobal hepatic DTC involvement might be a contraindication for the resection of solid liver metastasis.