Clinical Cancer Research Meeting Calendar Advances in Breast Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by de Jong, K. P.
Right arrow Articles by van den Berg, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by de Jong, K. P.
Right arrow Articles by van den Berg, A.
Clinical Cancer Research Vol. 11, 4067-4073, June 1, 2005
© 2005 American Association for Cancer Research

Imaging, Diagnosis, Prognosis

P53 Mutation Analysis of Colorectal Liver Metastases: Relation to Actual Survival, Angiogenic Status, and p53 Overexpression

Koert P. de Jong1, Annette S.H. Gouw2, Paul M.J.G. Peeters1, Marian Bulthuis2, Lorian Menkema2, Robert J. Porte1, Maarten J.H. Slooff1, Harry van Goor2 and Anke van den Berg2

Authors' Affiliations: 1 Division of Hepato-Pancreato-Biliary Surgery and Liver Transplantation, Department of Surgery and 2 Department of Pathology and Laboratory Medicine, University Medical Center Groningen, Groningen, the Netherlands

Requests for reprints: Koert P. de Jong, Division of Hepato-Pancreato-Biliary Surgery and Liver Transplantation, University Medical Center Groningen, P.O. Box 30001, 9700 RB Groningen, the Netherlands. Phone: 31-50-3612896; Fax: 31-50-3614873; E-mail: k.p.de.jong{at}chir.umcg.nl.


   Abstract
Top
 Abstract
Patients and Methods
 Results
 Discussion
 References
 
Purpose: To correlate TP53 mutations with angiogenic status of the tumor and prognosis after liver surgery in patients with colorectal liver metastases and to correlate immunohistochemical staining of p53 protein with TP53 gene mutations.

Experimental Design: Tumors of 44 patients with surgically treated colorectal liver metastases were analyzed for (a) TP53 mutations using denaturing gradient gel electrophoresis followed by sequencing, (b) microvessel density using the hot spot overlap technique, (c) apoptotic rate in tumor cells and endothelial cells of tumor microvessels using double immunostaining for anti–cleaved caspase 3 and anti-CD34, and (d) expression of p53 protein using immunohistochemistry.

Results: TP53 mutations were detected in 36% of the metastases and occurred more frequently in liver metastases from left-sided colon tumors than from right-sided colon tumors (P = 0.04). In metastases with TP53 mutations, microvessel density was higher compared with tumors with wild-type p53. Endothelial cell apoptosis was not different in tumor microvessels from TP53-mutated versus nonmutated tumors. The 5-year actual survival was not influenced by TP53 mutational status, microvessel density, or endothelial cell apoptotic rate of the tumors. Based on immunohistochemical p53 overexpression, the positive and negative predictive values of TP53 mutations were 61% and 82%.

Conclusions: In patients with surgically treated colorectal liver metastases, TP53 mutations and angiogenic status did not influence prognosis. Immunohistochemistry is not a reliable technique for detecting TP53 mutations.

Key Words: prognostic factors • partial hepatectomy • apoptosis

Mutations in the TP53 tumor suppressor gene play a central role in many types of cancer (1). Therefore, TP53 alterations are frequently reported in relation to prognosis and response to treatment. Colorectal cancer, being one of the more frequent types of cancer in the Western world, has been frequently studied with regard to p53 (over)expression. The presumed relation between aberrant p53 expression, prognosis, and angiogenesis in primary colorectal cancer has been studied extensively although with contradicting results. However, studies on TP53 mutational status of colorectal liver metastases in relation to prognosis are infrequently found.

In cell lines, it was found that wild-type p53 has antiangiogenic effects whereas mutated p53 has proangiogenic effects (2, 3). It is, therefore, interesting to evaluate TP53 mutational status and angiogenic properties in the clinical setting in patients with resected colorectal liver metastases.

Basically, alterations in p53 can be analyzed by mutation analysis or immunohistochemical staining. Mutations in the TP53 gene can either result in the production of a stable p53 protein, which can be detected immunohistochemically, or production of truncated p53 protein. The latter type of mutations (null mutations) will result in false-negative immunohistochemical staining for the p53 protein and to the erroneous conclusion that no mutation is present. Therefore, it is questionable whether immunohistochemistry correlates with the actual TP53 mutation status. This is relevant because immunohistochemistry is frequently used as the sole method of analyzing dysregulation of p53 (4).

In the present study, we analyzed colorectal liver metastases for the presence of TP53 mutations, microvessel density, rate of endothelial cell apoptosis, and overexpression of p53 protein. We hypothesized that (a) angiogenic properties, such as microvessel density and rate of apoptotic endothelial cells in the tumor, are associated with TP53 mutations, (b) TP53 mutations are not associated with prognosis after liver surgery for colorectal liver metastases, and (c) p53 overexpression detected with immunohistochemistry is a poor predictor of TP53 mutations.


   Patients and Methods
Top
 Abstract
 Patients and Methods
Results
 Discussion
 References
 
Patients. In patients referred for surgical treatment of colorectal liver metastases, preoperative analysis (computed tomography scan of thorax and abdomen and bone scan) were done to rule out extrahepatic tumors. Colonoscopy with random biopsies of the anastomosis was done to rule out other primary colon tumors and anastomotic recurrences. Only patients who underwent radical (microscopic free, R0) primary tumor resections were eligible for liver surgery. An exploratory laparotomy was done if extrahepatic tumors were absent. The abdomen was inspected and lymph nodes of the hepatoduodenal ligament and suspicious lesions were analyzed by frozen section. Intraoperative ultrasound of the liver was done and if no extrahepatic tumor growth was found, surgery was done obtaining tumor-free margins of at least 1 cm. Recorded clinical data include sex, site of the primary tumor (right sided: cecum, ascending, and transverse colon; left sided: descending colon, sigmoid, and rectum), and disease-free and overall survival. Intervals between date of resection of primary tumor and diagnosis of liver metastases, as well as between partial liver resection and diagnosis of recurrent disease, were recorded. Follow-up included clinical, biochemical (carcinoembryonic antigen test), thoracic X-ray, and liver ultrasound evaluation every 3 months for the first 2 years, and every 6 months thereafter. Recurrences were classified as liver, abdominal not liver, thoracic, skeletal, or combinations. Recurrent disease was treated according to standard clinical practice and included surgery, chemotherapy, and/or radiotherapy.

Histology. All procedures and use of (anonymized) tissue were done according to recent national guidelines. Frozen and paraffin-embedded specimens of 44 consecutive patients, operated before April 1999, were retrieved from our tumor bank. One representative tissue block of the metastasis was used from each patient. Tumors were classified according to the stage of the primary tumor (Unio Internationale Contra Cancrum tumor-node-metastasis system), and histologic grading (well, moderately, and poorly differentiated) was done according to the WHO classification (5). Carcinoembryonic antigen serum levels were determined using an automated assay (IMX, Abbot, Hoofddorp, the Netherlands).

Immunohistochemistry. Monoclonal antibodies for p53 (BP53 and DO7, Neomarkers, Union City, CA) were used. After deparaffinization, sections were preheated 3 x 15 minutes at 115°C and endogenous peroxidase was blocked with 0.03% H2O2. Sections were then incubated with either DO7 (1:400) or BP53 (1:800) for 2 hours at room temperature followed by incubation with peroxidase-conjugated rabbit–anti-mouse and peroxidase conjugated goat–anti-rabbit (1:100). The staining reaction was developed using diaminobenzidine and counterstaining with hematoxylin. Sections incubated with PBS only served as negative controls and were consistently negative. Multitissue blocks of gastrointestinal carcinomas served as positive controls.

To assess microvessel density and endothelial cell apoptosis rate, double immunostaining with anti–cleaved caspase 3 and anti-CD34 was done.3 Briefly, deparaffinized slides were pretreated by microwave heating followed by incubation with anti–cleaved caspase 3 rabbit polyclonal antibody (Cell Signaling Technology, Beverly, MA; www.cellsignal.com) for 1 hour at room temperature. The staining reaction was reached by subsequent labeling with peroxidase-labeled goat–anti-rabbit and rabbit–anti-goat antibody (DAKO, Glostrup, Denmark) followed by diaminobenzidine reaction. The slides were then incubated with anti-CD34 antibody (clone QBEND 10, Immunotech, Marseilles, France), in a ready-to-use dilution for 1 hour at room temperature. CD34 staining reaction was visualized with alkaline phosphatase–conjugated goat–anti-mouse antibody (DAKO) and chromogen fast red substrate (Sigma, St. Louis, MO).

Quantification of p53 expression, microvessel density, and apoptosis. For p53 expression, 10 high-power fields per liver metastasis were scored semiquantitatively by two independent observers without knowledge of the clinical data (A.S.H. Gouw and K.P. de Jong). Only viable tumor tissue was evaluated. In 90% of the cases, the interobserver discrepancy was below 10%. When slides were assigned scores that differed by >10%, they were reevaluated until agreement was reached. Expression of p53 nuclear protein was scored semiquantitatively using five classes of staining: 0 (no immunostaining), 1 (<10% positive nuclei), 2 (10-40% positive nuclei), 3 (40-70% positive nuclei), and 4 (>70% positive nuclei). The tumor-liver interface was evaluated for vascular hot spots. Subsequently, at a x400 magnification, the microvessel density in the vascular hot spots was quantified using the Chalkley point overlap morphometric technique as described previously (6). Briefly, the Chalkley grid appears as a circle in the microscopic field and contains several dots. The ocular (containing the grid) of the microscope is rotated until the maximum number of vessels are in contact with a dot. This number is recorded. The number of apoptotic endothelial cells in the same hotspot areas was counted in the whole field of the circle of the Chalkley grid.

DNA isolation and TP53 gene mutation analysis. Tissue slides were screened for the percentage of tumor cells. Materials were selected if at least 50% of the sample consisted of tumor cells. In total, we analyzed 44 tumor samples for the presence of mutations in the TP53 gene. For the 25 cases of which frozen tissue was available, exons 2 to 11 were screened; for the 19 cases of which only paraffin-embedded tissue was available, we screened exons 5 to 8. DNA was isolated from tissue samples with DEXPAT (Takara Shuzo, Co., Ltd., Otsu, Japan) according to standard protocols. All analyses were carried out in duplicate.

Amplification of the TP53 gene. External and internal primer sequences used for the separate amplification of exons 2 to 11 of the TP53 gene were derived from published data (7). Genomic DNA of frozen tissue samples was directly amplified using only the internal primers. For DNA isolated from paraffin-embedded tissues, a nested PCR was done. The preamplification was carried out in 30 µL containing ±150 pg DNA, 0.2 mmol/L deoxynucleotide triphosphate, 1 unit Taq polymerase (Amersham Pharmacia Biotech, Roosendaal, the Netherlands), the reaction buffer provided by the manufacturer, and 150 ng of the external TP53 exons 5 to 8 primers. The PCR program consisted of 25 cycles with a denaturation step of 30 seconds at 94°C, an annealing step of 30 seconds at 61°C, and an elongation step of 45 seconds at 72°C. The first denaturation step lasted for 5 minutes and the final elongation step lasted for 7 minutes (PCR apparatus; GeneAmp 9700, Perkin-Elmer Applied Biosystems, Foster City, CA). Postamplification was done with 1 µL preamplification PCR product. The PCR conditions were the same as described above except that the internal primers with the GC clamps were used and that the number of cycles was increased to 30. An aliquot of 5 µL was analyzed on a 2% agarose gel.

Denaturing gradient gel electrophoresis. The denaturing gradient gel electrophoresis technique was done as described previously (7). Duplex formation of the PCR products was done by incubation at 94°C followed by renaturation at 63°C for 10 minutes, at 58°C for 15 minutes, and at 55°C for 15 minutes. Denaturing gradient gel electrophoresis analysis was done in an Ingeny PhorU-1 PCR apparatus (Ingeny, Goes, the Netherlands). The PCR products were electrophoresed in a 9% polyacrylamide (acrylamide/biacrylamide 37.5:1) gel containing a 35% to 75% urea-formamide denaturing gradient (100% urea-formamide = 7 mol/L urea per 40% deionized formamide). Electrophoresis was done at 110 V and 58°C for 15 hours. From the cases that showed an aberrant four-banded pattern of two homoduplexes and two heteroduplexes, the mutated homoduplex was excised, eluted in TE–4, and reamplified in a 75 µL reaction mix as described above using TP53 primer with a M13 tail (forward: 5'-cgacgttgtaaaacgacggccagt-3' and reverse: 5'-tttcacacaggaaacagctatgac-3'). The PCR products were purified using the "high pure PCR product amplification kit" according to the manufacturer (Roche, Mannheim, Germany) and subjected to fluorescence sequencing. For confirming the mutation, both strands were sequenced and compared with the germ line sequence as published in the Genbank (accession number U94788).

Statistics. Clinicopathologic variables of patients with nonmutated and noncausative mutated tumors were compared with mutated tumors using a {chi}2 test or Fisher exact test where appropriate. Correlation between variables was measured by calculating the Pearson correlation coefficient. Microvessel density, endothelial cell, and tumor cell apoptosis rate were compared in nonmutated/noncausative mutated versus mutated tumors using Students' t test. Overall and disease-free actual survival after primary tumor resection and after liver surgery were compared using the log-rank test with Yates's continuity correction. For these comparisons, microvessel density and endothelial cell apoptotic rate were categorized as "high" and "low" based on values above or below the median values of 3.33 and 0.33, respectively. A separate analysis of time to liver relapse and liver relapse-free survival after liver surgery and after resection of the primary tumor was done. Because only specimens of patients operated before April 1999 were used, we were able to present actual survival instead of estimated (actuarial) survival. The {kappa} statistic was computed to measure agreement between immunohistochemistry and mutation analysis. {kappa} values below 0.40 are considered to represent a fair to poor agreement, 0.41 to 0.60 as moderate agreement and >0.61 as good to very good agreement.


   Results
Top
 Abstract
 Patients and Methods
 Results
Discussion
 References
 
Baseline characteristics. In 17 (38.6%) of the 44 tumors, BP53 monoclonal antibody stained <10% of the tumor cell nuclei and were scored as negative. The remaining 27 tumors (61.4%) stained positive, with 5 (11.4%) between 10% and 40% positive nuclei, 11 (25%) between 40% and 70%, and 11 (25%) >70%. The numbers for DO7 immunostaining were 18 (40.9%) <10% of nuclei, 8 (18.2%) 10% to 40% nuclei, 15 (34.1%) 40% to 70% nuclei, and 3 (6.8%) >70% nuclei.

A deviant four-banded pattern, indicative for a TP53 gene mutation, was detected in 20 of the 44 tumor samples. In 16 tumors of 44 patients (36%), a DNA alteration was identified that affected the p53 protein (1 deletion, 1 splice site, 4 frame shifts, and 10 amino acid substitutions). In 11 of these tumors, abnormal p53 protein was produced that was detected by immunohistochemistry. In five tumors, null mutations were found of which three stained positive for p53 protein and two did not. In four patients, noncausative DNA alterations were found. The individual characteristics of the tumors are shown in Table 1. Table 2 shows that clinicopathologic characteristics were comparable in the patients with tumors with TP53 mutations versus patients with nonmutated tumors. Only 1 of 11 liver metastases originating from the right side of the colon showed mutations in TP53 compared with 15 of 33 metastases originating from the left side of the colon (P = 0.04). Carcinoembryonic antigen serum levels before the liver operation tended to be higher in patients with mutant p53 in the metastases compared with patients with wild-type p53 (P = 0.07).


View this table:
[in this window]
[in a new window]

  		Table 1. TP53 mutations in relation to immunohistochemistry in resected colorectal liver metastases

 

View this table:
[in this window]
[in a new window]

  		Table 2. Clinicopathologic characteristics of the two patient groups

 
Association of TP53 mutations and angiogenic characteristics. The mean microvessel density in tumors with TP53 mutations was 4.1 (SD1.2), which was higher than the mean microvessel density of 3.3 (SD 1.3) in nonmutated tumors (P < 0.05). This difference was not found (P > 0.2) if immunohistochemistry was used. This was irrespective of whether tumors were considered positive if >10% or >40% of tumor cell nuclei stained positive. The number of apoptotic endothelial cells in the tumor microvessels of nonmutated tumors was 0.3 (SD 0.3) versus 0.4 (SD 0.5) in tumors with TP53 mutations (P = 0.4). Also, the mean number of apoptotic tumor cells was not different (P = 0.2) in tumors with TP53 mutations (1.7, SD 0.9) versus nonmutated tumors (2.2, SD 1.4). These results suggest that in tumors with TP53 mutations, proangiogenic activity reflected by microvessel density is more pronounced than in tumors without TP53 mutations.

Association of TP53 mutations and angiogenic status with survival. All patients were included in the survival analysis until their date of death or until their survival time was 60 months after liver surgery. Only one patient with a tumor that was positive both for BP53 and DO7, but contained no TP53 mutation, was censored 33 months after her partial liver resection due to death of unrelated cause. Until her date of death, no signs of recurrent disease could be detected. All other surviving patients at 5 years were free of disease, except one. Overall actual survival data after primary tumor resection and after liver surgery are presented in Table 3. It can be concluded that none of the characteristics influenced survival significantly. The same holds true for liver relapse-free survival (P > 0.2, data not shown). Patients with tumors with both TP53 mutations and a high microvessel density (n = 13) had a median survival after liver surgery of 43 months (31% alive at 5 years), which was not different from 17 patients with nonmutated and low microvessel density tumors (median survival 33 months, 5 years survival 35%, P = 0.7). Also, no difference was found when the survival analysis was based on the date of primary tumor resection (data not shown).


View this table:
[in this window]
[in a new window]

  		Table 3. Actual survival in relation to TP53 mutational state, microvessel density, and endothelial cell apoptotic rate in patients with colorectal liver metastases

 
Association of TP53 mutations and angiogenic status with interval to recurrences. In the 31 patients who developed relapses (at any site) after liver surgery, the median time to detection was 9 months in patients with nonmutated tumors versus 8 months in patients with tumors containing mutated TP53 (P = 0.9). A separate analysis of the interval of liver relapses after liver surgery revealed median values of 12 and 9 months in patients with nonmutated versus mutated tumors (P = 0.7) Also, after resection of the primary tumor, no differences in liver relapse interval were detected with median values of 32 and 38 months (P = 0.7). Median interval of liver relapses after primary tumor resection was not different in low (24 months) versus high (38 months, P = 0.1) microvessel density. Also, no difference in liver relapse interval was found in low versus high apoptotic endothelial cell rate (27 and 38 months respectively, P = 0.5).

Predictive value of immunostaining for TP53 mutations. Analyzing all 44 tumors revealed that immunohistochemical staining of p53, using the two frequently used monoclonal antibodies (BP53 and DO7), correlated very weakly with TP53 mutations. The correlation coefficients were 0.2 (P = 0.17) for BP53 and 0.3 (P = 0.024) for DO7. The positive predictive value of positive immunostaining for the presence of TP53 mutations was 44% for BP53 and 50% for DO7, if tumors with >10% of the nuclei stained positive. For those cases, negative predictive values were 77% and 83%, respectively. Sensitivity and specificity were highest but still rather low for DO7 immunostaining (81% and 54%, respectively). The {kappa} value for DO7 was 0.31 and for BP53 0.19. If the cutoff value for positive immunostaining was set above 40% positive nuclei, the highest positive and negative predictive values were 61% (DO7) and 82% (BP53), respectively. The maximum value for sensitivity was obtained with BP53 (75%) and the maximum specificity was 75% for BP53. {kappa} values were 0.43 (DO7) and 0.36 (BP53). From these data, we conclude that immunohistochemistry is not a reliable technique to show the presence of TP53 mutations.


   Discussion
Top
 Abstract
 Patients and Methods
 Results
 Discussion
References
 
It is claimed that wild-type p53 has an antiangiogenic effect, possibly by up-regulation of the angiogenesis inhibitor thrombospondin (2). Additionally, mutant p53 was found to stimulate vascular endothelial growth factor production (3). Other evidence for the role of p53 in angiogenesis came from in vivo experiments in mice in which p53 deficient cell lines showed a less pronounced and slower response to antiangiogenic therapy compared with p53 wild-type cell lines (8).

In the present study, we analyzed TP53 mutational status and angiogenic characteristics, such as microvessel density and endothelial cell apoptotic rate, in patients who underwent surgery for colorectal liver metastases. We found that TP53 mutations were not of prognostic importance despite the fact that we detected a higher microvessel density in TP53 mutated versus nonmutated tumors. We separately analyzed actual survival in patients with high versus low microvessel density; however, microvessel density did not influence prognosis either.

Published series on the role of TP53 mutations in liver metastases in relation to prognosis show variable results (Table 4). An important observation was done in 103 patients with unresectable liver metastases treated with chemotherapy in which patients with TP53 stop mutations (compared with TP53 wild type) had a relative risk of dying of 3.14 (9). This finding was confirmed in another report in which an association between protein-truncating mutations and the occurrence of liver metastases and a tendency to a worse survival was described (10).


View this table:
[in this window]
[in a new window]

  		Table 4. Studies examining the association of TP53 mutation in colorectal liver metastases and prognostic variables

 
We could find only two other studies dealing with microvessel density in colorectal liver metastases in relation to prognosis (11, 12). In the former study on 71 patients with curative liver resections, low microvessel density was found to be an independent favorable prognostic variable. Confusing in that study, however, was that 15 of 71 patients were judged to have curative liver resections despite the presence of other distant metastases. The second study reports a worse survival in patients with high microvessel density liver metastases compared with patients with low microvessel density liver metastases (12). In this study, it is difficult to understand that after liver resection the development of recurrences in the liver remnant and/or distant metastases (the most frequent causes for death after liver resection) were not associated with microvessel density.

Unfortunately, the confusion on the prognostic importance of p53 and microvessel density is not less in published series on primary colorectal cancer. In a recent review, TP53 mutations were found to be associated with a worse survival in 14 studies, with a better survival in three studies and no association in eight studies (13). One of the reasons for this is probably that many studies are based on immunohistochemical analysis of p53. Additionally, some reports consider p53 overexpressed if >10% of the nuclei were positive, whereas others have their cutoff value at 50% (4, 14). In primary colorectal cancer, p53 overexpression was an independent risk factor for the development of liver metastases (15). In another study in 366 primary tumor samples from patients participating in an adjuvant chemotherapy trial, p53 overexpression was associated with a worse outcome in univariate but not in multivariate analysis (4). Also, the type of TP53 mutation was not of prognostic importance (16).

In primary colorectal cancers, conflicting results on p53 expression and angiogenic state are published. A positive correlation between p53 overexpression and microvessel density was found in two studies using immunohistochemistry on primary colorectal cancer in humans (14, 15). In the latter study, vascular endothelial growth factor expression was examined and it showed that patients with p53+/vascular endothelial growth factor+ tumors had a significantly lower survival compared with patients with p53–/vascular endothelial growth factor– tumors. The authors concluded that in primary colorectal cancer, p53 overexpression is associated with a proangiogenic state and, thus, with a higher chance of disease progression and a worse survival. A comparable conclusion was drawn in a study on 145 patients with colorectal cancer in which high microvessel density and p53 overexpression was associated with a worse survival compared with patients with tumors with low microvessel density and negative p53 immunostaining (17). This difference in survival was caused by a higher proportion of patients developing metastases in the high microvessel density/p53+ group compared with the low microvessel density/p53– group. This suggests that in primary tumors metastatic potential is associated with high microvessel density and p53 positivity. Quite contrasting results were reported on 173 patients with primary colorectal tumors; survival in patients with high microvessel density was significantly better than in patients with tumors with low microvessel density (18). In 22 primary tumors and corresponding liver metastases with p53 mutations, microvessel density was higher than in tumors with wild-type p53 (P < 0.01), whereas this difference was not found when p53 status was evaluated using immunohistochemistry (19). Survival analysis was not done in that study.

The present study also confirms that immunoreactivity of two frequently used monoclonal antibodies to p53 does not adequately predict the presence of TP53 mutations. The correlation between immunostaining and the occurrence of mutations in the p53 gene was rather weak as is reflected by low positive and negative predictive values of immunostaining. The {kappa} value was generally below 0.40 representing a fair to poor agreement. The maximum {kappa} value of 0.43 (moderate agreement) was reached when the cutoff value for positive immunostaining was set at >40% of tumor cell nuclei staining positive. In three other studies in patients with colorectal liver metastases, both immunohistochemistry and mutation analysis of p53 was done (19–21). In these studies, positive predictive values varied from 57% to 100% and negative predictive values from 100% to 71%. Our study revealed that 39% of tumors with p53 immunostaining do not have TP53 mutations and 17% of tumors despite absence of p53 immunostaining do have TP53 mutations. The 25 cases for which the whole p53 coding region was screened for mutations did not reveal mutations outside the hotspot region. Therefore, it is unlikely that mutations were missed in the 19 paraffin samples.

Our finding of a lower incidence of TP53 mutations in metastases derived from right versus left-sided colon cancers confirms findings by others in primary tumors (16, 22). In a previous study using immunohistochemistry, we did not find a difference in overexpression of p53 in right versus left-sided tumors (23). The idea that the anatomic site of colon cancer determines clinicopathologic behavior is based on several studies in which various molecular markers were found to be different for tumors in different anatomic locations (24).

The metastatic process is a rather complicated cascade of events (25). Several steps, such as penetration of a cancer cell from the primary tumor cell into the tumor vessels, transportation through the bloodstream, extravasation in the distant organ, establishment of a metastatic focus, resisting immune attacks, and creating a prosperous environment for invasion and growth, have to be taken before a distant metastasis can settle and survive. Therefore, it is unlikely that only one aspect of this cascade (microvessel density in the primary tumor) is such an important prognostic variable that by itself it can have an impact on prognosis. Moreover, if microvessel density in the primary tumor is of prognostic influence, it does not have to be true for microvessel density in already established metastases. It is very likely that angiogenesis is important for the growth of liver metastases (as it is for growth of the primary tumors), but the role of angiogenesis by itself on the prognosis is debatable.

An important advantage of the present study is the follow-up. None of the 44 patients included in the study were lost to follow-up for other reasons than death. Moreover, because all patients were operated before April 1999, a minimal follow-up period of 5 years was obtained. Therefore, the survival analysis is based on actual data instead of estimations that form the basis for actuarial survival. Of particular importance are the disadvantages of analyzing survival using the actuarial (Kaplan-Meier) method (e.g., high numbers lost to follow-up or the occurrence of competing risk events; ref. 26). Also, censored patients (those with missing information on survival time) can influence results, for instance by overestimation of survival (27, 28). All these drawbacks do not apply to the present study because survival is based on actual numbers.

From this study, in 44 patients with surgically treated colorectal liver metastases, we conclude that TP53 mutation analysis has no prognostic influence, that TP53 mutations are associated with a higher microvessel density, that microvessel density and endothelial cell apoptotic rate do not influence survival, and that predicting mutations of the TP53 gene by immunostaining for p53 protein is unreliable. Because of the central role of p53 in many cellular pathways, further studies with larger number of patients are needed to identify whether specific mutations have prognostic significance (29).


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

3 K.P. de Jong, et al. Endothelial cell apoptosis in tumor microvessels. An additional issue to the quantification of angiogenesis in solid human tumors, submitted for publication. Back

Received 11/22/04; revised 2/11/05; accepted 3/ 1/05.


   References
Top
 Abstract
 Patients and Methods
 Results
 Discussion
 References
 

  1. Steele RJ, Thompson AM, Hall PA, Lane DP. The p53 tumour suppressor gene. Br J Surg 1998;85:1460–7.[CrossRef][Medline]
  2. Dameron KM, Volpert OV, Tainsky MA, Bouck N. Control of angiogenesis in fibroblasts by p53 regulation of thrombospondin-1. Science 1994;265:1582–4.[Abstract/Free Full Text]
  3. Kieser A, Weich HA, Brandner G, Marme D, Kolch W. Mutant p53 potentiates protein kinase C induction of vascular endothelial growth factor expression. Oncogene 1994;9:963–9.[Medline]
  4. Garrity MM, Burgart LJ, Mahoney MR et al. Prognostic value of proliferation, apoptosis, defective DNA mismatch repair, and p53 overexpression in patients with resected Dukes' B2 or C colon cancer: a North Central Cancer Treatment Group Study. J Clin Oncol 2004;22:1572–82.[Abstract/Free Full Text]
  5. Jass JR, Sobin LH, editors. Histological typing of intestinal tumours. WHO International histological classification of tumours. 2nd ed. Berlin: Springer-Verlag; 1989. p. 32–3..
  6. Vermeulen PB, Gasparini G, Fox SB, et al. Second international consensus on the methodology and criteria of evaluation of angiogenesis quantification in solid human tumours. Eur J Cancer 2002;38:1564–79.
  7. Hayes VM, Bleeker W, Verlind E, et al. Comprehensive TP53-denaturing gradient gel electrophoresis mutation detection assay also applicable to archival paraffin-embedded tissue. Diagn Mol Pathol 1999;8:2–10.[CrossRef][Medline]
  8. Yu JL, Rak JW, Coomber BL, Hicklin DJ, Kerbel RS. Effect of p53 status on tumor response to antiangiogenic therapy. Science 2002;295:1526–8.[Abstract/Free Full Text]
  9. Etienne MC, Chazal M, Laurent-Puig P, et al. Prognostic value of tumoral thymidylate synthase and p53 in metastatic colorectal cancer patients receiving fluorouracil-based chemotherapy: phenotypic and genotypic analyses. J Clin Oncol 2002;20:2832–43.[Abstract/Free Full Text]
  10. Miyaki M, Iijima T, Yasuno M, et al. High incidence of protein-truncating mutations of the p53 gene in liver metastases of colorectal carcinomas. Oncogene 2002;21:6689–93.[CrossRef][Medline]
  11. Miyagawa S, Miwa S, Soeda J, Kobayashi A, Kawasaki S. Morphometric analysis of liver macrophages in patients with colorectal liver metastasis. Clin Exp Metastasis 2002;19:119–25.[CrossRef][Medline]
  12. Nanashima A, Ito M, Sekine I, et al. Significance of angiogenic factors in liver metastatic tumors originating from colorectal cancers. Dig Dis Sci 1998;43:2634–40.[CrossRef][Medline]
  13. Iacopetta B. TP53 mutation in colorectal cancer. Hum Mutat 2003;21:271–6.[CrossRef][Medline]
  14. Liang JT, Huang KC, Jeng YM, Lee PH, Lai HS, Hsu HC. Microvessel density, cyclo-oxygenase 2 expression, K-ras mutation and p53 overexpression in colonic cancer. Br J Surg 2004;91:355–61.[CrossRef][Medline]
  15. Kang SM, Maeda K, Onoda N, et al. Combined analysis of p53 and vascular endothelial growth factor expression in colorectal carcinoma for determination of tumor vascularity and liver metastasis. Int J Cancer 1997;74:502–7.[CrossRef][Medline]
  16. Soong R, Powell B, Elsaleh H, et al. Prognostic significance of TP53 gene mutation in 995 cases of colorectal carcinoma. Influence of tumour site, stage, adjuvant chemotherapy and type of mutation. Eur J Cancer 2000;36:2053–60.
  17. Vermeulen PB, Van den Eynden GG, Huget P, et al. Prospective study of intratumoral microvessel density, p53 expression and survival in colorectal cancer. Br J Cancer 1999;79:316–22.[CrossRef][Medline]
  18. Prall F, Gringmuth U, Nizze H, Barten M. Microvessel densities and microvascular architecture in colorectal carcinomas and their liver metastases: significant correlation of high microvessel densities with better survival. Histopathology 2003;42:482–91.[CrossRef][Medline]
  19. Kern A, Taubert H, Scheele J, et al. Association of p53 mutations, microvessel density and neoangiogenesis in pairs of colorectal cancers and corresponding liver metastases. Int J Oncol 2002;21:243–9.[Medline]
  20. Saw RP, Koorey D, Painter D, Gallagher PJ, Solomon MJ. p53, DCC and thymidylate synthase as predictors of survival after resection of hepatic metastases from colorectal cancer. Br J Surg 2002;89:1409–15.[CrossRef][Medline]
  21. Heide I, Thiede C, Sonntag T, et al. The status of p53 in the metastatic progression of colorectal cancer. Eur J Cancer 1997;33:1314–22.
  22. Hamelin RR, Laurent-Puig P, Olschwang S, et al. Association of p53 mutations with short survival in colorectal cancer. Gastroenterology 1994;106:42–8.[Medline]
  23. de Jong KP, Stellema R, Karrenbeld A, et al. Clinical relevance of expression of transforming growth factor {alpha}, epidermal growth factor receptor, p53 and Ki67 in colorectal liver metastases and corresponding primary tumors. Hepatology 1998;28:971–9.[CrossRef][Medline]
  24. Iacopetta B. Are there two sides to colorectal cancer? Int J Cancer 2002;101:403–8.[CrossRef][Medline]
  25. Chambers AF, Groom AC, MacDonald IC. Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer 2002;2:563–72.[CrossRef][Medline]
  26. Satagopan JM, Ben Porat L, Berwick M, Robson M, Kutler D, Auerbach AD. A note on competing risks in survival data analysis. Br J Cancer 2004;91:1229–35.[CrossRef][Medline]
  27. Leung KM, Elashoff RM, Afifi AA. Censoring issues in survival analysis. Annu Rev Public Health 1997;18:83–104.[CrossRef][Medline]
  28. Gudjonsson B. Survival statistics gone awry: pancreatic cancer, a case in point. J Clin Gastroenterol 2002;35:180–4.[CrossRef][Medline]
  29. Soussi T, Beroud C. Assessing TP53 status in human tumours to evaluate clinical outcome. Nat Rev Cancer 2001;1:233–40.[CrossRef][Medline]
  30. Russo A, Migliavacca M, Bazan V, et al. Prognostic significance of proliferative activity, DNA-ploidy, p53 and Ki-ras point mutations in colorectal liver metastases. Cell Prolif 1998;31:139–53.[CrossRef][Medline]
  31. Rosty C, Chazal M, Etienne MC, et al. Determination of microsatellite instability, p53 and K-RAS mutations in hepatic metastases from patients with colorectal cancer: relationship with response to 5-fluorouracil and survival. Int J Cancer 2001;95:162–7.[CrossRef][Medline]
  32. Tullo A, D'Erchia AM, Honda K, et al. Characterization of p53 mutations in colorectal liver metastases and correlation with clinical parameters. Clin Cancer Res 1999;5:3523–8.[Abstract/Free Full Text]
  33. Yang Y, Forslund A, Remotti H, et al. P53 mutations in primary tumors and subsequent liver metastases are related to survival in patients with colorectal carcinoma who undergo liver resection. Cancer 2001;91:727–36.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 CarcinogenesisHome page
D. G. Mollevi, T. Serrano, M. M. Ginesta, J. Valls, J. Torras, M. Navarro, E. Ramos, J. R. Germa, E. Jaurrieta, V. Moreno, et al.
Mutations in TP53 are a prognostic factor in colorectal hepatic metastases undergoing surgical resection
Carcinogenesis, June 1, 2007; 28(6): 1241 - 1246.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
E. Lopez-Knowles, S. Hernandez, M. Kogevinas, J. Lloreta, A. Amoros, A. Tardon, A. Carrato, S. Kishore, C. Serra, N. Malats, et al.
The p53 Pathway and Outcome among Patients with T1G3 Bladder Tumors.
Clin. Cancer Res., October 15, 2006; 12(20): 6029 - 6036.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by de Jong, K. P.
Right arrow Articles by van den Berg, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by de Jong, K. P.
Right arrow Articles by van den Berg, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online