Annals of Surgical Oncology Cite Track
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/s10434-006-9155-0
Annals of Surgical Oncology 14:1272-1284 (2007)
© 2007 Society of Surgical Oncology
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
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gasser, M.
Right arrow Articles by Waaga-Gasser, A. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gasser, M.
Right arrow Articles by Waaga-Gasser, A. M.

Original Article

Comparative Analysis of Predictive Biomarkers for Therapeutical Strategies in Colorectal Cancer

Martin Gasser, MD1, Christiane Gerstlauer2, Martin Grimm2, Marco Bueter, MD1, Tatiana Lebedeva, PhD3, Jens Lutz, MD4, Uwe Maeder, MD5, Carmen Ribas, MD6, Claudia Ribas, MD6, Ekaterina Nichiporuk2, Andreas Thalheimer, MD1, Uwe Heemann, MD4, Arnulf Thiede, MD1, Detlef Meyer, MD1 and Ana Maria Waaga-Gasser, PhD2

1 Department of Surgery I, University of Wuerzburg, Oberduerrbacherstr. 6, 97080, Wuerzburg, Germany
2 Department of Surgery I, Molecular Oncology and Immunology, University of Wuerzburg, Oberduerrbacherstr. 6, 97080, Wuerzburg, Germany
3 American Red Cross, New England Region, Dedham, MA 02026, USA
4 Department of Nephrology, University of Munich, Klinikum rechts der Isar, Ismaninger Str. 22, 81675, Munich, Germany
5 Interdisziplinäres Tumorzentrum, University of Wuerzburg, Josef-Schneiderstr. 2, 97080, Wuerzburg, Germany
6 Instituto de Pesquisas Medicas, Faculdade Evangelica do Parana, Rua Padre Anchieta 2770, 80730000 CuritibaParana, Brasil

Correspondence: Address correspondence and reprint requests to: Ana Maria Waaga-Gasser, PhD; E-mail: waaga-gasser{at}chirurgie.uni-wuerzburg.de


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Prognostic information regarding the risk of postoperative tumor recurrence defined by a profile of serological, morphological and/or molecular markers can have potential value, particularly for patients with colorectal carcinoma (CRC) of the International Union Against Cancer (UICC) stage II/III who may benefit from adjuvant chemotherapy after surgery.

Methods: A retrospective study of 783 patients with CRC (UICC I–III) including a subgroup analysis of 116 subjects was conducted to determine preoperative serum carcinoembryonic antigen (CEA), carbohydrate antigen (CA) 19-9, and p53 serum levels. In addition, protein and gene expression of p53, CEA, and adenomatous polyposis coli (APC) was assessed in the tumors of those patients. The values of all serological, morphological, and molecular parameters were correlated with clinicopathological characteristics for their predictive value of tumor recurrence over a mean follow-up period of 32 ± 6.2 months.

Results: Serum CEA but not CA 19-9 or p53 was a significant prognostic factor for disease-free survival, along with UICC and T/N stage. When comparing elevated CEA, CA 19-9, and p53 serum levels with expression of the markers in the tumors, their overall expression was found to be 61.3% in the serum versus 93.5% in the tumor in analyzed patients (n = 116). In particular, all patients in UICC stages I–III who demonstrated at least three elevated markers (CEA/CA 19-9/p53) in serum and/or in the tumor presented with tumor recurrence/metastases.

Conclusion: Overall increased p53, CEA, and CA 19-9 serum levels and their marker expression in the tumor may be used at the time of primary tumor removal for defining patients at risk for tumor recurrence.

Key Words: Prognostic marker • Colorectal carcinoma • p53 • CEA • CA 19-9


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
CRC is one of the leading causes of cancer-associated mortality world wide, and despite major advances in diagnosis and treatment of the disease, mortality has remained unchanged during the past 20 years.1,2 Currently, Dukes’ classification is the most commonly used predictor of prognosis for CRC patients; indeed, 30–40% of Dukes’ stages B [International Union Against Cancer (UICC) II] and C (UICC III) patients experience relapse and die of the disease.1,2 Identification of additional prognostic markers would help to define groups of patients with Dukes’ stage B and/or C CRC into patients with high and low risk of recurrence and to select appropriate supplemental treatment after surgery.

Numerous investigations have been performed addressing the possible predictive value of the tumor suppressor gene p53 and its mutations in patients with colon cancer. It has been demonstrated that in patients with locally advanced tumor, a mutant or increased expression of p53 is associated with a worse clinical outcome.3–11 However, several investigations have found either the opposite or no association between the expression of p53 and clinical outcome.12–14 The adenomatous polyposis coli (APC) gene and p53 drive the transition from normal epithelium through increasing adenomatous dysplasia to colorectal cancer. In some studies, the serum APC molecular marker was closely correlated with lymph node metastasis and TNM stage, but no significant difference was observed in the association of APC gene mutations in primary tumors with patients’ demographic data.8 Carcinoembryonic antigen (CEA) is a complex glycoprotein that is upregulated in approximately 90% of advanced colorectal cancers and contributes to the malignant characteristics of the tumor.1 It can be measured quantitatively in serum, where its preoperative levels have prognostic value. That was demonstrated in several studies for CEA serum levels > 5 ng/ml in both Dukes’ B and C patients (Table 1Go). Because the test for CEA levels is unreliable at early stages of CRC, measuring CEA levels is not suitable for population screening.


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

  		TABLE 1. Prognostic value of CEA and CA 19-9
 
Whereas elevated preoperative CEA serum levels have been associated with a poor prognosis, monitoring of CEA postoperatively may allow identification of patients with metastatic disease in whom surgical resection or other localized therapy might have a potential benefit. However, there is no clear difference in survival after resection of metastases of CRC between patients in whom second-look surgery was performed on the basis of elevated CEA levels and those with other laboratory or imaging abnormalities.15 There is also no clear consensus on the frequency or duration of CEA monitoring although the American Society of Clinical Oncology (ASCO) guidelines currently recommend it every 2–3 months for at least 2 years after diagnosis. In the follow-up of patients undergoing palliative therapy, the CEA level correlates well with response. Moreover, CEA is not only indicative of response but may also identify patients with stable disease who would profit from combination therapy in terms of survival and symptom relief.15 Carbohydrate antigen (CA 19-9) is often used in combination with CEA to manage patients with colorectal cancer even though there is insufficient evidence to support the use of CA 19-9 in such patients.

The aim of this study was to define a profile of relevant markers on which therapeutic decisions could be made with greater precision for given individuals with CRC. We examined the tumor suppressor gene p53 in comparison with APC, CEA, and CA 19-9 as markers that could have a potential value in defining patients with Dukes’ stage B (UICC II) and C (UICC III) who may benefit from adjuvant chemotherapy.


   PATIENTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Seven hundred and eighty-three patients (452 male and 331 female) with a mean age of 66.06 ± 5.4 years and histologically confirmed CRC treated at the University Hospital of Wuerzburg, Germany, were evaluated retrospectively for this study (Table 2Go). The UICC classification was used to determine CRC tumor stage. Only patients operated with curative intent in UICC stages I–III (Dukes’ A–C) were included in this study. Those with metastases (UICC IV, Dukes’ D) or secondary carcinoma at the time of diagnosis were excluded. Peripheral blood was obtained from the patients before surgery. Tumors were evaluated for location, stage, and differentiation grade (Table 2Go). Data concerning age, gender, level of wall infiltration, and lymph node metastasis were collected in our database (Table 2Go). All patients were followed up regularly at 3- to 6-month intervals, resulting in a completeness index (i.e., observed divided by expected follow-up time) of 0.96. The mean follow-up time was 4.9 ± 3.5 (median 3.9) years. Physical examinations, blood tests, chest radiography, abdominal ultrasonography, and computed tomography (CT) or magnetic resonance imaging (MRI) were carried out in accordance with the guidelines of the German tumor centers. In addition, out of the 783 studied patients, tumor tissue samples from 116 consecutive patients operated in our department between October 2001 and December 2003 were collected—with informed consent—immediately after surgical resection, frozen instantly in liquid nitrogen, and stored at –80°C until analyzed. Blood samples from these patients were separated on Lymphoprep according to the manufacturer’s instructions (Nycomed Pharma, Oslo, Norway) for CA 19-9 and p53 evaluation. Correlation between the development of clinical metastases/recurrence and CEA serum levels was analyzed during a mean follow-up of 6.5 ± 4 years whereas CA 19-9 and p53 serum levels and immunohistological data and gene expression were determined during a mean follow-up of 32 ± 6.2 months. The protocol used was approved by the local medical ethics committee.


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

  		TABLE 2. Clinicopathological characteristics
 
CEA, CA 19-9, and p53 Measurement in Peripheral Blood
Blood samples were collected from the patients before surgery. Preoperative serum levels of CEA and CA 19-9 were measured with the commercial solid-phase double-antibody sandwich immunoassay (Roche Laboratories, Basel, Switzerland). The upper limit of normal values were ≤5 ng/ml for CEA and ≤37 U/ml for CA 19-9. Preoperative p53 serum levels were performed by enzyme-linked immunosorbent assay (ELISA) (Fa. Biosource International, CA, USA), following the manufacturer’s protocol.

Humoral Response to p53
p53-specific immunoglobulin (Ig)G titers were quantified using the commercially available p53 ELISA Kit (Oncogene, Boston, MA, USA). Results for each serum sample were calculated first by determining the relative p53 autoimmune index, following the manufacturer’s calculation protocol (cut-off value: 0.035). The anti-p53 IgG titer (U/ml) of the positive serum samples was calculated at x50 dilution using the linear regression curve obtained by plotting the concentration of standards versus the absorbance. In all ELISA experiments, patient sera (n = 116) were tested in duplicate and the experiments repeated three times. Sera from healthy patients were used as controls (n = 35).

Immunohistochemistry
Representative biopsies of tumor samples with surrounding colorectal mucosa were fixed in 10% buffered formalin for histological examination. Paraffin sections were evaluated using hematoxylin and eosin (H&E). Additional biopsies were snap frozen and stored at –80°C for ribonucleic acid (RNA) extraction and reverse transcriptase polymerase chain reaction (RT-PCR) analysis. Cryostat sections fixed in paraformaldehyde-lysine-periodate were prepared for cell-surface-antigen staining. Monoclonal antibody (mAb) DO-7 (anti-p53, 35–45), CEA mAb, CA 19-9 mAb, isotype-matched mAbs or purified IgG1, and controls for residual endogenous peroxidase activity were included in each experiment (DakoCytomation, Heidelberg, Germany). The percentage of tumor cells staining for p53, APC, CEA, and CA 19-9 was estimated within 20 consecutive high-power fields (x40 magnification) and scored semiquantitatively as low (+), medium (++), or high (+++). Blinded research was performed. The slides were read independently by at least three observers using light microscopy. In the event of interobserver discrepancy, the observers met and arrived at a consensus agreement.

RNA Extraction
RNA was extracted from at least 10 µg of homogenized tumor tissue using the RNA extraction kit (Qiagen, Hilden, Germany), washed in diethyl pyrocarbonate (DEPC)-75% ethanol before being dissolved in DEPC-water, and stored at –70°C until further analysis. The amount of total RNA was determined by measuring absorbance at 260 nm. The purity of the total RNA was established by confirming that the 260:280-nm ratio was within the 1.8–2.0 range, indicating that the RNA preparations were free of protein contaminants.

Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) for Detection of p53 Mutations
cDNA was prepared from 2 µg of total RNA of 24 patients using the Promega AccessQuick RT-PCR System (Promega, WI, USA). The procedure was performed according to the standard RT-PCR protocol using avian myeloblastosis virus (AMV) reverse transcriptase. Two sets of primers amplifying the p53 coding region spanning exons 4–9 were used. The primer sequences were: for exons 4–6, 5'-TGT CCC CGG ACG ATA TTG AAC-3' and 5'-TTC CTT CCA CTC GGA TAA GAT GC-3', amplicon size 465 bp; and for exons 5–9, 5'-GCT CAG ATA GCG ATG GTC TGG C-3' and 5'-TCT CGG AAC ATC TCG AAG CG-3', amplicon size 484 bp. RT-PCR conditions for the first reaction were 48°C for 45 min; followed by 2 min at 95°C; followed by 40 cycles of 94°C for 30 s, 60°C for 60 s, and 68°C for 20 s; followed by 68°C for 5 min. Before sequencing, all PCR products were purified following the QIAquick PCR purification kit protocol (Qiagen, Hilden, Germany) using vacuum manifold. PCR products were blindly sequenced on an ABI Prism 373 fluorescent dye terminator (PE/Applied Biosystems, Foster City, CA, USA).

Real-Time Polymerase Chain Reaction (Real-Time PCR) for p53, CEA, and APC Genes in Tumor Specimens
mRNA expression of p53, CEA, and APC was analyzed in colorectal tumor specimens by real-time PCR (n = 46). RNA was extracted as described above. cDNA was prepared using 2 µg of heat-denatured RNA. The following primer sequences were used (Sigma Genosys, Woodlands, TX): for GAPDH, 5'- ATC CCA TCA CCA TCT TCC AGG-3' and 5'-CGC CCC ACT TGA TTT TGG-3'; for APC, 5'-TAT GGA AGC CGG GAA GGA TC-3' and 5'-AGG ACT GCA CTC TCC AGA ACG-3'; for CEA, 5'-GCC CGC ATA CAG TGG TCG-3' and 5'-ATC AGC AGG GAT GCA TTG G-3'; and for p53, 5'-CCA GAA AAC CTA CCA GGG CA-3' and 5'-GAA TGC AAG AAG CCC AGA CG-3'. Primers were designed using the Primer Express software for primer design to amplify short segments of 50–150 base pairs of target cDNA. Optimum primer concentration was determined by titration. Real-time quantitative PCR was performed in a two-step RT-PCR using SYBR Green PCR Master Mix (PE Biosystems, Foster City, CA, USA) with 100 ng cDNA and 300 nM of primers in a total reaction volume of 50 µl. PCR thermal cycling conditions were as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 60 s. Gene-specific products were continuously measured by an ABI Prism 7700 sequence detector (Applied Biosystems), and relative quantification was done following the manufacturer’s directions. All samples were assayed in duplicate and normalized during data analysis by a passive reference dye provided in the SYBR Green PCR Master Mix to compensate for well-to-well fluorescence variations. The average threshold cycle (Ct) value was calculated as the cycle number at which fluorescence of reporter reaches a fixed threshold. The difference ({delta}Ct) between the average Ct values of the samples in the target wells and those of the housekeeping gene, GAPDH, was assessed, followed by the calculation of the difference between the average {delta}Ct values of the samples for each target and the {delta}Ct value of the control sample for that target ({delta}{delta}Ct). The relative quantification value, fold change, is expressed as 2–{delta}{delta}Ct.

Statistical Analysis
The chi-square test was used for univariate analysis between CEA and clinicopathological data. The Kaplan–Meier method was performed to calculate the cumulative survival rates and to plot survival curves; the log-rank test was used to find statistical differences between curves. Tumor-related survival was calculated from the time of surgery to the last contact or death. Relapse-free time was calculated from the date of surgery to the date of first tumor relapse. Multivariate analysis by Cox regression was then performed to determine the most important predictors of survival among all possible variables. P values < 0.05 were considered statistically significant and values < 0.01 highly significant. All statistical analysis was conducted using the SPSS 12.0 for Windows statistical software package. The Mann–Whitney U test was performed to compare CEA, CA 19-9, and p53 serum levels and immunostaining, gene expression (p53, APC, CEA, CA 19-9), and the development of postoperative metastases/recurrence in the 116 patients. Significance was assumed at P < 0.05.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Clinicopathological characteristics of the 783 patients and the prognostic factor CEA are summarized in Table 2Go. Pathologic stages were defined according to UICC classification.16

Clinicopathological Characteristics and Prognostic Factors
Preoperatively elevated CEA serum levels (≥5 ng/ml) were significantly associated with UICC staging (P = 0.002), level of wall infiltration (P < 0.001), and lymph node metastases (P = 0.030), as determined by univariate analysis (chi-square test). UICC staging (P < 0.0001; Fig. 1aGo), level of wall infiltration (P < 0.0001), lymph node metastasis (P < 0.0001), CEA serum level (P = 0.0003; Fig. 1bGo), and T and N stage (P < 0.0001; Fig. 1c and dGo, respectively) were found to be significant prognostic factors for tumor-related survival. In contrast, gender, age, localization, grading, CA 19-9 serum level (P = 0.2180), and p53 serum level (P = 0.8655) measured as single parameters did not prove to be significant prognostic factors for tumor-related survival. Although all the patients with UICC stage III and serum levels > 5 ng/ml revealed a worse prognosis for tumor-related survival (P = 0.0044), those with UICC stage IIIA (pT1/T2 pN1) showed a significantly better survival rate (P = 0.0410; Fig. 2aGo). Moreover, patients with UICC stage IIIA and < 5 ng/ml serum levels demonstrated a better prognosis for tumor-related survival than those with > 5 ng/ml and better than patients with UICC IIIB (pT3/T4 pN1) and C (pT1-T4 pN2) (P = 0.0410; Fig. 2b, c, and dGo, respectively). By multivariate Cox analysis, UICC staging (P < 0.001) and CEA (P = 0.002) were found to be independent prognostic factors in colorectal patients.


Figure 1
View larger version (60K):
[in this window]
[in a new window]

  		FIG. 1. a–d. Clinicopathological characteristics and prognostic factors. International Union Against Cancer (UICC) staging (a, P < 0.001), carcinoembryonic antigen (CEA) serum level (b, P = 0.003), and T and N stage (c and d, respectively; P < 0.001) were found to be significant prognostic factors.

 

Figure 2
View larger version (47K):
[in this window]
[in a new window]

  		FIG. 2. a–d. Clinicopathological characteristics and prognostic factors. Patients of International Union Against Cancer (UICC) stage III with preoperative carcinoembryonic antigen (CEA) serum levels > 5 ng/ml showed a worse prognosis (a, P = 0.0044); those at UICC stage IIIA demonstrated a significantly better survival rate (b, P = 0.0410). Patients with UICC stage IIIA and CEA serum levels < 5 ng/ml had a better prognosis than those with CEA serum levels > 5 ng/ml and better than patients with UICC IIIB and C (b, c, and d, respectively; P = 0.0410).

 
Relapse-free survival was significantly influenced by UICC staging (P < 0.0001), the level of wall infiltration (P < 0.0001), lymph node metastases (P < 0.0001), and by CEA serum level (P = 0.0436). Colon carcinomas showed a lower relapse rate than those of the rectum (P = 0.0210), despite the fact of additional total mesorectal excision. A prognostic significance of gender, age, grading, serum CA 19-9, and p53 as single parameters was not observed (data not shown).

Humoral Response to p53
In the series of the 116 consecutively analyzed patients, the relationship between p53-specific antibodies in serum and the expression of p53 in the tumor was studied; the results were then correlated with disease stage (UICC stages I–III). We found 41.3% (n = 48) of patients that displayed p53-specific IgG serum antibodies; in 68 patients, antibodies were either not detectable or were below the critical range. No correlation was found between the levels of free p53 protein and patient antibody levels (data not shown). Immunohistochemical analysis of p53 expression showed that tumors of UICC stages II and III had significantly stronger staining intensity than those of earlier stages (UICC I, P < 0.05, Fig. 3Go). The presence of p53-specific IgG antibodies correlated with strong p53 staining intensity (80–100% expression of positive cells; P < 0.05), suggesting a correlation between intracellular accumulation of p53 in tumor cells and humoral response to p53. In contrast, the presence of a humoral response showed no correlation with levels of free p53 protein in sera.


Figure 3
View larger version (477K):
[in this window]
[in a new window]

  		FIG. 3. Immunohistochemical analysis of p53 expression. Tumors in patients of International Union Against Cancer (UICC) stages II and III showed significantly stronger staining intensity than those in UICC stage I (P < 0.005). The percentage of tumor cells staining for adenomatous polyposis coli (APC), carcinoembryonic antigen (CEA), carbohydrate antigen (CA) 19-9, and p53 was estimated within 20 consecutive high-power fields (x40 magnification) and was scored as low, medium, or high.

 
p53 Mutational Status
The p53 mutational status was analyzed in 24 out of 116 randomly chosen patients with CRC. Five different point mutations located at five different sites along the p53 sequence were identified in the tumor biopsies [Base pair (bp) 82 C:G, bp 300 T:C, bp 366 A:G, bp 686 T:C, bp 711 T:C). p53 mutations showed no significant correlation with the presence of p53-specific antibodies. Remarkably, 20% of the analyzed tumors (n = 5 out of 24 patients) were found to contain mutations that coincided with the development of metastases in these patients (Table 3Go).


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

  		TABLE 3. p53 mutations in colorectal cancer patients

 
Correlation between the Detection of CEA, CA 19-9, and p53 in the Serum and in the Tumor with Pathological Parameters
Of the 116 patients with analysis of tumor tissue, 61.3% (71/116) showed at least one elevated bio-marker—CEA, CA 19-9 or p53—in their serum while 93.5% expressed one of the markers within their tumors (Table 4Go). In addition, serum CEA, CA 19-9, and p53 were detected in 57.7%, 29.6%, and 75%, respectively, of patients with tumors simultaneously expressing these markers (Table 4Go). Overall, 65.7% (71/108) of patients showed expression of at least one marker in the serum and in the tumor (Table 4Go). A stage-related increase in gene expression levels of APC, CEA, or p53 and in the protein expression of APC, CEA, CA 19-9, and p53 was observed (Table 5Go, Fig. 3Go). APC, a well-investigated gene, was highly expressed in our study, even in the early stage of the disease (UICC I–III: 83.3–100%, data not shown). Interestingly, even in the early tumor stages (UICC I), a remarkable gene expression of CEA and p53 was detected (in 66.7% and 50.0% of tumors, respectively). Overall, the presence of at least one marker alteration in each tumor was observed in 50% (3/6) of UICC I, 73.7% (14/19) of UICC II, and 85.7% (18/21) of UICC stage III of the disease (Table 5Go). The association of the presence of p53 in the serum and its expression in the tumor (immunostaining and gene expression) for CRC patients with pathologic parameters (UICC grading) is depicted in Table 6Go. In contrast to the expression of p53 in the tumor, no correlation was observed between the presence of p53 in the serum and UICC stage or differentiation (P > 0.05).


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

  		TABLE 4. Detection of CEA, CA 19-9, and p53 in the serum and their expression in the tumor (immunostaining)
 

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

  		TABLE 5. Genetic profiling of the biomarkers within the tumor as a function of tumor stage (Real Time PCR, n=46)
 

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

  		TABLE 6. Correlation between the detection of p53 in the serum and in the tumor (immunostaining/gene expression) with pathological parameters
 
With the exception of CEA, CA 19-9 and p53 on their own showed no correlation between elevated serum levels and/or expression in the tumor and postoperative recurrence/metastases. Interestingly, all patients who showed elevated serum levels and/or expression in their tumor (protein or gene expression) in all three parameters (CEA, CA 19-9, and p53) developed recurrence/metastases during the follow-up (36 ± 6.6 months) independently of their UICC stage (UICC I–III). Table 7Go shows ten representative patients out of 116 analyzed individuals, five of them with detectable p53 mutations (Table 3Go) and tumor recurrence and five who failed to demonstrate positivity for all three markers (CEA, CA 19-9, and p53). They revealed tumor-free survival > 56 months after R0 resection, indicating that patients with negative CEA serum levels could benefit from analysis of these biomarkers.


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

  		TABLE 7. Correlation between elevated serum levels and expression in the tumor of CEA, CA 19-9 and p53 with postoperative recurrence/metastases
 
These data indicate that at the time of primary tumor removal, p53 and CA 19-9 may be used in combination with CEA for defining patients at risk for tumor recurrence. Those patients may particularly benefit from adjuvant therapies. However, patients with elevated p53 or CA 19-9, as independent serum prognostic markers, seem not to be at increased risk compared with patients with elevated CEA serum levels (> 5 ng/ml).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
To identify groups of patients who may need adjuvant treatment, molecular staging and correlation with clinical data may be helpful in classifying histologically similar tumors. Colon cancer develops through a multistep process, with an accumulation of multiple genetic alterations that are often the cause of a form of genomic instability. The two best known mechanisms of genomic instability are chromosomal instability (CIN) and microsatellite instability (MSI). The CIN phenotype is found in approximately 85% of sporadic colon cancers and is characterized by aneuploidy, multiple chromosomal rearrangements, and an accumulation of tumor suppressor genes such as p53 and APC. The MSI phenotype is associated with small insertions and deletions, mainly in repetitive sequences (microsatellites), and is found in approximately 15% of cases. This instability, often referred to as high-frequency MSI (MSI-H), is caused by defects of the mismatch repair system, which is involved in repairing DNA errors that arise during DNA replication. Clear-cut correlations between the somatic genetic alterations in tumors and the clinical behavior of the tumor are rare. Only a few markers, such as MSI-H and p53, seem to have a prognostic value.17 Mutations in the p53 gene are associated with aggressive tumor growth and subsequent reduced survival time whereas MSI-H seems to be correlated with a favorable outcome. In general, predicting biologic behavior of—in particular—stage III colon cancers is difficult and remains a great clinical problem. In this study, we particularly analyzed the significance of the tumor suppressor gene p53 compared with the known antigens CEA and CA 19-9 as putative indicative marker for better predicting the progression of colorectal cancer. The aim was to identify early those patients who are at risk for tumor relapse after curative surgery of their primary tumor.

Currently, UICC or Dukes’ stages are probably seen as the main prognostic factors in CRC. However, other clinicopathological factors, such as lymphovascular infiltration, cellular differentiation, DNA ploidy, and oncogene expression, may be of prognostic relevance.18 Besides CEA, which is routinely used in colorectal cancer patients after surgery for the primary tumor, further prognostic markers should be defined for clinical usage. The prognostic value of CEA has been well studied (Table 1Go). However, normal serum CEA levels have been demonstrated in a significant number of patients, even in advanced tumor stages. CA 19-9 has been shown to be an insignificant parameter in terms reliably predicting the course for a specific patient with colorectal cancer. Our study shows the significance of UICC stage as the strongest prognostic factor in patients with colorectal cancer, which corresponds to the published data. In a study with 358 patients diagnosed with CRC, a relationship between preoperative CEA serum levels and survival was reported.19 The study found that the recurrence rate was higher in patients with Dukes’ B (UICC II) and C (UICC III) disease who had preoperative CEA serum levels > 5 ng/ml. Moreover, another study with 425 patients has shown that the preoperative CEA level significantly affected survival rates in stages II and III patients.20 Further, the authors have also proven that Dukes’ B (UICC II) and C (UICC III) stages are high-risk groups.21–24 Those patients had a worse clinical outcome when having elevated serum levels of tumor markers according to the respective cutoff values (Table 1Go). Other investigators reported that higher preoperative serum levels of tumor markers are associated with poorer prognosis only in Dukes’ C (UICC III) disease25–27 whereas in a study with 563 patients, it was emphasized that Dukes’ C1 seems to be an especially high-risk group.28,29 It was also pointed out that Dukes’ C patients with involvement of four or more lymph nodes have a worse 5-year survival rate than those with CEA levels > 10 ng/ml.30

Our current study performed on 783 patients confirms the observation that CEA levels ≥ 5 ng/ml are of significant prognostic value in stage III patients. As proven by others, it is also of great importance to subdivide patients with positive nodes in Dukes’ C (UICC III) on the basis of the depth of tumor penetration.28–30,18 In our study, UICC stage IIIA patients (pT1/T2 pN1) with elevated serum CEA levels (> 5 ng/ml) were specifically identified as a group at risk for tumor relapse. Moreover, patients with UICC stage IIIA and < 5 ng/ml serum levels demonstrated a better prognosis than those with > 5 ng/ml and better than those with UICC IIIB (pT3/T4 pN1) and C (pT1–T4 pN2). No such correlation was observed for CA 19-9 serum values when measured alone. In contrast to the serum levels, the results of CA 19-9 immunohistochemistry and gene expression analysis in the tumor showed, similar to CEA, a stage-dependent correlation with the course of the disease and confirmed UICC III patients as a high-risk group. However, in other studies, the authors could only find a significant influence of abnormal CEA levels (> 5 ng/ml) in UICC II patients but not for patients in UICC III already with tumor-infiltrated lymph nodes.31 This latter observation was confirmed by others for both markers CEA (CEA ≥ 5 ng/ml) and CA 19-9 (CA 19-9 ≥ 37 U/ml). In stage II of colorectal cancer, patients with elevated levels of both CEA and CA 19-9 have a significantly poorer prognosis than patients displaying normal CEA and CA 19-9 levels.32 Others demonstrated that CEA and CA 19-9 have prognostic impact in all patient groups without finding remarkable differences between stages.33–37,18 The results of these studies do not allow using the markers very specifically; they do not point out a high-risk group of patients who might benefit from a more aggressive adjuvant therapy to prolong their survival or, on the other hand, to protect patients of lower risk from overtreatment. There are several reasons that may account for such confusing and even contradictory results. First, some reports had an insufficient sample size. Second, different statistical methods were used and may have resulted in different conclusions. Third, different definitions of abnormal CEA levels or different methods for CEA detection were used.

Mutations of the p53 gene are the most common genetic alteration, known to occur in up to 60% of patients with colorectal cancer. Overexpression of p53 in tumor cells has been reported to indicate poor prognosis.38,39 Therefore, evaluation of the clinical relevance of p53 as a prognostic factor seems to be of importance.

Several investigators showed, by using immunohistochemistry, a significant association of positive p53 staining of the tumor tissue with worse clinical outcome.3–5 Two of them strongly suggested the complementary use of p53 protein expression and serum CEA level. Yamamura et al. proved that the combination of p53 overexpression, lymph node metastasis, venous invasion, and CEA > 5 ng/ml is an excellent prognostic indicator for recurrence.4 A strong association of positive p53 immunostaining and elevated CEA levels and a low cumulative disease-free survival has also been demonstrated.5 Other studies have used a combination of immunohistochemistry and gene expression analysis, showing that patients with p53 alterations had a worse prognosis. However, this correlation was only proven for p53 accumulation but not for p53 gene overexpression. In addition, a significant correlation with tumor stage was observed.6,7 Only p53 gene mutations seemed to be significantly associated with an occurrence of distal CRC.7 Gene analysis has shown a strong correlation between p53 gene mutation and impaired survival.8,11 However, an association between elevated p53 serum levels and clinical outcome has not always been observed.12–14 Regarding p53 levels in the serum, results comparable to our study group were seen by Shim et al.13 No stage-dependent correlation was detected between the increase of p53 in patients’ serum levels and their UICC stage. Furthermore, in our study, the presence of p53-specific IgG antibodies correlated with strong p53 staining intensity, suggesting a correlation between the intra-cellular accumulation of p53 in the tumor cells and the humoral response to p53, in accordance with previous studies.9–10 Interestingly, although in our data p53 levels in the serum had no significant influence on disease-free survival, a correlation between p53 gene and protein expression in tumor and UICC stage was found in these patients, which indicates that p53 gene mutations and p53 protein expression in the tumor are predicative biomarkers of a worse clinical outcome.

Altogether, our results indicate that the combination of CEA serum levels (> 5 ng/ml) and elevated values of CA 19-9 and p53 protein in the serum in combination with positive CEA, CA 19-9, and p53 gene and protein expression provide the most prognostic significance. All patients who independently of their UICC stage (UICC I–III) showed at least three elevated parameters (CEA, CA 19-9, and p53) in their serum and/or in the tumor exhibited recurrence/metastases during the follow-up. These data indicate that at the time of primary tumor removal, p53 and CA 19-9 may be used in combination with CEA in the serum and in the tumor for defining patients at risk for tumor recurrence. To determine whether those patients may particularly benefit from adjuvant therapies has to be analyzed in further studies.


   ACKNOWLEDGMENTS
 
This work is supported by The Deutsche Bun-desstiftung Umwelt (DBU, 16011). The authors thank Mrs. Ulrike Faber and Mrs. Helene Greb for their assistance in the preparation of this manuscript and Mrs. Andrea Trumpfheller, Mrs. Sabine Mueller, and Mrs. Mariola Dragan for their technical support.


   FOOTNOTES
 
Martin Gasser and Christiane Gerstlauer are Co-first authors

Received for publication June 4, 2006. Accepted for publication June 5, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Wiggers T, Arends JW, Schutte B, et al. A multivariate analysis of pathologic prognostic indicators in large bowel cancer. Cancer 1988; 61(2):386–95.[CrossRef][Medline]
  2. Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in human cancers. Science 1991; 253(5015):49–53.[Abstract/Free Full Text]
  3. Allegra CJ, Paik S, Colangelo LH, et al. Prognostic value of thymidylate synthase, Ki-67, and p53 in patients with Dukes’ B and C colon cancer: a National Cancer Institute-National Surgical Adjuvant Breast and Bowel Project collaborative study. J Clin Oncol 2003; 21(2):241–50.[Abstract/Free Full Text]
  4. Yamamura T, Matsuzaki H, Suda T, et al. Clinicopathological variables and p53 overexpression as a combined prognosticator for hematogenic recurrence in colorectal cancer. J Surg Oncol 1999; 70(1):1–5.[Medline]
  5. Diez M, Pollan M, Muguerza JM, et al. Time-dependency of the prognostic effect of carcinoembryonic antigen and p53 protein in colorectal adenocarcinoma. Cancer 2000; 88(1):35–41.[CrossRef][Medline]
  6. Caldes T, Iniesta P, Vega FJ, et al. Comparative survival analysis of p53 gene mutations and protein accumulation in colorectal cancer. Oncology 1998; 55(3):249–57.[CrossRef][Medline]
  7. Okubo R, Masuda H, Nemoto N. p53 mutation found to be a significant prognostic indicator in distal colorectal cancer. Oncol Rep 2001; 8(3):509–14.[Medline]
  8. Hsieh JS, Lin SR, Chang MY, et al. APC, K-ras, and p53 gene mutations in colorectal cancer patients: correlation to clinicopathologic features and postoperative surveillance. Am Surg 2005; 71(4):336–43.[Medline]
  9. Soussi T. p53 Antibodies in the sera of patients with various types of cancer: a review. Cancer Res 2000; 60:1777–88.[Abstract/Free Full Text]
  10. Zalcman G, Tredaniel J, Schlichtholz B, et al. Prognostic significance of serum p53 antibodies in patients with limited-stage small cell lung cancer. Int J Cancer 2000; 89:1–6.[CrossRef][Medline]
  11. Hamelin R, Laurent-Puig P, Olschwang S, et al. Association of p53 mutations with short survival in colorectal cancer. Gastroenterology 1994; 106(1):42–8.[Medline]
  12. Watanabe T, Wu TT, Catalano PJ, et al. Molecular predictors of survival after adjuvant chemotherapy for colon cancer. N Engl J Med 2001; 344(16):1196–206.[Abstract/Free Full Text]
  13. Shim KS, Kim KH, Park BW, et al. Increased serum levels of mutant p53 proteins in patients with colorectal cancer. J Korean Med Sci 1998; 13(1):44–8.[Medline]
  14. Adrover E, Maestro ML, Sanz-Casla MT, et al. Expression of high p53 levels in colorectal cancer: a favourable prognostic factor. Br J Cancer 1999; 81(1):122–6.[CrossRef][Medline]
  15. Goldstein MJ, Mitchell EP. Carcinoembryonic antigen in the staging and follow-up of patients with colorectal cancer. Cancer Invest 2005; 23(4):338–51.[CrossRef][Medline]
  16. Wittekind Ch, Meyer HJ, Bootz F. TNM Klassification maligner Tumoren. Sixth Edition Berlin, New York: Springer Verlag; 2002.
  17. Westra JL, Plukker JT, Buys CH, Hofstra RM. Genetic alterations in locally advanced stage II/III colon cancer: a search for prognostic markers. Clin Colorectal Cancer 2004; 4(4):252–9.[Medline]
  18. Wang WS, Lin JK, Chiou TJ, et al. CA19-9 as the most significant prognostic indicator of metastatic colorectal cancer. Hepatogastroenterology 2002; 49(43):160–4.[Medline]
  19. Wanebo HJ, Rao B, Pinsky CM, et al. Preoperative carcinoembryonic antigen level as a prognostic indicator in colorectal cancer. N Engl J Med 1978; 299(9):448–51.[Abstract]
  20. Chu DZ, Erickson CA, Russell MP, et al. Prognostic significance of carcinoembryonic antigen in colorectal carcinoma. Serum levels before and after resection and before recurrence. Arch Surg 1991; 126(3):314–6.[Abstract/Free Full Text]
  21. Wolmark N, Fisher B, Wieand HS, et al. The prognostic significance of preoperative carcinoembryonic antigen levels in colorectal cancer. Results from NSABP (National Surgical Adjuvant Breast and Bowel Project) clinical trials. Ann Surg 1984; 199(4):375–82.[Medline]
  22. Carriquiry LA, Pineyro A. Should carcinoembryonic antigen be used in the management of patients with colorectal cancer? Dis Colon Rectum 1999; 42(7):921–9.[CrossRef][Medline]
  23. Behbehani AI, Al-Sayer H, Farghaly M, et al. Prognostic significance of CEA and CA 19-9 in colorectal cancer in Kuwait. Int J Biol Markers 2000; 15(1):51–5.[Medline]
  24. Turoldo A, Balani A, Scaramucci M, et al. [Prognostic importance of preoperative CEA in patients with colorectal cancer]. Chir Ital 2002; 54(4):469–76.[Medline]
  25. Goslin R, Steele G Jr., Macintyre J, et al. The use of preoperative plasma CEA levels for the Stratification of patients after curative resection of colorectal cancers. Ann Surg 1980; 192(6):747–51.[Medline]
  26. Lewi H, Blumgart LH, Carter DC, et al. Pre-operative carcinoembryonic antigen and survival in patients with colorectal cancer. Br J Surg 1984; 71(3):206–8.[Medline]
  27. Allende T, Garcia Muniz JL, Vizoso F, et al. [Preoperative serum levels of the carcinoembryonic antigen (CEA) and prognosis in colorectal cancer]. Rev Esp Med Nucl 2001; 20(5):358–64.[Medline]
  28. Steele G Jr., Ellenberg S, Ramming K, et al. CEA monitoring among patients in multi-institutional adjuvant G.I. therapy protocols. Ann Surg 1982; 196(2):162–9.[Medline]
  29. Wang WS, Lin JK, Chiou TJ, et al. Preoperative carcinoembryonic antigen level as an independent prognostic factor in colorectal cancer: Taiwan experience. Jpn J Clin Oncol 2000; 30(1):12–6.[Abstract/Free Full Text]
  30. Moertel CG, O’Fallon JR, Go VL, et al. The preoperative carcinoembryonic antigen test in the diagnosis, staging, and prognosis of colorectal cancer. Cancer 1986; 58(3):603–10.[CrossRef][Medline]
  31. Staab HJ, Anderer FA, Brummendorf T, et al. Prognostic value of preoperative serum CEA level compared to clinical staging. I. Colorectal carcinoma. Br J Cancer 1981; 44(5):652–62.[Medline]
  32. Chen CC, Yang SH, Lin JK, et al. Is it reasonable to add preoperative serum level of CEA and CA19-9 to staging for colorectal cancer? J Surg Res 2005; 124(2):169–74.[CrossRef][Medline]
  33. Lindmark G, Bergstrom R, Pahlman L, Glimelius B. The association of preoperative serum tumour markers with Dukes’ stage and survival in colorectal cancer. Br J Cancer 1995; 71(5):1090–4.[Medline]
  34. Carpelan-Holmstrom M, Haglund C, Lundin J, et al. Pre-operative serum levels of CA 242 and CEA predict outcome in colorectal cancer. Eur J Cancer 1996; 32A(7):1156–61.[CrossRef]
  35. Forones NM, Tanaka M. CEA and CA 19-9 as prognostic indexes in colorectal cancer. Hepatogastroenterology 1999; 46(26):905–8.[Medline]
  36. Reiter W, Stieber P, Reuter C, et al. Multivariate analysis of the prognostic value of CEA and CA 19-9 serum levels in colorectal cancer. Anticancer Res 2000; 20(6D):5195–8.[Medline]
  37. Young Tabusso F, Payet Meza E, Celis Zapata J, et al. [Value of the preoperative carcinoembryonic antigen as an independent prognostic factor in cancer of the colon and rectum]. Rev Gastroenterol Peru 2002; 22(3):213–20.[Medline]
  38. Goh HS, Yao J, Smith DR. p53 point mutation and survival in colorectal cancer patients. Cancer Res 1995; 55(22):5217–21.[Abstract/Free Full Text]
  39. Yamaguchi A, Kurosaka Y, Fushida S, et al. Expression of p53 protein in colorectal cancer and its relationship to short-term prognosis. Cancer 1992; 70(12):2778–84.[CrossRef][Medline]




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
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gasser, M.
Right arrow Articles by Waaga-Gasser, A. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gasser, M.
Right arrow Articles by Waaga-Gasser, A. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS