To evaluate both the clinicopathological and prognostic significance of p53 protein, the expression of p53 protein was immunohistochemically examined by use of CM1, PAb1801, DO7, and DO1 antibodies on paraffin-embedded colorectal adenocarcinomas from 293 patients. Overexpression of the p53 protein was present in 49% of the samples studied with CM1, 18% with PAb1801, 30% with DO7, and 44% with DO7. The p53 overexpression, as detected by any of the antibodies, tended to associate with distal colorectal, nonmucinous, DNA nondiploid, and higher proliferatively active tumors (P < 0.05) but was irrespective of the patient’s gender, age, tumor growth pattern, or Dukes’ stage (P ≥ 0.05). The p53 protein that was detected by CM1 in either the nucleus (P = 0.007) or the cytoplasm (P = 0.0005) was an independent prognostic indicator. DO1 staining correlated with poor prognosis in the patients with Dukes’ B tumor (P = 0.02). However, neither PAb1801 nor DO7 staining could predict prognosis (P > 0.05). p53-positive staining by any of the antibodies predicted significantly poor prognosis compared with p53-negative reactivity (P = 0.007). These results suggest that there was essentially no difference in the significance of p53 overexpression as detected by any of the four antibodies with regard to clinicopathological variables. However, CM1 was the best antibody for predicting prognosis in this series of colorectal cancer patients.

The alteration of the p53 tumor suppressor gene is a common genetic event in colorectal cancer and as such has been strongly linked to development of the cancer (1). Primary antibodies and the immunohistochemical procedures used to detect the p53 gene product vary considerably. Furthermore the majority of the investigators in this field have used only one of the antibodies for analyzing the association of immunostaining with clinicopathological factors and prognosis, and also incorporated various numbers of colorectal cancers with different characteristics and with different criteria for staining, making the results of these studies quite controversial (Refs. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34; Table 1). It is difficult to evaluate which antibodies and immunostaining procedures are the best to detect p53 protein in predicting clinicopathological and prognostic significance.

In the present study, p53 expression was immunohistochemically detected by using CM1, PAb1801, DO7, and DO1 antibodies on formalin-fixed, paraffin-embedded archival colorectal adenocarcinomas from 293 patients with an 11- to 25-year follow-up. The aim of this study was to identify whether there were differences in the various antibodies for detecting p53 protein to predict clinicopathological significance and prognostic importance in the cancer patients.

Patients and Pathological Data.

Paraffin-embedded tissue blocks were collected from 293 patients with primary colorectal adenocarcinoma, diagnosed at the Department of Pathology, Linköping University, between 1972 and 1986. None of the patients received preoperative radiotherapy or chemotherapy. The patients were followed up from 11 to 25 years, and 134 patients died due to colorectal cancer. The patient’s gender, age, tumor location, and Dukes’ stage were taken from both surgical and pathological records. The mean age was 69 years (range, 33–93 years). Proximal tumors included tumors in the appendix, the ascending colon, and the transverse colon, and distal tumors were located on the descending colon, the sigmoid colon, and the rectum. Both the growth pattern and the histological type were classified by two investigators as recommended by Morson (35) and Hermanek (36). DNA ploidy was measured with flow cytometry, and the S-phase fraction was calculated as previously described (37). The positive cases with proliferating cell nuclear antigen previously were detected by using immunohistochemistry (38).

Antibodies.

Monoclonal antibodies PAb1801 (Oncogene Science, Inc., Manhasset, NY), DO7 (Dako, Glostrup, Denmark), DO1 (Oncogene Science), and polyclonal antibody CM1 (Novocastra Laboratories Ltd., Newcastle upon Tyne, United Kingdom) were used in this study to detect both mutant and wild-type p53 proteins. CM1, PAb1801, DO7, and DO1 recognize epitope located between amino acids 1 and 393, 32 and 79, 19 and 26, and 37 and 45, respectively.

Immunohistochemical Assay.

Serial sections taken from paraffin-embedded tissue were deparaffinized by xylene and rehydrated in graded ethanols. Endogenous peroxidase activity was blocked by treating with 1–3% H2O2 in methanol for 20 min. To carry out the immunostaining of DO7 and DO1 antibodies the slides were pretreated in a citrate buffer medium (pH 6.0) using a microwave oven at 650 W for 2 × 6 min to unmask the epitopes. Following a short rinse in PBS, the sections were incubated with 10% normal swine or rabbit serum for 20 min to block nonspecific immunostaining. After removal of the blocking solution, the primary antibodies (CM1 and PAb1801 in 1:100 for 30 min, DO7 in 1:25, and DO1 in 1.5 μg/ml for 2 h) were applied. Subsequently, the sections were incubated with swine anti-rabbit or rabbit anti-mouse immunoglobulins (Dako) for 30 min, followed by rabbit or mouse peroxidase-antiperoxidase (Dako) for an additional 30 min. The peroxidase reaction was performed by adding a 0.05% 3,3-diaminobenzidine tetrahydrochloride solution (Sigma Chemical Co., St. Louis, MO) in PBS, which contained 0.02% H2O2, for a total of 8 min. The sections were counterstained with light hematoxylin, dehydrated in ethanols, cleared in xylene, and mounted under a coverslip. The sections known to positively stain were included in each run, receiving either a primary antibody as a positive control, or the isotypes MOPC-21 for IgG1, UPC 10 for IgG2a, MOPC141 for IgG2b (Sigma), or PBS was used as negative controls.

The slides were examined and scored independently by two investigators who had no clinical or pathological information. The p53 expression was considered to be positive when the tumor cells were stained, irrespective of the percentage and intensity of the positive cells. The positive cells that were located either on the margins of sections or in the poor morphological areas were not counted.

Statistical Analysis.

The χ2 test (39) was used to test the association of p53 expression with other factors. Cox’s proportional hazards model (40) was used to estimate and test the relationship between p53 expression and prognosis. The survival curves were computed according to the method of Kaplan and Meier (41). All p values that are cited are two-sided.

p53 Expression Detected by CM1, PAb1801, DO7, and DO1 Antibodies.

The percentage of p53 expression that was detected by the antibodies CM1, PAb1801, DO7, and DO1 in colorectal cancer are presented in Table 2. All four of the antibodies gave a distinct nuclear staining in the tumor cells, while CM1 also gave cytoplasmic staining and staining in both the nucleus and the cytoplasm (Fig. 1). The frequency of occurrence of positive staining in CM1 was the highest (49%) while that in PAb1801 was the lowest (18%). In 150 cases p53 overexpression was displayed (61%) with at least one of the antibodies (57 cases stained with one antibody, 29 with two, 30 with three, and 34 with four), and 95 (39%) were negative with all four of the antibodies. The distribution of cases that were stained with any two of the antibodies is shown in Fig. 2. The most concordant immunoreactivity was between PAb1801 and DO7 (215 versus 49 cases), and the least was between CM1 and PAb1801 (193 versus 96 cases). There was heterogeneity in the percentage of reactive cells and in the intensity of staining between cases and cells of the same case.

p53 Expression in Relation to Clinicopathological and Other Variables.

Table 3 displays the relationships between p53 overexpression and clinicopathological factors. p53 positive staining, as detected by any of the antibodies, tended to be more frequently seen in distal tumor, nonmucinous carcinoma (except CM1 staining), DNA nondiploid, and higher proliferative active (S-phase fraction and proliferating cell nuclear antigen) tumors. No significant relationships were found between the p53 expression and either gender, tumor location, Dukes’ stage, or growth pattern (P ≥ 0.05).

p53 Expression in Relation to Survival Rate.

The relationship between p53 overexpression and survival in the patients with Dukes’ stage A–C tumor was analyzed. CM1 staining, either in the nucleus irrespective of cytoplasmic expression (P = 0.007) or in the cytoplasm irrespective of nuclear expression (P = 0.0005), was a strong prognostic factor independent of gender, age, tumor location, Dukes’ stage, growth pattern, and differentiation. DO1 staining predicted prognosis only in the patients with Dukes’ B tumors (P = 0.02). However, neither PAb1801 nor DO1 staining was related to survival in either the entire series of patients or in the subgroups containing either different tumor location or Dukes’ stage (P > 0.05).

In univariate Cox analysis, p53 positive staining by any of the antibodies predicted significantly poor prognosis when compared to p53-negative reactivity (P = 0.007; Fig. 3). In multiple analyses that included gender, age, tumor location, Dukes’ stage, growth pattern, and differentiation, p53-positive staining reached a statistical borderline of prognosis (P = 0.08). The patients were further categorized into three groups based on the p53 staining: negative; staining with any three of the antibodies; and staining with all four of the antibodies. The latter two groups of the patients had similar 5-year survival rates (57 and 52%, respectively), which were significantly lower than that of the negative group (77%, P = 0.02).

The majority of the previous studies on p53 expression in colorectal cancer were performed in different series of cases (including various numbers of cases, tumor locations, stages, differentiation, pre- or posttreatments, and follow-up periods); these studies were also carried out using different antibodies and various immunohistochemical procedures (including a variety of fixations, detecting systems, and criteria of the immunostaining evaluation). These previous results between p53 overexpression and clinicopathological variables or prognosis are controversial (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, Table 1). Therefore, it is difficult to evaluate which antibody is the most valuable for predicting clinicopathological and prognostic significance. To gain more knowledge about this matter, we analyzed the four most commonly used antibodies in the same series of formalin-fixed, paraffin-embedded archival tissue from colorectal cancer patients with a long follow-up period. We revealed several variations in the occurrence of the p53 overexpression that was detected between these four different antibodies. The differences were mainly dependent on the primary antibodies and the microwave pretreatment. Since various p53 mutations in different tumors can result in the formation of a changed or truncated protein, or in a p53 protein which has altered its conformation by binding to other proteins or macromolecules, or because of changed temperature (42, 43), the antibodies that recognize different regions of p53 protein would not detect all of the proteins. Our data clearly show that CM1 had the highest frequency of p53 protein staining, following by DO1, DO7, and PAb1801. This is probably due to the fact that CM1 recognizes the whole segment of p53 protein (amino acids 1–393). Both DO7 (amino acids 19–26) and DO1 (amino acids 37–45) bind shorter segments of the protein, but since the sections were pretreated by microwave that enhanced immunolabeling (44), they provided more staining frequency than the PAb1801. Although the PAb1801 is capable of recognizing a longer segment, between amino acids 32 and 79, the sections did not received microwave irradiation, thus resulting in less positive staining. The frequencies of p53 overexpression in other studies were between 25 and 69% in CM1, 49 and 61% in DO1, 30 and 72% in DO7, and 44 and 72% in PAb1801 (Refs. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34; Table 1). The frequencies of p53 positivity observed in our study were slightly less than that found in other studies. One probable reason for this is that the positive stained cells in the areas with poor morphology and in the margin of sections were not counted.

In the present study, frequencies of p53 overexpression detected by different antibodies were different but the frequencies determined by any two of the antibodies were mainly identical, as similar reports by others (45, 46, 47). Genetic analyses have shown that the majority of p53 mutations reported thus far in colorectal tumors are clustered within exons 5–8 and are located in four highly conserved domains, although the mutations may occur throughout the whole gene (1, 48). It is tempting to speculate that most of the tumor cells presented similar mutations and conformations of p53 proteins, which can be recognized by most of the antibodies.

In this study, p53 overexpression detected by any of the four antibodies was more common in the left colon and rectum, nonmucinous carcinomas, DNA nondiploid, and higher proliferative active tumors, which is a trend similar to that found in many previous reports (2, 3, 4, 6, 11, 18, 21, 24, 25, 26, 29, 31, 32). These findings further supported the fact that p53 protein is involved in the initiation and progression of left and rectal cancers, cellular differentiation, and proliferative activity.

We have demonstrated that CM1 staining, either in the nucleus or in the cytoplasm, and DO1 staining in Dukes’ B tumor was very indicative of a strongly unfavorable prognosis in the cancer patients. To the best of our knowledge, there are only two other groups who have used the CM1 antibody to study the prognostic significance of cytoplasmic staining in colorectal cancer patients, and the authors also found that cytoplasmic staining is an effective prognostic factor (4, 7). Three other groups show the prognostic significance of nuclear staining with CM1 (2, 3, 5) although the results conflicted with those of other authors (4, 6, 7, 8). It is apparent that DO1 is not a commonly used antibody compared with the other three antibodies. However, analysis of DO1 staining in stages IB and II of colorectal cancer reveal the prognostic significance of the p53 protein (32). When taken together with our results of DO1 staining, it suggests that DO1 staining provided a prognostic value in the early stages of the tumors. A review of studies incorporated PAb1801 staining indicates that it was widely used in colorectal cancers, but its prognostic significance was limited (Refs. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34; Table 1). We were unable to find that PAb1801 was related to prognosis in this series and another separate series of 163 colorectal cancers from our laboratory (49). However, DO7 antibody was better at evaluating survival rates in colorectal cancer patients in other studies (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34), but we were not able to conclusively prove this.

In the present study, no differences were found between the four antibodies for determination of the clinicopathological values including gender, age, tumor location, Dukes’ stage, growth pattern, and histological type. However, CM1 seems to be the best antibody to detect p53 protein, which may be used as a strong prognostic factor. Therefore, it is necessary to select primary antibody and optimize procedures to provide more information, which can be used for refining prognostic indicators to evaluate tumor aggressiveness and select patients for adjuvant therapies.

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.

        
1

This study was supported by grants from the Cancer Foundation and FORSS, Sweden.

Fig. 1.

p53 positive expression is immunohistochemically detected by four antibodies, CM1 (A), PAb1801 (B), DO7 (C), and DO1 (D) in series of sections from the same tumor. Original magnification, ×200.

Fig. 1.

p53 positive expression is immunohistochemically detected by four antibodies, CM1 (A), PAb1801 (B), DO7 (C), and DO1 (D) in series of sections from the same tumor. Original magnification, ×200.

Close modal
Fig. 2.

Comparison of p53 immunoreactivity stained with any two of the four antibodies. ▪, tumors with identical immunoreactivity either positive or negative with both of the two antibodies; □, tumors stained with only one of the two antibodies.

Fig. 2.

Comparison of p53 immunoreactivity stained with any two of the four antibodies. ▪, tumors with identical immunoreactivity either positive or negative with both of the two antibodies; □, tumors stained with only one of the two antibodies.

Close modal
Fig. 3.

Kaplan-Meier survival curves for the patients with colorectal adenocarcinoma. The survival rate in the patients with p53-positive tumor, stained by any one of the four studied antibodies, was significantly lower than that in the patients with p53-negative staining tumor (P = 0.007).

Fig. 3.

Kaplan-Meier survival curves for the patients with colorectal adenocarcinoma. The survival rate in the patients with p53-positive tumor, stained by any one of the four studied antibodies, was significantly lower than that in the patients with p53-negative staining tumor (P = 0.007).

Close modal
Table 1

Review of relationships of p53 expression with clinicopathological variables and prognosis in colorectal cancer

AuthorNo.Positive(%)SexAgeSiteStageGrowth patternGrade/typeDNA ploidyProliferative activityPrognosis
CM1            
 Starzyska et al. (2) (1992) 107 46 a − − 
 Auvinen et al. (3) (1994) 144 69 − − 
 Bosari et al. (4) (1994) 197 33 (n) − − − (n) − (n) − 
  50 (c) − − + (c) − (c) + 
 Diez et al. (5) (1995) 61 44 − − 
 Tomoda and Kakéji (6) (1995) 144 25 − − − − − − 
 Flamini et al. (7) (1996) 96 28 (n) − − − − − (n) 
 Langlois et al. (8) (1997) 74 36 − − 
 Starzyska et al. (9) (1997) 102 46 − − − − − − 
PAb1801            
 Purdie et al. (10) (1991) 86 47 − − − − − 
 Yamaguchi et al. (11) (1992) 100 61 − − 
 Yamaguchi et al. (12) (1993) 203 60 − − − − 
 Bosari et al. (4) (1994) 206 46 − − − 
 Darmon et al. (13) (1994) 34 65 − − − − 
 Nathanson et al. (14) (1994) 84 62 − − − 
 Suzuki et al. (15) (1994) 67 51 − − − − − 
 Zeng et al. (16) (1994) 107 47 (n) − − − − − + (n) 
  57 (c) − (c) 
 Grewal et al. (17) (1995) 66 52 − − − − 
 Lazaris et al. (18) (1995) 60 47 − − 
 Belluco et al. (19) (1996) 50 72 − − − − − 
 Fante et al. (20) (1996) 150 47 − − − − − − − 
 Paradiso et al. (21) (1996) 71 44 − − − − − 
 Smith et al. (22) (1996) 100 62 − − − − − 
DO7            
 Dix et al. (23) (1994) 100 64 
 Kim et al. (24) (1995) 471 45 − − − − 
 Mulder et al. (25) (1995) 109 72 − − − − 
 Flamini et al. (7) (1996) 96 28 (n) − − − − − (n) 
  19 (c) − − − − + (c) 
 Lanza et al. (26) (1996) 204 61 − − − − − − − 
 Leahy et al. (27) (1996) 66 47–52 − − − − 
 Pereira et al. (28) (1997) 80 30 − − − 
 Soong et al. (29) (1997) 556 30 − − − 
 Kaklamanis et al. (30) (1998) 224 64 − − − − − 
 Kressner et al. (31) (1999) 190 48 − − − − − 
DO1            
 Baretton et al. (32) (1996) 101 59 − − − − − 
 Brett et al. (33) (1996) 59 49 − 
 Poller et al. (34) (1997) 250 61 − − − − − − − 
AuthorNo.Positive(%)SexAgeSiteStageGrowth patternGrade/typeDNA ploidyProliferative activityPrognosis
CM1            
 Starzyska et al. (2) (1992) 107 46 a − − 
 Auvinen et al. (3) (1994) 144 69 − − 
 Bosari et al. (4) (1994) 197 33 (n) − − − (n) − (n) − 
  50 (c) − − + (c) − (c) + 
 Diez et al. (5) (1995) 61 44 − − 
 Tomoda and Kakéji (6) (1995) 144 25 − − − − − − 
 Flamini et al. (7) (1996) 96 28 (n) − − − − − (n) 
 Langlois et al. (8) (1997) 74 36 − − 
 Starzyska et al. (9) (1997) 102 46 − − − − − − 
PAb1801            
 Purdie et al. (10) (1991) 86 47 − − − − − 
 Yamaguchi et al. (11) (1992) 100 61 − − 
 Yamaguchi et al. (12) (1993) 203 60 − − − − 
 Bosari et al. (4) (1994) 206 46 − − − 
 Darmon et al. (13) (1994) 34 65 − − − − 
 Nathanson et al. (14) (1994) 84 62 − − − 
 Suzuki et al. (15) (1994) 67 51 − − − − − 
 Zeng et al. (16) (1994) 107 47 (n) − − − − − + (n) 
  57 (c) − (c) 
 Grewal et al. (17) (1995) 66 52 − − − − 
 Lazaris et al. (18) (1995) 60 47 − − 
 Belluco et al. (19) (1996) 50 72 − − − − − 
 Fante et al. (20) (1996) 150 47 − − − − − − − 
 Paradiso et al. (21) (1996) 71 44 − − − − − 
 Smith et al. (22) (1996) 100 62 − − − − − 
DO7            
 Dix et al. (23) (1994) 100 64 
 Kim et al. (24) (1995) 471 45 − − − − 
 Mulder et al. (25) (1995) 109 72 − − − − 
 Flamini et al. (7) (1996) 96 28 (n) − − − − − (n) 
  19 (c) − − − − + (c) 
 Lanza et al. (26) (1996) 204 61 − − − − − − − 
 Leahy et al. (27) (1996) 66 47–52 − − − − 
 Pereira et al. (28) (1997) 80 30 − − − 
 Soong et al. (29) (1997) 556 30 − − − 
 Kaklamanis et al. (30) (1998) 224 64 − − − − − 
 Kressner et al. (31) (1999) 190 48 − − − − − 
DO1            
 Baretton et al. (32) (1996) 101 59 − − − − − 
 Brett et al. (33) (1996) 59 49 − 
 Poller et al. (34) (1997) 250 61 − − − − − − − 
a

+, significance; −, no significance; x, no data; n, nuclear staining; c, cytoplasmic staining.

Table 2

Expression of p53 with antibodies CM1, PAb1801, DO7, and DO1 in colorectal adenocarcinoma

p53 expression
AntibodyNo.NegativePositive
CM1 293 149 (51)a 144 (49) 
PAb1801 289 238 (82) 51 (18) 
DO7 265 186 (70) 79 (30) 
DO1 257 144 (56) 113 (44) 
Four 245 95 (39) 150 (61) 
p53 expression
AntibodyNo.NegativePositive
CM1 293 149 (51)a 144 (49) 
PAb1801 289 238 (82) 51 (18) 
DO7 265 186 (70) 79 (30) 
DO1 257 144 (56) 113 (44) 
Four 245 95 (39) 150 (61) 
a

Numbers in parentheses, percentage.

Table 3

Expression of p53 in relation to patient’s gender, age, tumor location, growth pattern, Dukes’ stage, histological type, DNA ploidy,S-phase fraction, and proliferating cell nuclear antigen in colorectal adenocarcinoma

% showing p53 expression
CategoryCM1 (n = 293)1801 (n = 289)DO7 (n = 265)DO1 (n = 257)
Gender P = 0.68 P = 0.16 P = 0.31 P = 0.87 
 Male 40 21 27 43 
 Female 38 15 32 44 
Age (yr) P = 0.19 P = 0.09 P = 0.25 P = 0.12 
 <70 35 14 26 39 
 ≥70 42 21 33 49 
Location P = 0.2 P = 0.02 P = 0.02 P = 0.35 
 Proximal 36 14 25 43 
 Distal 43 25 38 49 
Dukes’ stage P = 0.41 P = 0.05 P = 0.46 P = 0.75 
 A 53 37 45 57 
 B 33 11 22 37 
 C 40 18 35 42 
 D 38 14 25 49 
Growth pattern P = 0.47 P = 0.97 P = 0.9 P = 0.48 
 Expanding 33 16 30 40 
 Infiltrative 38 16 29 45 
Histological type P = 0.55 P = 0.02 P = 0.02 P = 0.02 
 Nonmucinous 38 19 32 47 
 Mucinous 44 10 10 23 
DNA P = 0.06 P = 0.06 P = 0.001 P = 0.002 
 Diploid 33 13 20 34 
 Tetraploid 29 10 35 39 
 Aneuploid 45 21 39 54 
S-phase fraction P = 0.13 P = 0.55 P = 0.002 P = 0.02 
 <10% 31 13 21 35 
 ≥10% 41 16 41 52 
Proliferating cell nuclear antigen P = 0.006 P = 0.002 P = 0.0005 P = 0.11 
 Negative 28 17 37 
 Positive 45 23 37 48 
% showing p53 expression
CategoryCM1 (n = 293)1801 (n = 289)DO7 (n = 265)DO1 (n = 257)
Gender P = 0.68 P = 0.16 P = 0.31 P = 0.87 
 Male 40 21 27 43 
 Female 38 15 32 44 
Age (yr) P = 0.19 P = 0.09 P = 0.25 P = 0.12 
 <70 35 14 26 39 
 ≥70 42 21 33 49 
Location P = 0.2 P = 0.02 P = 0.02 P = 0.35 
 Proximal 36 14 25 43 
 Distal 43 25 38 49 
Dukes’ stage P = 0.41 P = 0.05 P = 0.46 P = 0.75 
 A 53 37 45 57 
 B 33 11 22 37 
 C 40 18 35 42 
 D 38 14 25 49 
Growth pattern P = 0.47 P = 0.97 P = 0.9 P = 0.48 
 Expanding 33 16 30 40 
 Infiltrative 38 16 29 45 
Histological type P = 0.55 P = 0.02 P = 0.02 P = 0.02 
 Nonmucinous 38 19 32 47 
 Mucinous 44 10 10 23 
DNA P = 0.06 P = 0.06 P = 0.001 P = 0.002 
 Diploid 33 13 20 34 
 Tetraploid 29 10 35 39 
 Aneuploid 45 21 39 54 
S-phase fraction P = 0.13 P = 0.55 P = 0.002 P = 0.02 
 <10% 31 13 21 35 
 ≥10% 41 16 41 52 
Proliferating cell nuclear antigen P = 0.006 P = 0.002 P = 0.0005 P = 0.11 
 Negative 28 17 37 
 Positive 45 23 37 48 
1
Hollstein M., Sidransky D., Vogelstein B., Harris C. C. p53 mutations in human cancers.
Science (Washington DC)
,
253
:
49
-53,  
1991
.
2
Starzynska T., Bromley M., Ghosh A., Stern P. L. Prognostic significance of p53 overexpression in gastric and colorectal carcinoma.
Br. J. Cancer
,
66
:
558
-562,  
1992
.
3
Auvinen A., Isola J., Visakorpi T., Koivula T., Virtanen S., Hakama M. Overexpression of p53 and long-term survival in colon carcinoma.
Br. J. Cancer
,
70
:
293
-296,  
1994
.
4
Bosari S., Viale G., Bossi P., Maggioni M., Coggi G., Murray J. J., Lee A. K. Cytoplasmic accumulation of p53 protein: an independent prognostic indicator in colorectal adenocarcinomas.
J. Natl. Cancer Inst.
,
86
:
681
-687,  
1994
.
5
Diez M., Enriquez J. M., Camunas J., Gonzalez A., Gutierrez A., Muguerza J. M., Ruiz A., Granell J. Prediction of recurrence in B—C stages of colorectal cancer by p53 nuclear overexpression in comparison with standard pathological features.
Eur. J. Surg. Oncol.
,
21
:
635
-639,  
1995
.
6
Tomoda H., Kakeji Y. Immunohistochemical analysis of p53 in colorectal cancer regarding clinicopathological correlation and prognostic significance.
J. Surg. Oncol.
,
58
:
125
-128,  
1995
.
7
Flamini G., Curigliano G., Ratto C., Astone A., Ferretti G., Nucera. P., Sofo L., Sgambato A., Boninsegna A., Crucitti F., Cittadini A. Prognostic significance of cytoplasmic p53 overexpression in colorectal cancer: an immunohistochemical analysis.
Eur. J. Cancer
,
32
:
802
-806,  
1996
.
8
Langlois N. E., Lamb J., Eremin O., Heys S. D. Apoptosis in colorectal carcinoma occurring in patients aged 45 years and under: relationship to prognosis, mitosis, and immunohistochemical demonstration of p53, c-myc and bcl-2 protein products.
J. Pathol.
,
182
:
392
-397,  
1997
.
9
Starzynska T., Bromley M., Marlicz K., Roberts S. A., Ucinski M., Stern P. L. Accumulation of p53 in relation to long-term prognosis in colorectal carcinoma.
Eur. J. Gastroenterol. Hepatol.
,
9
:
183
-186,  
1997
.
10
Purdie C. A., O’Grady J., Piris J., Wyllie A. H., Bird C. C. p53 expression in colorectal tumors.
Am. J. Pathol.
,
138
:
807
-813,  
1991
.
11
Yamaguchi A., Kurosaka Y., Fushida S., Kanno M., Yonemura Y., Miwa K., Miyazaki I. Expression of p53 protein in colorectal cancer and its relationship to short-term prognosis.
Cancer (Phila.)
,
70
:
2778
-2784,  
1992
.
12
Yamaguchi A., Nakagawara G., Kurosaka Y., Nishimura G., Yonemura Y., Miyazaki I. p53 immunoreaction in endoscopic biopsy specimens of colorectal cancer, and its prognostic significance.
Br. J. Cancer
,
68
:
399
-402,  
1993
.
13
Darmon E., Cleary K. R., Wargovich M. J. Immunohistochemical analysis of p53 overexpression in human colonic tumors.
Cancer Detect. Prev.
,
18
:
187
-195,  
1994
.
14
Nathanson S. D., Linden M. D., Tender P., Zarbo R. J., Jacobsen G., Nelson L. T. Relationship among p53, stage, and prognosis of large bowel cancer.
Dis. Colon Rectum
,
37
:
527
-534,  
1994
.
15
Suzuki H., Matsumoto K., Koide A., Tada T., Fujino I., Okuda A., Shigemori C. Correlation of p53 with the clinicopathologic features and prognosis of colorectal adenocarcinoma.
Surg. Today
,
24
:
85
-87,  
1994
.
16
Zeng Z. S., Sarkis A. S., Zhang Z. F., Klimstra D. S., Charytonowicz E., Guillem J. G., Cordon-Cardo C., Cohen A. M. p53 nuclear overexpression: an independent predictor of survival in lymph node-positive colorectal cancer patients.
J. Clin. Oncol.
,
12
:
2043
-2050,  
1994
.
17
Grewal H., Guillem J. G., Klimstra D. S., Cohen A. M. p53 nuclear overexpression may not be an independent prognostic marker in early colorectal cancer.
Dis. Colon Rectum
,
38
:
1176
-1181,  
1995
.
18
Lazaris A. C., Theodoropoulos G. E., Anastassopoulos P., Nakopoulou L., Panoussopoulos D., Papadimitriou K. Prognostic significance of p53 and c-erbB-2 immunohistochemical evaluation in colorectal adenocarcinoma.
Histol. Histopathol.
,
10
:
661
-668,  
1995
.
19
Belluco C., Guillem J. G., Kemeny N., Huang Y., Klimstra D., Berger M. F., Cohen A. M. p53 nuclear protein overexpression in colorectal cancer: a dominantpredictor of survival in patients with advanced hepatic metastases.
J. Clin. Oncol.
,
14
:
2696
-2701,  
1996
.
20
Fante R., Di Gregorio C., Losi L., Roncucci L., Ponz, de Leon M. Clinico-pathological correlation and prognostic significance of nuclear p53 protein in colorectal cancer: Colorectal Cancer Study Group of the University and Health Care District of Modena.
Ital. J. Gastroenterol.
,
28
:
205
-210,  
1996
.
21
Paradiso A., Rabinovich M., Vallejo C., Machiavelli M., Romero A., Perez J., Lacava J., Cuevas M. A., Rodriquez R., Leone B., Sapia M. G., Simone G., De Lena M. p53, and PCNA expression in advanced colorectal cancer: response to chemotherapy and long-term prognosis.
Int. J. Cancer
,
69
:
437
-441,  
1996
.
22
Smith D. R., Ji C. Y., Goh H. S. Prognostic significance of p53 overexpression and mutation in colorectal adenocarcinomas.
Br. J. Cancer
,
74
:
216
-223,  
1996
.
23
Dix B. R., Robbins P., Soong R., Jenner D., House A. K., Iacopetta B. J. The common molecular genetic alterations in Dukes’ B and C colorectal carcinomas are not short-term prognostic indicators of survival.
Int. J. Cancer
,
59
:
747
-751,  
1994
.
24
Kim Y. W., Lee S., Park J. H., Yoon T. Y., Park Y. K., Yang M. H. An immunohistochemical study of the expression of p53 protein in colon cancer.
J. Korean Med. Sci.
,
10
:
176
-182,  
1995
.
25
Mulder J. W., Baas I. O., Polak M. M., Goodman S. N., Offerhaus G. J. Evaluation of p53 protein expression as a marker for long-term prognosis in colorectal carcinoma.
Br. J. Cancer
,
71
:
1257
-1262,  
1995
.
26
Lanza G., Jr., Maestri I., Dubini A., Gafa R., Santini A., Ferretti S., Cavazzini L. p53 expression in colorectal cancer: relation to tumor type, DNA ploidy pattern and short-term survival.
Am. J. Clin. Pathol.
,
105
:
604
-612,  
1996
.
27
Leahy D. T., Salman R., Mulcahy H., Sheahan K., O’Donoghue D. P., Parfrey N. A. Prognostic significance of p53 abnormalities in colorectal carcinoma detected by PCR-SSCP and immunohistochemical analysis.
J. Pathol.
,
180
:
364
-370,  
1996
.
28
Pereira H., Silva S., Juliao R., Garcia P., Perpetua F. Prognostic markers for colorectal cancer: expression of P53 and BCL2.
World J. Surg.
,
21
:
210
-213,  
1997
.
29
Soong R., Grieu F., Robbins P., Dix B., Chen D., Parsons R., House A., Iacopetta B. p53 alterations are associated with improved prognosis in distal colonic carcinomas.
Clin. Cancer Res.
,
3
:
1405
-1411,  
1997
.
30
Kaklamanis L., Savage A., Whitehouse R., Doussis-Anagnostopoulou I., Biddolph S., Tsiotos P., Mortensen N., Gatter K. C., Harris A. L. Bcl-2 protein expression: association with p53 and prognosis in colorectal cancer.
Br. J. Cancer
,
77
:
1864
-1869,  
1998
.
31
Kressner U., Inganas M., Byding S., Blikstad I., Pahlman L., Glimelius B., Lindmark G. Prognostic value of p53 genetic changes in colorectal cancer.
J. Clin. Oncol.
,
17
:
593
-599,  
1999
.
32
Baretton G. B., Vogt M., Muller C., Diebold J., Schneiderbanger K., Schmidt M., Lohrs U. Prognostic significance of p53 expression, chromosome 17 copy number, and DNA ploidy in non-metastasized colorectal carcinomas (stages IB and II).
Scand. J. Gastroenterol.
,
31
:
481
-489,  
1996
.
33
Brett M. C., Pickard M., Green B., Howel-Evans A., Smith D., Kinsella A., Poston G. p53 protein overexpression and response to biomodulated 5-fluorouracil chemotherapy in patients with advanced colorectal cancer.
Eur. J. Surg. Oncol.
,
22
:
182
-185,  
1996
.
34
Poller D. N., Baxter K. J., Shepherd N. A. p53 and Rb1 protein expression: are they prognostically useful in colorectal cancer? Br.
J. Cancer
,
75
:
87
-93,  
1997
.
35
Morson B. C. Colour atlas of gastrointestinal pathology Curran R. C. eds. .
Oxford Colour Atlases of Pathology
,
: Oxford University Press Oxford, United Kingdom  
1988
.
36
Hermanek P. Colorectal carcinoma: histopathological diagnosis and staging.
Baillières Clin. Gastroenterol.
,
3
:
511
-529,  
1989
.
37
Sun X-F., Carstensen J. M., Stål O., Wingren S., Hatschek T., Zhang H., Nordenskjöld B. Prognostic significance of DNA ploidy and S-phase fraction in colorectal adenocarcinoma.
Diagn. Oncol.
,
3
:
121
-126,  
1993
.
38
Sun X. F., Carstensen J. M., Stal O., Zhang H., Nordenskjold B. Proliferating cell nuclear antigen (PCNA) in relation to ras, c-erbB-2, p53, clinico-pathological variables and prognosis in colorectal adenocarcinoma.
Int. J. Cancer
,
69
:
5
-8,  
1996
.
39
Armitage P., Berry G. Statistical Methods in Medical Research Blackwell Scientific Publications Oxford, United Kingdom  
1987
.
40
Cox D. R. Regression models and life tables.
J. R. Statist. Soc. B.
,
34
:
187
-220,  
1972
.
41
Kaplan E., Meier P. Nonparametric estimation from incomplete observations.
J. Am. Statist. Assoc.
,
53
:
457
-481,  
1958
.
42
Baas I. O., Mulder J-W. R., Offerhaus J. A. An evaluation of six antibodies for immunohistochemistry of mutant p53 gene product in archival colorectal neoplasms.
J. Pathol.
,
172
:
5
-12,  
1994
.
43
Klotzsche O., Etzrodt D., Hohenberg H., Bohn W., Deppert W. Cytoplasmic retention of mutant tsp53 is dependent on an intermediate filament protein (vimentin) scaffold.
Oncogene
,
16
:
3423
-3434,  
1998
.
44
Shi S-R., Key E. M., Kalra L. K. Antigen retrieval in formalin-fixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections.
J. Histochem. Cytochem.
,
39
:
741
-748,  
1991
.
45
Kressner U., Lindmark G., Gerdin B., Pahlman L., Glimelius B. Immunohistological p53 staining is of limited value in the staging and prognostic prediction of colorectal cancer.
Anticancer Res.
,
16
:
951
-957,  
1996
.
46
Caldes T., Iniesta P., Vega F. J., de Juan C., Lopez J. A., Diaz-Rubio E., Fernandez C., Cerdan J., Balibrea J. L., Benito M. Comparative survival analysis of p53 gene mutations and protein accumulation in colorectal cancer.
Oncology
,
55
:
249
-257,  
1998
.
47
Nitti D., Belluco C., Montesco M. C., Bertorelle R., Da Pian P. P., Fassina A., Ninfo V., Chieco-Bianchi L., Lise M. Nuclear p53 protein expression in resected hepatic metastases from colorectal cancer: an independent prognostic factor of survival.
Eur. J. Cancer
,
34
:
851
-855,  
1998
.
48
Frebourg T., Friend S. H. Cancer risks from germline p53 mutations.
J. Clin. Invest.
,
90
:
1637
-1641,  
1992
.
49
Adell G., Stål O., Klintenberg C., Sjödahl R., Norrdenskjöld B., Sun X-F. p53 status: an indicator for the effect of preoperative radiotherapy of rectal cancer.
Radiother. Oncol.
,
51
:
169
-174,  
1999
.