| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
Department of Pathology [U. M., W. E. G.] and Medical Statistics Section, Department of Medicine [H. L. W.], University of Alabama at Birmingham, Birmingham, Alabama 35294
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSMultivariate Survival AnalysesDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSMultivariate Survival AnalysesDISCUSSIONREFERENCES |
|---|
Differential expressions of MUC1 and MUC2 have been reported in colorectal adenomas (4, 5, 6) and CRCs2 (4 , 7 , 8) . Several earlier studies in colorectal neoplasia have demonstrated that higher levels of expression of MUC1 (9, 10, 11) and/or MUC2 (12, 13, 14) in CRCs were correlated with increased incidence of regional lymph node metastasis and liver metastasis.
A few studies, in colorectal neoplasia (11 , 15) as well as in other human malignancies (16 , 17) , have suggested that increased expression of the core peptide of MUC1 is associated with poor prognosis. The role of MUC2 in predicting the clinical outcome is more controversial. On the basis of studies in colorectal neoplasia (12, 13, 14) , increased expression of MUC2 was considered likely to be a predictor of poor patient survival; however, in studies of pancreatic adenocarcinomas (18 , 19) , biliary carcinomas (19 , 20) , and gastric carcinomas (21) , increased MUC2 expression was associated with better patient prognosis.
New prognostic biomarkers are needed to characterize the aggressiveness of tumors, because the most commonly used indicators of prognosis, the grades and pathological stages of tumors, do not predict adequately either the clinical course of most malignancies or the biological characteristics of specific tumors. Although expression of MUC1 and MUC2 has been associated with the aggressiveness of CRCs, their prognostic usefulness in colorectal neoplasia has not been evaluated adequately. In addition, several recent studies have indicated that the ethnicity of the patient population should be considered in the evaluation of prognostic significance of molecular markers in colorectal neoplasia (22 , 23) . Therefore, we evaluated the prognostic usefulness of expression of MUC1 and MUC2 in CRCs collected from African-Americans and Caucasians, the two major ethnic patient populations of the United States. The combination of biomarkers has been reported to be useful prognostically in colorectal neoplasia (15 , 24) . The archival tissues used in this study were analyzed previously for nuclear accumulation of p53 (p53nac); therefore, we also correlated the concomitant expression of MUC1 or MUC2 and p53nac with patient survival.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSMultivariate Survival AnalysesDISCUSSIONREFERENCES |
|---|
Immunohistochemistry.
Selected paraffin blocks representative of both tumor and normal tissue
of each case were sectioned at a 5-µm thickness and mounted on
Superfrost/Plus slides (Fisher Scientific, Pittsburgh, PA). The
immunohistochemical staining procedure was carried out as described
earlier (22
, 24)
. Expression of MUC1 and MUC2 was detected
using the anti-MUC1 monoclonal antibody, DF3 (final dilution, 1:200;
Novocastra Laboratories, Ltd., Newcastle upon Tyne, United Kingdom),
and anti-MUC2 monoclonal antibody, Ccp58 (final dilution, 1:200.
Novocastra) after antigen recovery by boiling tissue sections in
citrate buffer for 5 min. The remainder of the staining procedure and
the protocol for the antigen recovery were described in detail
elsewhere (24)
. In contrast, antigen recovery techniques
are not recommended for the detection of p53nac
in CRCs to evaluate its prognostic importance (24)
.
Assessment of MUC1 and MUC2 Expression and p53nac.
The assessment of MUC1 and MUC2 expression and p53nac was
performed by two authors (U. M. and W. E. G.) to limit the bias. A
semiquantitative ISS for MUC1 and MUC2 was obtained as described
previously (24
, 25) . Immunostaining intensity of
individual cells was scored on a scale of 0 (no staining) to +4
(strongest staining), and each observer estimated the proportion of
cells stained at each intensity. The immunostaining scores of the two
authors were combined to obtain an average ISS. A cutoff value of the
ISS of 
0.5 plus at least 25% of tumor cells immunostaining were
required to classify a tumor as positive for expression of either the
MUC1 or MUC2 antigens.
The colorectal tumors were classified as positive for p53nac if malignant cells demonstrated p53nac in 10% or more nuclei without antigen recovery (22 , 24) .
Statistical Analysis.
The 
2
test (26)
was used to
assess the univariate association of baseline characteristics with MUC1
and MUC2 expression. The possible combinations of MUC1 or MUC2
expression with p53nac were evaluated in four
subgroups of tumors. The period from the date of resection to the date
of death or last contact (if alive) was used for survival analysis.
Outcome analysis was based on patients who were alive or had died of
CRC as described previously (22)
. The log-rank test was
used to compare Kaplan-Meier survival curves based on status of
expression of MUC1 or MUC2 and/or p53nac
(27)
. A Cox proportional hazards model (28)
was used to compare the survival of patients with and without MUC1 or
MUC2 expression after adjusting for the status of
p53nac and other confounding variables. The
clinical confounding variables of CRC used in the analyses were pT, pN,
and M components of Tumor-Node-Metastasis stage, age, sex, tumor
location, tumor size, and differentiation. A stepwise model-building
procedure was used to determine the significant factors in predicting
patient survival with CRC. Hazard ratios and 95% confidence intervals
were calculated to identify the risk factors. The above analyses were
performed separately for African-American and Caucasian patients. All
tests were two-sided, and P < 0.05 was considered
statistically significant.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSMultivariate Survival AnalysesDISCUSSIONREFERENCES |
|---|
. The mean age at surgery (67.2 ±
13.0 versus 64.3 ± 12.3), the proportion of patients
alive at the last follow-up (18 of 58, 31% versus 33 of
108, 31%), the stage of the disease at diagnosis, and the median
survival of patients, based on Kaplan-Meier survival analysis (11.37
years versus 12.03 years), were similar in African-Americans
and Caucasians. In African-Americans, 53% (31 of 58) of tumors
were located in the proximal colorectum, and 47% (27 of 58) of tumors
were located in the distal colorectum, whereas in Caucasians, the
respective distribution of tumors in the colorectum was 48% (52 of
108) and 52% (56 of 108); however, this difference is not
statistically different (Table 1)
. These two ethnic groups are not
significantly different with regard to the incidence of tumors positive
for expression of MUC1, MUC2, and p53nac and
other clinicopathological and histological features of CRCs; however,
the rate of relapse was higher in Caucasians as compared with
African-Americans (P =0 .032; Table 1
).
|
a), except for an occasional
weak to low level of staining of superficial columnar cells (in 10
cases) and focal staining of cells at the base of the crypt in
transitional mucosa adjacent to invasive tumors (in 4 cases). MUC1
expression was prominently detected along the apical (luminal) surface
of malignant cells forming glandular structures. If the tumor was
positive for the MUC1 expression, the intraluminal areas of malignant
glands demonstrated strong MUC1 immunostaining (Fig. 1
b).
|
. In African-Americans and
Caucasians, there was no significant association between MUC1
expression and patient demographics and size, location, stage,
differentiation, and histological type of the tumor or
p53nac; however, there was a trend in the
association between MUC1 expression and cancer relapse at follow-up in
Caucasian patients (P = 0.060; Table 2
). In the complete population
(i.e., both racial subgroups together), however, the
phenotypic expression of MUC1 was higher in advanced stage tumors and
was significantly associated with tumor penetration into the serosal
wall (pT, P = 0.045) and distant metastasis (M,
P = 0.044; data not shown). In African-Americans and
Caucasians, the incidence of MUC1 expression was similar in CRCs
collected from proximal and distal colorectum (Table 2)
.
|
c). The immunostaining was relatively stronger in the
epithelial cells located in the base of the crypts than the more
superficial epithelial cells in transitional epithelium adjacent to
tumors (Fig. 1
c). In contrast to the benign normal mucosa
away from the tumor, the expression of MUC2 was more extensive in the
superficial epithelial cells of the transitional epithelium than in the
epithelial cells located in the base of the crypts. MUC2 expression was
observed in the majority of tumors; however, the percentage of positive
malignant cells in a tumor varied among MUC2-positive tumors. The
staining pattern of MUC2 in malignant cells was cytoplasmic and
perinuclear, as observed in benign normal tissue (Fig. 1
d).
No MUC2 expression was observed in the extracellular connective tissue.
The extent and intensity of MUC2 expression increased as the content of
mucin in tumors increased (Fig. 2)
.
|
. The expression of MUC2 in CRCs collected
from African-Americans and Caucasians and its association with
different clinicopathological and demographic characteristics are shown
in Table 2
. In both the ethnic groups, the expression of MUC2 was
associated with the histological subtype of the tumor. In
African-Americans and Caucasians, all mucinous CRCs were immunopositive
for MUC2 expression; in contrast, 68% (26 of 38) and 61% (46 of 75)
of nonmucinous CRCs, respectively, were positive for MUC2 expression
(Table 2)
. There was a significant inverse association between MUC2
positivity and p53nac in both African-American
(P = 0.004) and Caucasian (P = 0.011)
patients (Table 2)
.
Univariate Survival Analysis
Association between MUC1 Expression and Patient Survival.
Univariate Kaplan-Meier survival analysis on MUC1 expression in
CRCs and patient survival showed no significant association in the
complete patient population (n = 166; log rank,
P = 0.208; data not shown). Because our prior studies
in CRC indicated that the usefulness of biomarkers in predicting
patient survival may be restricted to certain ethnic groups of patients
(22
, 29) , we analyzed the prognostic significance of MUC1
expression in African-American and Caucasian patients separately. The
Kaplan-Meier curves based on MUC1 expression in CRCs and the race of
patients are shown in Fig. 3
.
Immunohistochemical expression of MUC1 (0.5 ISS) in CRCs was
correlated with poor prognosis in Caucasian patients (log rank,
P = 0.023) but not in African-American (log rank,
P = 0.305; Fig. 3
, A and B,
respectively). In contrast, African-American patients with CRCs
exhibiting MUC1 expression had a better survival probability than
patients with CRCs that did not express MUC1; however, this association
was not statistically significant (Fig. 3
B).
|
c). Patients with CRCs
demonstrating MUC1 expression and p53nac
demonstrated the lowest probability of overall survival, and the best
probability of overall survival was observed in patients with CRCs that
did not demonstrate either expression of MUC1 or
p53nac (log rank, P = 0.004, Fig. 3
C). However, in African-Americans, survival curves for
these four combinations of MUC1/p53nac were not
significantly different from each other (log rank, P =
0.253), and these two biomarkers together did not demonstrate
prognostic usefulness (Fig. 3
D).
Association between MUC2 Expression and Patient Survival.
The expression of MUC2 in CRCs did not correlate with overall patient
survival in univariate Kaplan-Meier analysis performed either on the
complete population, which included African-Americans and Caucasians
(log rank, P = 0.906), or on African-Americans (log
rank, P = 0.851) and Caucasians (log rank,
P = 0.969) separately (data not shown). Univariate
analysis on the combined effect of expression of MUC2 and
p53nac on patient survival probability did not
demonstrate statistically significant differences in African-American
(log rank, P = 0.878) or in Caucasian (log rank,
P = 0.240) patient populations (data not shown).
|
Multivariate Survival Analyses |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSMultivariate Survival AnalysesDISCUSSIONREFERENCES |
|---|
; however, expression of MUC2 was not
an independent prognostic indicator. The combination of MUC1 expression
and p53nac increases the risk of dying from CRC
to 5.15 times for Caucasian patients with CRCs positive for MUC1
expression and p53nac as compared with patients
negative for both MUC1 expression and p53nac.
|
. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSMultivariate Survival AnalysesDISCUSSIONREFERENCES |
|---|
The majority of CRCs secrete some amount of intracellular and/or extracellular mucin. Mucins are large glycoproteins (Mr >200,000) composed of oligometric polypeptide backbones to which are attached numerous linear or branched chains of one to 20 monosaccharides. These polypeptide backbones also are called "apomucins" and have been designated as MUC antigens. The mucins play an important role in the protective lining of epithelial surfaces. Several studies have reported that during the course of carcinogenesis, the biosynthesis of MUC antigens, particularly MUC1 and MUC2, has been altered with regard to the rate of synthesis and the extent of their glycosylation. Furthermore, a recent study (31) demonstrated that vaccination with the MUC1 tandem repeat peptide elicits immune responses in chimpanzees and suggested that this vaccination holds promise to be a safe and effective anticancer vaccine.
In a recent review, Nauhausen (30) described the importance of ethnic differences in cancer risk and in the underlying genetic variations. In addition, this review suggested that the knowledge gained by studying the effect of a single, frequent genetic alteration in a well-defined population can be applied to larger populations, and these findings can be useful in designing effective prevention and treatment strategies. Furthermore, several recent studies in colorectal neoplasia have demonstrated that the prognostic value of some molecular markers depends on the patient population studied (22 , 23) .
Multiple studies have reported that increased expression of either MUC1 of MUC2 is associated with aggressiveness of tumors in several human malignancies (16) , including colorectal neoplasia (11, 12, 13 , 32 , 33) . However, only a few studies have evaluated the prognostic usefulness of the phenotypic expression of MUC1 and MUC2 in colorectal neoplasia (11 , 15 , 34) ; moreover, to our knowledge, there are no studies regarding differences in their prognostic values based on ethnic groups.
Although a similar proportion of CRCs collected from African-American and Caucasian patients were positive for MUC1 expression, we observed a significant association between expression of MUC1 and poor outcome of only Caucasian patients by both univariate and multivariate analyses in this study. In contrast, African-Americans with CRCs exhibiting MUC1 expression had a better probability of survival; however, this association was not statistically significant. This may be the reason, however, that when the complete population (Caucasian and African-American patients together) was analyzed, the MUC1 expression was not associated with patient survival and showed no prognostic value. Furthermore, two recent studies in Japanese populations (11 , 15) also reported that MUC1 expression in colorectal tumors was an independent prognostic marker by both univariate and multivariate analyses. These studies together with others suggest that the specificity of the prognostic importance of MUC1 expression in colorectal tumors may be related to race of the patient and also suggest that the ethnicity of patient should be considered in the evaluation of importance of MUC1 expression in predicting prognosis of patients with CRCs.
Combined expression of MUC1 and p53nac, which showed no correlation with each other biologically, resulted in a more useful prognostic predictor for Caucasian but not for African-Americans patients. Caucasian patients with CRCs positive for MUC1 expression and p53nac had a significantly poorer prognosis than patients with CRCs negative for MUC1 expression and negative for p53nac. The combination of MUC1 expression and p53nac increases the risk of dying from CRC to 5.15 times for patients with CRCs positive for both MUC1 expression and p53nac as compared with patients negative for both MUC1 expression and p53nac.
In this study, we observed a significant association between MUC2 immunoreactivity and the histological type of the CRC in both ethnic groups. The proportion of tumors and extent of immunostaining for MUC2 expression was higher in mucinous CRCs than in nonmucinous CRCs, as observed in other studies (7 , 10 , 32) . Hanski et al. (12) also reported low mRNA levels of MUC2 in nonmucinous colorectal carcinomas and correlated this down-regulation of the MUC2 gene with a decrease in the metastatic potential of these tumors. In addition, Cho et al. (13) reported significantly higher mRNA and apomucin protein levels of MUC2 in mucinous colorectal carcinomas and correlated the increased MUC2 levels with extensive local invasion and distant metastasis. Furthermore, a recent study by Sternberg et al. (14) provided direct evidence of down-regulation of MUC2 expression (both protein and mRNA levels) by antisense transfection resulted in decrease in hepatic metastasis of colon cancer cells. These results suggest that the level of MUC2 expression may be associated with the histological type and aggressiveness of CRCs.
In the current study, we could not analyze the prognostic usefulness of MUC1 or MUC2 based on the anatomical location of the CRCs together with the ethnicity because of the small sample size of the subgroups. As shown in our recent study (22) , p53nac was correlated significantly with poor survival of Caucasian patients with CRCs located in the proximal colon but not in African-American patients; however, p53nac was not of prognostic value in patients of either race with CRCs of the distal colorectum. Consequently, future studies in larger patient populations should determine whether the prognostic value of expression of MUC1 and MUC2 antigens differ significantly within these two racial groups based on different anatomical sites in the colorectum. The demonstration of the prognostic usefulness of MUC1 expression in the Caucasian patient population but not in the combined population further emphasizes the importance of evaluating potential racial differences in the prognostic usefulness of molecular markers.
In summary, we report that stronger phenotypic expression of the MUC1 antigen in a subgroup of CRCs from Caucasian patients was associated with a poorer clinical outcome than that of Caucasian patients with CRCs that expressed little or no MUC1 antigen. Thus, strong phenotypic expression of the MUC1 antigen is likely to identify an aggressive subgroup of CRCs. In addition, the prognostic usefulness of MUC1 in Caucasian patients was increased when combined with p53nac. The combination of these two biomarkers may prove useful in selecting or in developing novel therapeutic regimens for Caucasian patients with colorectal adenocarcinomas.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 To whom requests for reprints should be
addressed, at Department of Pathology, University of Alabama at
Birmingham, University Station, Birmingham, AL 35294. Phone:
(205) 934-4214; Fax: (205) 975-7128; E-mail: grizzle{at}path.uab.edu 
2 The abbreviations used are: CRC, colorectal
adenocarcinoma; p53nac, nuclear accumulation of p53; ISS,
immunostaining score.
Received 3/ 2/00; revised 7/17/00; accepted 7/17/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSMultivariate Survival AnalysesDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A S Mall Analysis of mucins: role in laboratory diagnosis J. Clin. Pathol., September 1, 2008; 61(9): 1018 - 1024. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A Lugli, I Zlobec, K Baker, P Minoo, L Tornillo, L Terracciano, and J R Jass Prognostic significance of mucins in colorectal cancer with different DNA mismatch-repair status J. Clin. Pathol., May 1, 2007; 60(5): 534 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E Levi, D S Klimstra, N V Adsay, A Andea, and O Basturk MUC1 and MUC2 in pancreatic neoplasia J. Clin. Pathol., May 1, 2004; 57(5): 456 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Baldus, S. P. Monig, S. Huxel, S. Landsberg, F.-G. Hanisch, K. Engelmann, P. M. Schneider, J. Thiele, A. H. Holscher, and H. P. Dienes MUC1 and Nuclear {beta}-Catenin Are Coexpressed at the Invasion Front of Colorectal Carcinomas and Are Both Correlated with Tumor Prognosis Clin. Cancer Res., April 15, 2004; 10(8): 2790 - 2796. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Manne, N. C. Jhala, J. Jones, H. L. Weiss, C. Chatla, S. Meleth, C. Suarez-Cuervo, and W. E. Grizzle Prognostic Significance of p27kip-1 Expression in Colorectal Adenocarcinomas Is Associated with Tumor Stage Clin. Cancer Res., March 1, 2004; 10(5): 1743 - 1752. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ookawa, T. Kudo, S. Aizawa, H. Saito, and S. Tsuchida Transcriptional Activation of the MUC2 Gene by p53 J. Biol. Chem., December 6, 2002; 277(50): 48270 - 48275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Bach, D. Schrag, O. W. Brawley, A. Galaznik, S. Yakren, and C. B. Begg Survival of Blacks and Whites After a Cancer Diagnosis JAMA, April 24, 2002; 287(16): 2106 - 2113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Manne, B. D. Gary, D. K. Oelschlager, H. L. Weiss, A. R. Frost, and W. E. Grizzle Altered Subcellular Localization of Suppressin, a Novel Inhibitor of Cell-Cycle Entry, Is an Independent Prognostic Factor in Colorectal Adenocarcinomas Clin. Cancer Res., November 1, 2001; 7(11): 3495 - 3503. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
