Clinical Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium Stand Up to Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Manne, U.
Right arrow Articles by Grizzle, W. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manne, U.
Right arrow Articles by Grizzle, W. E.
Clinical Cancer Research Vol. 10, 1743-1752, March 2004
© 2004 American Association for Cancer Research

Molecular Oncology, Markers, Clinical Correlates

Prognostic Significance of p27kip-1 Expression in Colorectal Adenocarcinomas Is Associated with Tumor Stage

Upender Manne1, Nirag C. Jhala1, Jennifer Jones1, Heidi L. Weiss2, Chakrapani Chatla1, Sreelatha Meleth2, Catalina Suarez-Cuervo1 and William E. Grizzle1

1 Department of Pathology and 2 Biostatistics Unit of Comprehensive Cancer Center, University of Alabama at Birmingham, Birmingham, Alabama


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Purpose: Although the decreased expression of p27kip-1, a cyclin-dependent kinase inhibitor, has been correlated with advanced tumor stage and short survival of patients with colorectal adenocarcinomas (CRCs), its prognostic value based on the tumor site, tumor stage, and patient ethnicity was not assessed. Therefore, in this study, we investigated whether the prognostic value of p27kip-1 expression varies with the tumor site, tumor stage and patient ethnicity.

Experimental Design: We evaluated 206 (85 African Americans and 121 Caucasians) archival tissue specimens of first primary CRCs for immunohistochemical expression of p27kip-1, and its prognostic significance was analyzed using univariate Kaplan-Meier and multivariate Cox regression survival methods.

Results: Although, similar proportion of CRCs with decreased p27kip-1 expression was observed in all stages (range, 26–36%), the decreased p27kip-1 expression has been shown as a marker of poor prognosis only for patients with stage III tumors both in univariate (log-rank test, P = 0.014) and multivariate (hazard ratio = 3.2, 95% confidence interval = 1.3–7.7; P = 0.01) survival analyses. The decreased expression of p27kip-1 was associated with a high histologic grade (P = 0.016) in stage II CRCs, and with distal tumors (P = 0.001), tumor invasion (P = 0.044), and with local recurrence (P = 0.008) in stage III CRCs.

Conclusions: No prognostic significance was found for p27kip-1 expression in stages I, II, or IV CRCs, and its prognostic value was not associated with either ethnicity or tumor location. These studies suggest that decreased expression of p27kip-1 is an indicator of poor prognosis and aids in identifying a subset of patients with aggressive forms of stage III CRCs.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Colorectal adenocarcinoma (CRC) is the third leading cause of cancer associated deaths in both men and women in the United States (1) . The stage of the tumor at the time of diagnosis remains one of the powerful indicators of aggressiveness of this disease and aids in predicting the survival of patients. However, pathological stage of CRC may not be the best indicator of clinical outcome because groups of patients with tumors of identical stage have different treatment responses and clinical outcomes. Recent studies, however, have reported that the molecular changes that are associated with the aggressiveness of CRC have different biological consequences based on patient age, race, and ethnicity, as well as the anatomical location of the tumor in the colorectum even after controlling for the tumor stage (2, 3, 4, 5, 6, 7, 8) .

Aberrations in the molecular components of cell cycle checkpoints are a common feature of many human malignancies, and several of these molecules are known to have prognostic significance in CRC. Of these, p27kip-1, a cyclin-dependent kinase inhibitor that regulates progression of cells from G1 into S phase in a cell cycle is being increasingly recognized as an important factor for determining the biological behavior of invasive tumors. Decreased expression of p27kip-1 has been correlated with advanced tumor stage and short patient survival in several human cancers, including CRC (9, 10, 11) . Both biochemical and immunohistochemical evaluation of p27kip-1 expression in human CRCs have shown diverse results in relation to its prognostic value (Table 1)Citation Citation .


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

  		Table 1 Review of description of p27kip-1 expression and its prognostic value in colorectal adenocarcinomas

 

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

  		Table 1A Continued

 
In a recent study, it was noted that both benign (histologically normal) and neoplastic tissues expressed variable amounts of the p27kip-1 protein, as assessed by Western blot analyses; however, these differences were not statistically significant (12) . In another study, no relationships were found between p27kip-1 expression and gender, age, tumor location, growth pattern, and expression of other molecular markers, including p53, p73, and DCC (13) ; however, in a subset of patients, it was reported that the lack of p27kip-1 expression is an indicator of short survival for patients with early-stage CRCs (Duke Stage B) located in the proximal colon (13 , 14) . Additional larger studies are required to evaluate the biological consequences of expression of p27kip-1 in CRCs in relation to tumor stage, anatomical location of the tumor, and patient ethnicity.

In this study, we evaluated a large CRC patient population [total = 206 (85 African Americans and 121 Caucasians)] for phenotypic expression of p27kip-1 to assess its prognostic significance based on tumor stage, anatomical location, and patient ethnicity.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Formalin-fixed archival tissue blocks of CRCs from 504 patients (204 African Americans and 300 non-Hispanic Caucasians) were collected randomly from archives of the University of Alabama at Birmingham and the Veterans Administration Hospitals. All these patients underwent surgery for adenocarcinoma of the colorectum with curative or palliative intent. The selection of archival tissues was restricted to patients with first primary CRCs resected between 1981 and 1993. Because of limitations of the resources, for the current study, subsets of 206 patients with CRCs (85 African Americans and 121 Caucasians) were selected randomly from the initial group of 504 without the knowledge of clinical outcome. The pathological characteristics of CRCs, patient demographics, and patient follow-up information were collected from the pathology reports, patient charts, and tumor registries, respectively. Tissue sections stained with H&E were evaluated to determine the grades of the tumors, which were categorized as either well, moderately, or poorly differentiated as suggested by WHO (15) . Of these CRCs, 41 were classified as well, 134 as moderately, and 31 as poorly differentiated. Tumor stage was based on the classification of the American Joint Committee on Cancer classification (16 , 17) . There were 42 stage I, 69 stage II, 53 stage III, and 42 stage IV patients in this study. The anatomical locations of the tumors and their grouping into proximal and distal colonic and the rectal tumors were performed as described in earlier studies (18 , 19) .

During our initial process of selection, patients with multiple primaries, with multiple malignancies, or with family or personal histories of cancer were excluded; therefore, our study population consists only patients with sporadic first primary CRCs. We included patients who have undergone surgery alone as a therapeutic intervention, but we excluded patients who received pre- or postsurgical chemo- or radiation therapy in our study to control for treatment bias.

Immunohistochemical Staining.
Paraffin sections (5-µm thickness) of blocks representative of both tumor and benign tissues of each case were mounted on Superfrost/Plus slides (Fisher Scientific, Pittsburgh, PA). The immunohistochemical staining procedure was carried out as described earlier (2, 3, 4) . Specifically, the tissues were incubated for 1 h at room temperature with monoclonal antibody to p27kip-1 (clone 1B4; Novacostra, Newcastle upon Tyne, United Kingdom) after antigen recovery by microwave boiling in citrate buffer (pH 7.4) for 10 min. The remainder of the staining procedure and the protocol for the antigen retrieval were described in detail elsewhere (2, 3, 4) . Hematoxylin was used as a counterstain. A multitissue control block with colonic lymph nodes and benign colonic epithelium was used as the positive control.

Assessment of p27kip-1 Staining.
The staining assessment was performed by two authors (U. M. and N. J.) independently but together to ensure consistency in evaluation. The proportion of p27kip-1-expressing cells varied from 0 to 100%, and the intensity of nuclear staining also varied from weak to strong. Therefore, the percentage of cells at each intensity of staining were recorded on a scale of 0 (no staining) to +4 (strongest staining) by two authors independently. If there was a disagreement in their assessment, they resolved before combining the individual scores. The immunostaining scores of the two authors were combined to obtain an average percent positive score as well as staining intensity for each case. Using this method of staining assessment, we evaluated expression of p27kip-1 in benign colonic epithelium away and adjacent to invasive malignant lesion. A cutoff value of immunostaining staining score of >=0.7 plus at least 50% of malignant cells immunostaining was used to classify tumors with increased expression (>=50% cells positive plus >=0.7 staining intensity score) or with decreased expression (<50% positivity or <0.7 intensity score) of p27kip-1 antigen. This cutoff value was the median value of p27kip-1 staining in the benign epithelium (average of away and adjacent to invasive lesion). Staining in lymphocytes and uninvolved colonic epithelium within the tissue sections served as internal positive controls for the expression of p27kip-1.

Statistical Analyses.
The association between p27kip-1 and clinicopathological or biological characteristics was analyzed using the {chi}2 test (20) . Ps were calculated, and significance was assessed at an {alpha} level of 0.05. The median follow-up period of the complete study population of 206 patients was 4.7 years (range, <1–18 years). The period from the date of resection to the date of death or last contact (if alive) was used for survival analyses. Outcome analyses were based on patients who were alive or had died of CRC as described previously (2, 3, 4, 5, 6) . Univariate overall survival was obtained using Kaplan-Meier estimates (21) . The log-rank test was used to compare Kaplan-Meier survival curves based on the status of p27kip-1 expression. Separate multivariate Cox regression models (22) were built for patients with stages I, II, III, and IV CRCs, and the survival of patients was compared with and without p27kip-1 expression after adjustments for confounding variables. The clinical confounding variables of CRC used in these analyses were pT, pN, and M components of Tumor-Node-Metastasis stage, age, sex, ethnicity, tumor location, tumor size, and differentiation. A stepwise model-building procedure was used to determine the significant factors in predicting survival related to CRC. The interactions between p27kip-1 expression with other significant variables were tested for significance. Hazard ratios and 95% confidence intervals were calculated to identify the risk factors.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Demographic and Clinicopathological Characteristics of the Patient Population.
The clinical, pathological, and biological features of the 206 patients are reported in Table 2Citation . Patients with different tumor stages (I, II, III, and IV) were evenly distributed in this population. The distribution of CRCs in the colorectum was 43, 37, and 20% in the proximal colon, the distal colon, and the rectum, respectively. Mean age at the time of surgery was 65.4 years (range, 26.0–70.0 years), and the median survival was 5.58 years (95% confidence interval = 3.59–9.0). There was a predominance of males in our study (124 of 206, 60%) because the majority of patients treated at the Veterans Administration Hospital of Birmingham were males. At the last follow-up, the proportion of patients alive was 49% (101 of 206), dead because of colorectal neoplasia was 32% (66 of 206), and died because of other causes was 19% (Table 2)Citation .


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

  		Table 2 Clinicopathological characteristics and expression of p27Kip-1 in colorectal adenocarcinomas

 
p27kip-1 Expression and Its Association with Different Clinicopathological Parameters.
As observed in several other studies (Table 1)Citation Citation , the expression of p27kip-1 was predominantly in the nucleus; however, in a small proportion of cells (~20%) in a tumor tissue section p27kip-1 was localized in the cytoplasm. For this study, we considered only distinct nuclear expression of p27kip-1. Increased phenotypic expression of p27kip-1 was detected in 70% (145 of 206) of CRCs (Table 2)Citation .

The analysis of correlation between the proportions of CRCs with increased expression of p27kip-1 with different clinicopathological parameters in each tumor stage group was shown in Table 3Citation . The incidence of CRCs with decreased p27kip-1 expression was 36, 28, 30, and 26% in stages I, II, III, and IV, respectively. There was a significant association between p27kip-1 expression and the degree of histological tumor differentiation in patients with stage II CRCs, which is that its expression was gradually decreased from well to poorly differentiated CRCs ({chi}2, P = 0.016; Table 3Citation ). In stage III patient group, the expression of p27kip-1 was significantly associated with the anatomical location of CRCs, the increased p27kip-1 was observed in CRCs located in the proximal colon (91%), and in the rectum (77%) as compared with the distal colon CRCs (39%; {chi}2, P = 0.001). In this patient group, the extent of expression of p27kip-1 decreased as the extent of invasion of tumor into the bowel increased (pT1 to pT4, {chi}2, P = 0.044). Also in the stage III cases, the decreased expression of p27kip-1 was significantly associated with an increasing incidence of recurrence ({chi}2 P = 0.008) and with clinical outcome; specifically, a higher proportion of patients with increased p27kip-1 expression were alive at the time of last follow-up ({chi}2, P = 0.018; Table 3Citation ).


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

  		Table 3 Association of p27kip-1 expression with selected clinicopathological characteristics of patients with colorectal adenocarcinomas in relation to their tumor stage

 
We also used a multiple logistic regression model to simultaneously evaluate the association of each parameter with p27kip-1 in all four tumor stage groups (stages I, II, III, and IV). This multivariate logistic analysis will address the issue of multiple testing for each individual parameter, and only one overall model is used to evaluate the clinical parameters simultaneously. In this analysis of association of variables with p27kip-1 positivity, we found similar results as reported in Table 3Citation ; specifically, tumor differentiation (P = 0.008) remained significantly associated with increased expression of p27kip-1 in patients with stage II CRCs (data not shown). Despite a small number of patients in stage III group, we still observed significance or an indication of trend of the association of p27kip-1 expression with patient outcome (P = 0.008), tumor invasion (pT; P = 0.062), and tumor location (P = 0.095; data not shown).

Survival Analyses.
Kaplan-Meier univariate survival analysis on the complete patient population (n = 206) demonstrated no significant differences in their overall survival between the patient groups with decreased or increased levels of phenotypic p27kip-1 expression (log rank, P = 0.152; data not shown). Because, this study population has a considerable number of African American patients, we also conducted survival analyses to assess the significance of p27kip-1 expression based on patient race; the analyses of clinical outcome demonstrated that in either of the racial groups, the cumulative survival curves of low and increased levels of p27kip-1 expression did not show statistically significant differences (log rank, P = 0.273 and P = 0.183, for African Americans or Caucasians, respectively; data not shown). Univariate survival analyses based on p27kip-1 expression and on tumor location also did not show significant differences for all three anatomical sites (log rank, P = 0.377, P = 0.248 and P = 0.631, for proximal, distal, and rectal tumors, respectively; data not shown).

We also analyzed the significance of p27kip-1 expression in the complete study population based on the tumor stage and found that patients with tumors exhibiting decreased p27kip-1 expression had significantly shorter overall survival only in the stage III patient-group (log rank, P = 0.014; Fig. 1CCitation ). Similar analyses in stage I, stage II, and stage IV groups did not show statistically significant differences between patients with decreased or increased levels of p27kip-1 expression (Fig. 1A, B, and D)Citation .



View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Kaplan-Meier analysis showing the overall survival of colorectal adenocarcinoma patients categorized according to the stage of the tumor and status of p27kip-1 expression. The statistical significance of the difference between curves of p27kip-1 high expressors (increased expression) p27kip-1 low expressors (decreased expression) was compared in stage I (A), stage II (B), stage III (C), and stage IV (D) patient groups. Ps were calculated by the log-rank test.

 
The multivariate regression models built separately for each tumor stage demonstrated that decreased expression of p27kip-1 was an independent indicator of poor prognosis only for patients with stage III CRCs. The hazard ratios and 95% confidence interval for stage III was 3.2 and 1.3–7.7 (P = 0.01), respectively (Table 4)Citation . No prognostic significance was found for p27kip-1 expression in stages I, II, or IV CRCs (Table 4)Citation . In multivariate regression model, which included all groups of patients (stage I–IV together), the decreased p27kip-1 expression was an independent marker of poor overall survival; however, it was only marginally significant (hazard ratio 0.7, 95% confidence interval, 0.4–1.0; P = 0.058). In this overall model, older age, African American race, high tumor grade, bowel wall invasion, regional lymph node invasion, and distant metastasis were independent predictors of overall survival (Table 4)Citation .


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

  		Table 4 Cox regression multivariate analysis to evaluate independent prognostic value of p27kip-1 expression in colorectal adenocarcinomas in relation to tumor stage

 

   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Our study population had similar proportions of CRCs with the decreased levels of p27kip-1 expression in all stages (stages I–IV); however, decreased expression of p27kip-1 was associated significantly with short patient survival of patients with stage III tumors only. Our findings also demonstrated that stage III CRCs, which were located in the distal colon, are less likely to express p27kip-1 at high levels and that low levels of p27kip-1 expression were correlated significantly with the extent of invasion of tumor to other adjacent organs (pT4) and with local recurrence. These results suggest that p27kip-1 expression may be used to identify more aggressive phenotypes of stage III CRCs.

Several abnormalities in the molecular features of cell cycle checkpoints have been implicated in several human malignancies, including CRC; however, the prognostic values of many of these cell cycle antigens are controversial in CRC. For example, wild-type p53 eliminates genetically unstable cells that enter the cell cycle, and a mutant p53 will allow these cells to progress to a malignant phenotype. The role of p53 in colorectal neoplasia has been well studied; however, its value in predicting the clinical outcome has been controversial. Nevertheless, some recent studies from others and our laboratories evaluated its prognostic value based on tumor location and ethnicity and have demonstrated that nuclear accumulation of p53 is an independent prognostic indicator of patients with proximal tumors (23, 24, 25) , specifically, for Caucasian patients (3 , 7 , 8) . These novel strategies of assessing the predictive value of a molecular marker of colorectal neoplasia would yield valuable information, which will be useful in identifying aggressive phenotypes.

Increased expression of p27kip-1in mammalian cells induces a G1 block of the cell cycle (26 , 27) and high levels of p27kip-1 found in quiescent cells suggest that expression of p27kip-1 also plays a role in maintaining cells in G0 (28 , 29) . As with other Kip proteins, p27kip-1 has a nuclear localization signal in its -COOH terminus. Loss of p27kip-1 expression may result in the development and/or progression of tumors; however, this loss of expression does not appear to result from gene mutation (30, 31, 32) . p27kip-1 is posttranslationally regulated by an ubiquitin-proteasome-dependent degradation pathway, and p27kip-1 also is regulated by cell-cell contacts (33) . p27kip-1 expression can also be regulated by transcriptional mechanism. In CRCs, reduced expression of p27kip-1 in the metachronous metastases compared with the corresponding primary tumor suggests that the down-regulation of p27kip-1 in circulating malignant cells may accentuate the ability of these cells to metastasize to liver (34) . Also, accumulation of p27kip-1 has been implicated in the differentiation of different varieties of cells (35 , 36) . In CRCs, Loda et al. (9) demonstrated that absent or low levels of p27kip-1 are caused by increased protease-mediated degradation rather than from altered gene expression.

Variable levels of p27kip-1 expression have been observed in benign and malignant epithelial components of the colorectum (9) . The role of this molecular marker in assessing the aggressiveness of CRCs and in predicting the clinical outcome of patients with colorectal neoplasia has been examined by several investigators and the majority of these studies, which are cited in the Medline or PubMed (until the end of July 2003) are listed in Table 1Citation Citation . This review table clearly indicates discrepancy of prognostic significance of p27kip-1 expression in colorectal neoplasia. Several studies reported that the lack of p27kip-1 expression was associated with short patient survival (9 , 11 , 13 , 14 , 37, 38, 39, 40) ; however, such an association was not found in some other studies (41, 42, 43, 44, 45) . In some studies, it was suggested that p27kip-1 was an independent predictor of patients specifically with early stage CRCs (stages I and II; Refs. 9 , 13 , 39 ), whereas, studies by Tenjo et al. (11) observed that p27kip-1 expression as an independent prognostic marker for patients with stage III CRCs.

In our study, all patients have received a uniform treatment of surgical resection either with curative or palliative intents and none of them were treated with chemotherapeutic agents such as 5-fluorouracil or with any other neoadjuvant or adjuvant therapies. Adjuvant chemotherapies were not common practices at the beginning of our study in 1981, but use of adjuvant therapy for CRC increased in 1988. Although, the Food and Drug Administration has approved leucovorin for use in combination with 5-fluorouracil to prolong survival in the palliative treatment of patients only with advanced CRC (stage IV) in 1952, the adjuvant treatment in combination with 5-fluorouracil after surgical resection in patients with Dukes’ Stage C colon cancer was approved only in 1990.3 Therefore, we did not identify many patients who received chemotherapy during our study period and as stated in methods section, we excluded patients who received pre- or postsurgical treatment. Thus, using our study population, we evaluated the prognostic value of p27kip-1 expression without treatment biases. Also, we assessed its prognostic value based on race, the tumor stage, and tumor site.

In this study, similar proportions of CRCs with the decreased levels of p27kip-1 expression were noted in all stages (26–36%). Similar to studies by Palmqvist et al. (14) , we also did not see a significant association between p27kip-1 expression and the tumor differentiation; however, when stratified by tumor stage, our study found that in stage II CRCs, the decreased expression of p27kip-1 was associated with poorer histological differentiation. This association, however, did not impact overall survival of patients with stage II CRCs. In stage III CRCs, p27kip-1 expression was significantly correlated with tumor location, i.e., distal tumors exhibited decreased levels of p27kip-1 expression and the extent of p27kip-1 expression decreased as the invasion of tumor into the lower layers of the bowel wall increased (pT1 to pT4). Similar to our findings, Ciaparrone et al. (46) also found a significant correlation between p27kip-1 expression and tumor grade with well and moderately differentiated CRCs expressing higher p27kip-1, whereas the poorly differentiated CRCs had significantly lower p27kip-1 expression. Furthermore, we observed that tumors, which exhibited decreased expression of p27kip-1, had higher incidence of local recurrence in patients with stage III CRCs, and such an association also was reported in prostatic carcinoma (47) .

Our study demonstrated in both univariate and multivariate analyses that the increased expression of p27kip-1 was associated with improved survival of patients with stage III CRCs but not with stages I, II, or IV tumors. These results are in contrast to some of the previously published studies where it was shown that decreased p27kip-1 expression was associated with poor prognosis of patients with stage II tumors (9 , 13) tumors without nodal involvement (stage I plus II; Ref. 39 ). However, similar to our findings, studies by Tenjo et al. (11) observed that p27kip-1 is an independent prognostic marker for patients with stage III CRCs. Although, the exact reasons for these contradictory results are not known, the admixture of our patient population for race and for the tumor site may be affecting these findings. Also, it is possible that in our study, we used monoclonal antibody clone 1B4 and established rigorous standards of evaluation as well as cutoff values (>=50% cells with immunostaining score value >= 0.7), and only nuclear p27kip-1 expression was considered in staining evaluation. Whereas, studies by Zhang et al. (13) used a different antibody clone (SX53G8), different cutoff values for categorizing tumors as high expressors of p27kip-1 (>10%), considered only cytoplasmic p27kip-1 staining for survival analyses and suggested that decreased expression of p27kip-1 was associated with poor survival of patients with Dukes B tumors in univariate analysis. Unlike studies of Zhang et al. (13) and Kobayashi et al. (42) , ours as well as several other studies (Table 1)Citation Citation have used nuclear p27kip-1 staining for survival analyses. The higher cutoff values used in our study to categorize CRCs into high and low expressors are similar to other investigators; additionally, we used more rigorous standards of staining evaluation that included two independent observers to reduce individual bias. Besides these technical variations, other confounders of patient survival, including the adjuvant treatment therapies received and racial admixture specifically in studies of the United States, were not clearly described in the majority of earlier studies. Moreover, these contradictory findings also suggest that the future larger studies should consider the tumor stage and the anatomical location of tumor in the colorectum, race, and ethnicity in evaluating the prognostic value of p27kip-1 expression in colorectal neoplasia. This has led us to propose that molecular and demographic features should be added to the staging of colorectal tumors (7 , 8) .

It is well established in the literature that patients with higher tumor stage (stages III and IV) have a poorer prognosis than stage I or II and staging is used in clinical practice for therapeutic decision making. To add to this concept, multivariate survival analyses of our study also noted significant differences in patient outcome based on age (>=65 years), tumor differentiation (poor differentiation), and ethnicity (African Americans) even after controlling for pathological stage of the tumor (Table 4)Citation .

In this study, we have shown both in univariate as well as in multivariate analyses that decreased expression of p27kip-1 is a significant predictor of poor clinical outcome of stage III patients who received surgery alone as a treatment regimen. Our findings also suggest that in stage III tumors, p27kip-1 expression will determine local recurrence. Also, these higher stage tumors exhibit underlying molecular heterogeneity that might determine the behavior of selected tumors. In our studies, the prognostic value of p27kip-1 expression was not associated with race or the anatomical location of the tumor. As demonstrated in our prior studies (2, 3, 4, 5, 6, 7, 8) , the use of multiple molecular markers, including p27kip-1 expression, may help in identifying more aggressive tumors and in predicting the clinical outcome, thus, might help clinical oncologists in designing or selecting more aggressive therapies for subgroups of patients with colorectal cancer.


   ACKNOWLEDGMENTS
 
We thank the staff of the Tissue Procurement Facility at University of Alabama at Birmingham for their technical help, and we also thank Dr. Sameera Vohra for her assistance in manuscript preparation. We thank Libby Chambers, Cindy Legge, and Lisa Jinright for their secretarial support.


   FOOTNOTES
 
Grant support: National Cancer Institute Grants 1RO1 CA98932-01, 1RO3 CA097542-01, and CA86359-04, which are funded by NIH.

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.

Requests for reprints: Upender Manne, Assistant Professor, Department of Pathology, University of Alabama at Birmingham, 565-LHRB, Building 701, 19th Street South, Birmingham, AL, 35294-0007. Phone: (205) 934-4276; Fax: (205) 975-9927; E-mail: manne{at}path.uab.edu

3 Internet address: http://www.fda.gov/cder/cancer/druglistframe.htm. Back

Received 8/13/03; revised 11/14/03; accepted 11/21/03.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. American Cancer Society. . Cancer Facts and Figures, 1-52, American Cancer Society Atlanta, GA 2003.
  2. Manne U., Myers R. B., Moron C., Poczatek R. B., Dillard S., Weiss H., Brown D., Srivastava S., Grizzle W. E. Prognostic significance of Bcl-2 expression and p53 nuclear accumulation in colorectal adenocarcinoma. Int. J. Cancer, 74: 346-358, 1997.[CrossRef][Medline]
  3. Manne U., Weiss H. L., Myers R. B., Danner O. K., Moron C., Srivastava S., Grizzle W. E. Nuclear accumulation of p53 in colorectal adenocarcinoma: prognostic importance differs with race and location of the tumor. Cancer (Phila.), 83: 2456-2467, 1998.
  4. Manne U., Weiss H. L., Grizzle W. E. Racial differences in the prognostic usefulness of MUC1 and MUC2 in colorectal adenocarcinomas. Clin. Cancer Res., 6: 4017-4025, 2000.[Abstract/Free Full Text]
  5. Manne U., Weiss H. L., Grizzle W. E. Bcl-2 expression is associated with improved prognosis in patients with distal colorectal adenocarcinomas. Int. J. Cancer, 89: 423-430, 2000.[CrossRef][Medline]
  6. Manne U., Gary B. D., Oelschlager D. K., Weiss H. L., Frost A. R., Grizzle W. E. Altered subcellular localization of suppressin, a novel inhibitor of cell-cycle entry, is an independent prognostic factor in colorectal adenocarcinomas. Clin. Cancer Res., 7: 3495-3503, 2001.[Abstract/Free Full Text]
  7. Grizzle W. E., Manne U., Jhala N. C., Weiss H. L. Molecular characterization of colorectal neoplasia in translational research. Arch. Pathol. Lab. Med., 125: 91-98, 2001.[Medline]
  8. Grizzle W. E., Manne U., Weiss H. L., Jhala N., Talley L. Molecular staging of colorectal cancer in African-American and Caucasian patients using phenotypic expression of p53, Bcl-2, MUC-1 AND p27(kip-1). Int. J. Cancer, 97: 403-409, 2002.[CrossRef][Medline]
  9. Loda M., Cukor B., Tam S. W., Lavin P., Fiorentino M., Draetta G. F., Jessup J. M., Pagano M. Increased proteasome-dependent degradation of the cyclin-dependent kinase inhibitor p27 in aggressive colorectal carcinomas. Nat. Med., 3: 231-234, 1997.[CrossRef][Medline]
  10. Liu D. F., Ferguson K., Cooper G. S., Grady W. M., Willis J. p27 cell-cycle inhibitor is inversely correlated with lymph node metastases in right-sided colon cancer. J. Clin. Lab. Anal., 13: 291-295, 1999.[CrossRef][Medline]
  11. Tenjo T., Toyoda M., Okuda J., Watanabe I., Yamamoto T., Tanaka K., Ohtani M., Nohara T., Kawasaki H., Tanigawa N. Prognostic significance of p27Kip1 protein expression and spontaneous apoptosis in patients with colorectal adenocarcinomas. Oncology, 58: 45-51, 2000.[CrossRef][Medline]
  12. Sgambato A., Ratto C., Faraglia B., Merico M., Ardito R., Schinzari G., Romano G., Cittadini A. R. Reduced expression and altered subcellular localization of the cyclin-dependent kinase inhibitor p27(Kip1) in human colon cancer. Mol. Carcinog., 26: 172-179, 1999.[CrossRef][Medline]
  13. Zhang H., Sun X. F. Loss of p27 expression predicts poor prognosis in patients with Dukes’ B stage or proximal colorectal cancer. Int. J. Oncol., 19: 49-52, 2001.[Medline]
  14. Palmqvist R., Stenling R., Oberg A., Landberg G. Prognostic significance of p27Kip1 expression in colorectal cancer: a clinicopathological characterization. J. Pathol., 188: 18-23, 1999.[CrossRef][Medline]
  15. WHO. . International Histological Classification of Disease of Tumors, Histological Typing of Intestinal Tumors, Vol. 15: WHO Geneva, Switzerland 1976.
  16. Beahrs O. H. Staging of cancer of the colon and rectum. Cancer (Phila.), 70: 1393-1396, 1992.
  17. Bearss O. H., H. D., Hutter R. V. P., Kennedy B. J. Manual for Staging Cancer. American Joint Committee on Cancer. New York: J. B. Lippincott, 1992.
  18. Dinning J. P., Hixson L. J., Clark L. C. Prevalence of distal colonic neoplasia associated with proximal colon cancers. Arch. Intern. Med., 154: 853-856, 1994.[Abstract]
  19. Loffeld R., Putten A., Balk A. Changes in the localization of colorectal cancer: implications for clinical practice. J. Gastroenterol. Hepatol., 11: 47-50, 1996.[Medline]
  20. Fleiss J. . Statistical Methods for Rates and Proportions, John Wiley and Sons New York, NY 1981.
  21. Kaplan E., Meier P. Non-parametric estimation from incomplete observations. J. Am. Stat. Assoc., 53: 457-481, 1958.[CrossRef]
  22. Cox D. R. Regression models and life tables. J. R. Stat. Soc., 34: 187-220, 1972.
  23. Sun X. F., Carstensen J. M., Zhang H., Arbman G., Nordenskjold B. Prognostic significance of p53 nuclear and cytoplasmic overexpression in right and left colorectal adenocarcinomas. Eur. J. Cancer, 32A: 1963-1967, 1996.
  24. Diez M., Medrano M., Muguerza J. M., Ramos P., Hernandez P., Villeta R., Martin A., Noguerales F., Ruiz A., Granell J. Influence of tumor localization on the prognostic value of p53 protein in colorectal adenocarcinomas. Anticancer Res., 20: 3907-3912, 2000.[Medline]
  25. Samowitz W. S., Curtin K., Ma K. N., Edwards S., Schaffer D., Leppert M. F., Slattery M. L. Prognostic significance of p53 mutations in colon cancer at the population level. Int. J. Cancer, 99: 597-602, 2002.[CrossRef][Medline]
  26. Toyoshima H., Hunter T. p27, a novel inhibitor of G1 cyclin-Cdk protein kinase activity, is related to p21. Cell, 78: 67-74, 1994.[CrossRef][Medline]
  27. Polyak K., Lee M. H., Erdjument-Bromage H., Koff A., Roberts J. M., Tempst P., Massague J. Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell, 78: 59-66, 1994.[CrossRef][Medline]
  28. Hengst L., Reed S. I. Translational control of p27Kip1 accumulation during the cell cycle. Science (Wash. DC), 271: 1861-1864, 1996.[Abstract]
  29. Sandhu C., Garbe J., Bhattacharya N., Daksis J., Pan C. H., Yaswen P., Koh J., Slingerland J. M., Stampfer M. R. Transforming growth factor ß stabilizes p15INK4B protein, increases p15INK4B-cdk4 complexes, and inhibits cyclin D1-cdk4 association in human mammary epithelial cells. Mol. Cell. Biol., 17: 2458-2467, 1997.[Abstract]
  30. Morosetti R., Kawamata N., Gombart A. F., Miller C. W., Hatta Y., Hirama T., Said J. W., Tomonaga M., Koeffler H. P. Alterations of the p27KIP1 gene in non-Hodgkin’s lymphomas and adult T-cell leukemia/lymphoma. Blood, 86: 1924-1930, 1995.[Abstract/Free Full Text]
  31. St. Croix B., Florenes V. A., Rak J. W., Flanagan M., Bhattacharya N., Slingerland J. M., Kerbel R. S. Impact of the cyclin-dependent kinase inhibitor p27Kip1 on resistance of tumor cells to anticancer agents. Nat. Med., 2: 1204-1210, 1996.[CrossRef][Medline]
  32. Tanaka C., Yoshimoto K., Yang P., Kimura T., Yamada S., Moritani M., Sano T., Itakura M. Infrequent mutations of p27Kip1 gene and trisomy 12 in a subset of human pituitary adenomas. J. Clin. Endocrinol. Metab., 82: 3141-3147, 1997.[Abstract/Free Full Text]
  33. Schulze A., Zerfass-Thome K., Berges J., Middendorp S., Jansen-Durr P., Henglein B. Anchorage-dependent transcription of the cyclin A gene. Mol. Cell. Biol., 16: 4632-4638, 1996.[Abstract]
  34. Thomas G. V., Szigeti K., Murphy M., Draetta G., Pagano M., Loda M. Down-regulation of p27 is associated with development of colorectal adenocarcinoma metastases. Am. J. Pathol., 153: 681-687, 1998.[Abstract/Free Full Text]
  35. Durand B., Gao F. B., Raff M. Accumulation of the cyclin-dependent kinase inhibitor p27/Kip1 and the timing of oligodendrocyte differentiation. EMBO J., 16: 306-317, 1997.[CrossRef][Medline]
  36. Robker R. L., Richards J. S. Hormone-induced proliferation and differentiation of granulosa cells: a coordinated balance of the cell cycle regulators cyclin D2 and p27Kip1. Mol. Endocrinol., 12: 924-940, 1998.[Abstract/Free Full Text]
  37. Gunther K., Jung A., Volker U., Meyer M., Brabletz T., Matzel K. E., Reymond M. A., Kirchner T., Hohenberger W. p27(kip1) expression in rectal cancer correlates with disease-free survival. J. Surg. Res., 92: 78-84, 2000.[CrossRef][Medline]
  38. Yao J., Eu K. W., Seow-Choen F., Cheah P. Y. Down-regulation of p27 is a significant predictor of poor overall survival and may facilitate metastasis in colorectal carcinomas. Int. J. Cancer, 89: 213-216, 2000.[CrossRef][Medline]
  39. Belluco C., Esposito G., Bertorelle R., Del Mistro A., Fassina A., Vieceli G., Chieco-Bianchi L., Nitti D., Lise M. Absence of the cell cycle inhibitor p27Kip1 protein predicts poor outcome in patients with stage I-III colorectal cancer. Ann. Surg. Oncol, 6: 19-25, 1999.[Abstract]
  40. Rossi H. A., Liu Q., Banner B., Hsieh C. C., Savas L., Savarese D. The prognostic value of invariant chain (II) and Her-2/neu expression in curatively resected colorectal cancer. Cancer J., 8: 268-275, 2002.[Medline]
  41. Cheng J. D., Werness B. A., Babb J. S., Meropol N. J. Paradoxical correlations of cyclin-dependent kinase inhibitors p21waf1/cip1 and p27kip1 in metastatic colorectal carcinoma. Clin. Cancer Res., 5: 1057-1062, 1999.[Abstract/Free Full Text]
  42. Kobayashi M., Shiraishi T., Tonouchi H., Miki C., Kusunoki M. 5-FU improves p27-related poor prognosis in patients with Astler-Coller B2-C colorectal carcinoma. Oncol. Rep., 9: 29-33, 2002.[Medline]
  43. McKay J. A., Douglas J. J., Ross V. G., Curran S., Loane J. F., Ahmed F. Y., Cassidy J., McLeod H. L., Murray G. I. Analysis of key cell-cycle checkpoint proteins in colorectal tumours. J. Pathol., 196: 386-393, 2002.[CrossRef][Medline]
  44. Schwandner O., Bruch H. P., Broll R. p21, p27, cyclin D1, and p53 in rectal cancer: immunohistology with prognostic significance?. Int. J. Colorectal Dis., 17: 11-19, 2002.[CrossRef][Medline]
  45. Hoos A., Nissan A., Stojadinovic A., Shia J., Hedvat C. V., Leung D. H., Paty P. B., Klimstra D., Cordon-Cardo C., Wong W. D. Tissue microarray molecular profiling of early, node-negative adenocarcinoma of the rectum: a comprehensive analysis. Clin. Cancer Res., 8: 3841-3849, 2002.[Abstract/Free Full Text]
  46. Ciaparrone M., Yamamoto H., Yao Y., Sgambato A., Cattoretti G., Tomita N., Monden T., Rotterdam H., Weinstein I. B. Localization and expression of p27KIP1 in multistage colorectal carcinogenesis. Cancer Res., 58: 114-122, 1998.[Abstract/Free Full Text]
  47. Cote R. J., Shi Y., Groshen S., Feng A. C., Cordon-Cardo C., Skinner D., Lieskovosky G. Association of p27Kip1 levels with recurrence and survival in patients with stage C prostate carcinoma. J. Natl. Cancer Inst. (Bethesda), 90: 916-920, 1998.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Clin. Cancer Res.Home page
M. S. Lyall, S. R. Dundas, S. Curran, and G. I. Murray
Profiling Markers of Prognosis in Colorectal Cancer
Clin. Cancer Res., February 15, 2006; 12(4): 1184 - 1191.
[Abstract] [Full Text] [PDF]

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

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