Clinical Cancer Research The Future of Cancer Research: Science and Patient Impact
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 Lloyd, J. M.
Right arrow Articles by Hardingham, J. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lloyd, J. M.
Right arrow Articles by Hardingham, J. E.
Clinical Cancer Research Vol. 12, 417-423, January 2006
© 2006 American Association for Cancer Research

Imaging, Diagnosis, Prognosis

Identification of Early-Stage Colorectal Cancer Patients at Risk of Relapse Post-Resection by Immunobead Reverse Transcription-PCR Analysis of Peritoneal Lavage Fluid for Malignant Cells

Julia M. Lloyd1,2, Cassandra M. McIver1,3, Sally-Anne Stephenson1, Peter J. Hewett2, Nicholas Rieger2 and Jennifer E. Hardingham1,3

Authors' Affiliations: Departments of 1 Haematology-Oncology and 2 Surgery, The Queen Elizabeth Hospital, Woodville, South Australia, Australia and 3 Department of Physiology, Adelaide University, Adelaide, South Australia, Australia

Requests for reprints: Jennifer E. Hardingham, Department of Haematology-Oncology, Basil Hetzel Institute, The Queen Elizabeth Hospital, 28 Woodville Road, Woodville, Australia 5011. Phone: 61-8-82226142; Fax: 61-8-82226144; E-mail: jenny.hardingham{at}nwahs.sa.gov.au.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: Colorectal cancer patients diagnosed with stage I or II disease are not routinely offered adjuvant chemotherapy following resection of the primary tumor. However, up to 10% of stage I and 30% of stage II patients relapse within 5 years of surgery from recurrent or metastatic disease. The aim of this study was to determine if tumor-associated markers could detect disseminated malignant cells and so identify a subgroup of patients with early-stage colorectal cancer that were at risk of relapse.

Experimental Design: We recruited consecutive patients undergoing curative resection for early-stage colorectal cancer. Immunobead reverse transcription-PCR of five tumor-associated markers (carcinoembryonic antigen, laminin {gamma}2, ephrin B4, matrilysin, and cytokeratin 20) was used to detect the presence of colon tumor cells in peripheral blood and within the peritoneal cavity of colon cancer patients perioperatively. Clinicopathologic variables were tested for their effect on survival outcomes in univariate analyses using the Kaplan-Meier method. A multivariate Cox proportional hazards regression analysis was done to determine whether detection of tumor cells was an independent prognostic marker for disease relapse.

Results: Overall, 41 of 125 (32.8%) early-stage patients were positive for disseminated tumor cells. Patients who were marker positive for disseminated cells in post-resection lavage samples showed a significantly poorer prognosis (hazard ratio, 6.2; 95% confidence interval, 1.9-19.6; P = 0.002), and this was independent of other risk factors.

Conclusion: The markers used in this study identified a subgroup of early-stage patients at increased risk of relapse post-resection for primary colorectal cancer. This method may be considered as a new diagnostic tool to improve the staging and management of colorectal cancer.

Colorectal cancer remains a problem in terms of morbidity and mortality despite steady improvements in therapeutic regimens. Colorectal cancer is the second most common solid organ cancer in males and females in developed countries with an incidence worldwide of almost a million cases annually (1). In Australia, the overall mortality rate at 5 years post-diagnosis remains at ~58% (2). Early detection of colorectal cancer is correlated with high survival rates as 90% of patients that are diagnosed with TNM stage I tumors are alive 5 years after the initial diagnosis, compared with a steady decrease in 5-year survival rates in later stages, down to 7% to 10% in stage IV patients (3). Patients diagnosed with early-stage colorectal cancer (TNM stage I or II) undergo surgical tumor resection with curative intent, yet up to 30% of these patients suffer recurrent or metastatic disease within 5 years of surgery. This suggests that viable tumor cells with proliferative and perhaps metastatic potential had been shed from the primary tumor site either before removal of the tumor or during surgical resection. Such occult tumor cells are generally considered to be responsible for treatment failure in colorectal cancer patients (4). Currently, the use of adjuvant chemotherapy (5-fluorouracil and leucovorin) is the standard of care for resected stage III colon cancer. Its use in stage II (node negative) patients, however, has been controversial because there has been no direct evidence from randomized controlled trials to support the routine use of adjuvant chemotherapy for patients with stage II colon cancer (5). Hence, it is imperative that accurate staging of the disease is achieved so that appropriate treatment decisions can be made. In particular, the detection of occult disseminated tumor cells may identify early-stage patients at risk of developing recurrent or metastatic disease and who would benefit from adjuvant therapy post-resection of the primary tumor.

Here we used the technique of immunobead reverse transcription-PCR (RT-PCR; refs. 6, 7) to detect tumor-associated expression markers. The method uses magnetic beads coated with an epithelial-specific monoclonal antibody to isolate epithelial-derived tumor cells from blood or other samples, followed by RT-PCR. We have previously shown, using epithelial-specific markers cytokeratin (CK) 19, CK20, mucin 1, and mucin 2, that in stage III patients, epithelial cells detected in peripheral blood were predictive of shorter disease-free survival, suggesting that this technique has the potential to more accurately stratify patients into different prognostic groups. However, 4 of 34 patients with benign inflammatory bowel disease were also positive (8), suggesting that a search for more appropriate tumor markers was warranted.

In this study, we selected a panel of five molecular markers for the RT-PCR analysis of the immunobead samples: CK20, carcinoembryonic antigen (CEA), EphB4, laminin-5 {gamma}2 chain (LAM{gamma}2), and matrilysin (MAT) or matrix metalloproteinase 7 (Table 1). CK20 and CEA have been used by us and others as markers of colorectal tumor cells (8–14). Previously, using relative RT-PCR, we found EphB4 to be overexpressed by ≥2-fold in colon tumors compared with adjacent normal mucosa in 39 of 62 (63%) cases, suggesting it may be useful as a colon tumor marker (15). Similarly, we found LAM{gamma}2 to be overexpressed in 40 of 50 (80%) colon tumors.4 Using immunohistochemistry, LAM{gamma}2 overexpression was found to be a prognostic factor for poorer outcome in 103 cases of colorectal cancer (16). MAT is unique among the matrix metalloproteinases as it is localized to epithelial tumor cells rather than to the stroma, making it a good candidate tumor marker (17). This marker was shown to be overexpressed in 30 of 30 colon tumors compared with normal mucosa (18).


View this table:
[in this window]
[in a new window]
  		Table 1. Marker panel with respective sense and antisense primers, probe sequences, annealing temperatures, and MgCl2 concentration

 
The aims of this study were thus to use a panel of tumor-associated markers to determine the presence of disseminated colorectal tumor cells in peripheral blood and peritoneal lavage samples in early-stage colorectal cancer and to determine if detection of such cells was an independent prognostic variable for risk of recurrent or metastatic disease following curative resection for colorectal cancer.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Patients and controls. One hundred forty-six patients (80 males and 66 females) diagnosed with early-stage (TNM stage I or II) primary colorectal cancer were recruited into the study between January 1997 and June 2003. The median age was 74 years (range, 43-95 years) with 46 patients diagnosed as stage I and 100 as stage II. Patients with nonmalignant surgical colorectal disorders were used as negative controls to assess the tumor specificity of the candidate markers. This group consisted of 70 patients admitted for resection for inflammatory bowel disease (n = 42), diverticulitis (n = 20), and other nonmalignant bowel conditions (n = 8). As further negative controls for the panel of markers, peripheral blood samples from patients undergoing venesection for polycythemia vera were collected. Informed consent was obtained from all patients and the research protocol was approved by the institutional Ethics of Human Research Committee. The samples collected from surgical cases were (a) peripheral blood (20 mL) collected before surgery and again the morning after the surgical procedure, (b) NaCl (0.9% w/v) lavage samples (50 mL) aspirated from the tumor bed and the pelvic floor regions before manipulation of the bowel, and (c) after completion of the resection, further 50 mL NaCl lavage samples collected from the same regions. If the tumor was situated in the rectum, the samples were sourced from the pelvis, as was the case for benign controls. All peripheral blood and lavage samples were collected in dipotassium EDTA and processed within 2 hours.

Cell lines used as positive controls. Colon cancer cell lines were used as positive controls for RT-PCR assessment of candidate marker expression and added to normal blood samples to monitor the sensitivity of immunobead RT-PCR. The LIM cell lines 1215, 1863, 1899, 2099, 2405, and 2412 were kindly provided by Dr. R. Whitehead (Ludwig Institute for Cancer Research, Melbourne, Australia); cell lines HT-29, SW48, SW480, and SW620 were purchased from the American Type Culture Collection (Rockville, MD). The cells were maintained in RPMI 1640 (Life Technologies, Grand Island, NY), pH 7.4, supplemented with 100 units/mL penicillin, 100 µg/mL streptomycin, 200 µg/mL L-glutamine, and 10% FCS (JRH Biosciences, Lenexa, KS), and cultured at 37°C and 5% CO2 in air. For RT-PCR, RNA was extracted using Tri Reagent (Sigma, St. Louis, MO) according to the protocol of the manufacturer. As control cells for monitoring the sensitivity of immunobead isolation, 10-fold serial dilutions were prepared in normal blood.

Immunobead RT-PCR. Immunomagnetic beads (4 x 106; Dynabeads Epithelial Enrich, Dynal, Oslo, Norway) were added to whole blood or to the peritoneal lavage cell pellet resuspended in 10-mL sterile PBS (Dulbecco's modified). The tubes were mixed for 2 hours at 8°C and the bead-cell pellet isolated using a magnetic field. The supernatant was removed and the samples were washed twice in PBS, removing the tubes from the magnets and mixing each time before the bead cell pellet was isolated once more using a magnetic field. The bead-rosetted cells were lysed to release RNA with 0.1% IGEPAL detergent (Sigma), 10 mmol/L DTT, and 10 units of RNasin (Promega, Madison, WI) in a 15-µL volume. The samples were stored at –80°C overnight. Total RNA was reverse transcribed for 60 minutes at 37°C in a 30-µL reaction volume containing 1x First-Strand Buffer [50 mmol/L Tris-HCl (pH 8.3), 75 mmol/L KCl, 3 mmol/L MgCl2], 200 units of M-MLV reverse transcriptase (both from Life Technologies, Bethesda, MD), 750 ng pD(N)6 random hexamers (Pharmacia, Uppsala, Sweden), and 0.6 mmol/L of each deoxynucleotide triphosphate (Roche, Basel, Switzerland). The reverse transcription reaction was terminated with 3-minute incubation at 70°C.

PCR primers were designed for each marker to cross intron-exon boundaries to avoid amplification of contaminating genomic DNA (Table 1). Genomic DNA samples were checked by PCR with each set of primers to confirm that no products were amplified. PCR for each marker was done separately using the MasterCycler (Eppendorf, Hamburg, Germany). The final reaction volume was 50 µL containing a 5-µL aliquot of cDNA in 10x PCR buffer [10 mmol/L Tris-HCl (pH 8.3), 50 mmol/L KCl, 1.5 mmol/L MgCl2, 0.01% w/v gelatin (Qiagen, Melbourne, Australia)], 200 µmol/L of each deoxynucleotide triphosphate, 100 ng of each primer, 0.1 unit of HotStar Taq polymerase (Qiagen), and an additional 1 or 0.5 mmol/L MgCl2 (Qiagen) for CK20 and EphB4 reactions, respectively (Table 1). Cycling times were 15 minutes at 95°C, 40 seconds at 95°C, 40 seconds at the annealing temperature for each set of primers (Table 1), and 40 seconds extension at 74°C for 45 cycles, with a final extension cycle of 6 minutes at 74°C. The PCR-negative (no target) controls consisted of a 5-µL aliquot of RT master mix in PCR master mix and PCR master mix alone. The positive controls were samples of cell line cDNA known to express each marker. PCR products were run on a 1.5% agarose gel, transferred onto nylon membrane (Hybond N+, Amersham, Aylesbury, United Kingdom), and hybridized to a [32P]ATP end-labeled internal oligo probe. Autoradiographs were exposed for 24 to 72 hours. As a quality control, the sensitivity of the immunobead RT-PCR technique was checked to ensure that PCR products from 10 tumor cell line cells added into a 10-mL blood sample were detectable following the protocol.

Sample size calculation. The expected sample size required for statistical significance was calculated to be 146 patients (49 A-stage and 97 B-stage patients; Stata statistical software, version 7.0, 2001). This calculation was based on a P value of 0.05, a power of 0.8, an expected proportion stage I/stage II of 1:2, an expected 5-year survival proportion of 0.9 for stage I and 0.67 for stage II, and a hazard ratio of 3.8 from our previous pilot study.4

Statistical analysis. Clinical follow-up was obtained through the Cancer Registry database at this institution for the patients enrolled in the study. The database included disease stage based on histopathologic, radiologic, and clinical findings (TNM staging system), tumor differentiation, date and site of locoregional recurrence or distant secondary tumor, postoperative therapy, and date and cause of death. Univariate analysis of 10 variables (Table 2) was done using the product-limit method of Kaplan-Meier (19) and the log-rank test (20). Deaths not due to colorectal cancer were treated as censored observations. The required significance was P < 0.05. Variables were included in a Cox proportional hazards model (21) to test for independent prognostic variables. Associations were tested using Fisher's exact test. Statistical analysis was done using SPSS version 11.0 and GraphPad Prism version 4.02 for Windows (GraphPad Software, San Diego, CA; http://www.graphpad.com).


View this table:
[in this window]
[in a new window]
  		Table 2. Univariate analyses of survival using Kaplan-Meier method and log-rank statistic

 

   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
From the original cohort of 146 patients, 2 stage I and 16 stage II patients were excluded from the immunobead RT-PCR assay due to having received preoperative chemotherapy or radiotherapy, which may have affected tumor cell viability and RNA integrity. A further three patients were excluded from the survival analysis due to early death from postoperative complications at days 4, 16, and 20. Overall, there were 41 of 125 (32.8%) patients positive for one or more markers in at least one sample. This group was composed of 11 of 43 (26%) stage I and 30 of 82 (36.6%) stage II patients. Results for samples positive for individual markers in the various sample types (pre- and post-resection peripheral blood and lavage) are shown in Fig. 1 and Table 3A. MAT showed the highest frequency of detection, being positive in 26 colorectal cancer samples, followed by CEA in 20, CK20 in 7, EphB4 in 5, and LAM{gamma}2 in 4 samples. Four colorectal cancer patients were positive only for CK20, two only for EphB4, and two only for LAM{gamma}2. As a quality control measure, control cell line dilutions were also routinely tested and showed that the sensitivity of detection after autoradiography was at the level of 10 tumor cells per 10 mL of blood sample.


Figure 1
View larger version (12K):
[in this window]
[in a new window]
  		Fig. 1. Number of samples positive for each marker in colorectal cancer cases compared with benign control cases. X axis, numbers refer to sample type: 1, preoperative peripheral blood; 2, postoperative peripheral blood; 3, pre-resection lavage; 4, post-resection lavage.

 

View this table:
[in this window]
[in a new window]
  		Table 3. The number of samples positive for each marker in each sample type is shown

 
Ten variables were tested for their effect on survival outcome of the 125 patients in univariate analyses using the Kaplan-Meier method and log-rank test (Table 2). Results showed that patients who were positive for markers in post-resection peritoneal lavage samples had a mean survival of 43.9 months compared with 80.5 months for patients who were negative (P < 0.0001; Fig. 2). The tumor stage (P = 0.02) and the site of the primary tumor (colon versus rectum; P = 0.01) also had a significant effect on event-free survival in univariate analysis (Table 2). Further, the presence of disseminated marker-positive cells separated stage II patients into better or worse prognostic groups: Kaplan-Meier survival analysis for stage II patients showed a mean survival of 42.4 months for marker-positive patients compared with 78.0 months if marker negative (P < 0.0001).


Figure 2
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 2. Kaplan-Meier disease-specific survival functions: early-stage patients marker positive in post-resection lavage samples show a significantly poorer outcome compared with the marker-negative patients (P < 0.0001).

 
A multivariate analysis was done (Cox proportional hazards regression model) to test whether being positive in post-resection lavage samples was an independent prognostic factor for disease relapse. Using the backward likelihood ratio method, this factor was entered in the model with the other clinicopathologic variables of age, TNM stage, tumor site, tumor differentiation, postoperative adjuvant therapy, and gender. The Cox model showed that a positive marker result remained significant (hazard ratio, 6.2; 95% confidence interval, 1.9-19.6; P = 0.002) and was independent of the other risk factors as a prognostic marker of disease relapse. The log minus log plot of survival functions showed parallel curves, indicating that the regression analysis was valid.

Twenty-one patients, who were negative pre-resection, were positive post-resection. Ten of 21 (46.7%) relapsed post-surgery, 9 of 10 being positive for tumor markers in peritoneal lavage samples and 1 of 10 in peripheral blood; 5 patients developed distant metastases (median survival, 14.7 months; range, 4.9-31 months) and 5 developed local recurrences (median survival, 28.5 months; range 10.6-58.3 months).

Specificity of markers. Peripheral blood samples used as negative controls for the panel of candidate markers were analyzed by immunobead RT-PCR for each marker. Marker expression was not detected in 17 of 17 samples for CK20, in 47 of 47 for EphB4, in 57 of 57 for LAM{gamma}2, or in 57 of 57 for MAT. CEA was expressed in 2 of 41 samples. We found marker-positive samples in 12 of 70 (17.2%) patients having surgical resection for nonmalignant colorectal disorders (Table 3B): 1 patient (inflammatory bowel disease) was positive in pre-resection lavage fluid, 6 patients (4 diverticulitis, 2 inflammatory bowel disease) in post-resection lavage fluid, 3 (2 inflammatory bowel disease, 1 melanosis coli) in both pre- and post-resection lavage fluid, and 2 (1 inflammatory bowel disease, 1 diverticulitis) in post-resection peripheral blood. The comparison of marker-positive results in the four different sample types in colorectal cancer compared with benign controls is shown in Fig. 1. The association of marker-positive results with cancer samples (41 of 125) compared with benign samples (12 of 70) was significant (P = 0.02, Fisher's exact test).

Prediction of relapse. Of the 125 early-stage patients (median follow-up, 42.3 months; range, 2.1-85 months), 15 of 125 (12%) suffered disease relapse within the follow-up period, composed of 1 of 43 (2.3%) stage I and 14 of 82 (17%) stage II. Immunobead RT-PCR markers were positive in 12 of 15 (80%) of these patients (Table 4). There were 6 cases of distant metastases and 9 cases of locoregional recurrences. A positive marker result predicted 5 of 6 of the metastases and 7 of 9 of the locoregional recurrences, giving an overall sensitivity for prediction of relapse of 80%.


View this table:
[in this window]
[in a new window]
  		Table 4. Clinical characteristics of the 15 patients who had recurrent or metastatic disease post-resection

 
Immunobead RT-PCR markers were positive in 29 of 125 (23%) patients who did not relapse within the follow-up period (median, 25.3 months; range 4.6-61.7 months), giving a specificity of 77%. Of the 29 cases, 18 were marker positive in post-resection lavage samples, 13 in pre-resection lavage, 4 in postoperative peripheral blood, and 2 in preoperative peripheral blood (8 cases were positive in 2 sample types). MAT was positive in 17 samples, CEA in 13, EphB4 and LAM{gamma}2 in 3, and CK20 in 1. Overall, the association of recurrence or metastasis with marker-positive results (12 of 41) compared with marker negative results (3 of 84) was significant (P < 0.001, Fisher's exact test).


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
The aim of this study was to assess the suitability of the molecular markers CK20, CEA, EphB4, LAM{gamma}2, and MAT in identifying early-stage colorectal cancer patients at significant risk of developing recurrent or metastatic disease following curative resection of the primary tumor. We used the sensitive technique of immunobead RT-PCR to detect free marker-positive cells in peritoneal lavage fluid and peripheral blood in these patients both pre- and post-resection. Follow-up survival analysis has shown that the presence of such cells in post-resection peritoneal lavage fluid was associated with significantly shorter disease-free survival (P < 0.0001), and in multivariate analysis, this variable was independent of tumor stage, site of primary tumor, and tumor differentiation as a prognostic indicator.

The question of whether patients who were marker positive in peritoneal lavage samples were more likely to suffer local recurrence than distant metastasis was addressed. In our study, 36 of 125 (29%) patients were marker positive in peritoneal lavage samples (tumor bed area and pelvic floor region). Kaplan-Meier analysis showed that patients who were positive for tumor markers in cells isolated from these samples were significantly more likely to relapse (P < 0.0001). However, there was no difference in the number of locoregional recurrences (5 patients) compared with distant metastases (5) (P = 0.3, Fisher's exact test). Surgical trauma to the peritoneal surface of the bowel and adjacent tissues may promote the lodgment and growth of free tumor cells. Healing wound surfaces are rich in growth factors and cytokines and, combined with the transient immunosuppression occurring as a result of surgery, may facilitate growth of seeded tumor cells (22). Theoretically free tumor cells could also gain access to the bloodstream via damaged blood vessels before clot formation. Furthermore, release of proangiogenic factors by macrophages recruited to the wound site in the healing process may contribute to both the development of new tumor growth (23) and hematogenous spread (24).

Importantly, 21 patients, who were negative pre-resection, were positive post-resection, suggesting that tumor cells were shed as a result of surgical resection. Ten of 21 (47.6%) relapsed post-surgery with 9 of 10 positive for tumor markers in peritoneal lavage samples and 1 of 10 in peripheral blood. Several other studies have used cytology to identify free colorectal tumor cells in the peritoneal cavity. One such study found positive cytology for colorectal cancer cells in imprints of the peritoneal surface in 9.25% of cases (all stages) and this was correlated with death from disease in univariate (P = 0.008) and multivariate (P < 0.001) analyses (25). Vogel et al. (26) showed that 35.5% (32 of 90) of patients with colon cancer had positive cytology in peritoneal lavage samples. This was significantly associated with depth of invasion of the tumor and presence of metastases but not with unfavorable prognosis. In another study, positive cytology was found in 14 of 140 (10%) of patients undergoing curative resection, with all 14 patients developing local recurrences (27).

We found that more patients were marker positive in peritoneal lavage samples (36 of 125) compared with peripheral blood samples (8 of 125). Similarly, Bosch et al. (28) found 10 of 53 patients positive for tumor cells by cytology and immunocytochemistry in peritoneal lavage samples but only 3 of 53 positive in peripheral blood or mesenteric blood samples. The low frequency of tumor cells in peripheral blood is consistent with previous studies that have shown that tumor cells remain only transiently in the bloodstream. Most of the cells adhere to capillary beds within 4 hours and extravasate into the tissues with quite high efficiency (29), although the majority of these cells either undergo apoptosis (30) or remain dormant, sometimes for many years (31). Perhaps the use of multiple blood samples at different time points would improve the detection rate (32).

As there are no known tumor-specific markers for colorectal cancer, a panel of five markers that have been shown to be up-regulated in colonic epithelial tumors was used for RT-PCR analysis of the immunobead samples. Use of the immunobead technique for the prior isolation of epithelial cells has two advantages over RT-PCR using RNA from whole samples: first, the possibility of detecting free mRNA that may be present in the plasma or attached to blood cells of some cancer patients is avoided (33); second, the possibility of detecting "illegitimate" transcription from other cells (34) is substantially reduced. The tumor specificity of the CEA marker has been shown to improve when used in the immunobead RT-PCR method (12, 35) compared with the use of RNA from density gradient-separated blood cells in RT-PCR (9). As different molecular markers were expressed at different levels in any individual tumor, a panel of markers that we found to be overexpressed in most tumors was used to minimize the possibility of false-negative results.

There were 12 of 70 patients with nonmalignant disorders of the large bowel who were marker positive. A feature of inflammatory conditions, such as Crohn's disease and ulcerative colitis, is that they not only involve the process of shedding of epithelial cells, but the presence of inflammatory cytokines may be responsible for inducing much higher expression of cytokeratins and CEA in epithelial cells similar to the induction of CK19 and CEA expression in hematopoietic cells (36). The development of tumor markers with greater tumor specificity is thus an ongoing aim to reduce the incidence of false positive results.

In conclusion, we have shown that for a subgroup of patients with stage I and II colorectal cancer, detection of marker-positive cells by immunobead RT-PCR in samples of peritoneal lavage fluid taken during laparotomy was a significant risk factor for reduced survival post curative resection. This risk factor was independent of the established prognostic factors of tumor stage and site of primary tumor and may be useful in determining patients who would benefit from adjuvant chemotherapy.


   Acknowledgments
 
We thank the surgeons and theatre staff of The Queen Elizabeth Hospital Colorectal Unit for the provision of samples, M. Colbeck and V. Severin for help with the collection of follow-up data, and Janine Jones and Adrian Heard for advice on statistical analyses.


   Footnotes
 
Grant support: National Health and Medical Research Council (Australia), the Cancer Council of South Australia, and the Cancer and Bowel Research Trust (Australia).

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.

4 Unpublished data. Back

Received 7/ 6/05; revised 10/ 9/05; accepted 10/28/05.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Weitz J, Koch M, Debus J, Hohler T, Galle PR, Buchler MW. Colorectal cancer. Lancet 2005;365:153–65.[CrossRef][Medline]
  2. Australian Institute of Health and Welfare (AIHW) Australian Association of Cancer Registries (2001) http://www.aihw.gov.au/dataonline.cfm.
  3. Ries LAG, Eisner MP, Kosary CL, et al., editors. SEER Cancer Statistics Review, 1975–2002. Bethesda (MD): National Cancer Institute. http://seer.cancer.gov/csr/1975_2002/.
  4. Tsavellas G, Patel H, Allen-Mersh TG. Detection and clinical significance of occult tumor cells in colorectal cancer. Br J Surg 2001;88:1307–20.[CrossRef][Medline]
  5. Benson AB, III, Schrag D, Somerfield MR, et al. American Society of Clinical Oncology recommendations on adjuvant chemotherapy for stage II colon cancer. J Clin Oncol 2004;22:3408–19.[Abstract/Free Full Text]
  6. Hardingham JE, Kotasek D, Farmer B, et al. Immunobead-PCR: A technique for the detection of circulating tumor cells using immunomagnetic beads and the polymerase chain reaction. Cancer Res 1993;53:3455–8.[Abstract/Free Full Text]
  7. Eaton MC, Hardingham JE, Kotasek D, Sage RE, Dobrovic A. Immunobead RT-PCR: a sensitive method for detection of circulating tumor cells. Biotechniques 1997;22:100–5.[Medline]
  8. Hardingham JE, Hewett PJ, Sage RE, et al. Molecular detection of blood-borne epithelial cells in colorectal cancer patients and in patients with benign bowel disease. Int J Cancer 2000;89:8–13.[CrossRef][Medline]
  9. Castells A, Boix L, Gargallo L, Pique JM. Detection of colonic cells in peripheral blood of colorectal patients by means of reverse transcriptase and polymerase chain reaction. Br J Cancer 1998;78:1368–72.[Medline]
  10. Futamura M, Takagi Y, Koumura H, et al. Spread of colorectal cancer micrometastasis in regional lymph nodes by reverse transcriptase-polymerase chain reactions for carcinoembryonic antigen and cytokeratin 20. J Surg Oncol 1998;68:34–40.[CrossRef][Medline]
  11. Mori M, Mimori K, Ueo H, et al. Clinical significance of molecular detection of carcinoma cells in lymph nodes and peripheral blood by reverse transcription-polymerase chain reaction in patients with gastrointestinal or breast carcinomas. J Clin Oncol 1998;16:128–32.[Abstract/Free Full Text]
  12. Park S, Lee B, Kim I, et al. Immunobead RT-PCR versus regular RT-PCR amplification of CEA mRNA in peripheral blood. J Cancer Res Clin Oncol 2001;127:489–94.[CrossRef][Medline]
  13. Aoki S, Takagi Y, Hayakawa M, et al. Detection of peritoneal micrometastases by reverse transcriptase-polymerase chain reaction targeting carcinoembryonic antigen and cytokeratin 20 in colon cancer patients. J Exp Clin Cancer Res 2002;21:555–62.[Medline]
  14. Chen G, McIver CM, Texler MD, et al. Detection of occult metastases in lymph nodes from colorectal cancer patients: a multiple-marker reverse transcription-PCR study. Dis Colon Rectum 2004;47:679–86.[CrossRef][Medline]
  15. Stephenson SA, Slomka S, Douglas EL, Hewett PJ, Hardingham JE. Receptor protein tyrosine kinase EphB4 is up-regulated in colon cancer. BMC Mol Biol 2001;2:15–22.[CrossRef][Medline]
  16. Aoki S, Nakanishi Y, Akimoto S, et al. Prognostic significance of laminin-5 {gamma}2 chain expression in colorectal carcinoma: immunohistochemical analysis of 103 cases. Dis Colon Rectum 2002;45:1520–7.[Medline]
  17. Wilson CL, Matrisian LM. Matrilysin: an epithelial matrix metalloproteinase with potentially novel functions. Int J Biochem Cell Biol 1996;28:123–36.[CrossRef][Medline]
  18. Roeb E, Arndt M, Jansen B, Schumpelick V, Matern S. Simultaneous determination of matrix metalloproteinase (MMP)-7 MMP-1 -3 and -13 gene expression by multiplex PCR in colorectal carcinomas. Int J Colorectal Dis 2004;19:518–24.[CrossRef][Medline]
  19. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–81.[CrossRef]
  20. Peto R, Peto J. Asymptotically efficient rank invariant test procedures (with discussion). J R Stat Soc A 1972;135:185–9.
  21. Cox DR. Regression models and life tables. J R Stat Soc B 1972;34:187–220.
  22. Oosterling SJ, van der Bij GJ, van Egmond M, van der Sijp JRM. Surgical trauma and peritoneal recurrence of colorectal carcinoma. EJSO 2005;31:29–37.
  23. Crowther M, Brown NJ, Bishop ET, Lewis CE. Microenvironmental influence on macrophage regulation of angiogenesis in wounds and malignant tumors. J Leukoc Biol 2001;70:478–90.[Abstract/Free Full Text]
  24. Tanigawa N, Amaya H, Matsumura M, et al. Tumor angiogenesis and mode of metastasis in patients with colorectal cancer. Cancer Res 1997;57:1043–6.[Abstract/Free Full Text]
  25. Baskaranathan S, Phillips J, Mc Credden P, Solomon MJ. Free colorectal cancer cells on the peritoneal surface: correlation with pathological variables and survival. Dis Colon Rectum 2004;47:2076–9.[CrossRef][Medline]
  26. Vogel P, Ruschoff J, Kummel S, et al. Prognostic value of microscopic peritoneal dissemination: comparison between colon and gastric cancer. Dis Colon Rectum 2000;43:92–100.[CrossRef][Medline]
  27. Hase K, Ueno H, Kuranaga N, Utsunomiya K, Kanabe S, Mochizuki H. Intraperitoneal exfoliated cancer cells in patients with colorectal cancer. Dis Colon Rectum 1998;41:1134–40.[CrossRef][Medline]
  28. Bosch B, Guller U, Schnider A, et al. Perioperative detection of disseminated tumour cells is an independent prognostic factor in patients with colorectal cancer. Br J Surg 2003;90:882–8.[CrossRef][Medline]
  29. Luzzi KJ, MacDonald IC, Schmidt EE, et al. Multistep nature of metastatic inefficiency: dormancy of solitary cells after successful extravasation and limited survival of early micrometastases. Am J Pathol 1998;153:865–73.[Abstract/Free Full Text]
  30. Wong CW, Lee A, Shientag L, et al. Apoptosis: an early event in metastatic inefficiency. Cancer Res 2001;61:333–8.[Abstract/Free Full Text]
  31. Hunter KW. Host genetics and tumor metastasis. Br J Cancer 2004;90:752–5.[CrossRef][Medline]
  32. Wharton RQ, Jonas SK, Glover C, et al. Increased detection of circulating tumor cells in the blood of colorectal carcinoma patients using two reverse transcription-PCR assays and multiple blood samples. Clin Cancer Res 1999;5:4158–63.[Abstract/Free Full Text]
  33. Hasselmann DO, Rappl G, Rossler M, Ugerel S, Tilgen W, Reinbold U. Detection of tumor-associated circulating mRNA in serum, plasma and blood cells from patients with disseminated malignant melanoma. Oncol Rep 2001;8:115–8.[Medline]
  34. Chelly J, Concocordet JP, Kaplan JC, Khan A. Illegitimate transcription: transcription of any gene in any cell type. Proc Natl Acad Sci U S A 1989;86:2617–21.[Abstract/Free Full Text]
  35. Raynor M, Stephenson S-A, Walsh D, Pittman K, Dobrovic A. Optimisation of the RT-PCR detection of immunomagnetically enriched carcinoma cells. BMC Cancer 2002;2:14–8.[CrossRef][Medline]
  36. Jung R, Kruger W, Hosch S, et al. Specificity of reverse transcriptase polymerase chain reaction assays designed for the detection of circulating cancer cells is influenced by cytokines in vivo and in vitro. Br J Cancer 1998;78:1194–8.[Medline]



This article has been cited by other articles:


Home page
 Jpn J Clin OncolHome page
C. Shen, L. Hu, L. Xia, and Y. Li
Quantitative Real-time RT-PCR Detection for Survivin, CK20 and CEA in Peripheral Blood of Colorectal Cancer Patients
Jpn. J. Clin. Oncol., October 8, 2008; (2008) hyn105v1.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Surg. Oncol.Home page
S. Rekhraj, O. Aziz, S. Prabhudesai, E. Zacharakis, F. Mohr, T. Athanasiou, A. Darzi, and P. Ziprin
Can intra-operative intraperitoneal free cancer cell detection techniques identify patients at higher recurrence risk following curative colorectal cancer resection: a meta-analysis
Ann. Surg. Oncol., January 1, 2008; 15(1): 60 - 68.
[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 Lloyd, J. M.
Right arrow Articles by Hardingham, J. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lloyd, J. M.
Right arrow Articles by Hardingham, J. 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