| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Finsen Laboratory, Rigshospitalet, DK-2100 Copenhagen, Denmark [M. N. H-A., R. W. S., I. J. C., N. B.]; Department of Surgical Gastroenterology, Hvidovre Hospital, 2650 Hvidovre, DK-2650 Denmark [H. J. N.]; School of Biological Sciences, University of East Anglia, Norwich, NR4 7T3 United Kingdom [G. M.]; and Laboratory of Pathology, National Cancer Institute, Bethesda, Maryland 20892-1500 [W. S-S.]
ABSTRACT
The objective of the present study was to measure preoperative plasma tissue inhibitor of metalloproteinase (TIMP)-1 levels in colorectal cancer patients and relate these values to clinical and biochemical patient characteristics. TIMP-1 levels were determined by ELISA in EDTA plasma samples collected preoperatively from 588 colorectal cancer patients. Plasma TIMP-1 levels were distributed with a median value of 141.1 µg/liter (range, 53.7–788.7 µg/liter). Whereas no significant differences were found in the median plasma TIMP-1 levels among patients with Dukes’ stage A, B, and C disease, patients with Dukes’ stage D disease had significantly higher plasma TIMP-1 levels (P < 0.0001); however, high plasma TIMP-1 levels were not restricted to advanced disease. A relatively weak correlation between plasma TIMP-1 level and age was found (r = 0.35; P < 0.0001). There was no significant difference in TIMP-1 levels between males and females (P = 0.97). Univariate analysis demonstrated an increasing risk of mortality with increasing TIMP-1 levels [scored as the loge(TIMP-1); hazard ratio = 3.3; 95% confidence interval, 2.6–4.2; P < 0.0001]. Including covariates (Dukes’ stage, primary tumor location, gender, age, plasminogen activator inhibitor type 1, and soluble urokinase plasminogen activator receptor) in a multivariate analysis, TIMP-1 was retained in the final model (hazard ratio = 2.5; 95% confidence interval, 1.7–3.7; P < 0.0001). This study showed a highly significant association between preoperative plasma TIMP-1 levels and survival in colorectal cancer patients, with higher plasma TIMP-1 levels being associated with poor outcome. Independent of clinical parameters including Dukes’ stage, plasma TIMP-1 levels were found to strongly predict prognosis of colorectal cancer patients. Additional studies are needed to validate the clinical usefulness of plasma TIMP-1 measurements.
INTRODUCTION
Evaluation of the potential for a given tumor to be invasive and threaten patient survival by metastatic dissemination is a formidable task and is not yet reliably approachable by assessment of any known tumor markers. This problem, however, can perhaps be better approached through collection of data on those marker molecules specifically necessary for the invasive process. For example, aggressive malignant spread of tumor cells requires the secretion of proteolytic enzymes by either the tumor cells or the nonmalignant cells in the surrounding tumor stroma to cleave components of the extracellular matrix and basement membranes (1) . High tumor tissue levels of the serine proteinase uPA3 and its receptor (uPAR) have recently been shown to be correlated to with shorter survival in patients with colorectal cancer (2 , 3) after a similar correlation had been demonstrated in patients with breast cancer (4, 5, 6) . Recently, we have also demonstrated the prognostic significance of suPAR measured in preoperative plasma of colorectal cancer patients (7) . Surprisingly, high preoperative plasma levels of the main uPA inhibitor, PAI-1, have been shown to be associated with shorter survival of colorectal cancer patients (8) , and high tumor tissue levels of PAI-1 have in several studies been demonstrated to have prognostic significance in breast cancer (9, 10, 11, 12) .
The MMPs are a family of zinc-dependent, neutral pH optima, extracellular endopeptidases that can collectively degrade almost all components of the extracellular matrix (13, 14, 15) . The activity of these proteinases has been demonstrated in normal tissue development and in several pathological conditions. The activities of the MMPs are regulated in situ by a family of low molecular weight endogenous inhibitors referred to as the TIMPs. This family currently consists of four members, TIMP-1, TIMP-2, TIMP-3, and TIMP-4 (16, 17, 18, 19, 20) . Recent publications have reported that TIMP-1, in both free and complexed forms, is found in increased amounts in tumor stroma and basement membranes in colorectal cancer tissue (21, 22, 23, 24) and that high expression of TIMP-1 in these tissues is correlated to poor prognosis (25) . This finding is antithetical to the expected anti-invasive activity of this inhibitor (26 , 27) and suggests that TIMP-1 may have additional and as yet poorly characterized biological activities, such as growth stimulation (17 , 28, 29, 30) and inhibition of apoptosis (31 , 32) , which were recently demonstrated to be independent of the antiproteolytic activity of the inhibitor. Thus, by stimulating tumor cell growth and/or by inhibiting apoptosis leading to increased survival of malignant cells, high TIMP-1 levels in cancer patients may result in a short survival. It is of interest to note that a similar effect of TIMP-2 in apoptosis could not be demonstrated (32) .
In the present study, we measured TIMP-1 levels in preoperatively collected plasma samples from 588 colorectal cancer patients and related these values to patient outcome. This study is the first to demonstrate a highly significant association between preoperative plasma TIMP-1 levels and cancer patient survival independent of clinical parameters including Dukes’ stage.
PATIENTS AND METHODS
Patients.
A total of 588 patients undergoing elective surgery for colorectal
cancer were included in the study. Blood samples were obtained
preoperatively with informed consent from all patients in accordance
with the Helsinki Declaration, and the study was approved by the
Central National Ethical Committee. All patients had
histologically verified adenocarcinoma of the colon or rectum. The
diagnosis of colorectal cancer and its stage according to Dukes’
classification (33)
, with an added D group for cases with
metastatic or locally but not radically resected disease, was
established by analysis of the resected and immediately fixed (10%
phosphate-buffered formalin) bowel and biopsies from lymph nodes and
distant metastasis when present. Fifty-eight (10%) patients were
classified as having Dukes’ stage A disease, 218 (37%) patients were
classified as having Dukes’ stage B disease, 175 (30%) patients were
classified as having Dukes’ stage C disease, and 137 (23%) patients
were classified as having Dukes’ stage D disease. A total of 338
tumors were colon cancers, and 250 tumors were rectal cancers. Clinical
data such as age, gender, primary tumor localization, and survival
after surgery were registered. The median observation time of the
patients was 6.8 years (range, 5.7–7.9 years), and during the
observation period, 375 patients (64%) died. Seventeen deaths that
occurred within 1 month of surgery from postoperative complications
were censored. Recording of survival for all patients surviving 1 month
or more was based on death from all causes. The median age was 69 years
(range, 33–90 years), and there were 236 females and 352 males
represented in the patient material. Patients with Dukes’ stage A, B,
or C disease underwent complete resection of their tumors, whereas
patients with Dukes’ stage D disease had resection of their primary
tumor and distant metastases whenever possible. None of the patients
received adjuvant chemo- or radiotherapy. We have reported previously
on the association between plasma PAI-1 (8)
and plasma
suPAR (7)
content and survival in this patient
population.
Blood Samples.
Blood samples (5 ml) were collected preoperatively from all patients on
the day of their operation. Peripheral blood was drawn with minimal
stasis and collected in EDTA anticoagulant tubes (Becton Dickinson,
Mountain View, CA) in accordance with a previously described protocol
(34)
.
TIMP-1 ELISA.
TIMP-1 levels were measured in all EDTA plasma samples by a rigorously
tested sandwich ELISA described previously (34)
. In brief,
a sheep polyclonal anti-TIMP-1 antibody (35
, 36)
was used
for catching of the antigen in the ELISA wells. A murine
monoclonal anti-TIMP-1 IgG1 (MAC-15; Ref. 37
) was used for
detection of the antigen, and a rabbit antimouse
immunoglobulin/alkaline phosphatase conjugate (Dako, Glostrup, Denmark)
was used for the kinetic rate assay. Rate measurements were
collected automatically over a 1-h incubation period in a Ceres 900
plate reader (Biotek Instruments, Winooski, VT). A four-parameter
fitted standard curve calculated using Kineticalc II software was used
to determine the TIMP-1 concentration of each sample. Because
the monoclonal detection antibody MAC-15 binds both free TIMP-1 and
TIMP-1 in complex with MMPs (37)
, the total TIMP-1 content
of the measured sample captured by the sheep polyclonal anti-TIMP-1
antibody (35)
was determined by the ELISA. The intra-assay
coefficient of variation for 16 replicates of a control citrate plasma
pool measured on the same plate was 5.3%, and the inter-assay
coefficient of variation for 29 successive assays of the plasma pool
(on different days) was 6.2%.
Statistical Methods.
The SAS software package (version 6.12; SAS Institute, Cary, NC) was
used to manage patient data and perform statistical analyses. The
TIMP-1, PAI-1, and suPAR measurements were log transformed (natural
log) and treated as continuous variables for the uni- and multivariate
survival analyses. The end point for survival analysis was death from
all causes as obtained from the Danish Death Registry. The Kaplan-Meier
method was used to estimate survival probabilities, and the log-rank
test was used to test for equality of strata. For graphical
representation, the patients were stratified into groups based on the
TIMP-1 value, such that each stratum yielded an equal number of events
(deaths). This gave approximately equal statistical power to each
stratum and served to illustrate a progressive effect of increasing or
decreasing levels of TIMP-1. The Cox proportional hazards model was
applied for univariate analysis of continuous covariates and for
multivariate analysis. The assumption of proportional hazards was
verified graphically. Rank statistics were used to calculate
correlation coefficients and test hypotheses on location. Tests of
independence were done using the 
2
test. The
significance level was set to 5%.
RESULTS
TIMP-1 Levels in Patient Plasma.
Using kinetic rate ELISA, TIMP-1 levels were determined in all patient
plasma samples (n = 588). All of the plasma samples had
measurable levels of TIMP-1, and the median TIMP-1 value was found to
be 141.1 µg/liter (range, 53.7–788.7 µg/liter). The measured
TIMP-1 plasma values in the different Dukes’ stages are shown in Fig. 1
. The median TIMP-1 level for patients
with Dukes’ stage A disease was 120.7 µg/liter (mean, 137.6
µg/liter; SD, 64.8 µg/liter; range, 62.6–419.5 µg/liter). The
median TIMP-1 level for patients with Dukes’ stage B disease was 136.8
µg/liter (mean, 159.2 µg/liter; SD, 80.4 µg/liter; range,
53.7–549.8 µg/liter). The median TIMP-1 level for patients with
Dukes’ stage C disease was 131.6 µg/liter (mean, 144.6 µg/liter;
SD, 58.9 µg/liter; range, 58–410.5 µg/liter), and the median
TIMP-1 level for patients with Dukes’ stage D disease was 200.5
µg/liter (mean, 243.5 µg/liter; SD, 143.5 µg/liter; range,
81–788.7 µg/liter). TIMP-1 levels in Dukes’ stages A, B, and C were
not significantly different at the 5% level, whereas TIMP-1 levels in
patients with Dukes’ stage D were significantly higher
(P < 0.0001). However, as seen in Fig. 1
, a high
interindividual plasma TIMP-1 variability was observed in each of the
Dukes’ stages; therefore, higher TIMP-1 levels were not restricted to
advanced disease. A significant but rather weak correlation was found
between plasma TIMP-1 level and the age of the cancer patient
(r = 0.35; P < 0.0001), whereas there
was no significant difference in the levels of TIMP-1 between gender
(P = 0.94). The rank correlation between TIMP-1 and
PAI-1 plasma levels was 0.52 (n = 580;
P < 0.0001) and 0.53 between TIMP-1 and suPAR
(n = 588; P < 0.0001). A significant
difference was found between plasma TIMP-1 level and primary tumor
localization (P < 0.0001); patients with colon cancer
had significantly higher TIMP-1 plasma levels when compared with
patients with rectal cancer.
|
were calculated by stratifying
the patients into four groups according to the TIMP-1 level found in
each individual plasma sample, such that each stratum yielded an equal
number of deaths (death from all causes). These four strata were:
(a) TIMP-1 
118.7 µg/liter; (b) 118.7
µg/liter < TIMP-1 152.1 µg/liter; (c)
152.1 µg/liter < TIMP-1 223.4 µg/liter; and
(d) TIMP-1 > 223.4 µg/liter. The survival
probabilities at 24 and 48 months were 76% (95% CI, 72–84%) and
61% (95% CI, 54–68%) for stratum I and 34% (95% CI, 25–43%) and
27% (95% CI, 18–35%) for stratum IV. Thus, increasing TIMP-1 level
was associated with a continuously increasing risk of mortality.
|
and Dukes’
stage C (Fig. 4)
colon cancer patients
into three groups, respectively, according to the TIMP-1 level found in
each individual plasma sample, such that each stratum yielded an equal
number of deaths (death from all causes). The levels of TIMP-1 yielding
these three strata were as follows: (a) TIMP-1 143.2
µg/liter, 143.2 µg/liter < TIMP-1 193.4 µg/liter,
and TIMP-1 > 193.4 µg/liter for Dukes’ stage B colon cancer
patients; and (b) TIMP-1 133.0 µg/liter, 133.0
µg/liter < TIMP-1 173.2 µg/liter, and TIMP-1 >
173.2 µg/liter for Dukes’ stage C colon cancer patients.
|
|
and 6
illustrate patient survival for the
three patient strata in rectal cancer patients and colon cancer
patients, respectively. The levels of plasma TIMP-1 yielding these
strata were: (a) TIMP-1 115.8 µg/liter, 115.8
µg/liter < TIMP-1 151.7 µg/liter, and TIMP-1 >
151.7 µg/liter for rectal cancer patients; and (b)
TIMP-1 149.9 µg/liter, 149.9 µg/liter < TIMP-1 225.3 µg/liter, and TIMP-1 > 225.3 µg/liter for colon
cancer patients. For both rectal cancer patients and colon cancer
patients, the log-rank test revealed a statistically significant
difference (P < 0.0001) in survival in the three
patients strata, with a continuously increasing risk of mortality with
increasing TIMP-1 level. For univariate analysis of the individual
covariates, see Table 1
.
|
|
|
. Consistent with earlier studies, Dukes’
stage was significantly associated with survival; Dukes’ stage D
patients had a HR of 7.9 (95% CI. 5.9–10.7) compared to Dukes’ stage
B patients, and patients with Dukes’ stage C disease had a HR of 2.3
(95% CI, 1.8–3.1) compared to patients with Dukes’ stage B disease.
Patients with rectal cancer were shown to have a shorter survival than
patients with colon cancer in the multivariate model (HR = 1.4; 95%
CI, 1.1–1.8; P = 0.002). Age scored in years at entry was
not significantly correlated with survival (P = 0.06),
and neither was gender (P = 0.27). Higher plasma suPAR
was correlated with shorter survival (HR = 1.4; 95% CI, 1.0–1.9;
P = 0.08), whereas higher plasma PAI-1 levels were now
associated with longer survival (HR = 0.8; 95% CI, 0.6–0.9;
P = 0.005). Finally, the model showed that high plasma
TIMP-1 levels were independently associated with a shorter overall
survival (HR = 2.5; 95% CI, 1.7–3.7; P <
0.0001). Thus it was demonstrated that plasma TIMP-1 levels strongly
predicted the survival of colorectal cancer patients independent of
Dukes’ stage. Based on the plasma TIMP-1 values in the model, a
comparison of two hypothetical colorectal cancer patients having plasma
TIMP-1 values corresponding to the 25th and
75th percentile of the total TIMP-1 data
set (109.8 and 194.5 µg/liter, respectively) gave a HR value
of 1.7 (95% CI, 1.4–2.1). DISCUSSION
This study shows that TIMP-1 levels measured by ELISA in preoperatively obtained plasma samples are significantly associated with survival in patients with colorectal cancer independent of clinical parameters including Dukes’ stage; higher preoperative plasma TIMP-1 levels were found in patients with shorter overall survival. In a multivariate analysis, the plasma TIMP-1 level was found to be a statistically significant and independent variable for patient survival.
The measurements of plasma TIMP-1 levels in colorectal cancer patients were performed using a recently described kinetic rate ELISA (34) . This assay enabled specific detection of total TIMP-1 in plasma (EDTA, citrate, and heparin). The assay was tested rigorously, and requirements for sensitivity, specificity, stability, and good recovery were fulfilled. Using the TIMP-1 ELISA, we found in our first study (34) that TIMP-1 levels were elevated in plasma from patients with advanced colorectal cancer (Dukes’ stage D disease) and advanced breast cancer when compared with TIMP-1 levels measured in EDTA plasma from healthy individuals (median, 71.2 µg/liter; range, 58.0–91.8 µg/liter; n = 100). The present study demonstrates a high degree of interpatient variability, with some patients with early-stage colorectal cancer also having elevated plasma TIMP-1 levels.
Previously published data have shown that TIMP-1 mRNA levels in tumor tissue from patients with colorectal cancer (25) or breast cancer (38) or from patients with non-small cell lung cancer (39) were inversely associated with patient survival. Recent studies using ELISA measurements of total TIMP-1 in tumor tissue extracts from patients with breast cancer demonstrated that high protein levels of the inhibitor were also associated with short survival (40) . Furthermore, in an ELISA study by Zucker et al. (41) , it was found that patients suffering from advanced gastrointestinal cancer and patients with gynecological cancer had significantly increased plasma levels of MMP-9-TIMP-1 complex when compared with healthy donors. It was found in the same study that for patients with advanced gastrointestinal cancer (stage IV), increased plasma ratios of MMP-9:MMP-9-TIMP-1 complex were significantly associated with shorter survival. In a study of prostate cancer (42) , ELISA measurements of plasma MMP-1, MMP-3, and TIMP-1 showed significantly increased levels of MMP-3 and TIMP-1 in plasma from patients with metastatic disease when compared with the plasma levels found in patients with benign prostate hyperplasia, patients with localized prostate cancer, and healthy donors. However, to our knowledge, the present study including preoperatively collected plasma samples from 588 colorectal cancer patients is the first to demonstrate a strong association between total TIMP-1 levels measured by ELISA in preoperatively obtained plasma samples and survival of patients suffering from early-stage as well as advanced cancer.
The tissue-degrading processes that take place during cancer invasion and metastasis are dependent on the proteolytic activity of MMPs (13, 14, 15 , 43) . Several studies have demonstrated that high levels of these enzymes and their naturally occurring inhibitors are found in both tumor tissue (24 , 40 , 44) and blood (41 , 42 , 45) from cancer patients. Our findings of high TIMP-1 levels in plasma collected preoperatively from colorectal cancer patients add to the available information on proteinases and cancer. It could be speculated that the concomitant elevation of both proteinases and their inhibitors in a tumor reflects a high level of activation of the entire proteinase system (both enzymes and inhibitors) initiated during tumor growth and spread, followed by a leak of these proteins into the blood. A rise in the levels of proteinases and their inhibitors in cancer tissue is also seen in the uPA system (46) . For example, in studies of colorectal cancer tissue, high levels of both uPA and its inhibitor, PAI-1, have been found to be associated with shorter survival (46) . Likewise, it has been demonstrated that high levels of PAI-1 in blood from colorectal cancer patients were significantly associated with poor survival (8) . Moreover, it has recently been suggested that PAI-1 might be directly involved in promotion of tumor cell invasion through a mechanism involving facilitation of tumor angiogenesis (47) and/or inhibition of apoptosis (48) . Similarily, it has been demonstrated that TIMP-1 has growth-promoting activities by a growth factor-like activity (17 , 28 , 30) and/or by inhibiting apoptosis (31 , 32) , the latter of which involves a change in the regulation of apoptosis-mediating gene products. However, a study by Soloway et al. (48) using targeted mutagenesis of TIMP-1 in both tumor cells and host animals revealed that tumor cell TIMP-1 expression but not host TIMP-1 expression determined the invasive behavior of the tumor cells, i.e., loss of wild-type TIMP-1 expression conferred the invasive phenotype, suggesting a protective role of TIMP-1 in this experimental tumor model (48) .
In the present study, we have demonstrated a significant correlation between TIMP-1 and PAI-1 (rs = 0.52; P < 0.0001) and between TIMP-1 and suPAR (rs = 0.53; P < 0.0001) in blood from colorectal cancer patients. It has previously been shown that high PAI-1 plasma values were significantly but not strongly associated with poor prognosis in univariate analysis of the same patient material, whereas there was no significant association in multivariate analysis (8) . These results may suggest that these endogenous proteinase inhibitors may serve a similar function. Including PAI-1 in the present multivariate model showed that high values of TIMP-1 were still associated with poor prognosis (P < 0.0001), whereas low values of PAI-1 were now associated with poor prognosis (P = 0.005). This apparent interaction is poorly understood at present, although it is highly significant.
In conclusion, the present study has demonstrated a highly significant and Dukes’ stage-independent association between high preoperative plasma TIMP-1 levels and short survival in patients suffering from colorectal cancer. However, additional studies are needed to elucidate the biological significance of TIMP-1 in cancer progression as well as the potential clinical use of plasma TIMP-1 measurements.
ACKNOWLEDGMENTS
We thank Vibeke Jensen for technical assistance.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 Supported by the Danish Cancer Society, the
Danish Cancer Research Fund, the Biomed II (contract number
BMH4-CT96-0017), the Fabrikant Einar Willumsens Mindelegat, the
Foundation of 17.12.1981, the Dagmar Marshalls Fund, the
Wedell-Wedellsborgs Fund, the Aage & Johanne Louis-Hansen Fund, the
Danish Pharmacy Foundation of 1991, the Walter & Kristiane Christensen
Fund, and the RANX05 Colorectal Cancer Study Group. 
2 To whom requests for reprints should be
addressed, at Finsen Laboratory, Strandboulevarden 49, DK-2100
Copenhagen, Denmark. Phone: 45-35-45-56-06; Fax: 45-35-25-11-17;
E-mail: nils{at}finsenlab.dk
3 The abbreviations used are: uPA, urokinase
plasminogen activator; uPAR, uPA receptor; suPAR, soluble uPAR; TIMP,
tissue inhibitor of metalloproteinase; MMP, matrix metalloproteinase;
PAI-1, plasminogen activator inhibitor type 1; HR, hazard ratio; CI,
confidence interval.
Received 4/ 4/00; revised 8/14/00; accepted 8/18/00.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  S. O. Wurtz, S. Moller, H. Mouridsen, P. B. Hertel, E. Friis, and N. Brunner Plasma and Serum Levels of Tissue Inhibitor of Metalloproteinases-1 Are Associated with Prognosis in Node-negative Breast Cancer: A Prospective Study Mol. Cell. Proteomics, February 1, 2008; 7(2): 424 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-S. Kim, S. Y. Hwang, H.-Y. Kang, H. Sohn, S. Oh, J.-Y. Kim, J. S. Yoo, Y. H. Kim, C.-H. Kim, J.-H. Jeon, et al. Functional Proteomics Study Reveals That N-Acetylglucosaminyltransferase V Reinforces the Invasive/Metastatic Potential of Colon Cancer through Aberrant Glycosylation on Tissue Inhibitor of Metalloproteinase- Mol. Cell. Proteomics, January 1, 2008; 7(1): 1 - 14. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Varghese, M. Burness, H. Xu, T. Beresnev, J. Pingpank, and H. R. Alexander Site-Specific Gene Expression Profiles and Novel Molecular Prognostic Factors in Patients with Lower Gastrointestinal Adenocarcinoma Diffusely Metastatic to Liver or Peritoneum Ann. Surg. Oncol., December 1, 2007; 14(12): 3460 - 3471. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. M. Sorensen, P. Bystrom, I. J. Christensen, A. Berglund, H. J. Nielsen, N. Brunner, and B. Glimelius TIMP-1 Is Significantly Associated with Objective Response and Survival in Metastatic Colorectal Cancer Patients Receiving Combination of Irinotecan, 5-Fluorouracil, and Folinic Acid Clin. Cancer Res., July 15, 2007; 13(14): 4117 - 4122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Thaysen-Andersen, I. B. Thogersen, H. J. Nielsen, U. Lademann, N. Brunner, J. J. Enghild, and P. Hojrup Rapid and Individual-specific Glycoprofiling of the Low Abundance N-Glycosylated Protein Tissue Inhibitor of Metalloproteinases-1 Mol. Cell. Proteomics, April 1, 2007; 6(4): 638 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. V. Sorensen, C. Fenger, H. Winther, N. T. Foged, U. Lademann, N. Brunner, and P. A. Usher Characterization of Anti-TIMP-1 Monoclonal Antibodies for Immunohistochemical Localization in Formalin-fixed, Paraffin-embedded Tissue J. Histochem. Cytochem., October 1, 2006; 54(10): 1075 - 1086. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ruokolainen, P. Paakko, and T. Turpeenniemi-Hujanen Tissue Inhibitor of Matrix Metalloproteinase-1 Is Prognostic in Head and Neck Squamous Cell Carcinoma: Comparison of the Circulating and Tissue Immunoreactive Protein Clin. Cancer Res., May 1, 2005; 11(9): 3257 - 3264. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-S. Schrohl, M. N. Holten-Andersen, H. A. Peters, M. P. Look, M. E. Meijer-van Gelder, J. G. M. Klijn, N. Brunner, and J. A. Foekens Tumor Tissue Levels of Tissue Inhibitor of Metalloproteinase-1 as a Prognostic Marker in Primary Breast Cancer Clin. Cancer Res., April 1, 2004; 10(7): 2289 - 2298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bloomston, E. E. Zervos, and A. S. Rosemurgy II Matrix Metalloproteinases and Their Role in Pancreatic Cancer: A Review of Preclinical Studies and Clinical Trials Ann. Surg. Oncol., August 1, 2002; 9(7): 668 - 674. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. N. Holten-Andersen, I. J. Christensen, H. J. Nielsen, H. Lilja, G. Murphy, V. Jensen, N. Brunner, and T. Piironen Measurement of the Noncomplexed Free Fraction of Tissue Inhibitor of Metalloproteinases 1 in Plasma by Immunoassay Clin. Chem., August 1, 2002; 48(8): 1305 - 1313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Utsunomiya, Y. Hara, A. Kataoka, M. Morita, H. Arakawa, M. Mori, and S. Nishimura Cystatin-like Metastasis-associated Protein mRNA Expression in Human Colorectal Cancer Is Associated with Both Liver Metastasis and Patient Survival Clin. Cancer Res., August 1, 2002; 8(8): 2591 - 2594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. N. Holten-Andersen, I. J. Christensen, H. J. Nielsen, R. W. Stephens, V. Jensen, O. H. Nielsen, S. Sorensen, J. Overgaard, H. Lilja, A. Harris, et al. Total Levels of Tissue Inhibitor of Metalloproteinases 1 in Plasma Yield High Diagnostic Sensitivity and Specificity in Patients with Colon Cancer Clin. Cancer Res., January 1, 2002; 8(1): 156 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Yan and M. A. Moses A Case of Tumor Betrayal : Biphasic Effects of TIMP-1 on Burkitt's Lymphoma Am. J. Pathol., April 1, 2001; 158(4): 1185 - 1190. [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 |
