Purpose: Matrix metalloproteinase (MMP) belongs to a large group of proteases capable of breaking essentially all components of the extracellular matrix. They are implicated in all steps of tumorogenesis, cancer invasion, and metastasis. Among them, metalloproteinase type 1 (MMP-1) is implicated in tumor invasion and metastasis in different types of cancers including colorectal cancer in which its expression was correlated with poor prognosis. A polymorphism in the promoter region of the MMP-1 gene leads to a variation of its level of transcription.

Study Design:MMP-1 -1607ins/delG and MMP-3 - 1612 ins/delA promoter polymorphisms were genotyped by multiplex PCR from 201 colorectal cancer patients. The median follow-up of patients was 30 months. The MMP genotypes were correlated to clinical outcome.

Results: Patients with the -1607insG/-1607insG MMP-1 genotype had significantly worse specific survival than the others in the whole series (P < 0.04), in stage I to III patients (P < 0.001), and in patients stage I and II (P < 0.01). In multivariate analysis, MMP-1-1607insG allele showed to be an independent poor prognostic factor after adjustment on stage, age, and the use of adjuvant chemotherapy. MMP-3 polymorphism was not associated with survival.

Conclusions: In the subgroups of nondistant metatastic patients (stages I and II, and stages I-III), an inverse relation between the number of MMP-1-1607insG allele and survival was observed suggesting a gene dosage effect. Our results are consistent with the importance of MMP-1-1607ins/delG functional polymorphism in regulating transcription level and with the relationship between MMP-1 expression and cancer invasion, metastasis, and prognosis.

Matrix metalloproteinase (MMP) belongs to a large group of proteases, which includes over 22 known human zinc-dependent proteolytic enzymes. These are capable of breaking essentially all components of the extracellular matrix (1, 2). MMPs take part in high tissue turnover and remodeling, both physiologic and pathologic conditions such as cancer. MMPs are also implicated in all steps of tumorogenesis, cancer invasion, and metastasis (3). Tumor invasion and metastasis formation always begin with blood and lymphatic vessel infiltration. As these processes involve proteolysis of the extracellular matrix, MMPs are suggested to play a major role in tumor progression.

Colorectal cancer is the most common malignancy of the gastrointestinal tract. It is the third cause of cancer overall and the second leading cause of cancer-related death in the Europe and the United States with an incidence of 300,000 new cases (4). It still carries a relatively poor prognosis despite recent improvements in early diagnosis, surgical techniques, and chemotherapy. About 40% of the patients who undergo operative resections will die within 5 years due to recurrent disease or metastases. MMPs are essential for tumor cells to penetrate the basement membrane and colonize distant sites. Their role was shown in colorectal cancer (5, 6). Among MMPs, MMP-1 is the most ubiquitously expressed interstitial collagenase. Overexpression of MMP-1 was implicated in tumor invasion and metastasis in different types of cancers including colorectal cancer (7) and its expression in colorectal cancer tumor cells was correlated with poor prognosis (8). Rutter et al. (9) reported in melanoma a functional single nucleotide polymorphism 1607del/insG)in the MMP-1 promoter with a guanine (G)insertion. It creates an Ets binding site, 5-GGA-3, which increases the promoter activity. Thus, the MMP-1-1607insG allele has significantly higher transcriptional activity than the MMP-1-1607delG allele. An association between the presence of MMP-1-1607insG allele in the MMP-1 promoter and cancer susceptibility was described in ovarian (10) endometrial (11), and lung (12) cancers. Furthermore in melanoma (13), cervical cancer (14), and colorectal cancer (15), the presence of MMP1-1607insG allele was significantly associated with stage or invasiveness of the tumor.

MMP-3 is another important MMP interacting with tumor microenvironment. Its activities are complementary to MMP-1. It is located on the cluster gene 11q22, close to MMP-1 gene. A MMP-3-1612del/insA polymorphism in the promoter of the MMP-3 gene was described (16). In vitro analysis showed that the MMP-3-1612insA allele is associated with low transcription levels (17). In a case-control study, the frequency of the MMP-3-1612delA allele was higher in the breast cancer group than in controls and was more prevalent in the metastatic group than in the nonmetastatic one (18) suggesting a link between the presence of the MMP-3-1612delA allele with breast cancer susceptibility and prognosis. In colorectal cancer case-control studies, discrepant results were observed concerning the association between MMP-3 polymorphism and cancer risk (19).

It is noteworthy that the haplotypes MMP-1-1607insG/MMP-3-1612insA and MMP-1-1607delG /MMP-3-1612delA are more frequently found than expected (20, 21) leading to a preferred association of the higher and the lower transcriptional activity alleles for MMP-1 and MMP-3 genes, respectively.

To evaluate the role of the single nucleotide promoter functional polymorphisms of MMP-1 and MMP-3 genes on patient survival in colorectal cancer, we genotyped them in 201 colorectal cancer patients. We found that MMP-1-1607insG allele was associated with a worse prognosis independently from tumor stage.

Between April 1995 and May 2002, among the consecutive patients with colorectal cancer referred to the department of General and Oncological surgery of Georges Pompidou Hospital (tertiary care hospital), 357 newly diagnosed patients, without preoperative radiotherapy or previous cancer history were eligible. Among them, an informed consent was obtained from the 201 patients enrolled in this study. There were 96 males and 105 females, sex ratio was 1.1, and mean age 69 ± 11 years (range, 35-95 years). The median follow-up was 30 months (range, 3-87 months). Tumor locations were proximal in 63 cases (cephalad to the splenic flexure), distal colon in 98 cases, and rectal in 40 cases. According to the tumor-node-metastasis classification, 22, 63, 52, and 64 patients were, respectively, classified stage I, stage II, stage III, and stage IV. Chemotherapy regimens based on 5-fluorouracil were administrated in 48 patients in stages III and IV, depending on age and physical conditions. Seventy other patients received palliative chemotherapy for recurrence. Samples from tumor and nontumor colorectal tissues were harvested during surgery. Normal tissues were collected at least 5 cm distant from tumor. Samples were stored at −80°C until DNA extraction was done using the Qiamp tissue extraction kit (Qiagen, Courtaboeuf, France).

MMP-1 and MMP-3 Genotyping

Multiplex PCR was done on non tumor tissues allowing coamplification of MMP-1 and MMP-3 fragments using the following primers: 5′6FAM-CCCTCTTGAACTCACATGTTATG-3′ and 5′-ACTTTCCTCCCCTTATGGATTCC-3′ for MMP-1 and 5-TCCTCATATCAATGTGGCCAAA- 3′ and 5′6FAM-CGGCACCTGGCCTAAAGAC-3′ for MMP-3 as previously described (22). Briefly, 40 ng of genomic DNA were amplified by 0.5 units of HotStar Taq polymerase (Qiagen, Les Ulis, France), 200 nmol/L deoxynucleoside triphosphates, 2 mmol/L MgCl2, and 300 nmol/L of each primer on a GeneAmp PCR system 9700 (Applied Biosystems, Courtaboeuf, France). Fragments were separated after dilution on an ABI 310 genetic analyzer (Applied Biosystems).

To rule out classification errors from our MMP-1 and MMP-3 genotyping methods, first we genotyped MMP-1 on a subset of 20 controls with another published method (12). We reproduce individually the same genotyping results. Second, we validated our MMP-3 genotyping method by characterizing the 23 CEPH individuals, for whom MMP-3 genotype is available in the National Center for Biotechnology Information single nucleotide polymorphism database. We found identical genotyping data.

Statistical Analysis

Statistical tests were done with STATA software (STATA 7.0; College Station, TX). The χ2 or Fisher's two-tailed exact tests were used to determine differences in the repartition of the genetic polymorphisms analyzed among groups of patients according to tumor stage and tumor location and in the relationship between each categorical variable with MMP-1 and MMP-3 promoter polymorphisms.

Deviation from Hardy-Weinberg equilibrium was tested using χ2 test with 1 df.

Survival and End Points

The patients were followed every 3 months by clinical examination during the first three postoperative years and then every 6 months. An abdominal ultrasonography and a chest X-ray were done every 3 and 6 months, respectively, in nonmetastatic patients. Survival and progression end points were determined by reviewing medical records and interviewing patient's general physician or patient's family. None of the patients were lost of follow-up. All clinical data were reviewed by three of the authors (T.L., A.B., and F.Z.) without knowledge of molecular data.

Survival was calculated as the time from surgery to death from any cause for overall survival, to death from cancer for cancer-specific survival, to recurrence for disease-free survival, or to the date of last follow-up. For overall survival, all deaths, irrespective of cause, were considered events. For cancer-specific survival, only cancer-related deaths were considered; data concerning the patients who died from other causes or who were still alive at the end of our study were censored. Survival curves were constructed using the Kaplan-Meier method and compared using the log-rank test.

Multivariate analyses were done by Cox proportional hazards model. The tumor tumor-node-metastasis, age and tumor location, and MMP-1 and MMP-3 promoter polymorphisms were included in the multivariate Cox regressions.

Among the 201 patients included in this series, DNA extraction failed in 13 patients, and genotyping could not be achieved in 7 and 10 patients for MMP-1 and MMP-3 polymorphism, respectively. Histologic and clinical data from the patients are given according to MMP-1 and MMP-3 genotype in Table 1. There were no significant differences in the frequency of MMP-1 and MMP-3 genotypes according to gender, age, tumor location, and stage (Table 1).

Table 1

Genotype frequencies of MMP-1 and MMP-3 according to gender, age, tumor location, and stage

MMP-1-1607
MMP-3-1612
delG/delG, n (%)delG/insG, n (%)insG/insG, n (%)NDPdelA/delA, n (%)delA/insA, n (%)insA/insA, n (%)NDP
Gender           
    M 19 (20) 36 (40) 35 (40) 0.08 24 (25.5) 42 (44.6) 28 (29.9) 11 0.33 
    F 35 (35) 37 (37) 28 (28)  16 (18.2) 38 (43.2) 34 (38.6)  
Location           
    Right 14 (24.5) 20 (35) 23 (40.5) 0.50 14 (25) 23 (41) 19 (34) 0.66 
    Left 26 (28) 40 (42) 27 (28)  19 (21) 38 (41.7) 34 (37.3)  
    Rectum 14 (35) 13 (32.5) 13 (32.5)  7 (20) 19 (54.3) 9 (25.7)  
Stage     NS      
    I 6 (30) 7 (35) 7 (35) 0.92 4 (21) 5 (26) 10 (53) 0.47 
    II 15 (24.5) 22 (36) 24 (39.5)  10 (18) 26 (47) 19 (35)  
    III 16 (31.5) 20 (39) 15 (29.5)  10 (21) 23 (48) 15 (31)  
    IV 17 (29) 24 (41) 17 (30)  16 (27) 26 (43) 18 (30)  
Mean age 70 ± 12 66 ± 13 70 ± 11  0.09 69 ± 13 66 ± 13 70 ± 12  0.20 
MMP-1-1607
MMP-3-1612
delG/delG, n (%)delG/insG, n (%)insG/insG, n (%)NDPdelA/delA, n (%)delA/insA, n (%)insA/insA, n (%)NDP
Gender           
    M 19 (20) 36 (40) 35 (40) 0.08 24 (25.5) 42 (44.6) 28 (29.9) 11 0.33 
    F 35 (35) 37 (37) 28 (28)  16 (18.2) 38 (43.2) 34 (38.6)  
Location           
    Right 14 (24.5) 20 (35) 23 (40.5) 0.50 14 (25) 23 (41) 19 (34) 0.66 
    Left 26 (28) 40 (42) 27 (28)  19 (21) 38 (41.7) 34 (37.3)  
    Rectum 14 (35) 13 (32.5) 13 (32.5)  7 (20) 19 (54.3) 9 (25.7)  
Stage     NS      
    I 6 (30) 7 (35) 7 (35) 0.92 4 (21) 5 (26) 10 (53) 0.47 
    II 15 (24.5) 22 (36) 24 (39.5)  10 (18) 26 (47) 19 (35)  
    III 16 (31.5) 20 (39) 15 (29.5)  10 (21) 23 (48) 15 (31)  
    IV 17 (29) 24 (41) 17 (30)  16 (27) 26 (43) 18 (30)  
Mean age 70 ± 12 66 ± 13 70 ± 11  0.09 69 ± 13 66 ± 13 70 ± 12  0.20 

Abbreviation: ND, nondetermined.

MMP-1 and MMP-3 Genotypes

MMP-1-1607delG and MMP-1-1607insG allele frequencies were 0.48 and 0.52, respectively, and for MMP-3-1612delA and MMP-3-1612insA allele frequencies were 0.44 and 0.56, respectively. Genotype distribution of MMP-3 was consistent with Hardy-Weinberg equilibrium, whereas genotype distribution of MMP-1 was not (P < 0.002). The number of MMP-1 heterozygous patients was less than expected. Thereby, we were not able to calculate the linkage disequilibrium between MMP-1 and MMP-3 polymorphisms that was observed in previous published studies (21, 22).

Survival Analysis

In a first attempt, we considered the whole series including patients with metastasis (n = 190). The median survival for homozygous MMP-1-1607delG patients was not reached, the median survivals were 76 months and 42 months for heterozygous and homozygous MMP-1-1607insG patients, respectively (Fig. 1A, P < 0.09 log-rank test). A significant difference for specific survival between MMP-1-1607delG homozygous patients, heterozygous patients and MMP-1-1607insG homozygous was observed (Fig. 1B, P < 0.05 log-rank test).

Fig. 1

A, overall survival for MMP-1 genotypes; P < 0.09, log-rank test. B, specific survival for MMP-1 genotypes; P < 0.05, log-rank test.

Fig. 1

A, overall survival for MMP-1 genotypes; P < 0.09, log-rank test. B, specific survival for MMP-1 genotypes; P < 0.05, log-rank test.

Close modal

In a second analysis, we excluded stage IV patients. In these remaining patients (n = 132), significant differences were observed between MMP-1-1607delG homozygous patients, heterozygous patients and MMP-1-1607insG homozygous patients for overall survival (P = 0.007 log-rank test; Fig. 2A) as well as for specific survival (P < 0.001, log-rank test; Fig. 2B) and for disease-free survival (P = 0.005, log-rank test; Fig. 2C).

Fig. 2

A, overall survival of nonmetastatic patients for MMP-1 genotypes; P = 0.007, log-rank test. B, specific survival of nonmetastatic patients for MMP-1 genotypes; P < 0.001, log-rank test. C, disease-free survival of nonmetastatic patients for MMP-1 genotypes; P = 0.005, log-rank test.

Fig. 2

A, overall survival of nonmetastatic patients for MMP-1 genotypes; P = 0.007, log-rank test. B, specific survival of nonmetastatic patients for MMP-1 genotypes; P < 0.001, log-rank test. C, disease-free survival of nonmetastatic patients for MMP-1 genotypes; P = 0.005, log-rank test.

Close modal

A multivariate analysis was done. We found that MMP-1 genotype was an independent prognostic factor of overall survival (Table 2) after adjustment on stage age and the use of adjuvant chemotherapy.

Table 2

Survival analyses according to MMP-1 and MMP-3 genotypes

VariableCategoriesnEventsTime at riskCrude hazard ratio95% Confidence intervalPAdjusted hazard ratio95% Confidence intervalP
MMP-1           
Overall survival           
 -1607delG/-1607delG 37 1691     
 -1607delG/-1607insG 49 13 1840 2.45 0.87-6.89 0.088 3.8 1.3-11.0 0.014 
 -1607insG/-1607insG 46 20 1595 4.24 1.57-11.42 0.004 5.4 2-14.7 0.001 
Cancer-specific survival           
 -1607delG/-1607delG 37 1691     
 -1607delG/-1607insG 49 1840 1.6 0.38-6.73 0.51 2.09 0.48-9.02 0.32 
 -1607insG/-1607insG 46 16 1595 5.86 1.69-20.38 0.005 8.52 2.38-30.6 0.001 
Disease-free survival           
 -1607delG/-1607delG 37 1626      
 -1607delG/-1607insG 49 1713 1.09 0.33-3.59 0.88 1.25 0.37-4.22 0.71 
 -1607insG/-1607insG 46 16 1374 3.46 1.26-9.48 0.016 4.1 1.47-11.44 0.007 
MMP-3           
Overall survival           
 -1612delA/-1612delA 24 1071     
 -1612delA/-1612insA 54 18 1993 2.32 0.77-6.9 0.13 2.33 0.78-6.96 0.12 
 -1612insA/-1612insA 44 12 1550 2.15 0.68-6.7 0.18 2.36 0.74-7.5 0.14 
Cancer-specific survival           
 -1612delA/-1612delA 24 1071     
 -1612delA/-1612insA 54 10 1993 1.67 0.45-6.18 0.44 1.8 0.48-6.7 0.37 
 -1612insA/-1612insA 44 1550 0.52-7.6 0.3 2.18 0.56-8.4 0.26 
Disease-free survival           
 -1612delA/-1612delA 24 1046     
 -1612delA/-1612insA 54 11 1735 1.2 0.41-3.47 0.73 1.24 0.42-3.6 0.68 
 -1612insA/-1612insA 44 1450 0.33-3.16 0.95 1.07 0.34-3.3 0.9 
VariableCategoriesnEventsTime at riskCrude hazard ratio95% Confidence intervalPAdjusted hazard ratio95% Confidence intervalP
MMP-1           
Overall survival           
 -1607delG/-1607delG 37 1691     
 -1607delG/-1607insG 49 13 1840 2.45 0.87-6.89 0.088 3.8 1.3-11.0 0.014 
 -1607insG/-1607insG 46 20 1595 4.24 1.57-11.42 0.004 5.4 2-14.7 0.001 
Cancer-specific survival           
 -1607delG/-1607delG 37 1691     
 -1607delG/-1607insG 49 1840 1.6 0.38-6.73 0.51 2.09 0.48-9.02 0.32 
 -1607insG/-1607insG 46 16 1595 5.86 1.69-20.38 0.005 8.52 2.38-30.6 0.001 
Disease-free survival           
 -1607delG/-1607delG 37 1626      
 -1607delG/-1607insG 49 1713 1.09 0.33-3.59 0.88 1.25 0.37-4.22 0.71 
 -1607insG/-1607insG 46 16 1374 3.46 1.26-9.48 0.016 4.1 1.47-11.44 0.007 
MMP-3           
Overall survival           
 -1612delA/-1612delA 24 1071     
 -1612delA/-1612insA 54 18 1993 2.32 0.77-6.9 0.13 2.33 0.78-6.96 0.12 
 -1612insA/-1612insA 44 12 1550 2.15 0.68-6.7 0.18 2.36 0.74-7.5 0.14 
Cancer-specific survival           
 -1612delA/-1612delA 24 1071     
 -1612delA/-1612insA 54 10 1993 1.67 0.45-6.18 0.44 1.8 0.48-6.7 0.37 
 -1612insA/-1612insA 44 1550 0.52-7.6 0.3 2.18 0.56-8.4 0.26 
Disease-free survival           
 -1612delA/-1612delA 24 1046     
 -1612delA/-1612insA 54 11 1735 1.2 0.41-3.47 0.73 1.24 0.42-3.6 0.68 
 -1612insA/-1612insA 44 1450 0.33-3.16 0.95 1.07 0.34-3.3 0.9 

NOTE. n, no. patients; events, no. events.

Adjusted hazard ratio on age, use of adjuvant chemotherapy, and stage.

In a third analysis, we only included patients with stage I and II tumors (n = 81) in which the use adjuvant chemotherapy is matter of debate. Significant differences were observed among MMP-1-1607delG homozygous patients, MMP-1-1607delG heterozygous patients, and MMP-1-1607insG homozygous patients in overall survival (P < 0.02, log-rank test; Fig. 3A) as well as for specific survival (P < 0.01, Fig. 3B) and for disease-free survival (P < 0.03, log-rank test; Fig. 3C). The MMP-1 genotype was found to be an independent prognostic factor of survival after adjustment on age, stage, and the use of chemotherapy. Because no death occurred during the observations period time for MMP-1-1607delG homozygous patients taken as reference group, we were not able to calculate the odd ratio of the different genotypes.

Fig. 3

A, overall survival of stage I and stage II patients for MMP-1 genotypes; P < 0.02, log-rank test. B, specific survival of stage I and stage II patients for MMP-1 genotypes; P < 0.01, log-rank test. C, disease-free survival of stage I and stage II patients for MMP-1 genotypes; P < 0.03, log-rank test.

Fig. 3

A, overall survival of stage I and stage II patients for MMP-1 genotypes; P < 0.02, log-rank test. B, specific survival of stage I and stage II patients for MMP-1 genotypes; P < 0.01, log-rank test. C, disease-free survival of stage I and stage II patients for MMP-1 genotypes; P < 0.03, log-rank test.

Close modal

Similar analyses were done for MMP-3 genotypes. No significant differences were observed in overall, specific, or disease-free survival according to the different MMP-3 genotypes.

We investigated the influence of MMP-1 and MMP-3 promoter polymorphisms known to modulate MMP-1 and MMP-3 expression level (23), as prognostic factors in a series of 201 colorectal cancer patients. The frequencies of the different MMP-1 and MMP-3 alleles were consistent with those observed by others in Caucasian populations (12, 19). However, the distribution of MMP-1 genotypes in our series was not consistent with the Hardy-Weinberg equilibrium. This departure from the Hardy-Weinberg equilibrium either could be due to the association of the MMP-1 gene polymorphism with colorectal cancer (21) or could indicate the existence of stratification factors in our series of colorectal cancer patients. When we studied the MMP-1 genotypes distribution according to tumor stage, the distribution was consistent with the Hardy-Weinberg equilibrium in stages III and IV, whereas it was not in stages I and II, suggesting an association of one allele with prognosis. The MMP-1-1607insG and MMP-3-1612insA alleles and the MMP-1-1607delG and MMP-3-1612delA alleles where found in a linkage disequilibrium in several studies, the D′ coefficient varies from 0.50 to 0.92 according to the ethnic origin of the patients (21, 22). Our first intention was to investigate the MMP-1 and MMP-3 haplotypic effects on colorectal cancer patients. However, the absence of Hardy-Weinberg equilibrium for MMP-1 genotype distribution did not allow us to infer MMP-1 and MMP-3 haplotype frequencies. Thus, we separately investigated the role of each genotype on patient survival.

In 1996, Murray et al. (8) found that the MMP-1 protein expression in colorectal cancer cell was associated with a poor prognosis. This prognostic factor was independent of tumor-node-metastasis classification. Moreover, in 2001 Ghilardi et al. (15) found that the MMP-1-1607insG promoter allele may favor the tumor growth and the metastatic spread in colorectal cancer patients and then could be a factor of worse prognosis. In our study, considering the whole series, we found a significantly worse cancer-specific survival in the MMP-1-1607insG homozygous patients group as compared with both heterozygous and MMP-1-1607delG homozygous patients groups. When we excluded metastatic patients at the time of diagnosis, a significant inverse relation was observed between the number of MMP-1-1607insG allele and survival among colorectal cancer patients suggesting a gene dosage effect. This inverse relation was observed for overall, specific, and disease-free survivals. These results were confirmed in a multivariate analysis in which MMP-1 genotype was found to be an independent prognostic factor of overall survival after adjustment on stage, age, and the use of adjuvant chemotherapy (Table 2).

One of the most important issues of colon cancer patient management is the determination of prognostic factors in early stages of the disease in order to determine which patients are at risk of recurrence or poor prognosis and then would benefit from a chemotherapy regimen. When we focused on stages I and II, patients bearing the MMP-1-1607insG allele showed a poorevolution whereas no death was observed for the MMP-1-1607delG homozygous patients.

Concerning the different MMP-3 genotype patients groups, we were unable to show any relation between the different patient genotype groups and survival. In a previous study, we showed in head and neck cancer patients that MMP-3-1612insA homozygous patients had a better response to 5-fluorouracil cisplatin–based chemotherapy regimens (22). In the present series, our results seem to be independent to the administration of 5-fluorouracil–based adjuvant chemotherapy as shown in multivariate analyses.

Our results are consistent with the importance of MMP-1-1607ins/delG functional polymorphism in regulating transcription level and with the relationship between MMP-1 expression and cancer invasion, metastasis and prognosis described by others (13–15). Furthermore, it is interesting to underline that the genotype analysis was achieved in nontumor tissues suggesting that this prognostic factor could be determine before tumor staging, independently of the tumor characteristics themselves.

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
Nagase H, Woessner JF Jr. Matrix metalloproteinases.
J Biol Chem
1999
;
274
:
21491
–4.
2
Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry.
Circ Res
2003
;
92
:
827
–39.
3
Yoon SO, Park SJ, Yun CH, Chung AS. Roles of matrix metalloproteinases in tumor metastasis and angiogenesis.
J Biochem Mol Biol
2003
;
36
:
128
–37.
4
Midgley R, Kerr D. Colorectal cancer.
Lancet
1999
;
353
:
391
–9.
5
Leeman MF, Curran S, Murray GI. New insights into the roles of matrix metalloproteinases in colorectal cancer development and progression.
J Pathol
2003
;
201
:
528
–34.
6
Zucker S, Vacirca J. Role of matrix metalloproteinases (MMPs) in colorectal cancer.
Cancer Metastasis Rev
2004
;
23
:
101
–17.
7
Hewitt RE, Leach IH, Powe DG, Clark IM, Cawston TE, Turner DR. Distribution of collagenase and tissue inhibitor of metalloproteinases (TIMP) in colorectal tumours.
Int J Cancer
1991
;
49
:
666
–72.
8
Murray GI, Duncan ME, O'Neil P, Melvin WT, Fothergill JE. Matrix metalloproteinase-1 is associated with poor prognosis in colorectal cancer.
Nat Med
1996
;
2
:
461
–2.
9
Rutter JL, Mitchell TI, Buttice G, et al. A single nucleotide polymorphism in the matrix metalloproteinase-1 promoter creates an Ets binding site and augments transcription.
Cancer Res
1998
;
58
:
5321
–5.
10
Kanamori Y, Matsushima M, Minaguchi T, et al. Correlation between expression of the matrix metalloproteinase-1 gene in ovarian cancers and an insertion/deletion polymorphism in its promoter region.
Cancer Res
1999
;
59
:
4225
–7.
11
Ishioka S, Sagae S, Sugimura M, Nishioka Y, Kobayashi K, Kudo R. Clinical factors and biomarkers which affect a new universal grading system for ovarian epithelial carcinoma.
J Obstet Gynaecol Res
2001
;
27
:
313
–8.
12
Zhu Y, Spitz MR, Lei L, Mills GB, Wu X. A single nucleotide polymorphism in the matrix metalloproteinase-1 promoter enhances lung cancer susceptibility.
Cancer Res
2001
;
61
:
7825
–9.
13
Ye S, Dhillon S, Turner SJ, et al. Invasiveness of cutaneous malignant melanoma is influenced by matrix metalloproteinase 1 gene polymorphism.
Cancer Res
2001
;
61
:
1296
–8.
14
Nishioka Y, Sagae S, Nishikawa A, Ishioka S, Kudo R. A relationship between matrix metalloproteinase-1 (MMP-1) promoter polymorphism and cervical cancer progression.
Cancer Lett
2003
;
200
:
49
–55.
15
Ghilardi G, Biondi ML, Mangoni J, et al. Matrix metalloproteinase-1 promoter polymorphism 1G/2G is correlated with colorectal cancer invasiveness.
Clin Cancer Res
2001
;
7
:
2344
–6.
16
Ye S, Watts GF, Mandalia S, Humphries SE, Henney AM. Preliminary report: genetic variation in the human stromelysin promoter is associated with progression of coronary atherosclerosis.
Br Heart J
1995
;
73
:
209
–15.
17
Ye S, Eriksson P, Hamsten A, Kurkinen M, Humphries SE, Henney AM. Progression of coronary atherosclerosis is associated with a common genetic variant of the human stromelysin-1 promoter which results in reduced gene expression.
J Biol Chem
1996
;
271
:
13055
–60.
18
Ghilardi G, Biondi ML, Caputo M, et al. A single nucleotide polymorphism in the matrix metalloproteinase-3 promoter enhances breast cancer susceptibility.
Clin Cancer Res
2002
;
8
:
3820
–3.
19
Biondi ML, Turri O, Leviti S, et al. MMP1 and MMP3 polymorphisms in promoter regions and cancer. Clin Chem 2000;46: 2023–4.
20
Zinzindohoue F, Blons H, Hans S, et al. Single nucleotide polymorphisms in MMP1 and MMP3 gene promoters as risk factor in head and neck squamous cell carcinoma.
Anticancer Res
2004
;
24
:
2021
–6.
21
Hinoda Y, Okayama N, Takano N, et al. Association of functional polymorphisms of matrix metalloproteinase (MMP)-1 and MMP-3 genes with colorectal cancer.
Int J Cancer
2002
;
102
:
526
–9.
22
Blons H, Gad S, Zinzindohoue F, et al. Matrix metalloproteinase 3 polymorphism: a predictive factor of response to neoadjuvant chemotherapy in head and neck squamous cell carcinoma.
Clin Cancer Res
2004
;
10
:
2594
–9.
23
Ye S. Polymorphism in matrix metalloproteinase gene promoters: implication in regulation of gene expression and susceptibility of various diseases.
Matrix Biol
2000
;
19
:
623
–9.