|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2004.02.110 on July 12 2004 © 2004 American Society of Clinical Oncology. Upregulation of the Tissue Inhibitor of Metalloproteinase-1 Protein Is Associated With Progression of Human Non–Small-Cell Lung CancerRoswell Park Cancer Institute, State University of New York, Buffalo, NY; University of Pennsylvania, Philadelphia, PA; H. Lee Moffitt Cancer Center, Tampa, FL; and Tufts New England Medical Center, Boston, MA Address reprint requests to Dongfeng Tan, MD, Departments of Pathology and Cancer Genetics, Roswell Park Cancer Institute, State University of New York, Buffalo, Elm and Carlton Sts, Buffalo, NY 14263; e-mail: dongfeng.tan{at}roswellpark.org
PURPOSE: Tissue inhibitors of metalloproteinases (TIMPs) are naturally occurring inhibitors of matrix metalloproteinases (MMPs). It has been shown that TIMP-1 may be a multifunctional protein. Little is known about the role of TIMP-1 in progression and metastasis of human lung cancer (tumor inhibiting or tumor promoting), although studies using a variety of techniques have analyzed the expression of TIMP-1 mRNA and/or protein in human cancers. PATIENTS AND METHODS: We examined the expression of TIMP-1 protein by immunohistochemistry in patients (n = 160) with primary respectable (stage I to IIIA) non–small-cell lung cancer (NSCLC). RESULTS: Twenty-seven percent of the tumors (43 of 160) demonstrated elevated expression of this protein. We demonstrate that overexpression of TIMP-1 protein is associated with an adverse outcome. In addition, disease stage, patient's age, and performance status were all significantly related to survival. In multivariate analyses, patients with high TIMP-1 expression had a 90% increased risk of death when compared with those with low expression (relative risk, 1.92; 95% CI, 1.19 to 3.09; P = .008). TIMP-1 expression did not correlate with expression of MMP-2 and MMP-9. CONCLUSION: These results suggest that TIMP-1, independent of its inhibiting activity of MMPs, may have other function(s) critical for NSCLCs. The significance of our results is two-fold. The adverse outcome in patients with overexpression of TIMP-1 indicates its potential prognostic value in NSCLC. Thus, TIMP-1 overexpression may serve to help identify patients with particularly aggressive disease for adjuvant treatments. In addition, the TIMP-1 molecule may represent a novel therapeutic target for treatment of some NSCLCs.
Lung cancer is the leading cause of cancer-related death in men and women in the United States.1 Non–small-cell lung cancer (NSCLC) accounts for approximately 80% of all lung cancers. The 5-year survival of patients with NSCLC remains among the lowest of all major human cancers despite recent advances in surgical, radiation, and medical treatments.1 Therefore, there is intense interest in gaining a better understanding of the molecular and cellular processes involved in this aggressive disease. Reliable biomarkers to predict relapse and poor outcome are much needed to optimize medical management for patients at high risk for recurrence. Increasing evidence indicates that imbalance of matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) plays an important role in tumor invasion and metastasis.2-5 A series of studies have established that individual MMPs, including the most widely studied MMP-2, MMP-9, MMP-13, and MMP-14, contribute to tumor invasion and progression by breaking down the basement membrane and extracellular matrix (ECM).5,6 The main physiologic inhibitors of MMPs are TIMPs, which are a family of low molecular weight proteins capable of specific inhibition of the active forms of the MMPs. At least four TIMPs (TIMP-1, TIMP-2, TIMP-3, and TIMP-4) have been characterized.7,8 They are present in ECM and bind MMPs noncovalently in a 1:1 stoichiometric complex. The net balance between the TIMPs and MMPs generally correlates with tumorigenesis. However, studies have shown that TIMP expression may actually increase with tumor progression.9,10 TIMP-1, a 28.5-kd glycoprotein, is produced by various kinds of cells and found in tissue extracts and body fluids, suggesting that it is a fundamental and ubiquitous protein in humans. The original concept of TIMPs was that as inhibitors of MMPs, they would have anti-invasive and/or antimetastatic effects. Experimental studies reported an inverse correlation between TIMP-1 and metastatic potential in in vitro and in vivo models, supporting this concept.9,10 However, increasing evidence indicates that TIMPs are multifunctional, with apparent paradoxical effects on tumor progression.11,12 There is experimental evidence suggesting that TIMPs may function in a manner that promotes rather than suppresses tumor progression. TIMP-1 and TIMP-2 are both known to have growth-factor–like properties (erythroid potentiating activity), which have been separated from their MMP inhibitory functions by structure-function studies.11,12 TIMP-1 also was shown to have antiapoptotic activity.13,14 In human colorectal cancer, higher tissue levels of TIMP-1 mRNA have been observed in association with more (rather than less) invasive stages.15 Elevated TIMP-1 RNA also was associated with an aggressive subset of NSCLCs (n = 45) in a recent Japanese study.16 The mechanisms explaining a paradoxical effect of TIMPs in tumor progression are not fully understood and currently are under intense investigation. Analysis of TIMP-2–deficient mice revealed that TIMP-2 activates proMMP-2, which facilitates increased tumor growth and invasion, explaining in part why high levels of TIMP-2 expression in human tumors may correlate with a highly invasive phenotype.17 However, the role of TIMP-1 in cancer progression, particularly its prognostic value in NSCLC, remains largely unknown. The aim of this study was to examine the expression pattern of TIMP-l protein and explore its value as a prognostic indicator in NSCLC. Because TIMP-1 is associated with MMPs and the Bcl-2 family, relevant markers (MMP-2, MMP-9, and Bcl-2) were also studied. In addition, TIMP-1 protein expression was also correlated with p53, MCM-2, and Ki-67. A cohort of pathologically staged resected primary NSCLCs (n = 160) was studied to provide a relatively defined homogenous sample set to avoid confounding variables. In our study, we demonstrate that overexpression of TIMP-1 protein is associated with an adverse outcome in NSCLC. This association is independent of MMP-2, MMP-9, and Bcl-2. The findings are consistent with the hypothesis that TIMP-1 may have a role in actually promoting rather than inhibiting progression of some NSCLCs. In addition, our results suggest that TIMP-1 may represent a novel therapeutic target for the treatment of some NSCLCs.
Study Population This study was a retrospective analysis of patients diagnosed at Roswell Park Cancer Institute (Buffalo, NY) between March 29, 1995 and September 30, 1999. Inclusion criteria for this study were patients who had primary NSCLC, had tumor stages IA to IIIA, had received surgery as initial treatment modality, and had complete clinicopathologic data. One hundred sixty patients met these criteria. Surgical treatment consisted of six wedge resections, 123 lobectomies, and 31 pneumonectomies. One hundred three patients had mediastinal lymph node sampling, whereas 153 patients had hilar lymph node dissection. Patients were observed through March 6, 2002. The median patient follow-up time in this sample was 57 months. Clinicopathologic data collected included age, sex, smoking history, performance status (using the Eastern Cooperative Oncology Group scale), date of initial diagnosis, histopathologic diagnosis, grade of tumor differentiation, pathologic tumor stage, and date of death from NSCLC or last follow-up. Lung cancer was the cause of death in 88% of patients. Histologic diagnosis and grade of differentiation were assigned in accordance with the WHO criteria for lung and pleural tumors,18 and pathologic stage was based on the revised international system.19 This study was performed under an institutional review board–approved protocol to investigate molecular markers relevant to lung cancer pathogenesis.
Specimen Preparation and Histologic Examination
Antibodies and IHC
The IHC for Bcl-2 test was performed according to a standard procedure. In brief, antigen retrieval was performed using high pH Target Retrieval Solution (TRS; Dako) and trypsin digestion according to the Dako protocol. The primary antibody for Bcl-2 (clone 124; Dako; 1:100 dilution in phosphate-buffered saline) was incubated with antigen-retrieved tissue sections for 40 minutes. Mouse Envision (Dako) was used for detection. Slides were counterstained in hematoxylin for 3 minutes, dehydrated, and covered with a coverslip. Negative controls were treated identically except primary antibodies are omitted.
Statistical Analysis Follow-up time was calculated using the potential follow-up method. Overall patient survival was calculated from date of diagnosis to date of last follow-up examination (censored) or date of death (event). Differences in survival times between patient subgroups were analyzed using the log-rank statistic. Survival probabilities were calculated using the Kaplan-Meier method. Cox proportional hazards regression analysis was used to measure the association of clinicopathologic variables to overall survival. Statistical significance for model parameters was based on the likelihood ratio test. In all tests, statistical significance (two sided) was set at 5%, with no correction made for multiple testing.
Clinicopathologic Data The clinicopathologic features of the patients studied are summarized in Table 2. The median patient age was 65.2 years (range, 37.6 to 88.3 years). Eighty-seven patients (54%) were male. One hundred forty-three patients (90%) had a history of smoking. There were 109 patients with a performance score (PS) of 0, whereas 50 patients had a PS of 1, 2, or 3. Forty six patients (30%) had lymph node involvement, whereas 107 patients (70%) did not. Among the 156 patients with weight loss information available, weight loss was present in 20 patients (13%). Adenocarcinoma (n = 92) was the primary tumor type in this patient sample, whereas squamous cell carcinoma was the second most common type (32% of the patients). There were 71 patients (44%) who had a well or moderately well-differentiated histologic grade, and 89 patients (56%) whose grade was poor or undifferentiated. There were 100 patients (63%) with tumor stages IA to IB, and 60 patients (38%) who had tumor stages IIA to IIIA.
Expression Pattern of TIMP-1 in NSCLC We first analyzed the pattern of expression of TIMP-1 in tumor cells. The TIMP-1 immunoreactive products had a fine granular appearance and localized in the cytoplasm of tumor cells. Intensity of staining varied among different patients and in different areas of the same tumor. Heterogeneous staining was noted in all types of NSCLC. Some tumors showed focal patchy positive staining; others displayed more diffuse staining patterns in tumor nests. Overall, there were 117 patients (73%) with low (none or weak) TIMP-1 protein expression, and 43 patients (27%) who had high (moderate or strong) expression. In addition, expression of TIMP-1 protein also was noted in stromal cells (endothelial cells and fibroblasts). Generally, IHC staining in stromal cells was weak, with occasional overexpression (n = 14). There was no correlation between TIMP-1 expression in tumor cells and stromal cells (data not shown). Sections incubated with irrelevant rabbit antiserum consistently showed no immunostaining. Representative photographs of TIMP-1 expression are shown in Fig 1.
Association of Overexpression of TIMP-1 With Clinicopathologic Variables Clinicopathologic variables, stratified by TIMP-1 expression, are listed in Table 2. There was a trend seen between performance status and TIMP-1, in which those with a performance status of 0 also tended to have a low expression of TIMP-1 (P = .054). Tumors with different histologic subtypes displayed a different pattern of TIMP-1 expression. For instance, in squamous cell carcinoma, the peripheral areas more frequently stained positive, whereas central keratinized regions usually stained negative. However, no correlation was observed between TIMP-1 expression and tumor subtype or tumor differentiation. There was no association between any of the other clinicopathologic variables with TIMP-1 expression (Table 2).
Association of Overexpression of TIMP-1 With MMP-2 and MMP-9
Association of Overexpression of TIMP-1 With p53 and Bcl-2 We found that the relationship between p53 expression and TIMP-1 protein expression approached statistical significance (P = .05), whereby patients with low p53 expression also tended to have low expression of TIMP-1. These variables were relatively balanced between the low and high TIMP-1 protein expression groups. Bcl-2, with a clear cytoplasmic staining, was overexpressed in 63% of patients. The frequency of Bcl-2 expression in our study was in agreement with other studies using the same method.29 However, there was no association between TIMP-1 expression and Bcl-2. Representative photos of Bcl-2 are illustrated in Figures 2E and 2F.
Association of Overexpression of TIMP-1 With MCM-2 and Ki-67
Table 4 lists the results of the univariate ordinal logistic regression analysis. TIMP-1 was modeled as the response, which consisted of eight patients with no expression, 109 patients with weak expression, 38 patients with moderate expression, and five patients with strong expression. Patients with a PS
Association of TIMP-1 With Overall Survival The estimated survival distributions were calculated by the Kaplan-Meier method. Cox proportional hazards regression analysis was used to derive risk estimates related to overall patient survival for each of the clinicopathologic factors. Overall, those with low TIMP-1 expression had a median survival time of greater than 53.4 months, whereas those with high expression had a median survival time of 37.1 months. The differences in survival times between these two groups were highly significant (P = .001). Figure 3 illustrates patient survival over time, according to TIMP-1 expression. In univariate analysis, there were highly significant associations between survival and tumor stage, PS, TIMP-1 expression level, and patient age. Patients with stage II to IIIA had almost 2.5 times greater risk of death compared with the stage IA to IB patients (relative risk [RR], 2.42; 95% CI, 1.53 to 3.78; P = .0001). Those patients whose PS was 1, 2, or 3 had 2.5 times the risk of death compared with those patients whose PS was 0 (RR, 2.53; 95% CI, 1.61 to 3.96; P < .0001). Patients with high TIMP-1 expression had more than twice the risk of death (RR, 2.11; 95% CI, 1.33 to 3.36; P = .002). Patients whose age was greater than the median also had twice the risk of death (RR, 2.11; 95% CI, 1.33 to 3.34; P = .002). Univariate analyses of clinicopathologic variables associated with survival are summarized in Table 5.
In the multivariate analyses listed in Table 6, tumor stage, patient age, TIMP-1, and PS all remained significantly related to survival. Patients with stage IIA to IIIA had more than twice the risk of death compared with the stage IA to IB patients (RR, 2.36; 95% CI, 1.48 to 3.75; P = .0003). Those patients whose age was older than the median had twice the risk of death (RR, 2.02; 95% CI, 1.27 to 3.23; P = .003). Those with high TIMP expression had a 90% increased risk of death when compared with those with low expression (RR, 1.92; 95% CI, 1.19 to 3.09; P = .008). There was an 80% increased risk of death among those whose performance score was 1, 2, or 3, compared with those with a PS of 0 (RR, 1.80; 95% CI, 1.12 to 2.91; P = .02).
We demonstrate here that TIMP-1 protein is overexpressed in a subgroup of NSCLC patients. The association of TIMP-1 overexpression in primary NSCLC with a poor clinical outcome suggests a possible critical role of TIMP-1 in progression of NSCLC. In our study, TIMP-1 was primarily expressed in the tumor cells (Fig 1). This observation is consistent with previous studies.30-32 In contrast, TIMP-1 and TIMP-2 mRNA expression was found to localize predominantly in peritumoral stromal cells in one series.33 A discrepancy between the localization of mRNA transcript by in situ hybridization and TIMP-1 protein by IHC was reported in a recent study.34 In these reports, the mRNA transcripts were confined principally to stromal cells, whereas the immunostaining showed that the protein was localized in the tumor cells. These observations emphasize the importance of localization of TIMP-1 protein by IHC studies because it is difficult to extrapolate from the measurement of mRNA to the location of a protein, particularly when the protein is known to undergo post-translational modification. To date, a number of techniques have been used to detect the presence of TIMP-1 alteration. These approaches have included analyses of tissue sections, tissue lysates, and serum-based assays.9,10,25,26,33 A major advantage of IHC is the maintenance of the original tissue architecture, which permits analysis in its histologic context, in addition to the subcellular localization of TIMP-1 protein. Moreover, IHC is of value in analyzing archival tissue, whereas other methods require fresh tissue. This aspect, along with the general availability of reagents and low cost, accounts for the popularity of IHC in clinical evaluation.26,28,35-39 However, as with any technique, IHC has its limitations. The most significant disadvantage of IHC is the loss of sensitivity secondary to antigenic alterations caused by fixation procedures, in addition to the semiquantitative nature of IHC. One of the major findings in our study is that elevated expression of TIMP-1 protein is associated with decreased survival in patients with NSCLC. In our study, TIMP-1 expression, disease stage, age, and PS were all significantly associated with survival. In the multivariate analysis, the significance of TIMP-1 remained. Patients with overexpression of TIMP-1 have a 90% increased risk of death when compared with those without overexpression of TIMP-1. This is the first time the correlation between TIMP-1 protein overexpression and poor outcome in NSCLC has been shown in a series study using IHC techniques. It has been suggested recently that TIMP-1 RNA elevation has an adverse implication in clinical outcome of NSCLC.16 Elevated serum levels of TIMP-1 also have been shown to be associated with poor prognosis in lung cancer.40 Moreover, it has been demonstrated that high levels of TIMP-1 mRNA or protein are associated with reduced disease-free and overall survival in some other human tumors, including non-Hodgkin's lymphoma, colorectal, prostate, and breast cancers.15,27,35,41 Using IHC, Yano et al42 reported that urothelial cancer patients with high expression of TIMP-1 had a worse prognosis. Most recently, it has been shown that elevated TIMP-1, independent of MMP-2 and MMP-9, was associated with a poor survival and mortality risk for the patients with ovarian carcinoma.37 The concept that TIMP-1 prompts tumor progression is particularly intriguing given the fact that TIMPs have been well recognized as inhibitors of tumor growth and metastasis by inhibiting MMPs. It is unknown at present why the elevated expression of TIMP-1 is paradoxically associated with aggressive malignant cancer cells. One explanation for this observation is that the increased expression of TIMP-1 may be inversely related to the increased expression of MMPs during the tumor-mediated degradation of ECM.15 Other explanations include the possibility that TIMP-1 may have other functions that are independent of its MMP-inhibiting effect.4,6 In our study, we failed to demonstrate an association of TIMP-1 with MMP-2 and MMP-9. This might be explained partially by the fact that the level of TIMP and MMP proteins in tissue may not represent their actual biologic activities.43,44 It has been proposed that TIMP-1 has antiapoptotic activity,13,14 which may partially explain the paradoxical role in tumor progression. It has been observed in vitro that Bcl-2 overexpression induces TIMP-1 upregulation in breast epithelial cells, whereas it has no effect on TIMP-2 expression. Recombinant TIMP-1 complexed with proMMP-9 does not bind to the cell surface and fails to promote cell growth.45 We did not find a coexpression of these two markers, indicating that the linkage or correlation between TIMP-1 and Bcl-2 in human cancer may be more complex than that in the in vitro model. Therefore, in vivo systemic assessment of TIMP-1–mediated antiapoptotic effect needs to be investigated more thoroughly and explicitly. In addition, TIMP-1 overexpression inhibits apoptosis after the loss of cell adhesion (anoikis) in MCF10A cells, suggesting that the activity of TIMP-1 is not dependent on its ability to stabilize cell-matrix interactions.13 In our study, we also found elevated coexpression of p53 and TIMP-1 in human NSCLC. p53 was correlatively overexpressed in tumors with elevated TIMP-1 (P = .05), suggesting that TIMP-1 may play a role in the complex apoptotic-antiapoptotic process. However, we may observe individual phenomena occurring independently in cancer tissue because we failed to show an association of p53 with survival. Could the possible adverse prognostic value of TIMP-1 be related to its growth-promoting activity, as originally described in experimental studies?12 It has been shown that TIMP-1 has growth-promoting properties in a variety of cells.11,46 In our study, we were unable to demonstrate a significant correlation between MCM-2 and elevated TIMP-1 protein (P = .44). This may reflect the fact that TIMP-1 as a growth-promoting factor may have a more complicated effect in human cancer than in an experimental model. Nevertheless, we found that patients with elevated TIMP-1 and MCM-2 had a shorter survival compared with those with elevated TIMP-1 alone (data not shown). Finally, high expression of TIMPs may simply covariate with some other biologic factors of essential functional importance in lung tumor progression. The recent advances in our understanding of NSCLC tumorigenesis have uncovered a number of markers bearing a potential prognostic role in NSCLC. These markers include intercellular adhesion molecules (ie, e-cadherin/catenins, CEACAM138,47), growth factors and growth factor receptors (ie, epidermal growth factor receptor and Her-2/neu39,48), nuclear factors (MDMs, cyclins, Rb protein49,50), and angiogenic factors (ie, vascular endothelial growth factor, microvascular density51,52), among others.53 Therefore, the net effect of TIMP on tumorigenesis (tumor promotion or tumor suppression) may not only depend on the forms present in the local concentration of TIMPs, cellular and pericellular distribution, bioavailability of local concentration of TIMPs in microenvironment, and the time when the TIMP is presented to the tumor cells,54 but also on its interaction with other factors. The potentially complex protein-protein interactions in tumor cells warrants additional studies using laser capture microdissection, protein microarray, and cDNA gene profiling.55 Most recently, a synchronous overexpression of epidermal growth factor receptor and Her2/neu protein has been found to predict poor outcome in patients with stage I NSCLC, emphasizing important interactions that take place among different members of the ErbB family during tumor progression.39 The combined evaluation of such factors including TIMP-1, which may be biochemically linked, is likely to provide a more accurate prediction of patient outcome in the near future. Regardless of the mechanism(s) involved in TIMP-1 overexpression, the findings in this study are in agreement with the previous suggestion that lung cancer progression and invasion are likely influenced by the TIMP-1 phenotype, which is partially determined by TIMP-1 genotype of the tumor.56 If these findings are confirmed in prospective studies, two subgroups of NSCLC based on TIMP-1 protein phenotype (overexpression v no overexpression) may provide a useful biologic staging tool, which would predict survival in this aggressive malignancy. Such a prognostic classification and stratification based on this aspect of tumor biology might be crucial for clinical decision making. Another potentially important implication of our results is the role of TIMP-1 in medical management of patients with NSCLC. The association of elevated expression of TIMP-1 with an adverse clinical outcome in patients with NSCLCs raises a concern regarding some clinical trials using MMP inhibitors (MMPIs) or tissue inhibitor promoters as an adjuvant approach to treating advanced cancers, including NSCLCs.10,57,58 Many of these MMPIs (MMI270, for example) inhibit MMPs by increasing the TIMP-1 plasma level. To date, these trials have yielded disappointing results. An increased level of TIMP-1 may be among the potential rationales for the unsuccessful clinical trials of MMPI against cancer. MMPI, like TIMP, may function in favor of tumor growth by upregulation of angiogenic factors59 or by inhibiting the angiostatin- or endostatin-converting MMPs,60,61 and therefore playing a positive role in angiogenesis. Therefore, it can not be excluded that some MMPIs might have caused unwanted adverse effects in a subgroup of patients because of the increase in the TIMP-1 level,10 the latter potentially being invasion enhancing and/or metastasis promoting. If our results in this study and similar studies of other human cancers15,30,41 can be further validated by prospective studies, TIMP-1 protein may represent a novel therapeutic target for treatment of some NSCLCs. The following considerations of this study can be made. First, although a uniform set of patients (surgically respectable NSCLC) was used for the cohort, it is possible that a subset of these tumors may have been biologically more aggressive. For instance, we found that patient age was significantly related with survival and TIMP-1, as was PS. Second, TIMP-1 and relevant markers are heterogeneously expressed in lung cancer and evaluation of IHC has its own limitation because of its semiquantitative nature. Automatic scoring of IHC will improve the sensitivity of this methodology. Lastly, in a few studies investigating the prognostic value of the TIMP-1 protein in NSCLC, the cutoff criteria (10% of tumor cells stained positive) was arbitrary and may be biased. Consequently, there is a need for prospective studies to assess the potential independent prognostic value of measuring TIMP-1 protein in relation to other factors to optimize the management of operable primary NSCLC. Thus, our results might have significant clinical relevance with regard to prognosis and therapy. Conversely, our study suggests that TIMP-1 protein in tumor tissue could potentially serve as a tumor marker to predict aggressive behavior of NSCLC. This information also could serve as a tool for clinical decision making when selecting patients for adjuvant or neoadjuvant treatments of NSCLC. The elevated TIMP-1 in a more aggressive subset of NSCLCs raises a concern regarding the use of MMPIs as adjuvant treatment in some clinical trials. Our results may provide a basis to explore TIMP-1 as a therapeutic target. However, it should be noted that invasion and metastasis of lung cancer cells is likely to be complex, requiring other epigenetic and genetic factors62-65 in addition to the metalloproteinase cascade. Therefore, high expression of the TIMP-1 may simply covariate with some other (still unknown) biologic factors of essential functional importance in lung cancer pathogenesis.
The authors indicated no potential conflicts of interest.
We thank Amy Beck, Deborah Malik, Maisie Martinick, Joan Natella, Hong Zou, and Doughas Nixon and for their technical support. We thank Paul Soloway for reviewing the initial manuscript draft and providing valuable input.
Supported in part by a grant from the Roswell Park Cancer Institute and by shared resources of the Roswell Park Cancer Institute core grant NIH P30 CA016056-27. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Greenlee RT, Hill-Harmon BB, Murray T, et al: Cancer statistics, 2001. CA Cancer J Clin, 51:16–36, 2001 2. Cho NH, Hong KP, Hong SH, et al: MMP expression profiling in recurred stage IB lung cancer. Oncogene 23:845–851, 2004[CrossRef][Medline] 3. Egeblad M, Werb Z: New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2:161–173, 2002[Medline]
4. Nelson AR, Fingleton B, Rothenberg ML, et al: Matrix metalloproteinases: Biologic activity and clinical implications. J Clin Oncol 18:1135–1149, 2000 5. Stetler-Stevenson WG: Matrix metalloproteinases in angiogenesis: A moving target for therapeutic intervention. J Clin Invest 103:1237–1244, 1999[Medline]
6. Deryugina El, Soroceanu L, Strongin AY: Up-regulation of vascular endothelial growth factor by membrane-type 1 matrix metalloproteinase stimulates human glioma xenograft growth and angiogenesis. Cancer Res 62:580–588, 2002 7. Denhardt DT, Feng B, Edwards DR, et al: Tissue inhibitor of metalloproteinase (TIMP, aka EPA): Structure, control of expression and biological functions. Pharmacol Ther 59:329–336, 1993[CrossRef][Medline]
8. Uria JA, Ferrand AA, Velasco G, et al: Structure and expression in breast tumors of human TIMP-3, a new member of the metalloproteinase inhibitor family. Cancer Res 54:2091–2096, 1994 9. Jiang Y, Goldberg ID, Shi YE: Complex roles of tissue inhibitors of metalloproteinases in cancer. Oncogene 21:2245–2252, 2002[CrossRef][Medline]
10. Coussens LM, Fingleton B, Matrisian LM: Matrix metalloproteinase inhibitors and cancer: Trials and tribulations. Science 295:2387–2392, 2002 11. Stetler-Stevenson WG, Gersch N, Golde DW: Tissue inhibitor of metalloproteinase-2 (TIMP2) has erythroid-potentiating activity. FEBS Lett 296:231–234, 1992[CrossRef][Medline] 12. Hayakawa T, Yamashita K, Tanzanwa K, et al: Growth- promoting activity of tissue inhibitor of metalloproteinases-l (TIMP-l) for a wide range cells. FEBS Lett 298:29–32, 1992[CrossRef][Medline]
13. Li GY, Fridman R, Kim HR: Tissue inhibitor of metalloproteinase-1 inhibits apoptosis of human breast epithelial cells. Cancer Res 59:6267–6275, 1999 14. Guedez L, Stetler-Stevenson WG, Wolff L, et al: In vitro suppression of programmed cell death of B cells by tissue inhibitor of metalloproteinases-1. J Clin Invest 102:2002–2010, 1998[Medline] 15. Zeng ZS, Cohen AM, Zhang ZF, et al: Elevated tissue inhibitor of metalloproteinase 1 RNA in colorectal cancer stroma correlates with lymph node and distant metastases. Clin Cancer Res 1:899–906, 1995[Abstract] 16. Fong KM, Kida Y, Zimmerman PV, et al: TIMP-1 and adverse prognosis in non-small cell lung cancer. Clin Cancer Res 2:1369–1372, 1996[Abstract]
17. Wang Z, Juttermann R, Soloway PD: TIMP-2 is required for efficient activation of proMMP-2 in vivo. J Biol Chem 275:26411–26415, 2000 18. Travis WD, Colby TV, Corrin B, et al: World Health Organization Classification of Lung and Pleural Tumors. Berlin, Germany, Springer- Verlag, 1999
19. Mountain C: Revisions in the international system for staging lung cancer. Chest 111:1710–1717, 1997
20. Ramnath N, Hernandez F, Tan DF, et al: Mcm-2 is an independent predictor of survival in patients with non-small cell lung cancer. J Clin Oncol 19:4259–4266, 2001 21. Tan D, Li Q, Ramnath N, et al: Prognostic relevance of the expression of p53 oncoprotein in non-small cell lung cancer. Anticancer Res 23:1665–1672, 2003[Medline] 22. Tan D, Kirley S, Li Q, et al: Loss of Cables protein expression in human non-small cell lung cancer: A tissue microarray study. Hum Pathol 34:143–149, 2003[CrossRef][Medline]
23. Harvey SR, Hurd TC, Markus G, et al: Evaluation of urinary plasminogen activator, its receptor, matrix metalloproteinase-9, and von Willebrand factor in pancreatic cancer. Clin Cancer Res 9:4935–4943, 2003 24. Agresti A: Categorical Data Analysis. New York, NY, John Wiley & Sons, 1990, pp 306–347
25. Kumaki F, Matshi K, Kawai T, et al: Expression of matrix metalloproteinases in invasive pulmonary adenocarcinoma with bronchioloalveolar component and atypical adenomatous hyperplasia. Am J Pathol 159:2125–2135, 2001 26. Suzuki M, Lizasa T, Fujisawa T, et al: Expression of matrix metalloproteinases and tissue inhibitor of matrix metalloproteinase in non-small cell lung cancer. Invasion Metastasis 18:134–141, 1998[CrossRef][Medline] 27. Kitamura H, Oosawa Y, Kawano N, et al: Basement membrane patterns, gelatinase A and tissue inhibitor of metalloproteinase-2 expressions, and stromal fibrosis during the development of peripheral lung adenocarcinoma. Hum Pathol 30:331–338, 1999[CrossRef][Medline] 28. Kawano N, Osawa H, Ito T, et al: Expression of gelatinase A, tissue inhibitor of metalloproteinase-2, matrilysin, and trypsin (ogen) in lung neoplasms: An immunohistochemical study. Hum Pathol 28:613–622, 1997[CrossRef][Medline] 29. Martin B, Paesmans M, Berghmans T, et al: Role of Bcl-2 as a prognostic factor for survival in lung cancer: A systematic review of the literature with meta-analysis. Br J Cancer 89:55–64, 2003[CrossRef][Medline] 30. McCarthy K, Maguire T, McGreal G, et al: High levels of tissue inhibitor of metalloproteinase-1 predict poor outcome in patients with breast cancer. Int J Cancer 84:44–48, 1999[CrossRef][Medline] 31. Thomas P, Khokha R, Shepherd F, et al: Differential expression of matrix metalloproteinases and their inhibitors in non-small cell lung cancer. J Pathol 190:150–156, 2000[CrossRef][Medline] 32. Nakada M, Kita D, Futami K, et al: Roles of membrane type 1 matrix metalloproteinase and tissue inhibitor of metalloproteinase 2 in invasion and dissemination of human malignant glioma. J Neurosurg 94:464–473, 2001[Medline] 33. Urbanski SJ, Edwards DR, Mailand A, et al: Expression of metalloproteinases and their inhibitors in primary pulmonary carcinomas. Br J Cancer 66:1188–1194, 1992[Medline] 34. Nakagawa H, Yagihashi S: Expression of type IV collagen and its degrading enzymes in squamous cell carcinoma of lung. Jpn J Cancer Res 85:934–938, 1994[CrossRef][Medline]
35. Kallakury BV, Karikehalli S, Haholu A, et al: Increased expression of matrix metalloproteinases 2 and 9 and tissue inhibitors of metalloproteinase 1 and 2 correlate with poor prognosis variables in renal cell carcinoma. Clin Cancer Res 7:3113–3119, 2001
36. Trudel D, Fradet Y, Meyer F, et al: Significance of MMP-2 expression in prostate cancer: An immunohistochemical study. Cancer Res 63:8511–8515, 2003 37. Manenti L, Paganoni P, Floriani I, et al: Expression levels of vascular endothelial growth factor, matrix metalloproteinases 2 and 9 and tissue inhibitor of metalloproteinases 1 and 2 in the plasma of patients with ovarian carcinoma. Eur J Cancer 39:1948–1956, 2003
38. Bremnes RM, Veve R, Gabrielson E, et al: High-throughput tissue microassay analysis used to evaluate biology and prognostic significance of the e-cadherin pathway in non–small-cell lung cancer. J Clin Oncol 20:2417–2428, 2002
39. Onn A, Correa AM, Gilcrease M, et al: Synchronous overexpression of epidermal growth factor receptor and HER2-neu protein is a predictor of poor outcome in patients with stage I non-small cell lung cancer.Clin Cancer Res 10:136–143, 2004 (1 pt 1)
40. Ylisirnio S, Hoyhtya M, Makitaro R, et al: Elevated serum levels of type I collagen degradation marker ICTP and tissue inhibitor of metalloproteinase (TIMP) 1 are associated with poor prognosis in lung cancer. Clin Cancer Res 7:1633–1637, 2001 41. Still K, Robson CN, Autzen P, et al: Localization and quantification of mRNA for matrix metalloproteinase-2 (MMP-2) and tissue inhibitor of matrix metalloproteinase-2 (TIMP-2) in human benign and malignant prostatic tissue. Prostate 42:18–25, 2000[CrossRef][Medline] 42. Yano A, Nakamoto T, Mashimoto K, et al: Localization and expression of tissue inhibitor of metalloproteinase-1 in human urothelial cancer. J Urol 167:729–734, 2002[CrossRef][Medline]
43. Shofuda K, Moriyama K, Nishihashi A, et al: Role of tissue inhibitor of metalloproteinases-2 (TIMP-2) in regulation of pro-gelatinase A activation catalyzed by membrane-type matrix metalloproteinase-1 (MT1-MMP) in human cancer cells. J Biochem 124:462–470, 1998
44. Butler GS, Butler MJ, Atkinson SJ, et al: The TIMP2 membrane type 1 metalloproteinase "receptor" regulates the concentration and efficient activation of progelatinase A: A kinetic study. J Biol Chem 273:871–880, 1998 45. Luparello C, Avanzato G, Carella C, et al: Tissue inhibitor of metalloprotease (TIMP)-1 and proliferative behavior of clonal breast cancer cells. Breast Cancer Res Treat 54:235–244, 1999[CrossRef][Medline] 46. Hayakawa T, Yamashita K, Ohuchi E, et al: Cell growth-promoting activity of tissue inhibitor of metalloproteinases-2 (TIMP-2).J Cell Sci 107:2373–2379, 1994 (pt 9)[Abstract]
47. Laack E, Nikbakht H, Peters A, et al: Expression of CEACAM1 in adenocarcinoma of the lung: A factor of independent prognostic significance. J Clin Oncol 20:4279–4284, 2002 48. Koutsami MK, Gorgoulis VG, Kastrinakis NG, et al: Prognostic factors in non-small cell lung cancer. Anticancer Res 22:347–374, 2002[Medline]
49. Yoshida T, Tanaka S, Mogi A, et al: The clinical significance of cyclin B1 and Wee1 expression in non-small-cell lung cancer. Ann Oncol 15:252–256, 2004
50. Gessner C, Woischwill C, Schumacher A, et al: Nuclear YB-1 expression as a negative prognostic marker in non-small cell lung cancer. Eur Respir J 23:14–19, 2004 51. Chandrachud LM, Pendleton N, Chishoim DM, et al: Relationship between vascularity, age and survival in non-small cell lung cancer. Br J Cancer 76:1367–1375, 1997[Medline] 52. Strauss GM, Kwiatkowski DJ, Harpole DH, et al: Molecular and pathologic markers in stage I non-small cell carcinoma of the lung. J Clin Oncol 13:1265–1279, 1995[Abstract]
53. Bepler G, Gautam A, McIntyre LM, et al: Prognostic significance of molecular genetic aberrations on chromosome segment 11p15.5 in non–small-cell lung cancer. J Clin Oncol 20:1353–1360, 2002
54. Jiang Y, Wang M, Celiker MY, et al: Stimulation of mammary tumorigenesis by systemic tissue inhibitor of matrix metalloproteinase 4 gene delivery. Cancer Res 61:2365–2370, 2001
55. Liotta LA, Kohn EC, Petricoin EF: Clinical proteomics: Personalized molecular medicine. JAMA 286:2211–2214, 2001 56. Soloway P, Alexander CM, Werb Z, et al: Targeted mutagenesis of Timp-1 reveals that lung tumor invasion is influenced by Timp-1 genotype but not by that of the host. Oncogene 13:2307–2314, 1996[Medline]
57. Levitt NC, Eskens FALM, O'Byrne KJ, et al: Phase I and pharmacological study of the oral matrix metalloproteinase inhibitor, MMI270 (CGS27023A), in patients with advanced solid cancer. Clin Cancer Res 7:1912–1922, 2001
58. Macaulay VM, O'Byrne KJ, Sanders MP, et al: Phase I study of intrapleural batimastat(BB-94), a matrix metalloproteinase inhibitor, in the treatment of malignant pleural effusions. Clin Cancer Res 5:513–520, 1999
59. Kruger A, Soeltl R, Sopov I, et al: Hydroxamate-type matrix metalloproteinase inhibitor batimastat promotes liver metastasis. Cancer Res 61:1272–1275, 2001
60. Patterson BC, Sang QA: Angiostatin-converting enzyme activities of human matrilysin (MMP-7) and gelatinase B/type IV collagenase (MMP-9). J Biol Chem 272:28823–28825, 1997
61. Pozzi A, Moberg PE, Miles LA, et al: Elevated matrix metalloprotease and angiostatin levels in integrin alpha 1 knockout mice cause reduced tumor vascularization. Proc Natl Acad Sci U S A 97:2202–2207, 2000
62. Bosken CH, Wei Q, Amos CI, et al: An analysis of DNA repair as a determinant of survival in patients with non-small cell lung cancer. J Natl Cancer Inst 94:1091–1099, 2002
63. Nicholson SA, Okby NT, Khan MA, et al: Alterations of p14ARF, p53, and p73 genes involved in the E2F-1-mediated apoptotic pathways in non-small cell lung carcinoma. Cancer Res 61:5636–5643, 2001 64. Beer DG, Kardia SL, Huang CC, et al: Gene-expression profiles predict survival of patients with lung adenocarcinoma. Nat Med 8:816–824, 2002[Medline]
65. Zochbauer-Muller S, Fong KM, Virmani AK, et al: Aberrant promoter methylation of multiple genes in non-small cell lung cancers. Cancer Res 61:249–255, 2001 Submitted February 21, 2003; accepted March 30, 2004. Related Editorial
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
30 minutes to complete antigen unmasking. The Ventan automated stainer system (Ventana Medical Systems, Tucson, AZ) and enhanced 3,3-diaminoazobenzidine detection kit (Dako, Carpenteria, CA) were used to perform the avidin-biotin complex immunoenzymatic technique for visualization of macromolecule complexes and color chromagen binding. The primary antibody (monoclonal antibody, clone 102D1, dilution 1:100; Neomarkers Labvision, Fremont, CA) was incubated for 20 minutes. The slides then were washed, counterstained with Richard Alien hematoxylin, mounted with Crystal Mount (Biomed, Foster City, CA), and covered with a coverslip for microscopic evaluation. An irrelevant rabbit antiserum served as a negative control. The IHC slides were reviewed by three pathologists (I.A., J.B., and D.T.) who were blinded to clinical outcome data. Both the intensity and the proportion of tumor cells stained were evaluated. For TIMP-1 protein, four categories were recorded: 0, or occasional cell stained; 1, weak intensity, 1% to 10% of cells stained; 2, moderate intensity, 11% to 50% of cells stained; and 3, strong intensity, more than 50% of cells stained. Conflicting scores (n = 21) were reviewed under a conference microscope. The consensus scores were recorded. To explore the possible relationship of TIMP-1 with other tumor markers, MMP-2, MMP-9, and Bcl-2 were also studied. The monoclonal antibodies (Mabs) to MMP-2 and MMP-9 were purchased from R&D Systems (Minneapolis, MN). According to the manufacturer, MMP-9 Mab binds both pro- and active forms of human MMP-9 and shows no cross-reactivity with recombinant human MMP-1, MMP-2, or MMP-3. The details of the procedure for MMP-2 and MMP-9 have been described previously.
10%) for TIMP-1. Therefore, tumors with moderate to strong (> 10%) stains were designated as overexpression. p53 nuclear overexpression was designated if more than 10% tumor nuclei stained.
2 statistic. Because of the sparseness in the data set at the extreme levels of TIMP-1 expression, the results from the ordinal logistic regression analysis should be regarded as anecdotal. 
1 were 57% less likely to have a lower expression level of TIMP-1 compared with those with PS 0 (odds ratio, 0.43; 95% CI, 0.22 to 0.87; P = .02). Patients with p53 overexpression were 43% less likely to have a lower TIMP-1 expression level compared with those with a p53 baseline expression (odds ratio, 0.57; 95% CI, 0.29 to 1.11; P = .10). Similar results were obtained when logistic regression models were derived in which TIMP-1 was dichotomized as 0 or 1 v 2 or 3. A multivariate model was not feasible because of sparseness in the data set. 
