- Email: [email protected]
5G Polymorphism in Breast Cancer: A Meta-analysis
Archives of Medical Research 44 (2013) 39e45
ORIGINAL ARTICLE
Clinicopathological Significance of Plasminogen Activator Inhibitor-1 Promoter 4G/5G Polymorphism in Breast Cancer: A Meta-analysis Ju-Han Lee, Younghye Kim, Jung-Woo Choi, and Young-Sik Kim Department of Pathology, Korea University Ansan Hospital, Ansan, Republic of Korea Received for publication July 5, 2012; accepted December 5, 2012 (ARCMED-D-12-00344).
Background and Aims. Plasminogen activator inhibitor type 1 (PAI-1) is associated with poor prognosis in breast cancer. Transcriptional expression of the PAI-1 can be controlled by PAI-1 promoter 4G/5G polymorphism. However, the significance of PAI-1 promoter 4G/5G polymorphism in breast cancer patients is contentious. To address this controversy, we conducted a meta-analysis for the relationships between PAI-1 promoter polymorphism and clinicopathological characteristics of breast cancer. Methods. Relevant published studies were identified using a search of PubMed, Embase, and the ISI Web of Science. The effect sizes of PAI-1 promoter 4G/5G polymorphism on breast cancer risk, lymph node metastasis, histologic grade, and overall survival were calculated by odds ratio (OR) or hazard ratio. The effect sizes were combined using a random-effects model. Results. Individuals with 4G/4G genotype had a higher risk of breast cancer than those with the combined 4G/5G and 5G/5G genotypes (OR 5 1.388; p 5 0.031). Breast cancer patients with the 5G/5G genotype displayed lymph node metastasis more than patients with either the combined other genotypes (OR 5 1.495; p 5 0.027) or with the 4G/4G genotype (OR 5 1.623; p 5 0.018). However, the PAI-1 promoter 4G/5G polymorphism was not associated with histological grade or overall survival. Conclusions. PAI-1 promoter 4G/5G polymorphism is associated with a relatively increased risk of breast cancer development and lymph node metastasis. Ó 2013 IMSS. Published by Elsevier Inc. Key Words: PAI-1, Polymorphism, Breast cancer, Meta-analysis.
Introduction Plasminogen activator inhibitor-1 (PAI-1) is a 47-kDa glycoprotein and a member of the serine proteinase inhibitor superfamily (1e3). PAI-1 binds to either the urokinase-type plasminogen activator or tissue-type plasminogen activator and prevents fibrinolysis and excessive extracellular matrix degradation by hampering the switch of plasminogen to plasmin (1e3). PAI-1 overexpression is closely associated with poor prognosis in various cancers
Address reprint requests to: Young-Sik Kim, MD, PhD, Department of Pathology, Korea University Ansan Hospital, 516, Gojan-1 Dong, Danwon-Gu, Ansan-Si, Gyeonggi-Do 425-707, Republic of Korea; Phone: 82-31-412-5322; FAX: 82-31-412-5324; E-mail: [email protected]
arising in the breast, colon, and kidney (4e9). PAI-1 may promote cancer progression by modulating cell adhesion, stimulating cell proliferation and angiogenesis, and hindering the disproportionate degradation of the extracellular matrix (10e12). PAI-1 expression can be regulated by both external stimulating factors and genetic polymorphism. The polymorphic loci of the PAI-1 gene include a 3’ HindIII site, a CAn dinucleotide repeat in intron 3 and a 4G/5G insertion/deletion polymorphism at position 675 in the promoter (13e16). In vitro studies on the promoter 4G/ 5G polymorphism have shown that the transcription of PAI-1 is more elevated in the 4G allele than in the 5G allele (17). Clinically, high plasma PAI-1 levels in patients with myocardial infarction and diabetes are
0188-4409/$ - see front matter. Copyright Ó 2013 IMSS. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.arcmed.2012.12.002
40
Lee et al./ Archives of Medical Research 44 (2013) 39e45
Figure 1. Flowchart of article selection for meta-analysis.
related to the PAI-1 4G allele (17e20). Studies investigating the association between PAI-1 promoter 4G/5G polymorphism and breast cancer have yielded conflicting data (5,8,9,21e27). The inconsistent results might be explained by the facts that the effect of PAI-1 4G/5G polymorphism on breast cancer is negligible or compounded by other factors, and that each study involved a small number of cases. To address this controversy, we conducted a meta-analysis to determine the relationship between PAI-1 promoter 4G/5G polymorphism and breast cancer.
Materials and Methods Data Collection and Selection Criteria for Meta-analysis A comprehensive search of PubMed (http://www.ncbi.nlm. nih.gov/pubmed), Embase (http://www.embase.com) and the ISI Web of Science search interface (http://apps. isiknowledge.com) was conducted. The following keywords and their combinations were used: breast cancer, plasminogen activator inhibitor, PAI-1, polymorphism, and allele. In addition, the reference lists of the identified articles were manually searched. Overlapping articles or duplicate
Table 1. Characteristics of individual studies included in this meta-analysis Breast cancer
Control
Lymph node metastasis
Study
Country
Ethnicity
4G/4G
4G/5G
5G/5G
4G/4G
4G/5G
5G/5G
4G/4G
4G/5G
5G/5G
Palmirotta (8) Sternlicht (9) Yagmurdur (26) Lei (24) Minisini (25) Eroglu (23) B1asiak (22) Castell o (5) Alvarez-Millan (21) Zhang (27)
Italy UK Turkey Sweden Italy Turkey Poland Spain Spain China
Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Asian
25 790 35 322 56 12 31 24 12 320
42 1229 17 482 85 21 40 66 9 550
32 520 3 152 52 1 29 14 7 213
10 550 28 326 39 22 21
29 896 57 453 68 49 48
11 386 28 164 35 19 37
12 70 19
6 116 34a
1 46
17
24
20
6
4
4
a
The value indicates the number of combined 4G/5G and 5G/5G genotypes.
PAI-1 Promoter Polymorphism in Breast Cancer
Study name
Statistics for each study
41
Odds ratio and 95% CI
Odds Lower Upper ratio limit limit Z-Value p-Value Palmirotta R Sternlicht MD Yagmurdur MC Lei H Minisini AM Eroglu A Blasiak J
1.351 1.053 5.313 0.961 1.080 1.686 1.818 1.388
0.590 3.093 0.713 0.924 1.200 0.773 2.649 10.654 4.704 0.795 1.162 -0.408 0.667 1.748 0.311 0.719 3.952 1.202 0.960 3.444 1.836 1.031 1.870 2.161
0.476 0.439 0.000 0.683 0.756 0.230 0.066 0.031 0.01
0.1
1
4G/5G+5G/5G
10
100
4G/4G
Figure 2. Odd ratios with corresponding 95% confidence intervals of individual studies and pooled data for breast cancer risk between 4G/4G genotype and the other genotypes.
data were excluded by examining the authors’ names and affiliations for each publication. The following types of articles were included: 1) original articles demonstrating PAI-1 promoter 4G/5G polymorphism assessed only in primary breast cancer patients, 2) articles published before January 2012 in English, and 3) the most informative article when multiple articles were published by the same authors or groups. Articles lacking data or containing data inappropriate for meta-analysis, review articles without original data, and case reports were excluded.
heterogeneity. Sensitivity analyses were performed to examine the influence of each study on the pooled OR and HR by serially omitting an individual study and pooling the remaining studies. Publication bias was examined by funnel plots and Egger’s tests for the degree of asymmetry; p !0.05 was considered statistically significant. The pooled analysis was performed using Comprehensive Meta-analysis Software version 2.0 (Biostat, Englewood, NJ).
Data Pooling and Statistics Results
Odd ratios (ORs) or hazard ratios (HRs), or 95% confidence intervals (CIs) were calculated using meta-analysis (28e30). For studies without HRs for survival, we assessed HRs and CIs using an approximation method (30). The HRs or ORs were combined using a random-effects model (DerSimonian-Laird method). To identify and quantify interstudy heterogeneity, Q statistics (an adaptation of the c2 goodness-of-fit test) were calculated; p ! 0.10 was considered statistically significant. Meta-regression analyses were conducted to explore the sources of
By searching the databases concerning the association between PAI-1 promoter polymorphism and breast cancer, 24 relevant publications were identified. Of these investigations, ten articles (5,8,9,21e27) fulfilled the selection criteria (Figure 1). Seven studies reported the relationship of PAI-1 promoter 4G/5G polymorphism and breast cancer risk (8,9,22e26). Eight studies presented the association between PAI-1 promoter 4G/5G polymorphism and clinicopathological characteristics of breast cancer patients such
Table 2. Association between PAI-1 promoter 4G/5G polymorphism and breast cancer risk Population (No.) PAI-1 genotype 4G/4G 4G/5G 5G/5G 4G/4G 4G/4G 4G/5G
vs. vs. vs. vs. vs. vs.
4G/5Gþ5G/5G 4G/4Gþ5G/5G 4G/4Gþ4G/5G 5G/5G 4G/5G 5G/5G
Control
Cancer
3276 3276 3276 1676 2596 2280
3976 3976 3976 2060 3187 2705
Association OR (95% CI) 1.388 0.913 0.888 1.372 1.301 1.040
(1.031e1.870) (0.760e1.096) (0.677e1.164) (0.951e1.981) (0.989e1.710) (0.824e1.313)
Heterogeneity test p
Q
df
p
0.031 0.329 0.390 0.091 0.060 0.740
25.594 11.367 14.531 19.726 18.676 10.218
6 6 6 6 6 6
!0.001 0.078 0.024 0.003 0.005 0.116
42
Lee et al./ Archives of Medical Research 44 (2013) 39e45
Study name
Statistics for each study
Odds ratio and 95% CI
Odds Lower Upper ratio limit limit Z-Value p-Value Yagmurdur MC 0.958 0.079 11.673 -0.033
Lei H
1.635 Blasiak J 1.830 Alvarez-Millan JJ 1.333 1.623
1.039 0.635 0.204 1.086
2.573 5.271 8.708 2.426
0.973 0.034 0.263 0.764 0.018
2.124 1.120 0.300 2.365
0.01
0.1
1
4G/4G
10
100
5G/5G
Figure 3. Odd ratios and pooled data for lymph node metastasis between 5G/5G genotype and 4G/4G genotype.
as lymph node metastasis, histologic grade, and overall survival (Table 1) (5,9,21,22,24e27). Seven studies including 3,976 breast cancer patients and 3,276 control addressed the effects of PAI-1 promoter polymorphism on breast cancer development (8,9,22e26). The risk of breast cancer was significantly increased in 4G/4G genotype than in the combined 4G/5G and 5G/5G genotypes (OR 5 1.388; 95% CI: 1.031e1.870; p 5 0.031) (Figure 2). However, the comparisons among the other combinations of PAI-1 genotypes did not reveal any significant associations between PAI-1 promoter polymorphism and breast cancer risk (Table 2). There was statistical heterogeneity among these studies (Q 5 25.594, degrees of freedom (df) 5 6, p ! 0.001). The heterogeneity could be explained by the ratio of case size to control size ( p 5 0.034) and the number of control population ( p 5 0.04) by meta-regression analyses. Five studies reported the relationship of PAI-1 promoter 4G/5G polymorphism and lymph node metastasis and included 379 cases with and 750 cases without lymph node metastasis (21,22,24e26). The prevalence of lymph node metastasis in the 5G/5G genotype was significantly higher than in the 4G/4G genotype (OR 5 1.623; 95% CI: 1.086e2.426; p 5 0.018) (Figure 3) and in the combined 4G/4G and 4G/5G genotypes (OR 5 1.495; 95% CI: 1.048e2.131; p 5 0.027) (Table 3). There was no significant statistical heterogeneity among the studies (Q 5
0.263, df 5 3, p 5 0.967; Q 5 0.166, df 5 3, p 5 0.983, respectively). Four studies presented the histological grade of breast cancer according to the PAI-1 genotypes, which consisted of 627 cases of grades I and II and 304 cases of grade III (5,21,22,24). There was no significant association between histological grade and PAI-1 genotypes (Table 4). Moreover, three studies addressed the overall survival of breast cancer patients according to the PAI-1 genotypes (9,24,27). The investigations included 1,426 cases of the 4G/4G genotype, 2254 cases of the 4G/5G genotype, and 883 cases of the 5G/5G genotype. No significant relationship was found between PAI-1 genotypes and overall survival (Table 5). The other clinicopathological factors were not suitable for meta-analysis. Sensitivity analyses revealed that four studies (9,22,23,26) influenced the result of breast cancer risk between the 4G/4G genotype and the other genotypes (Figure 4). In addition, the study by Lei et al. (24) influenced the result of lymph node metastasis between the 4G/4G and 5G/5G genotypes. Funnel plots and Egger’s regression tests revealed no evidence of publication bias (Figure 5). Discussion A meta-analysis was conducted to estimate the relationships between PAI-1 promoter 4G/5G polymorphism and
Table 3. Association between PAI-1 promoter 4G/5G polymorphism and lymph node metastasis of breast cancer patients Patient (No.) PAI-1 genotype 5G/5G 4G/5G 4G/4G 5G/5G 5G/5G 4G/5G
vs. vs. vs. vs. vs. vs.
4G/4Gþ4G/5G 4G/4Gþ5G/5G 4G/5Gþ5G/5G 4G/4G 4G/5G 4G/4G
Association
No metastasis
Metastasis
651 651 750 336 409 557
326 326 379 176 221 225
OR (95% CI) 1.495 0.974 1.002 1.623 1.410 1.150
(1.048e2.131) (0.740e1.281) (0.640e1.566) (1.086e2.426) (0.966e2.058) (0.844e1.569)
Heterogeneity test p
Q
df
p
0.027 0.850 0.995 0.018 0.075 0.376
0.166 0.170 6.581 0.263 0.142 0.219
3 3 4 3 3 3
0.983 0.982 0.160 0.967 0.986 0.974
PAI-1 Promoter Polymorphism in Breast Cancer
43
Table 4. Association between PAI-1 promoter 4G/5G polymorphism and histologic grade of breast cancer patients Patient (No.) PAI-1 genotype 5G/5G 4G/5G 4G/4G 5G/5G 5G/5G 4G/5G
vs. vs. vs. vs. vs. vs.
Association
Grades I þ II
Grade III
627 540 540 264 371 445
304 287 287 151 189 234
4G/4Gþ4G/5G 4G/4Gþ5G/5G 4G/5Gþ5G/5G 4G/4G 4G/5G 4G/4G
OR (95% CI) 1.013 0.848 1.150 1.015 0.849 1.185
breast cancer risk or its clinicopathological characteristics. This pooled analysis revealed an increased incidence of breast cancer in the populations with 4G/4G genotype than in the populations with other genotypes. In addition, patients with 5G/5G genotype displayed more frequent lymph node metastasis than patients with the remaining genotypes. To the best of our knowledge, this is the first pooled analysis of breast cancer regarding PAI-1 promoter polymorphism. The prevalence of breast cancer in populations with the 4G/4G homozygous genotype was somewhat increased compared to populations with the other genotypes of the PAI-1 promoter. Our result disclosed that individuals with the 4G homozygous genotype may have a 39% increased breast cancer risk compared with the other genotypes. PAI-1 expression is regulated by polymorphisms in the PAI-1 promoter gene as well as numerous external stimuli in physiological and pathological conditions. The PAI-1 promoter contains an E-box motif (50 -CACGTG30 ) at 680 to 675 base pair, the last guanosine nucleotide of which is overlapped with the downward 4G/5G polymorphism nucleotides. PAI-1 transcription is regulated through the binding of upstream stimulatory factor (USF)-1, USF2, and BMAL2/CLOCK to the E-box motifs overlapping the 4G allele (transcriptional enhancer) or the 5G allele (transcriptional suppressor) (31e33). Therefore, the 4G/4G genotype has higher PAI-1 transcriptional activity and high PAI-1 plasma levels (17e19). This pooled analysis also revealed an association between the 4G allele and a higher risk of breast cancer, as compared to the other genotypes.
(0.604e1.700) (0.636e1.131) (0.848e1.561) (0.644e1.553) (0.566e1.272) (0.858e1.637)
Heterogeneity test p
Q
df
p
0.960 0.262 0.368 0.945 0.428 0.303
3.533 1.501 0.105 1.183 2.012 0.370
3 2 2 2 2 2
0.317 0.472 0.949 0.553 0.366 0.831
This meta-analysis showed that the prevalence of lymph node metastasis in patients with the 5G/5G homozygous type was higher than in patients with the other combined genotypes as well as the 4G/4G genotype. There has been considerable controversy with respect to the clinicopathological effects of PAI-1 polymorphism on breast cancer patients. No differences in histological grade and overall survival according to the PAI-1 genotypes were presently evident. However, Lei et al. (24) reported that 5G/5G genotype was correlated with lymph node metastasis and poor overall survival, whereas Alvarez-Millan et al. (21) described the association between 5G/5G genotype and well-differentiated phenotypes. In addition, Zhang et al. (27) reported that the 4G/4G homozygous type was associated with poor overall survival rates of breast cancer. Some studies using PAI1 deficient- and overexpressing mice revealed very little impact of PAI-1 on either carcinogenesis or metastasis, whereas another study indicated that PAI-1 functions differently according to its concentration around the tumor (34). Physiological concentrations of PAI-1 support in vivo cancer invasion and angiogenesis, whereas levels of PAI-1 that are higher than the physiological concentrations inhibit tumor angiogenesis (34). Our meta-analysis confirmed that 5G homozygous genotype has a high rate of lymph node metastasis. However, the role of either PAI-1 expression or PAI-1 promoter polymorphism in tumorigenesis and metastasis remains controversial. The present meta-analysis has several limitations. The genetic investigations contained a relatively small sample
Table 5. Association between PAI-1 promoter 4G/5G polymorphism and overall survival of breast cancer patients Association PAI-1 genotype 5G/5G vs. 4G/4G 5G/5G vs. 4G/5G 4G/5G vs. 4G/4G
Patient (No.) 2309 3137 3680
OR (95% CI) 1.237 (0.642e2.384) 1.245 (0.731e2.123) 1.086 (0.875e1.348)
Heterogeneity test p
Q
df
p
0.524 0.420 0.452
16.161 8.098 0.906
2 2 2
!0.001 0.017 0.636
44
Lee et al./ Archives of Medical Research 44 (2013) 39e45
Study name
Statistics with study removed
Odds ratio (95% CI) with study removed
Lower Upper Point limit limit Z-Value p-Value Palmirotta R Sternlicht MD Yagmurdur MC Lei H Minisini AM Eroglu A Blasiak J
1.401 1.610 1.055 1.628 1.475 1.369 1.338 1.388
1.018 0.987 0.944 1.032 1.048 0.999 0.977 1.031
1.929 2.626 1.180 2.569 2.078 1.875 1.834 1.870
2.069 1.908 0.948 2.093 2.227 1.955 1.815 2.161
0.039 0.056 0.343 0.036 0.026 0.051 0.070 0.031 0.01
0.1
4G/5G+5G/5G
1
10
100
4G/4G
Figure 4. Sensitivity analysis of meta-analysis for breast cancer risk between 4G/4G genotype and the other genotypes.
size, deficient control populations, and were heterogeneous in the control selection. Thus, a meta-analysis should be carried out using adequate inclusion and exclusion criteria for journal selection and needs to be interpreted considering the context of individual studies. In summary, this pooled analysis indicates that the PAI-1 promoter 4G/4G genotype is associated with mildly increased breast cancer risk and that the 5G/5G genotype is related to lymph node metastasis of breast cancer. However, further large-scale studies are required to
Figure 5. Funnel plot of meta-analysis for the breast cancer risk between 4G/5G genotype and 5G/5G genotype. Individual studies are represented by small circles.
elucidate the precise role of PAI-1 promoter 4G/5G polymorphism in breast cancer.
Acknowledgments This study was supported by a Korea University Grant. Conflicts of interest: None.
References 1. Schmitt M, Janicke F, Moniwa N, et al. Tumor-associated urokinasetype plasminogen activator: biological and clinical significance. Biol Chem Hoppe Seyler 1992;373:611e622. 2. Durand MK, Bodker JS, Christensen A, et al. Plasminogen activator inhibitor-I and tumour growth, invasion, and metastasis. Thromb Haemost 2004;91:438e449. 3. Stefansson S, McMahon GA, Petitclerc E, et al. Plasminogen activator inhibitor-1 in tumor growth, angiogenesis and vascular remodeling. Curr Pharm Des 2003;9:1545e1564. 4. Berger DH. Plasmin/plasminogen system in colorectal cancer. World J Surg 2002;26:767e771. 5. Castello R, Espana F, Vazquez C, et al. Plasminogen activator inhibitor-1 4G/5G polymorphism in breast cancer patients and its association with tissue PAI-1 levels and tumor severity. Thromb Res 2006; 117:487e492. 6. Choi JW, Lee JH, Park HS, et al. PAI-1 expression and its regulation by promoter 4G/5G polymorphism in clear cell renal cell carcinoma. J Clin Pathol 2011;64:893e897. 7. Harbeck N, Kates RE, Gauger K, et al. Urokinase-type plasminogen activator (uPA) and its inhibitor PAI-I: novel tumor-derived factors with a high prognostic and predictive impact in breast cancer. Thromb Haemost 2004;91:450e456. 8. Palmirotta R, Ferroni P, Savonarola A, et al. Prognostic value of presurgical plasma PAI-1 (plasminogen activator inhibitor-1) levels in breast cancer. Thromb Res 2009;124:403e408. 9. Sternlicht MD, Dunning AM, Moore DH, et al. Prognostic value of PAI1 in invasive breast cancer: evidence that tumor-specific factors are more important than genetic variation in regulating PAI1 expression. Cancer Epidemiol Biomarkers Prev 2006;15:2107e2114.
PAI-1 Promoter Polymorphism in Breast Cancer 10. Bajou K, Masson V, Gerard RD, et al. The plasminogen activator inhibitor PAI-1 controls in vivo tumor vascularization by interaction with proteases, not vitronectin. Implications for antiangiogenic strategies. J Cell Biol 2001;152:777e784. 11. Czekay RP, Aertgeerts K, Curriden SA, et al. Plasminogen activator inhibitor-1 detaches cells from extracellular matrices by inactivating integrins. J Cell Biol 2003;160:781e791. 12. Webb DJ, Thomas KS, Gonias SL. Plasminogen activator inhibitor 1 functions as a urokinase response modifier at the level of cell signaling and thereby promotes MCF-7 cell growth. J Cell Biol 2001;152:741e752. 13. Chen YQ, Sloan-Lancaster J, Berg DT, et al. Differential mechanisms of plasminogen activator inhibitor-1 gene activation by transforming growth factor-beta and tumor necrosis factor-alpha in endothelial cells. Thromb Haemost 2001;86:1563e1572. 14. Dong J, Fujii S, Li H, et al. Interleukin-6 and mevastatin regulate plasminogen activator inhibitor-1 through CCAAT/enhancer-binding protein-delta. Arterioscler Thromb Vasc Biol 2005;25:1078e1084. 15. Ha H, Oh EY, Lee HB. The role of plasminogen activator inhibitor 1 in renal and cardiovascular diseases. Nat Rev Nephrol 2009;5:203e211. 16. Swiatkowska M, Szemraj J, Cierniewski CS. Induction of PAI-1 expression by tumor necrosis factor alpha in endothelial cells is mediated by its responsive element located in the 4G/5G site. FEBS J 2005;272:5821e5831. 17. Grancha S, Estelles A, Tormo G, et al. Plasminogen activator inhibitor-1 (PAI-1) promoter 4G/5G genotype and increased PAI-1 circulating levels in postmenopausal women with coronary artery disease. Thromb Haemost 1999;81:516e521. 18. Burzotta F, Di Castelnuovo A, Amore C, et al. 4G/5G promoter PAI-1 gene polymorphism is associated with plasmatic PAI-1 activity in Italians: a model of gene-environment interaction. Thromb Haemost 1998;79:354e358. 19. Eriksson P, Kallin B, van’t Hooft FM, et al. Allele-specific increase in basal transcription of the plasminogen-activator inhibitor 1 gene is associated with myocardial infarction. Proc Natl Acad Sci USA 1995;92:1851e1855. 20. Nordt TK, Lohrmann J, Bode C. Regulation of PAI-1 expression by genetic polymorphisms. Impact on atherogenesis. Thromb Res 2001; 103(Suppl 1):S1eS5. 21. Alvarez-Millan JJ, Bocos C, Ferrin V, et al. PAI-1 promoter polymorphism modulates uPA-PAI complex accumulation by breast cancer cells. Oncology 2002;62:286e290.
45
22. Blasiak J, Smolarz B. Plasminogen activator inhibitor-1 (PAI-1) gene 4G/5G promoter polymorphism is not associated with breast cancer. Acta Biochim Pol 2000;47:191e199. 23. Eroglu A, Ulu A, Cam R, et al. Plasminogen activator inhibitor -1 gene 4G/5G polymorphism in patients with breast cancer. J BUON 2006; 11:481e484. 24. Lei H, Hemminki K, Johansson R, et al. PAI-1 -675 4G/5G polymorphism as a prognostic biomarker in breast cancer. Breast Cancer Res Treat 2008;109:165e175. 25. Minisini AM, Fabbro D, Di Loreto C, et al. Markers of the uPA system and common prognostic factors in breast cancer. Am J Clin Pathol 2007;128:112e117. 26. Yagmurdur MC, Atac FB, Tutar NU, et al. Prognostic value of the PAI-1 4G/5G polymorphism in invasive ductal carcinoma of the breast. Int Surg 2008;93:163e168. 27. Zhang X, Shu XO, Cai Q, et al. Functional plasminogen activator inhibitor-1 gene variants and breast cancer survival. Clin Cancer Res 2006;12:6037e6042. 28. Lee JH, Lee ES, Kim YS. Clinicopathologic significance of BRAF V600E mutation in papillary carcinomas of the thyroid: a meta-analysis. Cancer 2007;110:38e46. 29. Normand SL. Meta-analysis: formulating, evaluating, combining, and reporting. Stat Med 1999;18:321e359. 30. Parmar MK, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med 1998;17:2815e2834. 31. Chong NW, Codd V, Chan D, et al. Circadian clock genes cause activation of the human PAI-1 gene promoter with 4G/5G allelic preference. FEBS Lett 2006;580:4469e4472. 32. Dimova EY, Kietzmann T. The MAPK pathway and HIF-1 are involved in the induction of the human PAI-1 gene expression by insulin in the human hepatoma cell line HepG2. Ann NY Acad Sci 2006;1090:355e367. 33. Ma Z, Jhun B, Jung SY, et al. Binding of upstream stimulatory factor 1 to the E-box regulates the 4G/5G polymorphism-dependent plasminogen activator inhibitor 1 expression in mast cells. J Allergy Clin Immunol 2008;121:1006e1012.e1002. 34. Iwaki T, Urano T, Umemura K. PAI-1, progress in understanding the clinical problem and its aetiology. Br J Haematol 2012;157: 291e298.
ORIGINAL ARTICLE
Clinicopathological Significance of Plasminogen Activator Inhibitor-1 Promoter 4G/5G Polymorphism in Breast Cancer: A Meta-analysis Ju-Han Lee, Younghye Kim, Jung-Woo Choi, and Young-Sik Kim Department of Pathology, Korea University Ansan Hospital, Ansan, Republic of Korea Received for publication July 5, 2012; accepted December 5, 2012 (ARCMED-D-12-00344).
Background and Aims. Plasminogen activator inhibitor type 1 (PAI-1) is associated with poor prognosis in breast cancer. Transcriptional expression of the PAI-1 can be controlled by PAI-1 promoter 4G/5G polymorphism. However, the significance of PAI-1 promoter 4G/5G polymorphism in breast cancer patients is contentious. To address this controversy, we conducted a meta-analysis for the relationships between PAI-1 promoter polymorphism and clinicopathological characteristics of breast cancer. Methods. Relevant published studies were identified using a search of PubMed, Embase, and the ISI Web of Science. The effect sizes of PAI-1 promoter 4G/5G polymorphism on breast cancer risk, lymph node metastasis, histologic grade, and overall survival were calculated by odds ratio (OR) or hazard ratio. The effect sizes were combined using a random-effects model. Results. Individuals with 4G/4G genotype had a higher risk of breast cancer than those with the combined 4G/5G and 5G/5G genotypes (OR 5 1.388; p 5 0.031). Breast cancer patients with the 5G/5G genotype displayed lymph node metastasis more than patients with either the combined other genotypes (OR 5 1.495; p 5 0.027) or with the 4G/4G genotype (OR 5 1.623; p 5 0.018). However, the PAI-1 promoter 4G/5G polymorphism was not associated with histological grade or overall survival. Conclusions. PAI-1 promoter 4G/5G polymorphism is associated with a relatively increased risk of breast cancer development and lymph node metastasis. Ó 2013 IMSS. Published by Elsevier Inc. Key Words: PAI-1, Polymorphism, Breast cancer, Meta-analysis.
Introduction Plasminogen activator inhibitor-1 (PAI-1) is a 47-kDa glycoprotein and a member of the serine proteinase inhibitor superfamily (1e3). PAI-1 binds to either the urokinase-type plasminogen activator or tissue-type plasminogen activator and prevents fibrinolysis and excessive extracellular matrix degradation by hampering the switch of plasminogen to plasmin (1e3). PAI-1 overexpression is closely associated with poor prognosis in various cancers
Address reprint requests to: Young-Sik Kim, MD, PhD, Department of Pathology, Korea University Ansan Hospital, 516, Gojan-1 Dong, Danwon-Gu, Ansan-Si, Gyeonggi-Do 425-707, Republic of Korea; Phone: 82-31-412-5322; FAX: 82-31-412-5324; E-mail: [email protected]
arising in the breast, colon, and kidney (4e9). PAI-1 may promote cancer progression by modulating cell adhesion, stimulating cell proliferation and angiogenesis, and hindering the disproportionate degradation of the extracellular matrix (10e12). PAI-1 expression can be regulated by both external stimulating factors and genetic polymorphism. The polymorphic loci of the PAI-1 gene include a 3’ HindIII site, a CAn dinucleotide repeat in intron 3 and a 4G/5G insertion/deletion polymorphism at position 675 in the promoter (13e16). In vitro studies on the promoter 4G/ 5G polymorphism have shown that the transcription of PAI-1 is more elevated in the 4G allele than in the 5G allele (17). Clinically, high plasma PAI-1 levels in patients with myocardial infarction and diabetes are
0188-4409/$ - see front matter. Copyright Ó 2013 IMSS. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.arcmed.2012.12.002
40
Lee et al./ Archives of Medical Research 44 (2013) 39e45
Figure 1. Flowchart of article selection for meta-analysis.
related to the PAI-1 4G allele (17e20). Studies investigating the association between PAI-1 promoter 4G/5G polymorphism and breast cancer have yielded conflicting data (5,8,9,21e27). The inconsistent results might be explained by the facts that the effect of PAI-1 4G/5G polymorphism on breast cancer is negligible or compounded by other factors, and that each study involved a small number of cases. To address this controversy, we conducted a meta-analysis to determine the relationship between PAI-1 promoter 4G/5G polymorphism and breast cancer.
Materials and Methods Data Collection and Selection Criteria for Meta-analysis A comprehensive search of PubMed (http://www.ncbi.nlm. nih.gov/pubmed), Embase (http://www.embase.com) and the ISI Web of Science search interface (http://apps. isiknowledge.com) was conducted. The following keywords and their combinations were used: breast cancer, plasminogen activator inhibitor, PAI-1, polymorphism, and allele. In addition, the reference lists of the identified articles were manually searched. Overlapping articles or duplicate
Table 1. Characteristics of individual studies included in this meta-analysis Breast cancer
Control
Lymph node metastasis
Study
Country
Ethnicity
4G/4G
4G/5G
5G/5G
4G/4G
4G/5G
5G/5G
4G/4G
4G/5G
5G/5G
Palmirotta (8) Sternlicht (9) Yagmurdur (26) Lei (24) Minisini (25) Eroglu (23) B1asiak (22) Castell o (5) Alvarez-Millan (21) Zhang (27)
Italy UK Turkey Sweden Italy Turkey Poland Spain Spain China
Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Asian
25 790 35 322 56 12 31 24 12 320
42 1229 17 482 85 21 40 66 9 550
32 520 3 152 52 1 29 14 7 213
10 550 28 326 39 22 21
29 896 57 453 68 49 48
11 386 28 164 35 19 37
12 70 19
6 116 34a
1 46
17
24
20
6
4
4
a
The value indicates the number of combined 4G/5G and 5G/5G genotypes.
PAI-1 Promoter Polymorphism in Breast Cancer
Study name
Statistics for each study
41
Odds ratio and 95% CI
Odds Lower Upper ratio limit limit Z-Value p-Value Palmirotta R Sternlicht MD Yagmurdur MC Lei H Minisini AM Eroglu A Blasiak J
1.351 1.053 5.313 0.961 1.080 1.686 1.818 1.388
0.590 3.093 0.713 0.924 1.200 0.773 2.649 10.654 4.704 0.795 1.162 -0.408 0.667 1.748 0.311 0.719 3.952 1.202 0.960 3.444 1.836 1.031 1.870 2.161
0.476 0.439 0.000 0.683 0.756 0.230 0.066 0.031 0.01
0.1
1
4G/5G+5G/5G
10
100
4G/4G
Figure 2. Odd ratios with corresponding 95% confidence intervals of individual studies and pooled data for breast cancer risk between 4G/4G genotype and the other genotypes.
data were excluded by examining the authors’ names and affiliations for each publication. The following types of articles were included: 1) original articles demonstrating PAI-1 promoter 4G/5G polymorphism assessed only in primary breast cancer patients, 2) articles published before January 2012 in English, and 3) the most informative article when multiple articles were published by the same authors or groups. Articles lacking data or containing data inappropriate for meta-analysis, review articles without original data, and case reports were excluded.
heterogeneity. Sensitivity analyses were performed to examine the influence of each study on the pooled OR and HR by serially omitting an individual study and pooling the remaining studies. Publication bias was examined by funnel plots and Egger’s tests for the degree of asymmetry; p !0.05 was considered statistically significant. The pooled analysis was performed using Comprehensive Meta-analysis Software version 2.0 (Biostat, Englewood, NJ).
Data Pooling and Statistics Results
Odd ratios (ORs) or hazard ratios (HRs), or 95% confidence intervals (CIs) were calculated using meta-analysis (28e30). For studies without HRs for survival, we assessed HRs and CIs using an approximation method (30). The HRs or ORs were combined using a random-effects model (DerSimonian-Laird method). To identify and quantify interstudy heterogeneity, Q statistics (an adaptation of the c2 goodness-of-fit test) were calculated; p ! 0.10 was considered statistically significant. Meta-regression analyses were conducted to explore the sources of
By searching the databases concerning the association between PAI-1 promoter polymorphism and breast cancer, 24 relevant publications were identified. Of these investigations, ten articles (5,8,9,21e27) fulfilled the selection criteria (Figure 1). Seven studies reported the relationship of PAI-1 promoter 4G/5G polymorphism and breast cancer risk (8,9,22e26). Eight studies presented the association between PAI-1 promoter 4G/5G polymorphism and clinicopathological characteristics of breast cancer patients such
Table 2. Association between PAI-1 promoter 4G/5G polymorphism and breast cancer risk Population (No.) PAI-1 genotype 4G/4G 4G/5G 5G/5G 4G/4G 4G/4G 4G/5G
vs. vs. vs. vs. vs. vs.
4G/5Gþ5G/5G 4G/4Gþ5G/5G 4G/4Gþ4G/5G 5G/5G 4G/5G 5G/5G
Control
Cancer
3276 3276 3276 1676 2596 2280
3976 3976 3976 2060 3187 2705
Association OR (95% CI) 1.388 0.913 0.888 1.372 1.301 1.040
(1.031e1.870) (0.760e1.096) (0.677e1.164) (0.951e1.981) (0.989e1.710) (0.824e1.313)
Heterogeneity test p
Q
df
p
0.031 0.329 0.390 0.091 0.060 0.740
25.594 11.367 14.531 19.726 18.676 10.218
6 6 6 6 6 6
!0.001 0.078 0.024 0.003 0.005 0.116
42
Lee et al./ Archives of Medical Research 44 (2013) 39e45
Study name
Statistics for each study
Odds ratio and 95% CI
Odds Lower Upper ratio limit limit Z-Value p-Value Yagmurdur MC 0.958 0.079 11.673 -0.033
Lei H
1.635 Blasiak J 1.830 Alvarez-Millan JJ 1.333 1.623
1.039 0.635 0.204 1.086
2.573 5.271 8.708 2.426
0.973 0.034 0.263 0.764 0.018
2.124 1.120 0.300 2.365
0.01
0.1
1
4G/4G
10
100
5G/5G
Figure 3. Odd ratios and pooled data for lymph node metastasis between 5G/5G genotype and 4G/4G genotype.
as lymph node metastasis, histologic grade, and overall survival (Table 1) (5,9,21,22,24e27). Seven studies including 3,976 breast cancer patients and 3,276 control addressed the effects of PAI-1 promoter polymorphism on breast cancer development (8,9,22e26). The risk of breast cancer was significantly increased in 4G/4G genotype than in the combined 4G/5G and 5G/5G genotypes (OR 5 1.388; 95% CI: 1.031e1.870; p 5 0.031) (Figure 2). However, the comparisons among the other combinations of PAI-1 genotypes did not reveal any significant associations between PAI-1 promoter polymorphism and breast cancer risk (Table 2). There was statistical heterogeneity among these studies (Q 5 25.594, degrees of freedom (df) 5 6, p ! 0.001). The heterogeneity could be explained by the ratio of case size to control size ( p 5 0.034) and the number of control population ( p 5 0.04) by meta-regression analyses. Five studies reported the relationship of PAI-1 promoter 4G/5G polymorphism and lymph node metastasis and included 379 cases with and 750 cases without lymph node metastasis (21,22,24e26). The prevalence of lymph node metastasis in the 5G/5G genotype was significantly higher than in the 4G/4G genotype (OR 5 1.623; 95% CI: 1.086e2.426; p 5 0.018) (Figure 3) and in the combined 4G/4G and 4G/5G genotypes (OR 5 1.495; 95% CI: 1.048e2.131; p 5 0.027) (Table 3). There was no significant statistical heterogeneity among the studies (Q 5
0.263, df 5 3, p 5 0.967; Q 5 0.166, df 5 3, p 5 0.983, respectively). Four studies presented the histological grade of breast cancer according to the PAI-1 genotypes, which consisted of 627 cases of grades I and II and 304 cases of grade III (5,21,22,24). There was no significant association between histological grade and PAI-1 genotypes (Table 4). Moreover, three studies addressed the overall survival of breast cancer patients according to the PAI-1 genotypes (9,24,27). The investigations included 1,426 cases of the 4G/4G genotype, 2254 cases of the 4G/5G genotype, and 883 cases of the 5G/5G genotype. No significant relationship was found between PAI-1 genotypes and overall survival (Table 5). The other clinicopathological factors were not suitable for meta-analysis. Sensitivity analyses revealed that four studies (9,22,23,26) influenced the result of breast cancer risk between the 4G/4G genotype and the other genotypes (Figure 4). In addition, the study by Lei et al. (24) influenced the result of lymph node metastasis between the 4G/4G and 5G/5G genotypes. Funnel plots and Egger’s regression tests revealed no evidence of publication bias (Figure 5). Discussion A meta-analysis was conducted to estimate the relationships between PAI-1 promoter 4G/5G polymorphism and
Table 3. Association between PAI-1 promoter 4G/5G polymorphism and lymph node metastasis of breast cancer patients Patient (No.) PAI-1 genotype 5G/5G 4G/5G 4G/4G 5G/5G 5G/5G 4G/5G
vs. vs. vs. vs. vs. vs.
4G/4Gþ4G/5G 4G/4Gþ5G/5G 4G/5Gþ5G/5G 4G/4G 4G/5G 4G/4G
Association
No metastasis
Metastasis
651 651 750 336 409 557
326 326 379 176 221 225
OR (95% CI) 1.495 0.974 1.002 1.623 1.410 1.150
(1.048e2.131) (0.740e1.281) (0.640e1.566) (1.086e2.426) (0.966e2.058) (0.844e1.569)
Heterogeneity test p
Q
df
p
0.027 0.850 0.995 0.018 0.075 0.376
0.166 0.170 6.581 0.263 0.142 0.219
3 3 4 3 3 3
0.983 0.982 0.160 0.967 0.986 0.974
PAI-1 Promoter Polymorphism in Breast Cancer
43
Table 4. Association between PAI-1 promoter 4G/5G polymorphism and histologic grade of breast cancer patients Patient (No.) PAI-1 genotype 5G/5G 4G/5G 4G/4G 5G/5G 5G/5G 4G/5G
vs. vs. vs. vs. vs. vs.
Association
Grades I þ II
Grade III
627 540 540 264 371 445
304 287 287 151 189 234
4G/4Gþ4G/5G 4G/4Gþ5G/5G 4G/5Gþ5G/5G 4G/4G 4G/5G 4G/4G
OR (95% CI) 1.013 0.848 1.150 1.015 0.849 1.185
breast cancer risk or its clinicopathological characteristics. This pooled analysis revealed an increased incidence of breast cancer in the populations with 4G/4G genotype than in the populations with other genotypes. In addition, patients with 5G/5G genotype displayed more frequent lymph node metastasis than patients with the remaining genotypes. To the best of our knowledge, this is the first pooled analysis of breast cancer regarding PAI-1 promoter polymorphism. The prevalence of breast cancer in populations with the 4G/4G homozygous genotype was somewhat increased compared to populations with the other genotypes of the PAI-1 promoter. Our result disclosed that individuals with the 4G homozygous genotype may have a 39% increased breast cancer risk compared with the other genotypes. PAI-1 expression is regulated by polymorphisms in the PAI-1 promoter gene as well as numerous external stimuli in physiological and pathological conditions. The PAI-1 promoter contains an E-box motif (50 -CACGTG30 ) at 680 to 675 base pair, the last guanosine nucleotide of which is overlapped with the downward 4G/5G polymorphism nucleotides. PAI-1 transcription is regulated through the binding of upstream stimulatory factor (USF)-1, USF2, and BMAL2/CLOCK to the E-box motifs overlapping the 4G allele (transcriptional enhancer) or the 5G allele (transcriptional suppressor) (31e33). Therefore, the 4G/4G genotype has higher PAI-1 transcriptional activity and high PAI-1 plasma levels (17e19). This pooled analysis also revealed an association between the 4G allele and a higher risk of breast cancer, as compared to the other genotypes.
(0.604e1.700) (0.636e1.131) (0.848e1.561) (0.644e1.553) (0.566e1.272) (0.858e1.637)
Heterogeneity test p
Q
df
p
0.960 0.262 0.368 0.945 0.428 0.303
3.533 1.501 0.105 1.183 2.012 0.370
3 2 2 2 2 2
0.317 0.472 0.949 0.553 0.366 0.831
This meta-analysis showed that the prevalence of lymph node metastasis in patients with the 5G/5G homozygous type was higher than in patients with the other combined genotypes as well as the 4G/4G genotype. There has been considerable controversy with respect to the clinicopathological effects of PAI-1 polymorphism on breast cancer patients. No differences in histological grade and overall survival according to the PAI-1 genotypes were presently evident. However, Lei et al. (24) reported that 5G/5G genotype was correlated with lymph node metastasis and poor overall survival, whereas Alvarez-Millan et al. (21) described the association between 5G/5G genotype and well-differentiated phenotypes. In addition, Zhang et al. (27) reported that the 4G/4G homozygous type was associated with poor overall survival rates of breast cancer. Some studies using PAI1 deficient- and overexpressing mice revealed very little impact of PAI-1 on either carcinogenesis or metastasis, whereas another study indicated that PAI-1 functions differently according to its concentration around the tumor (34). Physiological concentrations of PAI-1 support in vivo cancer invasion and angiogenesis, whereas levels of PAI-1 that are higher than the physiological concentrations inhibit tumor angiogenesis (34). Our meta-analysis confirmed that 5G homozygous genotype has a high rate of lymph node metastasis. However, the role of either PAI-1 expression or PAI-1 promoter polymorphism in tumorigenesis and metastasis remains controversial. The present meta-analysis has several limitations. The genetic investigations contained a relatively small sample
Table 5. Association between PAI-1 promoter 4G/5G polymorphism and overall survival of breast cancer patients Association PAI-1 genotype 5G/5G vs. 4G/4G 5G/5G vs. 4G/5G 4G/5G vs. 4G/4G
Patient (No.) 2309 3137 3680
OR (95% CI) 1.237 (0.642e2.384) 1.245 (0.731e2.123) 1.086 (0.875e1.348)
Heterogeneity test p
Q
df
p
0.524 0.420 0.452
16.161 8.098 0.906
2 2 2
!0.001 0.017 0.636
44
Lee et al./ Archives of Medical Research 44 (2013) 39e45
Study name
Statistics with study removed
Odds ratio (95% CI) with study removed
Lower Upper Point limit limit Z-Value p-Value Palmirotta R Sternlicht MD Yagmurdur MC Lei H Minisini AM Eroglu A Blasiak J
1.401 1.610 1.055 1.628 1.475 1.369 1.338 1.388
1.018 0.987 0.944 1.032 1.048 0.999 0.977 1.031
1.929 2.626 1.180 2.569 2.078 1.875 1.834 1.870
2.069 1.908 0.948 2.093 2.227 1.955 1.815 2.161
0.039 0.056 0.343 0.036 0.026 0.051 0.070 0.031 0.01
0.1
4G/5G+5G/5G
1
10
100
4G/4G
Figure 4. Sensitivity analysis of meta-analysis for breast cancer risk between 4G/4G genotype and the other genotypes.
size, deficient control populations, and were heterogeneous in the control selection. Thus, a meta-analysis should be carried out using adequate inclusion and exclusion criteria for journal selection and needs to be interpreted considering the context of individual studies. In summary, this pooled analysis indicates that the PAI-1 promoter 4G/4G genotype is associated with mildly increased breast cancer risk and that the 5G/5G genotype is related to lymph node metastasis of breast cancer. However, further large-scale studies are required to
Figure 5. Funnel plot of meta-analysis for the breast cancer risk between 4G/5G genotype and 5G/5G genotype. Individual studies are represented by small circles.
elucidate the precise role of PAI-1 promoter 4G/5G polymorphism in breast cancer.
Acknowledgments This study was supported by a Korea University Grant. Conflicts of interest: None.
References 1. Schmitt M, Janicke F, Moniwa N, et al. Tumor-associated urokinasetype plasminogen activator: biological and clinical significance. Biol Chem Hoppe Seyler 1992;373:611e622. 2. Durand MK, Bodker JS, Christensen A, et al. Plasminogen activator inhibitor-I and tumour growth, invasion, and metastasis. Thromb Haemost 2004;91:438e449. 3. Stefansson S, McMahon GA, Petitclerc E, et al. Plasminogen activator inhibitor-1 in tumor growth, angiogenesis and vascular remodeling. Curr Pharm Des 2003;9:1545e1564. 4. Berger DH. Plasmin/plasminogen system in colorectal cancer. World J Surg 2002;26:767e771. 5. Castello R, Espana F, Vazquez C, et al. Plasminogen activator inhibitor-1 4G/5G polymorphism in breast cancer patients and its association with tissue PAI-1 levels and tumor severity. Thromb Res 2006; 117:487e492. 6. Choi JW, Lee JH, Park HS, et al. PAI-1 expression and its regulation by promoter 4G/5G polymorphism in clear cell renal cell carcinoma. J Clin Pathol 2011;64:893e897. 7. Harbeck N, Kates RE, Gauger K, et al. Urokinase-type plasminogen activator (uPA) and its inhibitor PAI-I: novel tumor-derived factors with a high prognostic and predictive impact in breast cancer. Thromb Haemost 2004;91:450e456. 8. Palmirotta R, Ferroni P, Savonarola A, et al. Prognostic value of presurgical plasma PAI-1 (plasminogen activator inhibitor-1) levels in breast cancer. Thromb Res 2009;124:403e408. 9. Sternlicht MD, Dunning AM, Moore DH, et al. Prognostic value of PAI1 in invasive breast cancer: evidence that tumor-specific factors are more important than genetic variation in regulating PAI1 expression. Cancer Epidemiol Biomarkers Prev 2006;15:2107e2114.
PAI-1 Promoter Polymorphism in Breast Cancer 10. Bajou K, Masson V, Gerard RD, et al. The plasminogen activator inhibitor PAI-1 controls in vivo tumor vascularization by interaction with proteases, not vitronectin. Implications for antiangiogenic strategies. J Cell Biol 2001;152:777e784. 11. Czekay RP, Aertgeerts K, Curriden SA, et al. Plasminogen activator inhibitor-1 detaches cells from extracellular matrices by inactivating integrins. J Cell Biol 2003;160:781e791. 12. Webb DJ, Thomas KS, Gonias SL. Plasminogen activator inhibitor 1 functions as a urokinase response modifier at the level of cell signaling and thereby promotes MCF-7 cell growth. J Cell Biol 2001;152:741e752. 13. Chen YQ, Sloan-Lancaster J, Berg DT, et al. Differential mechanisms of plasminogen activator inhibitor-1 gene activation by transforming growth factor-beta and tumor necrosis factor-alpha in endothelial cells. Thromb Haemost 2001;86:1563e1572. 14. Dong J, Fujii S, Li H, et al. Interleukin-6 and mevastatin regulate plasminogen activator inhibitor-1 through CCAAT/enhancer-binding protein-delta. Arterioscler Thromb Vasc Biol 2005;25:1078e1084. 15. Ha H, Oh EY, Lee HB. The role of plasminogen activator inhibitor 1 in renal and cardiovascular diseases. Nat Rev Nephrol 2009;5:203e211. 16. Swiatkowska M, Szemraj J, Cierniewski CS. Induction of PAI-1 expression by tumor necrosis factor alpha in endothelial cells is mediated by its responsive element located in the 4G/5G site. FEBS J 2005;272:5821e5831. 17. Grancha S, Estelles A, Tormo G, et al. Plasminogen activator inhibitor-1 (PAI-1) promoter 4G/5G genotype and increased PAI-1 circulating levels in postmenopausal women with coronary artery disease. Thromb Haemost 1999;81:516e521. 18. Burzotta F, Di Castelnuovo A, Amore C, et al. 4G/5G promoter PAI-1 gene polymorphism is associated with plasmatic PAI-1 activity in Italians: a model of gene-environment interaction. Thromb Haemost 1998;79:354e358. 19. Eriksson P, Kallin B, van’t Hooft FM, et al. Allele-specific increase in basal transcription of the plasminogen-activator inhibitor 1 gene is associated with myocardial infarction. Proc Natl Acad Sci USA 1995;92:1851e1855. 20. Nordt TK, Lohrmann J, Bode C. Regulation of PAI-1 expression by genetic polymorphisms. Impact on atherogenesis. Thromb Res 2001; 103(Suppl 1):S1eS5. 21. Alvarez-Millan JJ, Bocos C, Ferrin V, et al. PAI-1 promoter polymorphism modulates uPA-PAI complex accumulation by breast cancer cells. Oncology 2002;62:286e290.
45
22. Blasiak J, Smolarz B. Plasminogen activator inhibitor-1 (PAI-1) gene 4G/5G promoter polymorphism is not associated with breast cancer. Acta Biochim Pol 2000;47:191e199. 23. Eroglu A, Ulu A, Cam R, et al. Plasminogen activator inhibitor -1 gene 4G/5G polymorphism in patients with breast cancer. J BUON 2006; 11:481e484. 24. Lei H, Hemminki K, Johansson R, et al. PAI-1 -675 4G/5G polymorphism as a prognostic biomarker in breast cancer. Breast Cancer Res Treat 2008;109:165e175. 25. Minisini AM, Fabbro D, Di Loreto C, et al. Markers of the uPA system and common prognostic factors in breast cancer. Am J Clin Pathol 2007;128:112e117. 26. Yagmurdur MC, Atac FB, Tutar NU, et al. Prognostic value of the PAI-1 4G/5G polymorphism in invasive ductal carcinoma of the breast. Int Surg 2008;93:163e168. 27. Zhang X, Shu XO, Cai Q, et al. Functional plasminogen activator inhibitor-1 gene variants and breast cancer survival. Clin Cancer Res 2006;12:6037e6042. 28. Lee JH, Lee ES, Kim YS. Clinicopathologic significance of BRAF V600E mutation in papillary carcinomas of the thyroid: a meta-analysis. Cancer 2007;110:38e46. 29. Normand SL. Meta-analysis: formulating, evaluating, combining, and reporting. Stat Med 1999;18:321e359. 30. Parmar MK, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med 1998;17:2815e2834. 31. Chong NW, Codd V, Chan D, et al. Circadian clock genes cause activation of the human PAI-1 gene promoter with 4G/5G allelic preference. FEBS Lett 2006;580:4469e4472. 32. Dimova EY, Kietzmann T. The MAPK pathway and HIF-1 are involved in the induction of the human PAI-1 gene expression by insulin in the human hepatoma cell line HepG2. Ann NY Acad Sci 2006;1090:355e367. 33. Ma Z, Jhun B, Jung SY, et al. Binding of upstream stimulatory factor 1 to the E-box regulates the 4G/5G polymorphism-dependent plasminogen activator inhibitor 1 expression in mast cells. J Allergy Clin Immunol 2008;121:1006e1012.e1002. 34. Iwaki T, Urano T, Umemura K. PAI-1, progress in understanding the clinical problem and its aetiology. Br J Haematol 2012;157: 291e298.