Clinical Chemistry
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 53: 1280-1288, 2007. First published May 3, 2007; 10.1373/clinchem.2006.082800
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow 082800.Supplemental Data
Right arrow All Versions of this Article:
clinchem.2006.082800v1
53/7/1280    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sieuwerts, A. M.
Right arrow Articles by Foekens, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sieuwerts, A. M.
Right arrow Articles by Foekens, J. A.
Related Collections
Right arrow Cancer Diagnostics (since 2002)
(Clinical Chemistry. 2007;53:1280-1288.)
© 2007 American Association for Clinical Chemistry, Inc.

Cancer Diagnostics

Concentrations of TIMP1 mRNA Splice Variants and TIMP-1 Protein Are Differentially Associated with Prognosis in Primary Breast Cancer

Anieta M. Sieuwerts1,a, Pernille A. Usher2, Marion E. Meijer-van Gelder1, Mieke Timmermans1, John W.M. Martens1, Nils Brünner2, Jan G.M. Klijn1, Hanne Offenberg2 and John A. Foekens1

1 Department of Medical Oncology, Erasmus Medical Center, Rotterdam, The Netherlands.
2 Department of Veterinary Pathobiology, Faculty of Life Sciences, University of Copenhagen, Copenhagen, Frederiksberg C, Denmark.

aAddress correspondence to this author at: Erasmus MC, Josephine Nefkens Institute, Rm. Be 400, PO Box 2040, 3000 CA Rotterdam, The Netherlands. Fax 31-10-408-8377; e-mail a.sieuwerts{at}erasmusmc.nl.


   Abstract
Top
 Abstract
Introduction
 Patients and Methods
 Results
 Discussion
 References
 
Background: TIMP-1 protein is a prognostic factor for recurrence-free and overall survival (OS) time in breast cancer. We evaluated the prognostic value of TIMP1 mRNA and a novel TIMP1 mRNA splice variant in 1301 primary breast cancer patients.

Methods: We measured mRNA transcripts of full-length TIMP1 (TIMP1-v1) and the novel splice variant lacking exon 2 (TIMP1-v2) by use of real-time RT-PCR in frozen primary tumor samples. Transcript concentrations are correlated with histomorphological and biological factors, TIMP-1 protein, and distant metastasis-free survival (MFS) and OS time.

Results: TIMP1-v1 and TIMP1-v2 alone were not informative with respect to predicting prognosis. However, the PCR assay designed to measure the combination of v1 + v2 showed that high concentrations of this combination were associated with good prognosis. In Cox multivariate regression analysis, which also included the traditional prognostic factors, increasing concentrations were independently associated with prolonged MFS (P = 0.004) and OS (P = 0.048). Including TIMP-1 protein and TIMP1-v1+v2 mRNA together in the multivariate model revealed that protein and mRNA were both independently associated with prognosis, with hazard ratios pointing in opposite directions.

Conclusion: High concentrations of TIMP1-v1+2 mRNA are associated with good prognosis in patients with primary breast cancer. Since high concentrations of TIMP-1 protein are associated with poor prognosis, the presence of possible posttranscriptional mechanisms requires further investigation.


   Introduction
Top
 Abstract
 Introduction
Patients and Methods
 Results
 Discussion
 References
 
Tissue inhibitor of metalloproteinases-1 (TIMP-1) 1 is one of the naturally occurring inhibitors of matrix metalloproteinases (MMPs). A number of studies have demonstrated an association between high tumor-tissue concentrations of TIMP1 2 mRNA and TIMP-1 protein and a poor prognosis for breast cancer patients (1)(2)(3)(4)(5)(6)(7). However, TIMP-1 overexpression in malignant cells has also been associated with decreased proliferation (8) and with favorable clinical outcome, particularly in lymph node–negative (LNN) patients (8)(9), or has not been associated with breast cancer prognosis at all (10). One of the reasons for these contradictory reports might be the multifunctional roles ascribed to this protein. TIMP-1 not only inhibits MMPs (11)(12) but also affects cellular proliferation (13)(14), apoptosis (15)(16)(17), and angiogenesis (18), both dependent on and independent of its MMP-inhibiting function. Furthermore, TIMP-1 may exist in multiple molecular forms in the cancer environment and in circulation, e.g., in complex with other proteins or as differentially glycosylated variants. Cox regression analysis of recurrence-free survival in breast cancer patients suggested that a score based on both uncomplexed and total TIMP-1, reflecting the tumor level of TIMP-1/MMP complexes, would be a more precise estimate of prognosis than total TIMP-1 alone (19).

In analogy with free and complexed TIMP-1 protein, the prognostic value of TIMP-1 may be improved by detection of specific splice variants of TIMP1 mRNA. Furthermore, biological understanding of TIMP-1 protein and its gene might help in understanding the controversial findings about TIMP-1 with respect to tumor development and prognosis. To address this, we analyzed mRNA concentrations of the common full-length variant of TIMP1 (v1) and a newly discovered splice variant (v2) lacking exon 2 in a large cohort of 1301 primary breast tumors. We related TIMP1-v1 and v2 expression with histomorphological and clinical factors, mRNA expression of the proliferation marker Ki-67, and total TIMP-1 protein concentrations. Finally, we investigated whether mRNA expression of the TIMP1 splice variants adds to the prognostic value of total TIMP-1 protein.


   Patients and Methods
Top
 Abstract
 Introduction
 Patients and Methods
Results
 Discussion
 References
 
patients
A protocol for studying biological markers associated with disease outcome was approved by the medical ethics committee of the Erasmus Medical Center Rotterdam, The Netherlands (MEC 02.953). The present study, in which coded tumor tissues were used, was performed in accordance with the Code of Conduct of the Federation of Medical Scientific Societies in the Netherlands (http://www.fmwv.nl/). Tumor samples were originally submitted to our reference laboratory from 25 regional hospitals for measurements of steroid hormone receptors. Guidelines for primary treatment were similar for all hospitals. To avoid bias, tumors were selected from our tumor bank at the Erasmus Medical Center (Rotterdam, The Netherlands) by processing all available frozen tumor samples from female patients with breast cancer who entered the clinic during 1979–2001 from whom detailed clinical follow-up was available. Further inclusion criteria were as follows: >100 mg frozen tissue available, invasive breast cancer, no previous other cancer (except basal cell skin cancer or early-stage cervical cancer stage Ia/Ib), no 2nd primary breast tumor at first relapse, no adjuvant systemic treatment for the LNN patients, and >30% invasive tumor cell nuclei. Of the remaining samples, 8% were excluded because of poor RNA quality.

The remaining 1301 patients were treated either with breast-conserving surgery (44%) or with modified mastectomy (56%); 931 patients (72%) received adjuvant radiotherapy. During this period, 195 of the 620 lymph node–positive (LNP) patients did not receive adjuvant systemic therapy; 425 of the LNP patients were treated with adjuvant systemic therapy, of these patients 197 received hormonal therapy, 210 chemotherapy, and 18 received combination therapy. Routine postsurgical follow-up and definition of time to metastasis were as described (20). Median follow-up was 92 months (range 3–248 months). Six hundred sixty-nine (51%) patients developed a distant metastasis and count as events in the analysis for metastasis-free survival (MFS). Seventy-two patients (6%) died without evidence of disease and were censored at last follow-up in the analysis of MFS. Five hundred twenty-six patients (40%) died after a previous relapse. Thus, 598 patients (46%) were counted as events in the analysis of overall survival (OS). Tumor staging was according to the Union Internationale Contre le Cancer tumor node metastasis classification. Other relevant patient and tumor characteristics are listed in Table 1 (see also Table 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol53/issue7).


View this table:
[in this window]
[in a new window]

  		Table 1. Associations of TIMP1 mRNA variant concentrations with histomorphological and clinical factors.1

tissue processing and estimation of the amount of invasive tumor cells
We processed tissue and estimated amount of invasive tumor cells as described (21)(22). Only specimens with at least 30% of the nuclei of epithelial tumor cell origin and distributed uniformly over at least 70% of the hematoxylin-eosin–stained tissue section area were included. Furthermore, we dichotomized our tumor cohort at the median of 70% tumor cell nuclei in stroma-rich tumors (primary tumors containing ≥30% stromal components) and stroma-poor tumors (primary tumors containing at least 70% tumor cells).

rna isolation, CDNA synthesis, and quantification of specific MRNA species
RNA isolation, cDNA synthesis, quantification of specific mRNA species, and quality control checks were done as described in detail (21). We performed real-time RT-PCR in an ABI Prism 7700 Sequence Detection System (Applied Biosystems) and a Mx3000PTM Real-Time PCR System (Stratagene) using both Assay-on-Demand from Applied Biosystems and the intron-spanning forward and reverse primer combinations at the conditions shown in Supplemental Data Table 2. RT-PCR products generated with these primers by cell lines and cancer tissue samples were gel-purified and sequenced (AGOWA, Berlin, Germany), and sequences were confirmed by BLAST search as described (23). Primer sequences for ESR1, PGR, and the housekeeping genes have all been described, as have the PCR reactions and validations performed to ensure PCR specificity (21). To measure concentrations of the proliferation marker Ki-67, we used the Hs00606991 m1 Assay-on-Demand from Applied Biosystems. Concentrations of the target genes, expressed relative to our housekeeping set [low-abundance housekeeping gene hydroxymethylbilane synthase (HMBS, formerly porphobilinogen deaminase, PBGD), medium-abundance hypoxanthine phosphoribosyltransferase (HPRT), and high-abundance ß2-microglobulin (B2M)], were quantified as follows: mRNA target = 2(mean Ct housekeeping – mean Ct target), as described (21).

esr1 and pgr MRNA receptor status
We established mRNA cutpoints to define tumors as steroid hormone positive at 0.2 for estrogen receptor (ER) and 0.1 for progesterone receptor (PGR). We compared these mRNA cutoffs with the established protein cutpoints of 10 fmol/mg protein in the 1203 samples with known protein concentrations as established by ELISA. For ER, sensitivity and specificity were 93% and 72%, respectively, and positive and negative predictive accuracy were 90% and 81%. For PGR, sensitivity and specificity were 84% and 83%, respectively, and positive and negative predictive accuracy rates were 90% and 75%.

elisa
To compare TIMP1 mRNA concentrations with TIMP-1 protein, we used protein concentrations that were previously measured in cytosol preparations of the same tumors (6).

statistics
Computations were done with the use of the STATA statistical package, release 9 (STATA). We assessed differences in concentrations with the Mann–Whitney U-test or Kruskal–Wallis test. In these tests, patient and tumor characteristics were used as grouping variables. We tested the strengths of the associations between continuous variables with the Spearman rank correlation (rs). Variables were either log-transformed or Box-Cox–transformed to reduce the skewness. We investigated the prognostic values of the clinical and biological variables, with MFS and OS as the endpoints in the univariate, multivariate, and interaction analyses, with the use of the Cox proportional hazards model. The hazard ratio (HR) and its 95% CI were derived from these results. The proportionality assumption was investigated with a test based on the Schoenfeld residuals. We used Kaplan-Meier survival plots and log-rank tests for trend to assess the differences in time of the predicted high-risk, intermediate-risk, and low-risk groups of patients. All tests were 2-sided, and P <0.05 was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Patients and Methods
 Results
Discussion
 References
 
correlations between real-time timp1 pcr assays
In our initial screening using combinations of primers located 2 exons apart to cover the 6 exons of TIMP1, we measured mRNA transcripts of 2 variants of TIMP1 in a set of cultured cell lines and primary breast tumors: full-length TIMP1 (v1) and a novel variant lacking exon 2 (v2). The identification of these mRNA variants was confirmed by gel electrophoresis and sequence analysis and was further evaluated by real-time RT-PCR with the assays shown in Supplemental Data Table 2 in a representative selection of 180 primary breast tumors and various cell lines (Table 2 ). In these analyses, only EVSA-T and CAMA-1 cells lacked expression of TIMP1-v2 (Table 2 ). Although with our TaqMan probe–based TIMP1-v2 assay we were unable to detect v2 transcripts above baseline in ZR75.1 and T47-D cells within 45 amplification rounds, gel analysis showed that a faint 134-bp product representative for TIMP1-v2 was produced by these cells when amplified in 45 cycles with our SYBR-based TIMP1-v1+2 assay. The highest expression of TIMP1 mRNA was measured in a primary breast tumor–derived fibroblast strain (24) (Table 2 , 19T).


View this table:
[in this window]
[in a new window]

  		Table 2. mRNA concentrations of TIMP1 variants.1

All tumors readily expressed TIMP1-v1 mRNA; for only 16 of 1301 tumors were we unable to detect v2 transcripts within 45 amplification rounds. Because TIMP-1 protein overexpression has been inversely associated with cell proliferation (8), we matched our TIMP1 PCR data with those of the proliferation marker Ki-67 measured in the same preparations. The strength of the associations between TIMP1-v1, TIMP1-v2, TIMP1-v1+2, and Ki-67 mRNA are summarized in Table 3 . Whereas TIMP1-v1+2 showed a statistically significant association with TIMP1-v1, no correlation was observed with TIMP1-v2 (Spearman rs = 0.51 and 0.04, respectively). TIMP1-v1+2, compared with TIMP1-v1 and TIMP1-v2 separately, showed the strongest (inverse) correlation with Ki-67 (rs = –0.31, –0.09, and 0.01, respectively).


View this table:
[in this window]
[in a new window]

  		Table 3. Spearman correlations between mRNA concentrations measured by real-time RT-PCR.

associations with histomorphological and clinical factors
Associations of mRNA expression of TIMP1-v1, TIMP1-v2, and TIMP1-v1+2 with patient and tumor characteristics are shown in Table 1Up . Most notable are the opposing results of the TIMP1-v2 assay with the TIMP1-v1 and TIMP1-v1+2 assays. Whereas concentrations assessed with the TIMP1-v2 assay were negatively associated with age, and were lower in postmenopausal patients, results were the opposite for the other 2 TIMP1 assays. Furthermore, only for the TIMP1-v2 assay were concentrations significantly lower in the group of LNN patients compared with the group of LNP patients, and higher in the stroma-rich tumors compared with the group of stroma-poor tumors. On the other hand, concentrations measured with the TIMP1-v1 and TIMP1-v1+2 assays were significantly higher in ER-positive tumors compared with ER-negative tumors. No such difference was observed for concentrations measured with the TIMP1-v2 assay. Also with respect to grade and tumor size, only the TIMP1-v1 and TIMP1-v1+2 assays showed a relation—higher concentrations in the prognostically more favorable tumors. Finally, for all TIMP1 variants, concentrations were higher in infiltrating lobular carcinoma compared with infiltrating ductal carcinoma.

univariate and multivariate analysis for mfs and os
To assess a possible relationship of TIMP1 mRNA with prognosis, we first performed Cox univariate analyses for MFS and OS as a function of continuous TIMP1 mRNA concentrations. Because the proportional hazards assumption was violated for the total follow-up time, we restricted our exploration of the relationships of TIMP1 with MFS to the first 5 years of follow-up, with 571 failures in the 1301 patients. In these analyses, only concentrations measured with the TIMP1-v1+2 assay were significantly associated with MFS (HR 0.80, P <0.001) and OS (HR 0.85, P <0.001). Next, the TIMP1 variants were separately introduced to the base multivariate model that included the factors of age, menopausal status, nodal status, tumor size, grade, ER, and PGR (Table 4 ). Of the 3 TIMP1 assays, only the assay measuring both variant 1 and variant 2 (TIMP1-v1+2) contributed significantly to the multivariate model for MFS (HR 0.89, P = 0.004) (Table 4 ) and OS (HR 0.92, P = 0.048) (data not shown in a table). Adding Ki-67 and adjuvant systemic therapy or radiotherapy to the multivariate model that included TIMP1-v1+2 did not alter the coefficients of TIMP1-v1+2. To visualize the prognostic value of TIMP1-v1+2 in Kaplan-Meier curves, we divided mRNA concentration curves into 3 equal parts (low, intermediate, and high). These curves are shown in Fig. 1A for MFS and Fig. 1B for OS.


View this table:
[in this window]
[in a new window]

  		Table 4. Cox univariate and multivariate analysis for 5-year MFS as a function of TIMP1 variants in primary breast tumors from 1301 breast cancer patients.


Figure 1
View larger version (17K):
[in this window]
[in a new window]

  		Figure 1. Relationship between TIMP1-v1+2, divided in 3 equal parts in high, intermediate, and low concentrations, and MFS (A) and OS (B) in 1301 primary breast cancer patients. Patients at risk are indicated.

nodal and er status
We next performed exploratory Cox univariate analyses for MFS and OS as a function of mRNA expression in the clinically relevant subgroups of LNN, LNP, ER+, and ER– patients. Only those analyses that gave significant results after concentrations were entered as transformed continuous variables are shown in Table 5 . TIMP1-v1 mRNA concentrations alone were not significantly associated with nodal status and steroid hormone receptor status. But the following 2 divergent observations between the various assays are notable. First, whereas concentrations measured with the TIMP1-v2 assay were associated with good prognosis only in the subgroup of LNP patients, the association of increasing concentrations measured with the TIMP1-v1+2 assay and good prognosis were independent of nodal status. Second, whereas increasing concentrations of TIMP1-v1+2 were associated with good prognosis exclusively in the group of ER+ tumors, concentrations measured with the TIMP1-v2 assay were associated with good prognosis only in the group of ER– tumors.


View this table:
[in this window]
[in a new window]

  		Table 5. Cox univariate analysis for distant MFS and OS before and after categorizing tumors according to histomorphological and clinical criteria.

correlations between timp-1 Mrna and protein
To compare TIMP1 mRNA with total TIMP-1 protein, we made use of protein concentrations that were previously measured with ELISA in cytosol preparations of the same tumors (6). In the 839 tumors with both measures, the highest correlation between total protein and mRNA was observed for the real time RT-PCR assay able to measure TIMP1-v1 (rs = 0.34, P <0.001). The strength of the association was lower for the PCR assay able to measure both variant 1 and 2 (TIMP1-v1+2) (rs = 0.28, P <0.001) and inverse for the TIMP1-v2 assay (rs = –0.11, P = 0.01).

To ensure that our cohort of 1301 patients did not differ from the cohort of 2984 patients with protein data, we repeated all analyses for the overlapping cohort of 839 patients. We divided the protein concentrations in 3 equal parts to classify the tumors at the protein level as TIMP-1 low, intermediate, or high. High and intermediate vs low concentrations of TIMP-1 protein were significantly associated with shorter MFS in univariate analysis (HR intermediate vs low 1.56, HR high vs low 1.37, P = 0.002) and multivariate analysis (HR intermediate vs low 1.42, HR high vs low 1.42, P = 0.010), which is in agreement with the original study, in which high tumor tissue concentrations of TIMP-1 protein were identified as an independent marker of poor prognosis in 2984 primary breast cancers (6).

Next, we similarly compared the prognostic value of mRNA concentrations measured with our TIMP1-v1, v1+2, and v2 PCR assays in these 839 patients. Only increasing concentrations of TIMP1-v1+2 mRNA were significantly associated with a prolonged MFS in univariate analysis (HR intermediate vs low 0.81, HR high vs low 0.59, P <0.001) and multivariate analysis (HR intermediate vs low 0.89, HR high vs low 0.66, P = 0.020).

Finally, we explored a potential interaction between TIMP-1 protein and TIMP1 mRNA with respect to MFS. No such interaction was observed (P = 0.56). Including log-transformed continuous concentrations of both TIMP-1 protein and TIMP1-v1+2 mRNA to the base multivariate model for MFS revealed that TIMP-1 protein and TIMP1-v1+2 mRNA were both independently associated with prognosis, with HRs pointing in opposite directions (HR 1.36, 95% CI 1.16–1.61, P <0.001, n = 839 for protein and HR 0.78, 95% CI 0.69–0.87, P <0.001, n = 839 for mRNA).


   Discussion
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
References
 
Many research groups have investigated the link between prognosis in breast cancer and TIMP-1 protein or mRNA measured in primary tumors and TIMP-1 protein in serum/plasma, with conflicting results (1)(2)(3)(4)(5)(6)(7)(8)(9)(10). In agreement with earlier studies on TIMP1 mRNA (9)(10), but in contrast to other studies (1)(3)(4), we were unable to confirm that high concentrations of TIMP1 mRNA were associated with poor prognosis similar to results for TIMP-1 protein. However, mRNA transcript concentrations cannot always be compared with protein concentrations. This was confirmed in this study, in which we found a rather poor correlation between full-length TIMP-1 mRNA and TIMP-1 protein. In our view, the apparently contrasting findings between the prognostic value of TIMP1 mRNA expression and TIMP-1 protein suggests that key regulators of TIMP-1 protein involved in an adverse outcome act posttranscriptionally. Such regulatory mechanisms affecting protein concentrations, activity, and stability can act at the level of mRNA translation, protein folding, glycosylation, and (proteosomal) protein degradation. This possibility needs further investigation.

Our main purpose was to investigate the potential prognostic value of TIMP1 mRNA and a newly discovered splice variant to gain more knowledge on the biology of TIMP-1 in breast cancer. To address this, our retrospective study that included RNA preparations from tumor tissue obtained from 1301 patients suffering from primary breast cancer is, to the best of our knowledge, the largest study performed on the mRNA concentrations of TIMP1 to date. Because of different assay conditions, absolute values of real-time RT-PCR assays can be compared only within an assay, and values from different assays, such as our 2 assays measuring individual transcripts of TIMP-v1 and TIMP-v2, cannot simply be added. With our multiplex TIMP1-v1+2 assay that measures both transcripts in the same reaction with the same primer pairs, we corrected as much as possible for such differences in assay conditions. However, by measuring both transcripts in the same reaction, we cannot exclude that the shorter (v2) variant was favored relative to the larger (v1) variant.

All tumors expressed full-length TIMP1-v1 mRNA; we were unable to detect v2 transcripts for only 16 of 1301 tumors. Sequence analysis of TIMP1-v2 has already revealed that this variant lacking exon 2, if translated, is probably a soluble, intracellular protein lacking part of the region that directs the main inhibitory MMP activity (unpublished data). It is therefore unlikely that a putative protein of TIMP1-v2 forms complexes with MMPs, and thus it probably exhibits a biological function different from full length TIMP-1—indeed, our data suggest that. We found that TIMP1-v1 mRNA concentrations increase with age, are higher in ER/PGR-positive tumors, and are higher in smaller-sized and moderately differentiated to well differentiated tumors. In contrast to some reports (3)(4)(9), but in agreement with another report (10), TIMP1-v1 mRNA concentrations were not different in our cohort of 620 LNP patients compared with the group of 681 LNN patients. These discrepancies might be due to the relatively small sample sizes in the earlier studies [n = 30 LNN and 24 LNP (3); n = 49 LNN and 66 LNP (4)].

Separate evaluation of TIMP1-v2 in our patient cohort revealed a strong inverse correlation with age and no correlation with ER, PGR, grade, and tumor size. In addition, only for TIMP1-v2, concentrations were higher in stroma-rich compared with stroma-poor primary breast tumors. The lack of a correlation between ER and TIMP1-v2 concentrations suggests that TIMP1-v2, unlike TIMP1-v1, is regulated by an ER-independent mechanism. Moreover, the lower TIMP1-v2 and higher TIMP1-v1 mRNA concentrations in the older age group support our hypothesis that v2 is regulated by a different mechanism. Another observation we made is the relatively strong negative correlation between the proliferation marker Ki-67 and TIMP1-v1+2. No such correlation was observed for the TIMP1 assays able to measure the variants separately. This finding suggests that only the combined action of full-length TIMP-1 and its del-2 variant are effectively able to downregulate proliferation or monitor-reduced proliferation.

We recently raised the hypothesis that high concentrations of total TIMP-1 protein are not necessarily associated with poor prognosis but that the association depends on the ratio of uncomplexed/total TIMP-1 (19). In analogy with this, our present study shows that TIMP1-v1 mRNA and TIMP1-v2 mRNA alone were not associated with prognosis. However, our real-time RT-PCR assay developed to measure both transcripts at the same time revealed that high mRNA concentrations of the combination of both variants were associated with low tumor aggressiveness. Whether changing the balance between full-length TIMP-1 and its variant lacking exon 2 has potential as a possible therapeutic approach to reduce tumor aggressiveness remains to be investigated. To establish this, and since it is only the actual protein that is biologically active, variant-specific immunohistochemistry and a quantitative assay (ELISA) able to measure the putative del-2 protein in relation to full-length TIMP-1 protein are required.

In conclusion, this retrospective study on a large cohort of primary breast cancers provides evidence that the combined expression of full-length TIMP1-v1 mRNA and its v2 variant lacking exon 2 are associated with low tumor aggressiveness. This splice variant-dependent association might help our understanding of the role of TIMP-1 with respect to breast cancer.


   Acknowledgments
 
Grant funding/support: Partly supported by the Danish Cancer Society and the Danish Medical Research Counsel.

Financial disclosures: None declared.

Acknowledgments: This work is the result of an EORTC-PathoBiology Group collaboration. We thank Miranda Arnold, Anneke Goedheer, Roberto Rodriguez-Garcia, Anita Trapman, Vanja de Weerd, and Henk Portengen for their technical support.


   Footnotes
 
1 Nonstandard abbreviations: TIMP-1, tissue inhibitor of metalloproteinases-1; MMP, matrix metalloproteinase; LNN, lymph node–negative; LNP, lymph node–positive; MFS, metastasis-free survival; OS, overall survival; Ct, threshold cycle; ER, estrogen receptor; PGR, progesterone receptor; HR, hazard ratio.

2 Human genes: TIMP1, TIMP metallopeptidase inhibitor 1; TIMP1-v1, TIMP1 full-length variant; TIMP1-v2, TIMP1 variant lacking exon 2; ESR1, estrogen receptor 1 (formerly ER-{alpha}); PGR, progesterone receptor; HMBS, hydroxymethylbilane synthase, formerly porphobilinogen deaminase, PBGD; HPRT, hypoxanthine phosphoribosyltransferase; B2M, ß-2-microglobulin.


   References
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
 References
 

  1. Ree AH, Florenes VA, Berg JP, Maelandsmo GM, Nesland JM, Fodstad O. High levels of messenger RNAs for tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2) in primary breast carcinomas are associated with development of distant metastases. Clin Cancer Res 1997;3:1623-1628.[Abstract]
  2. McCarthy K, Maguire T, McGreal G, McDermott E, O’Higgins N, Duffy MJ. High levels of tissue inhibitor of metalloproteinase-1 predict poor outcome in patients with breast cancer. Int J Cancer 1999;84:44-48.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  3. Castello R, Estelles A, Vazquez C, Falco C, Espana F, Almenar SM, et al. Quantitative real-time reverse transcription-PCR assay for urokinase plasminogen activator, plasminogen activator inhibitor type 1, and tissue metalloproteinase inhibitor type 1 gene expressions in primary breast cancer. Clin Chem 2002;48:1288-1295.[Abstract/Free Full Text]
  4. Nakopoulou L, Giannopoulou I, Stefanaki K, Panayotopoulou E, Tsirmpa I, Alexandrou P, et al. Enhanced mRNA expression of tissue inhibitor of metalloproteinase-1 (TIMP-1) in breast carcinomas is correlated with adverse prognosis. J Pathol 2002;197:307-313.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  5. Schrohl AS, Christensen IJ, Pedersen AN, Jensen V, Mouridsen H, Murphy G, et al. Tumor tissue concentrations of the proteinase inhibitors tissue inhibitor of metalloproteinases-1 (TIMP-1) and plasminogen activator inhibitor type 1 (PAI-1) are complementary in determining prognosis in primary breast cancer. Mol Cell Proteomics 2003;2:164-172.[Abstract/Free Full Text]
  6. Schrohl AS, Holten-Andersen MN, Peters HA, Look MP, Meijer-van Gelder ME, Klijn JG, et al. Tumor tissue levels of tissue inhibitor of metalloproteinase-1 as a prognostic marker in primary breast cancer. Clin Cancer Res 2004;10:2289-2298.[Abstract/Free Full Text]
  7. Kuvaja P, Talvensaari-Mattila A, Paakko P, Turpeenniemi-Hujanen T. The absence of immunoreactivity for tissue inhibitor of metalloproteinase-1 (TIMP-1), but not for TIMP-2, protein is associated with a favorable prognosis in aggressive breast carcinoma. Oncology 2005;68:196-203.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  8. Nakopoulou L, Giannopoulou I, Lazaris A, Alexandrou P, Tsirmpa I, Markaki S, et al. The favorable prognostic impact of tissue inhibitor of matrix metalloproteinases-1 protein overexpression in breast cancer cells. Apmis 2003;111:1027-1036.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  9. Inoue H, Mimori K, Shiraishi T, Kataoka A, Sadanaga N, Ueo H, et al. Expression of tissue inhibitor of matrix metalloproteinase-1 in human breast carcinoma. Oncol Rep 2000;7:871-874.[Web of Science][Medline] [Order article via Infotrieve]
  10. Span PN, Lindberg RL, Manders P, Tjan-Heijnen VC, Heuvel JJ, Beex LV, et al. Tissue inhibitors of metalloproteinase expression in human breast cancer: TIMP-3 is associated with adjuvant endocrine therapy success. J Pathol 2004;202:395-402.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  11. Jiang Y, Goldberg ID, Shi YE. Complex roles of tissue inhibitors of metalloproteinases in cancer. Oncogene 2002;21:2245-2252.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  12. Wurtz SO, Schrohl AS, Sorensen NM, Lademann U, Christensen IJ, Mouridsen H, et al. Tissue inhibitor of metalloproteinases-1 in breast cancer. Endocr Relat Cancer 2005;12:215-227.[Abstract/Free Full Text]
  13. Luparello C, Avanzato G, Carella C, Pucci-Minafra I. Tissue inhibitor of metalloprotease (TIMP)-1 and proliferative behaviour of clonal breast cancer cells. Breast Cancer Res Treat 1999;54:235-244.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  14. Porter JF, Shen S, Denhardt DT. Tissue inhibitor of metalloproteinase-1 stimulates proliferation of human cancer cells by inhibiting a metalloproteinase. Br J Cancer 2004;90:463-470.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  15. Li G, Fridman R, Kim HR. Tissue inhibitor of metalloproteinase-1 inhibits apoptosis of human breast epithelial cells. Cancer Res 1999;59:6267-6275.[Abstract/Free Full Text]
  16. Lee SJ, Yoo HJ, Bae YS, Kim HJ, Lee ST. TIMP-1 inhibits apoptosis in breast carcinoma cells via a pathway involving pertussis toxin-sensitive G protein and c-Src. Biochem Biophys Res Commun 2003;312:1196-1201.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  17. Liu XW, Taube ME, Jung KK, Dong Z, Lee YJ, Roshy S, et al. Tissue inhibitor of metalloproteinase-1 protects human breast epithelial cells from extrinsic cell death: a potential oncogenic activity of tissue inhibitor of metalloproteinase-1. Cancer Res 2005;65:898-906.[Abstract/Free Full Text]
  18. Tomlinson J, Barsky SH, Nelson S, Singer S, Pezeshki B, Lee MC, et al. Different patterns of angiogenesis in sarcomas and carcinomas. Clin Cancer Res 1999;5:3516-3522.[Abstract/Free Full Text]
  19. Wurtz SO, Christensen IJ, Schrohl AS, Mouridsen H, Lademann U, Jensen V, et al. Measurement of the uncomplexed fraction of tissue inhibitor of metalloproteinases-1 in the prognostic evaluation of primary breast cancer patients. Mol Cell Proteomics 2005;4:483-491.[Abstract/Free Full Text]
  20. Wang Y, Klijn JG, Zhang Y, Sieuwerts AM, Look MP, Yang F, et al. Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet 2005;365:671-679.[Web of Science][Medline] [Order article via Infotrieve]
  21. Sieuwerts AM, Meijer-van Gelder ME, Timmermans M, Trapman AM, Garcia RR, Arnold M, et al. How ADAM-9 and ADAM-11 differentially from estrogen receptor predict response to tamoxifen treatment in patients with recurrent breast cancer: a retrospective study. Clin Cancer Res 2005;11:7311-7321.[Abstract/Free Full Text]
  22. Sieuwerts AM, Look MP, Meijer-van Gelder ME, Timmermans M, Trapman AM, Garcia RR, et al. Which cyclin E prevails as prognostic marker for breast cancer? Results from a retrospective study involving 635 lymph node-negative breast cancer patients. Clin Cancer Res 2006;12:3319-3328.[Abstract/Free Full Text]
  23. Holten-Andersen MN, Hansen U, Brunner N, Nielsen HJ, Illemann M, Nielsen BS. Localization of tissue inhibitor of metalloproteinases 1 (TIMP-1) in human colorectal adenoma and adenocarcinoma. Int J Cancer 2005;113:198-206.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
  24. Martens JW, Sieuwerts AM, Bolt-deVries J, Bosma PT, Swiggers SJ, Klijn JG, et al. Aging of stromal-derived human breast fibroblasts might contribute to breast cancer progression. Thromb Haemost 2003;89:393-404.[Web of Science][Medline] [Order article via Infotrieve]



The following articles in journals at HighWire Press have cited this article:


Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
T. A. O'Mara, J. A. Clements, and A. B. Spurdle
The Use of Predictive or Prognostic Genetic Biomarkers in Endometrial and Other Hormone-Related Cancers: Justification for Extensive Candidate Gene Single Nucleotide Polymorphism Studies of the Matrix Metalloproteinase Family and their Inhibitors
Cancer Epidemiol. Biomarkers Prev., September 1, 2009; 18(9): 2352 - 2365.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
T. van Agthoven, A. M. Sieuwerts, M. E. Meijer-van Gelder, M. P. Look, M. Smid, J. Veldscholte, S. Sleijfer, J. A. Foekens, and L. C.J. Dorssers
Relevance of Breast Cancer Antiestrogen Resistance Genes in Human Breast Cancer Progression and Tamoxifen Resistance
J. Clin. Oncol., February 1, 2009; 27(4): 542 - 549.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow 082800.Supplemental Data
Right arrow All Versions of this Article:
clinchem.2006.082800v1
53/7/1280    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sieuwerts, A. M.
Right arrow Articles by Foekens, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sieuwerts, A. M.
Right arrow Articles by Foekens, J. A.
Related Collections
Right arrow Cancer Diagnostics (since 2002)

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS