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

• DOI: 10.1074/mcp.m300019-mcp200
• Publication Date: 2003-07-26T23:20:31Z
• Authors: Anne-Sofie Schrohl; Ib Jarle Christensen; Anders N. Pedersen; Vibeke Jensen; Henning Mouridsen; Gillian Murphy; John A. Foekens; Nils Bruünner; Mads Nikolaj Holt...
• Keywords: Univariate analysis; Plasminogen activator inhibitor-1; Primary tumor

The purpose of this study was to investigate the association between tumor tissue levels total inhibitor metalloproteinases-1 (TIMP-1) and prognosis in patients with primary breast cancer analyze whether measurement TIMP-1 extracts added prognostic information that obtained from measurements urokinase-type plasminogen activator type 1 (PAI-1). An established sandwich enzyme-linked immunosorbent assay thoroughly validated for used determine 341 detergent-extracted samples cancer. median age 56 years (range, 29–75 years), 164 were lymph node-negative, 177 node-positive. follow-up time 8.5 7.3–11.3 during 153 experienced recurrence disease, 136 died. In univariate survival analysis, we found a significant level both shorter recurrence-free (p = 0.0004) overall 0.03). multivariate higher significantly independently predicted < 0.05, hazard ratios >1, comparing quartiles II–IV I). addition, PAI-1. conclusion, high are associated poor Furthermore adds However, further validation independent data sets is needed. present use adjuvant systemic therapy based on classical parameters (nodal status, size, grade malignancy, receptor age) (1.Goldhirsch A. Glick J.H. Gelber R.D. Coates A.S. Senn H.J. Meeting highlights: International Consensus Panel Treatment Primary Breast Cancer. Seventh Conference Adjuvant Therapy Cancer.J. Clin. Oncol. 2001; 19: 3817-3827Google Scholar). small proportion who not offered will still experience i.e. these undertreated. large would have remained free even without therapy, overtreated. To optimize an improvement methods identification at risk developing Proteolytic enzymes, their receptors, inhibitors all important processes invasion metastasis. (uPA) 1The abbreviations are: uPA, activator; PAI-1, 1; MMP, matrix metalloproteinase; TIMP, metalloproteinases; DBCG, Danish Cancer Cooperative Group; CI, confidence interval; HR, ratio; ELISA, assay. system, serine protease uPA mediates localized activation subsequent generation plasmin. Proteolysis mediated by plasmin then involved several such as cell migration (2.Andreasen P.A. Kjoller L. Christensen Duffy M.J. system metastasis: review.Int. J. 1997; 72: 1-22Google Both its principal (PAI-1) now well recognized markers and/or PAI-1 related (3.Look M.P. van Putten W.L. Harbeck N. I.J. Thomssen C. Kates R. Spyratos F. Ferno M. Eppenberger-Castori S. Sweep C.G. Ulm K. Peyrat J.P. Martin P.M. Magdelenat H. Brunner Duggan Lisboa B.W. Bendahl P.O. Quillien V. Daver Ricolleau G. Meijer-van Gelder M.E. Manders P. Fiets W.E. Blankenstein M.A. Broet Romain Daxenbichler Windbichler Cufer T. Borstnar Kueng W. Beex L.V. Klijn J.G. O'Higgins Eppenberger U. Janicke Schmitt Foekens J.A. Pooled analysis impact 8377 patients.J. Natl. Inst. 2002; 94: 116-128Google Scholar), two recently reached evidence clinical Scholar, 4.Hayes D.F. Trock B. Harris A.L. Assessing factors: when "statistically significant" clinically useful?.Breast Res. Treat. 1998; 52: 305-319Google 5.Janicke Prechtl Meisner Untch Selbmann H.K. Graeff Randomized chemotherapy trial high-risk, node-negative identified 1.J. 93: 913-920Google That explained demonstration proangiogenic effect (6.Bajou Noel Gerard Masson Holst-Hansen Skobe Fusenig N.E. Carmeliet Collen D. Foidart J.M. Absence host prevents vascularization.Nat. Med. 4: 923-928Google Scholar) antiapoptotic function molecule (7.Kwaan H.C. Wang Svoboda Declerck P.J. Plasminogen may promote tumour growth through inhibition apoptosis.Br. 2000; 82: 1702-1708Google classification leaves group that, although classified being measurements, do disease. Thus, additional needed complete stratification possible markers. As only one extracellular proteolytic systems, looked other systems alternative marker candidates. One metalloproteinases (MMPs) which interactions been established. For instance, MMP-3 capable activating pro-uPA (8.Orgel Schroder Hecker-Kia Weithmann K.U. Kolkenbrock Ulbrich cleavage pro-urokinase stromelysin-1.Clin. Chem. Lab. 36: 697-702Google also MMP-3-mediated separates receptor-binding domain enzymatically active part (9.Ugwu Van Hoef Bini Lijnen H.R. stromelysin-1 (MMP-3).Biochemistry. 37: 7231-7236Google inactivating means (10.Lijnen Arza Inactivation inhibitor-1 specific proteolysis (MMP-3).J. Biol. 275: 37645-37650Google MMPs plasminogen, similarly proteases can activate (11.Sternlicht M.D. Werb Z. How regulate behavior.Annu. Rev. Cell Dev. 17: 463-516Google make up family zinc-dependent endopeptidases cooperatively degrading most components (12.Egeblad New functions progression.Nat. 2: 161-174Google interact non-covalently members (TIMP) endogenous metalloproteinase inhibitors, thus inhibiting activity (13.Welgus H.G. Jeffrey J.J. Eisen A.Z. Roswit W.T. Stricklin G.P. Human skin fibroblast collagenase: interaction substrate inhibitor.Collagen Relat. 1985; 5: 167-179Google TIMPs assigned steps process metastasis, influencing progression four known (14.Docherty A.J. Lyons Smith B.J. Wright E.M. Stephens P.E. T.J. Murphy Reynolds Sequence human identity erythroid-potentiating activity.Nature. 318: 66-69Google It expressed range different types variety body fluids tissues (15.Welgus collagenase inhibitor. Comparative studies connective tissues, serum, amniotic fluid.J. 1983; 258: 12259-12264Google addition MMP-inhibitory function, suggested TIMP-1. These include antiapoptotic, growth-promoting, possibly antiangiogenic effects (16.Li Fridman Kim Tissue metalloproteinase-1 inhibits apoptosis epithelial cells.Cancer 1999; 59: 6267-6275Google 17.Lafleur Handsley M.M. Knauper Edwards D.R. Endothelial tubulogenesis within fibrin gels specifically requires membrane-type-matrix (MT-MMPs).J. Sci. 115: 3427-3438Google 18.Luparello Avanzato Carella Pucci-Minafra I. metalloprotease (TIMP)-1 proliferative behaviour clonal cells.Breast 54: 235-244Google 19.Fernandez H.A. Kallenbach Seghezzi Grossi E. Colvin Schneider Mignatti Galloway Inhibition endothelial gene transfer metalloproteinases-1.J. Surg. 156-162Google 20.Cornelius L.A. Nehring L.C. Harding Bolanowski Welgus Kobayashi D.K. Pierce R.A. Shapiro S.D. Matrix generate angiostatin: neovascularization.J. Immunol. 161: 6845-6852Crossref Google similar biological Having more complex role dissemination cells be expected. Especially idea purely protective TIMP-1, could anticipated should abandoned favor multifunctional one. tissue, mRNA transcripts detected stromal predominance peritumoral stroma invasive carcinomas (21.Brummer O. Athar Riethdorf Loning Herbst Matrix-metalloproteinases 1, 2, 3 2 benign malignant lesions: situ hybridization study.Virchows Arch. 435: 566-573Google 53 cancerous non-cancerous breast, than (22.Yoshiji Gomez D.E. Thorgeirsson U.P. Enhanced RNA expression cancer.Int. 1996; 69: 131-134Google Additionally value has investigated number studies. 34 patients, node-positive development distant metastases, poorer 5-year (23.Ree A.H. Florenes V.A. Berg Maelandsmo G.M. Nesland Fodstad High messenger RNAs (TIMP-1 TIMP-2) metastases.Clin. 3: 1623-1628Google Furthermore, 139 carcinomas, protein disease-free demonstrated using optimized cut-off point (24.McCarthy Maguire McGreal McDermott predict outcome 84: 44-48Google study, patient outcome. correlation clinicopathological (lymph node hormone menopausal age). particular, wanted uPA/PAI-1 measurements. included study. All underwent surgery period 1989–1993 Denmark histologically verified treatment protocols Group, DBCG-82 or DBCG-89 (25.Andersen K.W. Mouridsen H.T. Group (DBCG). A description register nation-wide program cancer.Acta 1988; 27: 627-647Google 26.Danish (DBCG)Danish 1977–1997. Secretariat, Copenhagen1998Google Patients metastases diagnosis previous malignancies excluded protocols. 10,918 registered DBCG period, provided unfixed frozen available. representative slight exceptions regarding size; younger medium-sized tumors slightly over-represented (27.Pedersen A.N. R.W. Briand Dano urokinase type-1 cancer: relation survival.Cancer 60: 6927-6934Google years). node-positive, node-negative; 109 premenopausal, 232 postmenopausal. 260 receptor-positive; receptor-positive if estrogen progesterone positive, >10 fmol/mg cytosol biochemistry >10% stained immunohistochemistry. During Recurrence defined appearance new lesions after confirmed biopsy relevant diagnostic procedures. follow-up, seen examination every months 1st year, 6 2–5, once year until 10 Recording death causes; Death Registry. Clinicopathological DBCG. surgical procedures mastectomy partial axillary dissection breast-conserving lumpectomy dissection. received local radiotherapy. Low did receive whereas randomized consisting endocrine described (26.Danish those patients), larger 5 cm, protocol invading chest wall, premenopausal ductal carcinoma showing II III malignancy irrespective size nodal status. 199 (58%) therapy. 33 (341) protocol, 21 remaining 308 178 whom Frozen extracted previously brief, stored −80 °C mechanically pulverized, suspended ice-cold, low pH, detergent-containing extraction buffer (28.Camiolo S.M. Siuta M.R. Madeja Improved medium tissue.Prep. Biochem. 1982; 12: 297-305Google centrifuged 105,000 × g h 4 °C. supernatant Before performing assay, thawed 37 Total content, concentrations concentration determined each extract. assays performed knowledge any data. Bradford method (29.Bradford rapid sensitive quantitation microgram quantities utilizing principle protein-dye binding.Anal. 1976; 248-254Google bovine serum albumin standard extract pool made 271 individual internal standard. pools plasma levels, respectively, references ELISA plate. (30.Holten-Andersen M.N. Nielsen Pedersen Hoyer-Hansen Quantitation healthy blood donors advanced cancer.Br. 80: 495-503Google extracts. immunoassay plates coated sheep polyclonal anti-TIMP-1 antibody (31.Hembry R.M. Immunolocalization cells. Characterization antiserum.J. 73: 105-119Google bound murine monoclonal (MAC 15) binds (32.Cooksley Hipkiss J.B. Tickle S.P. Holmes I.E. Docherty Lawson A.D. Immunoassays detection collagenase, stromelysin, enzyme-inhibitor complexes.Matrix. 1990; 10: 285-291Google 33.Holten-Andersen Maimonis Jensen Piironen antibodies Ligand Assay. 25: 87-90Google followed alkaline phosphatase-conjugated rabbit anti-mouse (DAKO, Glostrup, Denmark). Readings color taken min 60 405 nm. On plate, serial dilutions duplicates recombinant included, allowing determination control plate controls. diluted 1:101 sample dilution KinetiCalc software (Bio-Tek Instruments, Winooski, VT) calculation concentrations. extracts, therefore matrix. limit corresponding signal deviations above mean blank (sample only). linearity series ranging 5% (1:20) 0.04% (1:2560). recovery pool, 1:200 pool. Recovery calculated slope line illustrating representing pure buffer. Intra-assay variation 32 same interassay run 14 days. Tumor SAS® package (version 8.2, SAS Institute, Cary, NC) manage statistical analysis. normalized scored quartiles. end points cause. Kaplan-Meier estimate probabilities, log rank test equality strata. Cox proportional model Model done graphically Schoenfeld Martingale residuals Harrell et al. (34.Harrell Jr., F.E. Lee K.L. Mark D.B. Multivariable models: issues models, evaluating assumptions adequacy, measuring reducing errors.Stat. 15: 361-387Google Rank statistics calculate coefficients hypothesis location. p values less considered significant. Using treated 67 pg/ml. gave good down 0.08% (1:1280) Moreover 102%. Looking variation, intra-assay (32 duplicates) 7.3% (14 14.4%. contained measurable sample. (±S.D.) 17 (±1.23) ng/mg protein. 25.66 (±18.97) protein, 21.65 2.06–190.32 protein). distribution illustrated Fig. 1. status (Kruskall-Wallis test, 0.02) 0.01) shown Table no parameters; found. weak but correlations observed (Fig. B). positive Scholar).Table IPatient characteristics comparisons studied parametersn%Median (range)p valueaKruskal-Wallis comparison protein; values.ng/mg proteinLymph statusPositive1775221 (3–107)0.6Negative1644822 (2–190)Hormone statusbData available 331 only. Tumors (i.e. immunohistochemistry).Positive2607922 (5–113)0.6Negative712121 (3–190)Tumor size0–20 mm1002921 (5–163)20–50 mm1935723 (3–190)0.08>50 mm481420 (2–110)Tumor gradecDuctal only; 276 patients.Grade I823022 (5–70)Grade II1224422 (2–107)0.9Grade III722620 (3–190)Menopausal statusPremenopausal1093219 (6–58)0.02Postmenopausal2326823 (2–190)Age≤40 years21619 (8–41)0.0141–50 years852521 (6–190)51–60 years962819 (2–164)61–70 years1053123 (4–113)≥71 years341030 (5–