|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Plasma D-Dimer Levels in Operable Breast Cancer Patients Correlate With Clinical Stage and Axillary Lymph Node StatusFrom the Divisions of Medical and Radiation Oncology, Duke University Comprehensive Cancer Center, Durham, NC. Address reprint requests to Charles Greenberg, MD, Department of Medicine and Pathology, Duke University Medical Center, Durham, NC 27710; email green032{at}mc.duke.edu
PURPOSE: To investigate the relationship between preoperative plasma D-dimer levels and extent of tumor involvement in operable breast cancer patients. PATIENTS AND METHODS: A total of 140 preoperative plasma specimens were obtained from women scheduled to undergo diagnostic breast biopsies. Ninety-five patients in the initial group went on to undergo axillary lymph node dissection. Of the 140 patients from whom plasma samples were obtained, 102 were subsequently diagnosed with invasive breast carcinoma, nine were subsequently diagnosed with ductal carcinoma-in-situ, and 20 were subsequently diagnosed with benign breast disease. Plasma D-dimer levels were quantitated using a commercially available immunoassay kit (DIMERTEST; American Diagnostica, Greenwich, CT). The relationships between plasma D-dimer and other prognostic variables (tumor size, estrogen receptor, progesterone receptor, nuclear grade, histologic grade, lymphovascular invasion, and clinical stage grouping) were then examined using univariate and multivariate linear and logistic regression analyses.
RESULTS: Median plasma D-dimer levels were significantly higher in patients with invasive carcinoma than those patients with either benign breast disease or carcinoma-in-situ (P = .0001). A significant relationship existed between the presence of elevated D-dimer (> 100 ng/mL) and involved axillary lymph nodes ( CONCLUSION: Plasma D-dimer levels were markers of lymphovascular invasion, clinical stage, and lymph node involvement in operable breast cancer. This correlation suggests that detectable fibrin degradation, as measured by plasma D-dimer, is a clinically important marker for lymphovascular invasion and early tumor metastasis in operable breast cancer.
FIBRIN DEPOSITION and remodeling in the tumor extracellular matrix is an important initial step in tumor metastasis. For a tumor to successfully metastasize from its primary location, it must undergo several obligate steps, including the invasion into either the lymphatic or vascular lumen, transportation through the circulation, and establishment of viability in target tissues.1 Cross-linked fibrin in the extracellular matrix serves as a stable framework for endothelial cell migration during angiogenesis and tumor cell migration during invasion. Extracellular remodeling of fibrin is essential for angiogenesis in tumors,2 and activation of intravascular fibrin formation and degradation has been shown to occur in the plasma of breast cancer patients.3,4 In addition, other indicators of fibrinolytic pathway activation, such as levels of plasminogen activator inhibitor and urokinase plasminogen activator, have been shown to have prognostic significance in patients with breast cancer.5-8 D-Dimer, a fibrin degradation product, is produced when factor XIIIa, a cross-linked fibrin, is degraded by plasmin generated from plasminogen by the action of serine protease tissue plasminogen activator. Investigation of human D-dimer levels has been made easier through the production of a specific monoclonal antibody that does not recognize degradation of fibrinogen or noncross-linked fibrin. Elevated D-dimer levels can be detected in the circulation, and elevated levels have been detected in patients with disseminated intravascular coagulation,9,10 vaso-occlusive crisis in sickle cell disease,11 thromboembolic events,12-15 and myocardial infarction.16 D-Dimer levels are elevated in the plasma of various solid tumor patients, including lung,17,18 prostate,19,20 cervical,21-23 and colorectal cancer patients.24,25 In patients with colorectal cancer, D-dimer levels have been shown to correlate with depth of tumor invasion at the time of surgical excision.25 Plasma D-dimer levels have also been shown to directly correlate with other tumor markers, including CA-125 and carcinoembryonic antigen.21,25 In patients with operable breast cancer, D-dimer levels have been shown to be elevated at time of diagnosis26,27 and to decrease during adjuvant (epirubicin/cyclophosphamide) chemotherapy.28 There still exists a gap in our knowledge regarding the relationship between quantitative D-dimer levels and extent of disease involvement in primary breast cancer. Our results suggest that a tight correlation exists between early tumor metastasis, lymphovascular invasion, and plasma D-dimer levels in operable breast cancer patients.
Study Population One hundred forty female patients seen for diagnostic surgical breast procedures were enrolled onto this study. Patients were eligible for participation if they had undergone a single needle biopsy or needle aspirate of their breast lesion but were excluded if they had undergone any more invasive procedure (excision or incisional biopsy, lumpectomy, or mastectomy.) The study sample consists of two separate consecutive patient populations. The first population (31 patients) had banked frozen plasma available and had been seen at Duke University Medical Center Surgical Oncology clinic from May 1994 through October 1995. Based on the promising results from the pilot population, we went on to prospectively enroll 109 patients from May 1997 to November 1998. Informed consent was obtained from all patients by the Duke University Medical Center Institutional Review Board.
Quantitative D-Dimer Level Determination Quantitative D-dimer levels were obtained using the DIMERTEST immunoassay (American Diagnostica, Greenwich, CT). All samples were run in duplicate according to manufacturer’s recommendations. This commercially available D-dimer monoclonal antibody recognizes an epitope that is a specific product of cross-linked fibrin that has been subsequently degraded by plasmin. Therefore, the plasma D-dimer assay does not recognize degradation of fibrinogen or noncross-linked fibrin. D-Dimer levels greater than 100 ng/mL were considered to be elevated, because this is the upper limit of normal using a 90% confidence limit from the mean of values obtained from healthy volunteers in our own laboratory.
Histopathologic Characterization Patients who underwent axillary lymph node dissection had nodal tissue paraffin-embedded, hematoxylin and eosin stained, and microscopically examined for the presence of malignant cells by attending pathology staff at Duke University Medical Center. Subtotal involvement of the lymph node with malignant cells was considered positive for analysis purposes.
Clinical Staging
Statistical Analysis
Study analysis included information regarding (1) D-dimer levels as a continuous and dichotomous variable (
Patient Characteristics The characteristics of the study population are listed in Table 1. Of the 140 patients enrolled, nine were excluded from analysis after enrollment. Four patients were excluded after enrollment because of other concurrent conditions known to increase D-dimer levels, such as pregnancy (n = 2), chronic lymphocytic leukemia (n = 1), and recent (13 days prior) mechanical valve repair (n = 1). Two patients were excluded after enrollment because of bilateral breast malignancies, and three patients were excluded after enrollment because of already having undergone major breast surgical procedures (two patients had undergone lumpectomy) or because of having completed neoadjuvant chemotherapy (one patient). Three of the nine patients had elevated D-dimer levels (> 100 ng/mL).
Twenty-nine patients had benign breast disease at diagnosis. Seventy-seven patients (59%) underwent excisional biopsy as definitive surgical therapy. Fifty-four patients (41%) had modified radical mastectomy after biopsy. Sixty-one patients (73%) had axillary lymph node dissection with a median of 13 lymph nodes removed at the time of dissection. Thirty-three patients (35%) had involved lymph nodes, with the mean number of positive lymph nodes being 3.28 (range, one to 16 positive lymph nodes). Five patients (3%) had 10 or more involved lymph nodes. Five patients (3%) had five to nine involved lymph nodes. Three patients with 10 or more lymph nodes had preoperative documentation of palpable axillary lymph nodes. None of the patients with five to nine involved lymph nodes were noted preoperatively to have palpable lymph nodes. Of patients with malignant disease, mean tumor size was 1.94 cm (range, 0.2 to 10.5 cm). Seven patients (5%) had tumors larger than 5 cm. No patients had clinical or pathologic evidence of chest wall or skin involvement. Twenty-five patients (19%) had elevated D-dimer levels (> 100 ng/mL). The mean D-dimer level in the study population was 67.04 ng/mL (SD = 8.87 ng/mL); the median D-dimer level was 37.25 ng/mL.
D-Dimer and Underlying Breast Disease
D-Dimer and Histopathologic Variables Of all histopathologic variables examined, D-dimer levels correlated strongest with the number of positive lymph nodes in the pilot group (Spearman correlation = 0.43; P = .02), the prospective group (Spearman correlation = 0.51; P = .0001), and the combined group (Spearman correlation = 0.51; P = .0001) (Fig 2). In the pilot group, D-dimer levels directly correlated with extent of lymph node involvement but not with tumor size, estrogen receptor status, progesterone receptor status, or lymphovascular invasion. In the prospective group, D-dimer levels correlated with extent of lymph node involvement, estrogen receptor status (Spearman correlation = -0.32; P = .01), and lymphovascular invasion (Spearman correlation = 0.38; P = .02), but not with tumor size or progesterone receptor status. In the combined (pilot and prospective) group, D-dimer levels correlated with extent of lymph node involvement, tumor size (Spearman correlation = 0.26; P = .008), lymphovascular invasion (Spearman correlation = 0.47; P = .0002), and estrogen receptor status (Spearman correlation = -0.27; P = .019), but not with progesterone receptor status (Spearman correlation = 0.04; P = .73) (Table 2)
D-Dimer and Clinical Stage The results of correlation between D-dimer and clinical stage are listed in Table 3. No patients were stage IIIB or stage IV. There was a statistically significant difference in D-dimer levels based on clinical stage grouping (analysis of variance test; P = .002).
D-Dimer and Lymph Node Involvement When examined as a dichotomous variable (elevated = > 100 ng/mL; normal =  100 ng/mL), elevated D-dimer levels correlated strongly with the presence of positive lymph nodes ( 2 test; P = .001) (Table 4). This correlation was seen regardless of group examined (pilot group: 2 test, P = .02 and prospective group: 2 test, P = .001).
Using the cutoff of 100 ng/mL, plasma D-dimer levels had a positive predictive value of 0.79 and a negative predictive value of 0.80 for predicting the extent of lymph node involvement. Likewise, elevated plasma D-dimer levels had a high specificity of 0.92 and moderate sensitivity of 0.58. A receiver operating characteristic (ROC) curve for D-dimer values in predicting lymph node involvement is shown in Fig 3.
Regression Modeling in Predicting D-Dimer Levels and Lymph Node Involvement Presence of lymphovascular invasion and presence of positive lymph nodes were significant in predicting D-dimer levels in both univariate and multivariate linear regression models (Table 5). Elevated D-dimer levels and tumor size were significant in predicting the presence of positive lymph nodes in both univariate and multivariate logistic regression models (Table 6). Using logistic regression modeling, the probability of having involved lymph nodes can be predicted for a given tumor size. Table 7 lists the probability, according to D-dimer levels, of having positive lymph nodes for either a 1-cm primary tumor or a 2-cm primary tumor. Based on this model, a patient with a 2-cm tumor and an elevated level of D-dimer (200 ng/mL) was twice as likely (probability = 0.42) to have involved axillary lymph nodes as those patients with low levels of D-dimer (5 ng/mL; probability = 0.20).
For over 20 years, investigators have documented the importance of the fibrinolytic pathway in tumor angiogenesis and metastasis.1-8,32-34 Indicators of fibrinolytic pathway activation, such as levels of plasminogen activator inhibitor and urokinase plasminogen activator, have been shown to have prognostic significance in patients with breast cancer.5-8 Our study represents the first attempt to look at a product of fibrin degradation (D-dimer) as a specific marker for extent of disease in human breast cancer. Both of our study populations (pilot and prospective) were representative of mainstream clinical practices in terms of mean tumor size and lymph node involvement.35 The patient population was skewed toward an increased percentage (78%) of malignant disease than that population usually seen for routine diagnostic surgical procedures. By carrying out the study in a referral center and allowing for previous outside needle biopsy diagnosis, we enrolled a larger percentage of patients with already known malignant disease than if the study had taken place in the community setting. Previously, increased D-dimer levels have been demonstrated in a variety of intravascular and inflammatory conditions. Only 2% of patients in this study were prospectively excluded for concurrent conditions that could have elevated their D-dimer levels. Unrecognized conditions, such as asymptomatic venous thrombosis, certainly existed in our study population, and yet, D-dimer remained highly specific in detecting positive lymph nodes. The process of metastasis involves multiple tumor-host interactions. To survive, metastatic cancer cells must leave the primary tumor, migrate into the lymphovascular system, and establish a new blood supply at their metastatic site. Fibrin remodeling is almost certainly involved in all steps of metastasis and has been proven to play a crucial role in new vessel formation.36-39 This study confirms previous studies showing upregulated fibrinolytic activity (presence of plasma D-dimer) in malignant disease and increased levels of fibrinolytic activity (increased D-dimer levels) in metastatic disease. In addition, linear regression modeling showed a tight relationship between the presence of lymphovascular invasion and elevated D-dimer levels. This relationship suggests a possible, yet unproven, biologic mechanism for the entrance of D-dimer fragments into the circulation. Further research needs to be carried out looking at the mechanism by which fibrin degradation products enter the systemic circulation. Finally, research needs to be carried out looking at other levels of coagulation factors and their relationship with D-dimer levels. Lymph node status in postmenopausal patients with smaller tumors determines treatment recommendations. Determining lymph node status via standard lymph node dissection creates many significant long-term side effects, including pain, numbness, and lymphedema. Many attempts have been made to predict lymph node status without undergoing a full lymph node dissection. Tumor size has been shown to predict lymph node status in numerous large studies,35,40-42 and our combined study group results confirmed this finding. However, independently, our pilot study and prospective study failed to demonstrate a correlation between tumor size and axillary lymph node involvement. We believe this is because of the small number of patients in each study group and the large variance possible in tumor size. Patients with tumors larger than 3 cm face a greater than 50% chance of having axillary lymph node involvement. Accurate tumor size is not available prospectively, and clinical examination has been shown by many authors to be nonspecific.42,43 Many clinicians believe that premenopausal women with larger tumors should be spared the damaging effects of axillary lymph node dissection. However, in patients with smaller tumors (< 1 cm), the risks and long-term morbidity of axillary lymph node dissection versus the benefits of knowing a patient’s lymph node status are sometimes equivalent. Through the use of logistic regression modeling, our study shows that by holding tumor size constant at 1 cm, a healthy patient with an elevated D-dimer (200 ng/mL) is almost three times as likely to have positive lymph nodes as a patient with a low D-Dimer (5 ng/mL). This suggests that patients with small tumors and low D-dimer levels could possibly be spared the morbidity of an axillary lymph node dissection. Given its sensitivity for predicting positive lymph node involvement, an important role of D-dimer could be in combination with other predictive factors in determining whether or not axillary lymph node dissection is necessary. This study clearly supports a role for plasma D-dimer levels in predicting clinically (lymph node status) and biologically (lymphovascular invasion) relevant factors in operable breast cancer. D-dimer can not be used alone in predicting lymph node status given its low negative predictive value. We are now investigating plasma D-dimer levels as a prognostic marker for operable breast cancer and the usefulness of combining plasma D-dimer levels with sentinel lymph node biopsy. Given the ease with which plasma D-dimer levels can be obtained, we are looking into the utility of adding quantitative D-dimer levels into models for predicting lymphovascular invasion and axillary lymph node involvement.
Supported by grant no. CA-68438, a National Institutes of Health Specialized Program of Research Excellence grant in Breast Cancer at Duke University, Department of Defense grant no. 179717044, and National Cancer Institute grant no. CA-71753. We thank Elizabeth Wildermann for her dedicated assistance in plasma collection.
1. Fidler IJ, Ellis LM: The implications of angiogenesis for the biology and therapy of cancer metastasis. Cell 79:315-328, 1994[Medline] 2. Dvorak HF, Brown LF, Detmar M, et al: Vascular permea-bility factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 146:1029-1039, 1995[Abstract] 3. Grohndahl-Hansen J, Agerlin N, Munkholm-Larsen P: Sensitive and specific enzyme-linked immunosorbent assay of urokinase-type plasminogen activator and its application in plasma from patients with breast cancer. J Lab Clin Med 111:42-51, 1988[Medline] 4. Layer GT, Burnhard KH, Gaffney PJ, et al: Tissue plasminogen activators in breast cancer. Thrombosis Res 45:601-605, 1987[Medline]
5.
Spyratos F, Martin P-M, Hacene K, et al: Multiparametric prognostic evaluation of biological factors in primary breast cancer. Natl Cancer Inst 84:1266-1272, 1992
6.
Grondahl-Hansen J, Christensen IH, Rosenquist C, et al: High levels of urokinase-type plasminogen activator and its inhibitor PAI-1 in cytosolic extracts of breast carcinomas are associated with prognosis. Cancer Res 53:2513-2521, 1993 7. Janicke F, Schmitt M, Pache L, et al: Urokinase (uPa) and its inhibitor PAI-1 are strong and independent prognostic factors in node-negative breast cancer. Breast Cancer Res Treat 24:195-208, 1993[Medline] 8. Duffy MJ, Reilly D, McDermott E, et al: Urokinase plasminogen activator as a prognostic marker in different subgroups of patients with breast cancer. Cancer 74:2276-2280, 1994[Medline] 9. Wilde JT, Kitchen S, Kinsey S, et al: Plasma D-dimer levels and their relationship to serum fibrinogen/fibrin degradation products in hypercoaguable states. Br J Haematol 71:65-70, 1989[Medline] 10. Bick RL, Baker WF: Diagnostic efficacy of the D-dimer assay in disseminated intravascular coagulation (DIC). Thromb Res 65:785-790, 1992[Medline]
11.
Devine DV, Kinney TR, Thomas PF, et al: Fragment D-dimer levels: An objective marker of vaso-occlusive crisis and other complications of sickle cell disease. Blood 68:317-319, 1986 12. Rowbotham BJ, Carroll P, Whitaker AN, et al: Measurement of crosslinked fibrin derivatives-use in the diagnosis of venous thrombosis. Thromb Haemost 57:59-61, 1987[Medline] 13. Rowbotham BJ, Egerton-Vernon J, Whitaker AN, et al: Plasma cross linked fibrin degradation products in pulmonary embolism. Thorax 45:684-687, 1990[Abstract] 14. Carter CJ, Doyle DL, Dawson N, et al: Investigations into the clinical utility of latex D-dimer in the diagnosis of deep venous thrombosis. Thromb Haemost 69:8-11, 1993[Medline] 15. Ginsberg JS, Kearon C, Douketis J, et al: The use of D-dimer testing and impedance plethysmographic examination in patients with clinical implications of deep vein thrombosis. Arch Intern Med 157:1077-1081, 1997[Abstract]
16.
Francis CW, Connaghan DG, Scott WL, et al: Increased plasma concentration of crosslinked fibrin polymers in acute myocardial infarction. Circulation 75:1170-1175, 1987 17. Seitz R, Rappe N, Kraus M, et al: Activation of coagulation and fibrinolysis in patients with lung cancer: Relation to tumour stage and prognosis. Fibrinolysis 4:249-254, 1993
18.
Gabazza EC, Taguchi O, Yamakami T, et al: Evaluating prethrombotic state in lung cancer using molecular markers. Chest 103:196-200, 1993
19.
Nakashima J, Tachibana M, Ueno M, et al: Tumor necrosis factor and coagulopathy in patients with prostate cancer. Cancer Res 55:4881-4885, 1995 20. Okajima K, Kohno I, Tsuruta J, et al: Direct evidence for systemic fibrinogenolysis in a patient with metastatic prostatic cancer. Thrombosis Res 66:717-727, 1992[Medline] 21. Rella C, Coviello M, De Frenza N, et al: Plasma D-dimer measurement as a marker of gynecologic tumors: Comparison with CA 125. Tumori 79:347-51, 1993[Medline] 22. Olt GJ, Greenberg C, Synan I, et al: Preoperative assessment of fragment D-dimer as a predictor of postoperative venous thrombosis. Am J Obstet Gynecol 162:772-775, 1990[Medline] 23. Gadducci A, Baicchi U, Marrai R, et al: Pretreatment plasma levels of fibrinopeptide-A (FPA), D-dimer (DD), and von Willebrand factor (vWF) in patients with operable cervical cancer: Influence of surgical-pathological stage, tumor size, histologic type, and lymph node status. Gynecol Oncol 49:354-358, 1993[Medline] 24. Edwards CM, Warren J, Armstrong L, et al: D-Dimer: A useful marker of disease stage in surgery for colorectal cancer. Br J Surg 80:1404-1405, 1993[Medline] 25. Oya M, Akiyama Y, Yanagida T, et al: Plasma D-dimer level in patients with colorectal cancer: Its role as a tumor marker. Surg Today 28:373-378, 1998[Medline] 26. McCollough P, Douglas J, Lowe GDO, et al: In vivo measurement of fibrin formation and fibrinolysis in operable breast cancer. Thromb Haemost 61:318-321, 1989[Medline] 27. Mitter CG, Zielinski CC: Plasma levels of D-dimer: A crosslinked fibrin-degradation product in female breast cancer. J Cancer Res Clin Oncol 117:259-262, 1991[Medline] 28. Tempelhoff GF, Dietrich M, Hommel G, et al: Blood coagulation during adjuvant epirubicin/cyclophosphamide chemotherapy in patients with primary operable breast cancer. J Clin Oncol 14:2560-2568, 1996[Abstract] 29. Sheehan D, Hrapchak B: Theory and Practice of Histotechnology (ed 2). Columbus, OH,Battelle Press, 1980 30. Layfield LJ, Saria EA, Conlon DH, et al: Estrogen and progesterone status determined by the Ventana ES 320 automated immunohistochemical stainer and the CAS 200 Image Analyzer in 236 early stage breast carcinomas. J Surg Oncology 61:177-184, 1996[Medline] 31. American Joint Committee on Cancer: Cancer Staging Handbook (ed 5), in Fleming ID, Cooper JS, Henson DE, et al (eds). Philadelphia PA, Lippincott-Raven, 1998 32. Dvorak HF: Tumors: Wounds that do not heal. N Engl J Med 315:1650-1659, 1986[Medline] 33. Blood CA, Zetter BR: Tumor interactions with the vasculature: Angiogenesis and tumor metastasis. Biochim Biophys Acta 1032:89-118, 1990[Medline] 34. Ausprunk DH, Folkman J: Migration and proliferation of endothelial cells in preformed and newly formed blood vessels during tumor angiogenesis. Microvasc Res 14:53-65, 1977[Medline] 35. Carter C, Allen C, Henson D: Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases. Cancer 63:181-188, 1989[Medline] 36. Nagy JA, Brown LF, Senger DR, et al: Pathogenesis of tumor stroma generation: A critical role for leaky blood vessels and fibrin deposition. Biochim Biophys Acta 948:305-326, 1988 37. Nagy JA, Morgan ES, Herzberg KT, et al: Pathogenesis of ascites tumor growth: Angiogenesis, vascular remodeling, and stroma formation in the peritoneal lining. Res 55:376-385, 1995 38. Brown LF, Van De Water L, Harvey VS, et al: Fibrinogen influx and accumulation of cross-linked fibrin in healing wounds and in tumor stroma. Am J Path 130:455-465, 1988[Abstract] 39. Shoji M, Hancock WM, Abe K, et al: Activation of coagulation and angiogenesis in cancer. Am J Path 152:399-411, 1998[Abstract] 40. Nemoto T, Vana J, Bedwani R: Management and survival of female breast cancer: Results of a national survey by the American College of Surgeons. Cancer 51:1333-1338, 1983[Medline] 41. Fischer B, Slack N, Bross I: Cancer of the breast: Size of neoplasm and prognosis. Cancer 24:1071-1076, 1969[Medline] 42. Schottenfeld D, Nash A, Robbins G: Ten-year results of the treatment of primary operable breast carcinoma. Cancer 38:1001-1012, 1976[Medline] 43. Butcher H: Radical mastectomy for mammary carcinoma. Ann Surg 170:883-884, 1969[Medline] Submitted March 12, 1999; accepted September 16, 1999. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
