- Email: [email protected]
Strong significant correlation between MMP-9 and systemic symptoms in patients with localized renal cell carcinoma
ADULT UROLOGY
STRONG SIGNIFICANT CORRELATION BETWEEN MMP-9 AND SYSTEMIC SYMPTOMS IN PATIENTS WITH LOCALIZED RENAL CELL CARCINOMA NOZOMU KAWATA, YUSUKE NAGANE, TAKUMI IGARASHI, HITOSHI HIRAKATA, TAKETO ICHINOSE, TAKAHIKO HACHIYA, YUKIE TAKIMOTO, AND SATORU TAKAHASHI
ABSTRACT Objectives. To identify a relationship between clinical symptoms and matrix metalloproteinase (MMP)-2 and MMP-9, tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2, and membrane type MMP-1. Methods. Tumor samples from 232 patients with renal cell carcinoma with no distant metastasis were immunohistochemically stained for MMP-2 and MMP-9, TIMP-1 and TIMP-2, and membrane type MMP-1. The immunoreactivity of these factors was analyzed by semiquantitative multivariate analysis for correlation with clinical symptoms. Results. Patard’s criteria were used to classify symptoms at initial tumor clinical presentation, with three groups defined: S1, S2, and S3. The cancer-specific 5-year survival rate was 88.7%, 74.7%, and 67.6% for S1 (145 patients), S2 (69 patients), and S3 (18 patients), respectively (P ⫽ 0.0015). Multiple logistic regression analysis of preference was used to determine whether differences in the contribution of the symptoms were statistically significant. A maximal tumor diameter of 40 mm or greater and positive venous invasion were associated with a 262% and 281% increase in the odds of local symptoms, respectively. MMP-9 positive cases were associated with a 2979% increase in the odds of systemic symptoms with significance. Conclusions. This study found a strong significant correlation between the histopathologic expression of MMP-9 and the systemic symptoms of renal cell carcinoma. We propose the histopathologic measurement of MMP-9 as a useful tool for assessing the prognosis of patients with renal cell carcinoma with systemic symptoms. UROLOGY 68: 523–527, 2006. © 2006 Elsevier Inc.
T
he increased use of imaging technologies has led to a significant increase in the discovery of incidental renal cell carcinoma (RCC). In a recent study, approximately 15% to 48% of RCC cases were diagnosed incidentally through imaging.1 Another study reported that 20% of patients diagnosed with RCC present with paraneoplastic symptoms, and another 10% to 40% of patients develop paraneoplastic symptoms during the course of the disease.2 Blay et al.3 reported that an elevated immunosuppressive acidic protein (IAP) serum level is associated with inflammatory reaction in a tumor-bearing host and may thus correspond to the increase in serum From the Department of Urology, Nihon University School of Medicine, Tokyo, Japan Reprint requests: Nozomu Kawata, M.D., Department of Urology, Surugadai Hospital, Nihon University School of Medicine, 1-8-13 Kanda Surugadai, Chiyoda-ku, Tokyo 101-0062, Japan. E-mail: [email protected] Submitted: November 8, 2005, accepted (with revisions): March 30, 2006 © 2006 ELSEVIER INC. ALL RIGHTS RESERVED
interleukin-6 or C-reactive protein levels in patients with RCC.3 Kawata et al.4 demonstrated the effectiveness of C-reactive protein and IAP measurement as a means to detect high nuclear grade RCC preoperatively.4 It is well known that symptomatic RCC results in a poor prognosis and often exhibits postoperative recurrence and distant metastasis. In one study, the 5-year cancer-specific survival rate was significantly greater for incidental tumors than for symptomatic tumors (85.3% versus 62.5%).1 Multivariate analysis by Yaycioglu et al.5 found that symptomatic RCC correlated with a 517% increase in the odds of postoperative recurrence. Multivariate analysis of symptom classifications by Patard et al.6 found that the S classification, TNM stage, nuclear grade, fat invasion, and vein invasion are independent prognostic factors, suggesting that the pooling of information on symptoms and general health status may enhance the symptom-based prognosis of RCC.6 Symptomatic findings at presentation have been 0090-4295/06/$32.00 doi:10.1016/j.urology.2006.03.063 523
assumed to indicate an unfavorable prognosis, although, to our knowledge, the implications for prognosis have never been clarified. Although clinical stage and nuclear grade are recognized prognostic factors for RCC, Patard et al.6 assessed patients according to their symptoms at diagnosis. The patients were categorized into three groups: S1, having asymptomatic tumors; S2, having tumors with local symptoms; and S3, having tumors with systemic symptoms. Patard et al.6 found these criteria to be strongly associated with cancer-specific survival. Many earlier studies have examined the presenting symptoms in the context of incidentally diagnosed RCC.1,2 These studies hypothesized that incidentally discovered tumors are more likely to be diagnosed at a lower stage and therefore offer a more favorable prognosis. To our knowledge, no study has examined the relationship between symptoms and histopathologic analysis, together with the reasons for the occurrence of symptomatic RCC. It is well known that both matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) play an important role in the progression of RCC. MMPs are repeatedly implicated in metastasis, a sequential step following intravasation and extravasation, and have been extensively studied in malignant tumors for their multifunctional role in metastasis and angiogenesis.7 MMPs mediate extracellular matrix and basement membrane degradation during the early stage of tumor genesis, contributing to the formation of a microenvironment that promotes tumor growth. MMPs are also active in the later stages of cancer development in that they promote metastasis.8 Recent evidence has suggested that MMPs are also important in the early stage of tumor development before metastasis occurs.9 The present study examined the expression of MMP-2, MMP-9, TIMP-1, TIMP-2, and membrane type MMP-1 (MT-MMP-1) in RCC and evaluated their correlation with clinical symptoms, including systemic symptoms. MATERIAL AND METHODS Formalin-fixed, paraffin-embedded tissue samples were obtained from 232 patients who underwent surgery for RCC with neither distant metastasis nor advancement beyond Gerota’s fascia from January 1988 to December 2003 at Nihon University Itabashi Hospital and Surugadai Hospital. The average postoperative follow-up period was 51 ⫾ 45 months. The stage was confirmed by pathologic examination after surgery, and the tumors were pathologically staged according to the American Joint Committee on Cancer classification.10 Nuclear grade was determined using the criteria proposed by Fuhrman et al.11 Venous invasion was defined as tumor infiltration locally through the intact vessel wall, including the endothelium, leading to free extension of cancer cells to the lumen.12 524
FIGURE 1. Staining intensity classified using fourgrade scale (see text for details).
The clinical stage was determined using ultrasonography, computed tomography, bone scan, and chest x-ray. The maximal tumor diameter was determined from the pathologic specimens. Symptoms were classified using the criteria of Patard et al.6 on the basis of the initial tumor clinical presentation.
HISTOPATHOLOGIC EXAMINATION To determine the expression of MMP-2, MMP-9, TIMP-1, TIMP-2, and MT-MMP-1, 5-m serial sections were cut from a representative block of formalin-fixed, paraffin-embedded RCC tissue. The sections were deparaffinized with xylene and dehydrated in a series of graded ethanol baths. Endogenous peroxidase activity was blocked by 0.3% hydrogen peroxidase in absolute methanol for 30 minutes at room temperature. Anti-MMP-2, MMP-9, TIMP-1, TIMP-2, and MT-MMP-1 goat polyclonal antibody (1:50 dilution, Santa Cruz Biotechnology, Santa Cruz, Calif) was used as a primary antibody and incubated at 4°C for 18 hours using a standard avidin-biotin immunoperoxidase incubation method. The tissue-bound peroxidase was visualized by incubating the sections with 0.02% 3-3= diaminobenzidine in phosphate buffer (pH 7.2) containing 0.03% hydrogen peroxide for 5 minutes. The sections were lightly counterstained with Mayer’s hematoxylin. The staining intensity of MMP-2, MMP-9, TIMP-1, TIMP-2, and MTMMP-1 was classified using a four-grade scale, with 0 indicating the absence of immunostaining or faint membranous staining of rare tumor cells; 1⫹ indicating membranous staining in most tumor cells; 2⫹ indicating diffuse membranous and/or cytoplasmic staining in groups of tumor cells; and 3⫹ indicating significant cytoplasmic staining in most tumor cells (Fig. 1). For evaluation of immunohistochemical staining, staining intensities of 2⫹ and 3⫹ were considered strong expressions of each protein.
STATISTICAL ANALYSIS The correlation between protein expression and pathologic variables was assessed using chi-square univariate analysis. For statistical analysis, the nuclear grades were categorized into two groups (1 to 2 and 3 to 4). Survival curves for all univariate analyses were assessed using the Kaplan-Meier method and compared using the log-rank test. To determine potential clinicopathologic predictors of symptomatic RCC, we compared the qualitative results using multiple logistic regression analysis. The data were analyzed using the Statview J, version 5.0, software package (SAS Institute, Cary, NC). P ⬍0.05 was considered statistically significant. UROLOGY 68 (3), 2006
UROLOGY 68 (3), 2006
19 (13) 15 (22) 7 (39) 41 (18) 0.022 16 (11) 14 (20) 6 (33) 36 (16) 0.02 KEY: MTD ⫽ maximal tumor diameter; VI ⫽ venous invasion; MMP ⫽ matrix metalloproteinase; TIMP ⫽ tissue inhibitor of metalloproteinase; MT-MMP-1 ⫽ membrane type MMP-1. Data in parentheses are percentages.
114 (79) 59 (86) 17 (94) 190 (82) 0.12
MT-MMP-1 (Strong) TIMP-2 (Strong) TIMP-1 (Strong) MMP-9 (Strong)
23 (16) 17 (25) 15 (83) 55 (24) 0.0001 87 (60) 48 (70) 14 (78) 149 (64) 0.17 31 (21) 37 (54) 10 (56) 78 (34) 0.0001 18 (12) 21 (30) 5 (28) 44 (19) 0.0005 25 (17) 22 (32) 10 (56) 57 (25) 0.0008 72 (50) 56 (81) 13 (72) 141 (61) 0.0001 145 69 18 232
Table I presents the characteristics of the 232 patients treated by radical nephrectomy for RCC. Of the 232 patients evaluated in our study, 145 (62%) were treated for S1, 69 (30%) for S2, and 18 (8%) for S3. The median postoperative follow-up period was 44 months (range 1 to 150). The cancer-specific 5-year survival rate was 88.7% , 74.7%, and 67.6% for the S1, S2, and S3 patients, respectively (P ⫽ 0.0015). Table II presents the correlation between S classification and nine clinicopathologic factors. Of the 57 patients with tumor Stage T3 or worse, 10 were classified as S3 (P ⫽ 0.0008). Our study detected significant correlations for MMP-9, TIMP-2, and MT-MMP-1. Of 55 MMP-9 cases, 15 were S3 (P ⫽ 0.0001), 6 of 36 TIMP-2 cases were S3 (P ⫽ 0.02), and 7 of 41 MT-MMP-1 cases were S3 (P ⫽ 0.022). Multiple logistic regression analysis of preference was used to determine whether differences in the contribution of symptoms were statistically significant. Table III presents the correlations between the S2 criteria and nine clinicopathologic factors for the 214 non-S3 patients and the S3 criteria and nine clinicopathological factors for all 232 cases. These findings revealed that a maximal tumor diameter of 40 mm or more and positive venous invasion had a significant correlation with the S2 criteria. With respect to the contribution of
S1 S2 S3 Total P value
RESULTS
MMP-2 (Strong)
Data presented as numbers of patients, with percentages in parentheses, unless otherwise noted. * Data in parentheses are ranges. † S1, tumors discovered on routine examination or through ultrasonography, computed tomography, or other radiologic imaging performed as a result of patient complaints not related to a renal tumor; S2, tumors discovered through isolated local symptoms related to RCC, including hematuria or palpable mass; S3, tumors associated with altered health condition, including appetite loss, fatigue, weight loss, fever, or night sweats.
VI (Positive)
86 (37) 102 (44) 35 (15) 9 (4)
Nuclear Grade (>3)
136 (59) 39 (17) 57 (24)
Tumor Stage (>T3)
145 (62) 69 (30) 18 (8)
MTD (>40 mm)
60 (33–83) 59 (25–82)
Patients (n)
174 58
TABLE II. Correlation between symptoms and clinicopathologic factors using chi-square univariate test
Sex Male Female Average age* Men Women S classification† S1 S2 S3 Tumor stage T1 T2 T3 Nuclear grade 1 2 3 4
S Classification
TABLE I. Patient characteristics
525
TABLE III. Correlation between symptoms and clinicopathologic factors using multiple logistic regression analysis Factor
Odds Ratio* (S2 [69/214]/S3 [18/232])
P Value*
95% CI*
3.62/1.44 2.27/1.91 1.42/3.57 3.81/1.11 1.01/1.34 1.02/30.79 1.13/2.39 1.45/1.00 1.35/1.97
0.007/0.56 0.067/0.33 0.5/0.068 0.0004/0.82 0.96/0.69 0.95/0.001 0.78/0.45 0.47/0.99 0.56/0.34
1.71–7.92/0.40–5.13 0.94–5.49/0.18–2.24 0.31–1.76/0.90–14.12 1.82–7.98/0.22–3.28 0.48–2.13/0.16–3.32 0.36–2.62/6.52–145.34 0.46–2.76/0.24–23.03 0.52–4.06/0.27–3.69 0.47–3.88/0.12–2.09
†
MTD Nuclear grade Tumor stage VI† MMP-2 MMP-9† TIMP-1 TIMP-2 MT-MMP-1
KEY: CI ⫽ confidence interval; other abbreviations as in Table II. * Numerator indicates odds ratio of S2 criteria for 214 non-S3 patients; denominator indicates odds ratio of S3 criteria for all 232 patients. † MTD ⱖ40 mm and positive VI were associated with 262% and 281% increase in odds of local symptoms, respectively; positive MMP-9 was associated with 2979% increase in odds of systemic symptoms with statistical significance.
the S3 criteria, multivariate analysis revealed that only MMP-9 had a significant correlation. COMMENT In human tumors, the expression of MMPs has been reported as low in most benign elements and substantially greater in most malignancies, including carcinoma of the colon, breast, lung, pancreas, and prostate.13,14 Elevated MMP levels have also been found in the plasma and urine of patients with cancer of the colon, breast, prostate, and bladder.15 With respect to urinary tract tumors, serum MMP levels in patients with urothelial tumors had a significant correlation with cancer progression and poor prognosis.16 From these results, we can begin to develop a clearer understanding of the characteristics of symptomatic RCC by analyzing the relationship between MMPs and clinical symptoms. Kugler et al.17 found that MMP-2 and MMP-9 are elevated in RCC tumor tissue and their expression level correlated with tumor aggressiveness. In advanced tumors, the expression of MMPs was 4.8 times greater than in normal kidney tissue. MMP-9 was reported to be significantly elevated in tumor tissue from patients with RCC, and MMP-2 was unaffected.18 Fukata et al.19 reported that RCC primary specimens with distant metastasis showed greater levels of MMP-2 expression than did primary tumors without distant metastasis. They concluded that angiogenesis and MMPs were independent factors for the metastasis of RCC, especially to the lungs and lymph nodes. Although MMP-2 and MMP-9 have similar characteristics in terms of mediating extracellular matrix and basement membrane degradation, they differ in terms of their transcriptional regulation and extracellular activation and inhibition. For ex526
ample, macrophages, neutrophils, and keratinocytes express MMP-9, but not MMP-2.20 MMP-9 produced by tumor cells is involved in the breakdown of the connective tissue barriers when the cells migrate through the stroma and basement membrane of the blood vessels.21 Kim et al.22 demonstrated that MMPs, and most likely MMP-9, are required for intravasation. They concluded that breaching the vascular wall is a ratelimiting step for intravasation and, consequently, for metastasis and that interaction between urokinasetype plasminogen activator/urokinase-type plasminogen activator receptor and MMP-9 is required to complete this step. With respect to the relationship between MTMMP-1 and MMP-2, MT-MMP-1 induces the activation of MMP-2. Activation occurs through the MT-MMP-1-mediated proteolytic cleavage of a pro-peptide sequence from pro-MMP-2, which is inhibited by TIMP-2, but not TIMP-1.23 It is unclear why the patients with RCC and clinical symptoms had a less favorable prognosis than the patients with systemic symptoms. Systemic symptoms of RCC are believed to have a particularly strong correlation with an unfavorable prognosis. As shown by the results of our multiple logistic regression analysis (Table III), maximal tumor diameter, nuclear grade, tumor stage, and venous invasion did not have significant correlations with systemic symptoms. With respect to the MMP families, only MMP-9 had a strong correlation with systemic symptoms. Many of the symptoms presented by patients with RCC are paraneoplastic symptoms resulting from the elaboration by the tumor of various proteins. In particular, cachexia and malaise may result from the secretion of tumor necrosis factoralpha, interleukin-1, or interleukin-6 by the tumor or infiltrating immune cells.24 UROLOGY 68 (3), 2006
As discussed above, MMP-9 plays an important role in the prognosis of RCC, including branching of the vascular wall. These results suggest that the histopathologic expression of MMP-9 may cause systemic symptoms of RCC. We propose the histopathologic measurement of MMP-9 as a useful tool for assessing the prognosis of patients with RCC and systemic symptoms. CONCLUSIONS The results of our study have demonstrated strong significant relationships between MMP-9 and the systemic symptoms of RCC. Although this study involved only a small number of patients, we believe that measuring the level of MMP-9 expression can be readily extended to clinical situations. REFERENCES 1. Tsui K, Shvarts O, Smith RB, et al: Renal cell carcinoma: prognostic significance of incidentally detected tumors. J Urol 163: 426 – 433, 2000. 2. McDougal WS, and Garnick MB: Clinical signs and symptoms of renal cell carcinoma, in Vogelzang NJ, Scardino PT, Shipley WU, et al (Eds): Comprehensive Textbook of Genitourinary Oncology. Baltimore, Williams & Wilkins, 1995, Chapter IV, pp 111–115, 154 –159. 3. Blay JY, Rossi JF, Wijdenes J, et al: Role of IL-6 in the paraneoplastic inflammatory syndrome associated with renal cell carcinoma. Int J Cancer 72: 424 – 430, 1997. 4. Kawata N, Yamaguchi K, Hirakata H, et al: Immunosuppressive acidic protein detects high nuclear grade localized renal cell carcinoma. Urology 66: 736 –740, 2005. 5. Yaycioglu O, Roberts WW, Chan T, et al: Prognostic assessment of nonmetastatic renal cell carcinoma: a clinically based model. Urology 58: 141–145, 2001. 6. Patard JJ, Leray E, Cindolo L, et al: Multi-institutional validation of a symptom based classification for renal cell carcinoma. J Urol 172: 858 – 862, 2004. 7. Chambers AF, and Matrisian LM: Changing views of the role of matrix metalloproteinases in metastasis. J Natl Cancer Inst 89: 1260 –1270, 1997. 8. Egeblad M, and Werb Z: New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2: 161–174, 2002. 9. Overall CM, and Lopez-Otin C: Strategies for MMP inhibition in cancer: innovations for the post-trial era. Nat Rev Cancer 2: 657– 672, 2002.
UROLOGY 68 (3), 2006
10. Sobin LH, and Wittekind C: TNM Classification of Malignant Tumours, 5th ed. New York, John Wiley, 1997. 11. Fuhrman SA, Lasky LC, and Limas C: Prognostic significant of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 6: 655– 663, 1982. 12. Lang H, Lindner V, Saussine C, et al: Microscopic venous invasion: a prognostic factor in renal cell carcinoma. Eur Urol 38: 600 – 605, 2000. 13. Kawamata H, Kameyama S, Kawai K, et al: Marked acceleration of the metastatic phenotype of a rat bladder carcinoma cell line by the expression of the human gelatinase A. Int J Cancer 63: 568 –575, 1995. 14. Tsunezuka Y, Kinoh H, Takino T, et al: Expression of membrane-type matrix metalloproteinases 1 (MT1-MP) in tumor cells enhances pulmonary metastasis in an experimental metastasis assay. Cancer Res 56: 5678 –5683, 1996. 15. Zucker S, Hymowitz M, Conner C, et al: Measurement of matrix metalloproteinases and tissue inhibitors of metalloproteinases in blood and tissue: clinical and experimental applications. Ann NY Acad Sci 878: 212–227, 1999. 16. Gohji K, Fujimoto N, Komiyama T, et al: Evaluation of serum levels of matrix metalloproteinases-2 and 3 as new predictors of recurrence in patients with urothelial carcinoma. Cancer 78: 2379 –2387, 1996. 17. Kugler A, Hemmerlein B, Thelen P, et al: Expression of metalloproteinase 2 and 9 and their inhibitors in renal cell carcinoma. J Urol 160: 1914 –1918, 1998. 18. Lein M, Jung K, Laube C, et al: Matrix-metalloproteinases and their inhibitors in plasma and tumor tissue of patients with renal cell carcinoma. Int J Cancer 85: 801– 804, 2000. 19. Fukata S, Inoue K, Kamada M, et al: Levels of angiogenesis and expression of angiogenesis-related genes are prognostic for organ-specific metastasis of renal cell carcinoma. Cancer 103: 931–942, 2005. 20. Wilhelm SM, Collier IE, Marmer BL, et al: SV40-transformed human lung fibroblasts secrete a 92-kDa type IV collagenase which is identical to that secreted by normal human macrophages. J Biol Chem 264: 17213–17221, 1989. 21. Okada Y, Gonoji Y, Naka K, et al: Matrix metalloproteinase 9 (92-kDa gelatinase/type IV collagenase) from HT 1080 human fibrosarcoma cells: purification and activation of the precursor and enzymic properties. J Biol Chem 267: 21712–21719, 1992. 22. Kim J, Yu W, Kovalski K, et al: Requirement for specific proteases in cancer cell intravasation as revealed by a novel semiquantitative PCR-based assay. Cell 94: 353–362, 1998. 23. Gomez DE, Alonso DF, Yoshiji H, et al: Tissue inhibitors of metalloproteinases: structure, regulation and biological functions. Eur J Cell Biol 74: 111–122, 1997. 24. Laski ME, and Vugrin D: Paraneoplastic syndromes in hypernephroma. Semin Nephrol 7: 123–130, 1987.
527
STRONG SIGNIFICANT CORRELATION BETWEEN MMP-9 AND SYSTEMIC SYMPTOMS IN PATIENTS WITH LOCALIZED RENAL CELL CARCINOMA NOZOMU KAWATA, YUSUKE NAGANE, TAKUMI IGARASHI, HITOSHI HIRAKATA, TAKETO ICHINOSE, TAKAHIKO HACHIYA, YUKIE TAKIMOTO, AND SATORU TAKAHASHI
ABSTRACT Objectives. To identify a relationship between clinical symptoms and matrix metalloproteinase (MMP)-2 and MMP-9, tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2, and membrane type MMP-1. Methods. Tumor samples from 232 patients with renal cell carcinoma with no distant metastasis were immunohistochemically stained for MMP-2 and MMP-9, TIMP-1 and TIMP-2, and membrane type MMP-1. The immunoreactivity of these factors was analyzed by semiquantitative multivariate analysis for correlation with clinical symptoms. Results. Patard’s criteria were used to classify symptoms at initial tumor clinical presentation, with three groups defined: S1, S2, and S3. The cancer-specific 5-year survival rate was 88.7%, 74.7%, and 67.6% for S1 (145 patients), S2 (69 patients), and S3 (18 patients), respectively (P ⫽ 0.0015). Multiple logistic regression analysis of preference was used to determine whether differences in the contribution of the symptoms were statistically significant. A maximal tumor diameter of 40 mm or greater and positive venous invasion were associated with a 262% and 281% increase in the odds of local symptoms, respectively. MMP-9 positive cases were associated with a 2979% increase in the odds of systemic symptoms with significance. Conclusions. This study found a strong significant correlation between the histopathologic expression of MMP-9 and the systemic symptoms of renal cell carcinoma. We propose the histopathologic measurement of MMP-9 as a useful tool for assessing the prognosis of patients with renal cell carcinoma with systemic symptoms. UROLOGY 68: 523–527, 2006. © 2006 Elsevier Inc.
T
he increased use of imaging technologies has led to a significant increase in the discovery of incidental renal cell carcinoma (RCC). In a recent study, approximately 15% to 48% of RCC cases were diagnosed incidentally through imaging.1 Another study reported that 20% of patients diagnosed with RCC present with paraneoplastic symptoms, and another 10% to 40% of patients develop paraneoplastic symptoms during the course of the disease.2 Blay et al.3 reported that an elevated immunosuppressive acidic protein (IAP) serum level is associated with inflammatory reaction in a tumor-bearing host and may thus correspond to the increase in serum From the Department of Urology, Nihon University School of Medicine, Tokyo, Japan Reprint requests: Nozomu Kawata, M.D., Department of Urology, Surugadai Hospital, Nihon University School of Medicine, 1-8-13 Kanda Surugadai, Chiyoda-ku, Tokyo 101-0062, Japan. E-mail: [email protected] Submitted: November 8, 2005, accepted (with revisions): March 30, 2006 © 2006 ELSEVIER INC. ALL RIGHTS RESERVED
interleukin-6 or C-reactive protein levels in patients with RCC.3 Kawata et al.4 demonstrated the effectiveness of C-reactive protein and IAP measurement as a means to detect high nuclear grade RCC preoperatively.4 It is well known that symptomatic RCC results in a poor prognosis and often exhibits postoperative recurrence and distant metastasis. In one study, the 5-year cancer-specific survival rate was significantly greater for incidental tumors than for symptomatic tumors (85.3% versus 62.5%).1 Multivariate analysis by Yaycioglu et al.5 found that symptomatic RCC correlated with a 517% increase in the odds of postoperative recurrence. Multivariate analysis of symptom classifications by Patard et al.6 found that the S classification, TNM stage, nuclear grade, fat invasion, and vein invasion are independent prognostic factors, suggesting that the pooling of information on symptoms and general health status may enhance the symptom-based prognosis of RCC.6 Symptomatic findings at presentation have been 0090-4295/06/$32.00 doi:10.1016/j.urology.2006.03.063 523
assumed to indicate an unfavorable prognosis, although, to our knowledge, the implications for prognosis have never been clarified. Although clinical stage and nuclear grade are recognized prognostic factors for RCC, Patard et al.6 assessed patients according to their symptoms at diagnosis. The patients were categorized into three groups: S1, having asymptomatic tumors; S2, having tumors with local symptoms; and S3, having tumors with systemic symptoms. Patard et al.6 found these criteria to be strongly associated with cancer-specific survival. Many earlier studies have examined the presenting symptoms in the context of incidentally diagnosed RCC.1,2 These studies hypothesized that incidentally discovered tumors are more likely to be diagnosed at a lower stage and therefore offer a more favorable prognosis. To our knowledge, no study has examined the relationship between symptoms and histopathologic analysis, together with the reasons for the occurrence of symptomatic RCC. It is well known that both matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) play an important role in the progression of RCC. MMPs are repeatedly implicated in metastasis, a sequential step following intravasation and extravasation, and have been extensively studied in malignant tumors for their multifunctional role in metastasis and angiogenesis.7 MMPs mediate extracellular matrix and basement membrane degradation during the early stage of tumor genesis, contributing to the formation of a microenvironment that promotes tumor growth. MMPs are also active in the later stages of cancer development in that they promote metastasis.8 Recent evidence has suggested that MMPs are also important in the early stage of tumor development before metastasis occurs.9 The present study examined the expression of MMP-2, MMP-9, TIMP-1, TIMP-2, and membrane type MMP-1 (MT-MMP-1) in RCC and evaluated their correlation with clinical symptoms, including systemic symptoms. MATERIAL AND METHODS Formalin-fixed, paraffin-embedded tissue samples were obtained from 232 patients who underwent surgery for RCC with neither distant metastasis nor advancement beyond Gerota’s fascia from January 1988 to December 2003 at Nihon University Itabashi Hospital and Surugadai Hospital. The average postoperative follow-up period was 51 ⫾ 45 months. The stage was confirmed by pathologic examination after surgery, and the tumors were pathologically staged according to the American Joint Committee on Cancer classification.10 Nuclear grade was determined using the criteria proposed by Fuhrman et al.11 Venous invasion was defined as tumor infiltration locally through the intact vessel wall, including the endothelium, leading to free extension of cancer cells to the lumen.12 524
FIGURE 1. Staining intensity classified using fourgrade scale (see text for details).
The clinical stage was determined using ultrasonography, computed tomography, bone scan, and chest x-ray. The maximal tumor diameter was determined from the pathologic specimens. Symptoms were classified using the criteria of Patard et al.6 on the basis of the initial tumor clinical presentation.
HISTOPATHOLOGIC EXAMINATION To determine the expression of MMP-2, MMP-9, TIMP-1, TIMP-2, and MT-MMP-1, 5-m serial sections were cut from a representative block of formalin-fixed, paraffin-embedded RCC tissue. The sections were deparaffinized with xylene and dehydrated in a series of graded ethanol baths. Endogenous peroxidase activity was blocked by 0.3% hydrogen peroxidase in absolute methanol for 30 minutes at room temperature. Anti-MMP-2, MMP-9, TIMP-1, TIMP-2, and MT-MMP-1 goat polyclonal antibody (1:50 dilution, Santa Cruz Biotechnology, Santa Cruz, Calif) was used as a primary antibody and incubated at 4°C for 18 hours using a standard avidin-biotin immunoperoxidase incubation method. The tissue-bound peroxidase was visualized by incubating the sections with 0.02% 3-3= diaminobenzidine in phosphate buffer (pH 7.2) containing 0.03% hydrogen peroxide for 5 minutes. The sections were lightly counterstained with Mayer’s hematoxylin. The staining intensity of MMP-2, MMP-9, TIMP-1, TIMP-2, and MTMMP-1 was classified using a four-grade scale, with 0 indicating the absence of immunostaining or faint membranous staining of rare tumor cells; 1⫹ indicating membranous staining in most tumor cells; 2⫹ indicating diffuse membranous and/or cytoplasmic staining in groups of tumor cells; and 3⫹ indicating significant cytoplasmic staining in most tumor cells (Fig. 1). For evaluation of immunohistochemical staining, staining intensities of 2⫹ and 3⫹ were considered strong expressions of each protein.
STATISTICAL ANALYSIS The correlation between protein expression and pathologic variables was assessed using chi-square univariate analysis. For statistical analysis, the nuclear grades were categorized into two groups (1 to 2 and 3 to 4). Survival curves for all univariate analyses were assessed using the Kaplan-Meier method and compared using the log-rank test. To determine potential clinicopathologic predictors of symptomatic RCC, we compared the qualitative results using multiple logistic regression analysis. The data were analyzed using the Statview J, version 5.0, software package (SAS Institute, Cary, NC). P ⬍0.05 was considered statistically significant. UROLOGY 68 (3), 2006
UROLOGY 68 (3), 2006
19 (13) 15 (22) 7 (39) 41 (18) 0.022 16 (11) 14 (20) 6 (33) 36 (16) 0.02 KEY: MTD ⫽ maximal tumor diameter; VI ⫽ venous invasion; MMP ⫽ matrix metalloproteinase; TIMP ⫽ tissue inhibitor of metalloproteinase; MT-MMP-1 ⫽ membrane type MMP-1. Data in parentheses are percentages.
114 (79) 59 (86) 17 (94) 190 (82) 0.12
MT-MMP-1 (Strong) TIMP-2 (Strong) TIMP-1 (Strong) MMP-9 (Strong)
23 (16) 17 (25) 15 (83) 55 (24) 0.0001 87 (60) 48 (70) 14 (78) 149 (64) 0.17 31 (21) 37 (54) 10 (56) 78 (34) 0.0001 18 (12) 21 (30) 5 (28) 44 (19) 0.0005 25 (17) 22 (32) 10 (56) 57 (25) 0.0008 72 (50) 56 (81) 13 (72) 141 (61) 0.0001 145 69 18 232
Table I presents the characteristics of the 232 patients treated by radical nephrectomy for RCC. Of the 232 patients evaluated in our study, 145 (62%) were treated for S1, 69 (30%) for S2, and 18 (8%) for S3. The median postoperative follow-up period was 44 months (range 1 to 150). The cancer-specific 5-year survival rate was 88.7% , 74.7%, and 67.6% for the S1, S2, and S3 patients, respectively (P ⫽ 0.0015). Table II presents the correlation between S classification and nine clinicopathologic factors. Of the 57 patients with tumor Stage T3 or worse, 10 were classified as S3 (P ⫽ 0.0008). Our study detected significant correlations for MMP-9, TIMP-2, and MT-MMP-1. Of 55 MMP-9 cases, 15 were S3 (P ⫽ 0.0001), 6 of 36 TIMP-2 cases were S3 (P ⫽ 0.02), and 7 of 41 MT-MMP-1 cases were S3 (P ⫽ 0.022). Multiple logistic regression analysis of preference was used to determine whether differences in the contribution of symptoms were statistically significant. Table III presents the correlations between the S2 criteria and nine clinicopathologic factors for the 214 non-S3 patients and the S3 criteria and nine clinicopathological factors for all 232 cases. These findings revealed that a maximal tumor diameter of 40 mm or more and positive venous invasion had a significant correlation with the S2 criteria. With respect to the contribution of
S1 S2 S3 Total P value
RESULTS
MMP-2 (Strong)
Data presented as numbers of patients, with percentages in parentheses, unless otherwise noted. * Data in parentheses are ranges. † S1, tumors discovered on routine examination or through ultrasonography, computed tomography, or other radiologic imaging performed as a result of patient complaints not related to a renal tumor; S2, tumors discovered through isolated local symptoms related to RCC, including hematuria or palpable mass; S3, tumors associated with altered health condition, including appetite loss, fatigue, weight loss, fever, or night sweats.
VI (Positive)
86 (37) 102 (44) 35 (15) 9 (4)
Nuclear Grade (>3)
136 (59) 39 (17) 57 (24)
Tumor Stage (>T3)
145 (62) 69 (30) 18 (8)
MTD (>40 mm)
60 (33–83) 59 (25–82)
Patients (n)
174 58
TABLE II. Correlation between symptoms and clinicopathologic factors using chi-square univariate test
Sex Male Female Average age* Men Women S classification† S1 S2 S3 Tumor stage T1 T2 T3 Nuclear grade 1 2 3 4
S Classification
TABLE I. Patient characteristics
525
TABLE III. Correlation between symptoms and clinicopathologic factors using multiple logistic regression analysis Factor
Odds Ratio* (S2 [69/214]/S3 [18/232])
P Value*
95% CI*
3.62/1.44 2.27/1.91 1.42/3.57 3.81/1.11 1.01/1.34 1.02/30.79 1.13/2.39 1.45/1.00 1.35/1.97
0.007/0.56 0.067/0.33 0.5/0.068 0.0004/0.82 0.96/0.69 0.95/0.001 0.78/0.45 0.47/0.99 0.56/0.34
1.71–7.92/0.40–5.13 0.94–5.49/0.18–2.24 0.31–1.76/0.90–14.12 1.82–7.98/0.22–3.28 0.48–2.13/0.16–3.32 0.36–2.62/6.52–145.34 0.46–2.76/0.24–23.03 0.52–4.06/0.27–3.69 0.47–3.88/0.12–2.09
†
MTD Nuclear grade Tumor stage VI† MMP-2 MMP-9† TIMP-1 TIMP-2 MT-MMP-1
KEY: CI ⫽ confidence interval; other abbreviations as in Table II. * Numerator indicates odds ratio of S2 criteria for 214 non-S3 patients; denominator indicates odds ratio of S3 criteria for all 232 patients. † MTD ⱖ40 mm and positive VI were associated with 262% and 281% increase in odds of local symptoms, respectively; positive MMP-9 was associated with 2979% increase in odds of systemic symptoms with statistical significance.
the S3 criteria, multivariate analysis revealed that only MMP-9 had a significant correlation. COMMENT In human tumors, the expression of MMPs has been reported as low in most benign elements and substantially greater in most malignancies, including carcinoma of the colon, breast, lung, pancreas, and prostate.13,14 Elevated MMP levels have also been found in the plasma and urine of patients with cancer of the colon, breast, prostate, and bladder.15 With respect to urinary tract tumors, serum MMP levels in patients with urothelial tumors had a significant correlation with cancer progression and poor prognosis.16 From these results, we can begin to develop a clearer understanding of the characteristics of symptomatic RCC by analyzing the relationship between MMPs and clinical symptoms. Kugler et al.17 found that MMP-2 and MMP-9 are elevated in RCC tumor tissue and their expression level correlated with tumor aggressiveness. In advanced tumors, the expression of MMPs was 4.8 times greater than in normal kidney tissue. MMP-9 was reported to be significantly elevated in tumor tissue from patients with RCC, and MMP-2 was unaffected.18 Fukata et al.19 reported that RCC primary specimens with distant metastasis showed greater levels of MMP-2 expression than did primary tumors without distant metastasis. They concluded that angiogenesis and MMPs were independent factors for the metastasis of RCC, especially to the lungs and lymph nodes. Although MMP-2 and MMP-9 have similar characteristics in terms of mediating extracellular matrix and basement membrane degradation, they differ in terms of their transcriptional regulation and extracellular activation and inhibition. For ex526
ample, macrophages, neutrophils, and keratinocytes express MMP-9, but not MMP-2.20 MMP-9 produced by tumor cells is involved in the breakdown of the connective tissue barriers when the cells migrate through the stroma and basement membrane of the blood vessels.21 Kim et al.22 demonstrated that MMPs, and most likely MMP-9, are required for intravasation. They concluded that breaching the vascular wall is a ratelimiting step for intravasation and, consequently, for metastasis and that interaction between urokinasetype plasminogen activator/urokinase-type plasminogen activator receptor and MMP-9 is required to complete this step. With respect to the relationship between MTMMP-1 and MMP-2, MT-MMP-1 induces the activation of MMP-2. Activation occurs through the MT-MMP-1-mediated proteolytic cleavage of a pro-peptide sequence from pro-MMP-2, which is inhibited by TIMP-2, but not TIMP-1.23 It is unclear why the patients with RCC and clinical symptoms had a less favorable prognosis than the patients with systemic symptoms. Systemic symptoms of RCC are believed to have a particularly strong correlation with an unfavorable prognosis. As shown by the results of our multiple logistic regression analysis (Table III), maximal tumor diameter, nuclear grade, tumor stage, and venous invasion did not have significant correlations with systemic symptoms. With respect to the MMP families, only MMP-9 had a strong correlation with systemic symptoms. Many of the symptoms presented by patients with RCC are paraneoplastic symptoms resulting from the elaboration by the tumor of various proteins. In particular, cachexia and malaise may result from the secretion of tumor necrosis factoralpha, interleukin-1, or interleukin-6 by the tumor or infiltrating immune cells.24 UROLOGY 68 (3), 2006
As discussed above, MMP-9 plays an important role in the prognosis of RCC, including branching of the vascular wall. These results suggest that the histopathologic expression of MMP-9 may cause systemic symptoms of RCC. We propose the histopathologic measurement of MMP-9 as a useful tool for assessing the prognosis of patients with RCC and systemic symptoms. CONCLUSIONS The results of our study have demonstrated strong significant relationships between MMP-9 and the systemic symptoms of RCC. Although this study involved only a small number of patients, we believe that measuring the level of MMP-9 expression can be readily extended to clinical situations. REFERENCES 1. Tsui K, Shvarts O, Smith RB, et al: Renal cell carcinoma: prognostic significance of incidentally detected tumors. J Urol 163: 426 – 433, 2000. 2. McDougal WS, and Garnick MB: Clinical signs and symptoms of renal cell carcinoma, in Vogelzang NJ, Scardino PT, Shipley WU, et al (Eds): Comprehensive Textbook of Genitourinary Oncology. Baltimore, Williams & Wilkins, 1995, Chapter IV, pp 111–115, 154 –159. 3. Blay JY, Rossi JF, Wijdenes J, et al: Role of IL-6 in the paraneoplastic inflammatory syndrome associated with renal cell carcinoma. Int J Cancer 72: 424 – 430, 1997. 4. Kawata N, Yamaguchi K, Hirakata H, et al: Immunosuppressive acidic protein detects high nuclear grade localized renal cell carcinoma. Urology 66: 736 –740, 2005. 5. Yaycioglu O, Roberts WW, Chan T, et al: Prognostic assessment of nonmetastatic renal cell carcinoma: a clinically based model. Urology 58: 141–145, 2001. 6. Patard JJ, Leray E, Cindolo L, et al: Multi-institutional validation of a symptom based classification for renal cell carcinoma. J Urol 172: 858 – 862, 2004. 7. Chambers AF, and Matrisian LM: Changing views of the role of matrix metalloproteinases in metastasis. J Natl Cancer Inst 89: 1260 –1270, 1997. 8. Egeblad M, and Werb Z: New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2: 161–174, 2002. 9. Overall CM, and Lopez-Otin C: Strategies for MMP inhibition in cancer: innovations for the post-trial era. Nat Rev Cancer 2: 657– 672, 2002.
UROLOGY 68 (3), 2006
10. Sobin LH, and Wittekind C: TNM Classification of Malignant Tumours, 5th ed. New York, John Wiley, 1997. 11. Fuhrman SA, Lasky LC, and Limas C: Prognostic significant of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 6: 655– 663, 1982. 12. Lang H, Lindner V, Saussine C, et al: Microscopic venous invasion: a prognostic factor in renal cell carcinoma. Eur Urol 38: 600 – 605, 2000. 13. Kawamata H, Kameyama S, Kawai K, et al: Marked acceleration of the metastatic phenotype of a rat bladder carcinoma cell line by the expression of the human gelatinase A. Int J Cancer 63: 568 –575, 1995. 14. Tsunezuka Y, Kinoh H, Takino T, et al: Expression of membrane-type matrix metalloproteinases 1 (MT1-MP) in tumor cells enhances pulmonary metastasis in an experimental metastasis assay. Cancer Res 56: 5678 –5683, 1996. 15. Zucker S, Hymowitz M, Conner C, et al: Measurement of matrix metalloproteinases and tissue inhibitors of metalloproteinases in blood and tissue: clinical and experimental applications. Ann NY Acad Sci 878: 212–227, 1999. 16. Gohji K, Fujimoto N, Komiyama T, et al: Evaluation of serum levels of matrix metalloproteinases-2 and 3 as new predictors of recurrence in patients with urothelial carcinoma. Cancer 78: 2379 –2387, 1996. 17. Kugler A, Hemmerlein B, Thelen P, et al: Expression of metalloproteinase 2 and 9 and their inhibitors in renal cell carcinoma. J Urol 160: 1914 –1918, 1998. 18. Lein M, Jung K, Laube C, et al: Matrix-metalloproteinases and their inhibitors in plasma and tumor tissue of patients with renal cell carcinoma. Int J Cancer 85: 801– 804, 2000. 19. Fukata S, Inoue K, Kamada M, et al: Levels of angiogenesis and expression of angiogenesis-related genes are prognostic for organ-specific metastasis of renal cell carcinoma. Cancer 103: 931–942, 2005. 20. Wilhelm SM, Collier IE, Marmer BL, et al: SV40-transformed human lung fibroblasts secrete a 92-kDa type IV collagenase which is identical to that secreted by normal human macrophages. J Biol Chem 264: 17213–17221, 1989. 21. Okada Y, Gonoji Y, Naka K, et al: Matrix metalloproteinase 9 (92-kDa gelatinase/type IV collagenase) from HT 1080 human fibrosarcoma cells: purification and activation of the precursor and enzymic properties. J Biol Chem 267: 21712–21719, 1992. 22. Kim J, Yu W, Kovalski K, et al: Requirement for specific proteases in cancer cell intravasation as revealed by a novel semiquantitative PCR-based assay. Cell 94: 353–362, 1998. 23. Gomez DE, Alonso DF, Yoshiji H, et al: Tissue inhibitors of metalloproteinases: structure, regulation and biological functions. Eur J Cell Biol 74: 111–122, 1997. 24. Laski ME, and Vugrin D: Paraneoplastic syndromes in hypernephroma. Semin Nephrol 7: 123–130, 1987.
527