- Email: [email protected]
Matrix metalloproteinase inhibitor RO 28-2653 decreases liver metastasis by reduction of MMP-2 and MMP-9 concentration in BOP-induced ductal pancreatic cancer in Syrian Hamsters: Inhibition of matrix metalloproteinases in pancreatic cancer
ARTICLE IN PRESS
Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434 www.elsevier.com/locate/plefa
Matrix metalloproteinase inhibitor RO 28-2653 decreases liver metastasis by reduction of MMP-2 and MMP-9 concentration in BOP-induced ductal pancreatic cancer in Syrian Hamsters: Inhibition of matrix metalloproteinases in pancreatic cancer M. Kiliana,, J.I. Gregora, I. Heukampb, M. Hanela, M. Ahlgrimma, I. Schimkec, G. Kristiansend, A. Ommere, M.K. Walze, C.A. Jacobia, F.A. Wengere a
Clinic of General, Visceral, Vascular and Thoracic Surgery, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany b Clinic of Internal Medicine IV, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany c Clinic of Internal Medicine I, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany d Institute of Pathology, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany e Clinic of Surgery and Center of Minimal-Invasive Surgery, Kliniken Essen-Mitte, Henricistr. 92, University Duisburg-Essen, 45136 Essen, Germany Received 13 March 2006; accepted 25 August 2006
Abstract Background: Matrix metalloproteinases (MMP) are proteolytic enzymes which degrade the extracellular matrix and therefore play an important role in metastasis. However, the impact of MMP inhibitors (MMPI) on pancreatic cancer is still unclear. Thus we evaluated the influence of selective MMPI Ro 28-2653 on the incidence of liver metastases and the concentration of MMP-2 and MMP-9 in ductal pancreatic adenocarcinoma in Syrian hamster. Material and methods: One hundred and thirty male Syrian hamsters were randomised into 8 groups (Gr.1–3: n ¼ 15, Gr.4–8: n ¼ 17). Pancreatic cancer was induced by weekly subcutaneous injection of 10 mg N-nitrosobis-2-oxopropylamin (BOP)/kg body weight (Gr.4–8) while healthy control Gr. 1–3 received 0.5 ml sodium chloride 0.9%. Gr.1 and 4 had free access to a standard diet, Gr. 2, 3 and 5–8 received a diet rich in polyunsaturated fatty acids, which increases liver metastasis in this model. In week 17 oral therapy started: Gr.3 and 6: 60 mg Eudragit/kg body weight/d (vehicle of MMPI), Gr.7 and 8: 40 mg, respectively, 120 mg RO 282653/kg body weight/d; Gr.1, 2, 4, 5: no therapy. After 30 weeks all hamsters were sacrificed and histopathologically examined. Additionally concentrations of MMP-2 and MMP-9 were measured in non-metastatic liver and liver metastases. Results: Concentrations of MMP-2 and MMP-9 in liver metastases were decreased by high- and low-dose therapy with MMPI. Furthermore, the incidence of liver metastases was significantly reduced by low-dose therapy with Ro 28-2653. Conclusion: Low-dose therapy with Ro 28-2653 decreased liver metastasis due to an inhibition of MMP-2 and MMP-9 concentration in ductal pancreatic cancer. r 2006 Elsevier Ltd. All rights reserved.
1. Introduction Pancreatic cancer is the fifth leading cause for death of malignancies in Europe and the USA [1,2] with a Corresponding author. Tel.: +030 450622216; fax: +030 450522905. E-mail address: [email protected] (M. Kilian).
0952-3278/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.plefa.2006.08.004
median survival time less than 6 months [3–6]. Even after resection survival rates are low because of local recurrence and hepatic metastasis. Although several predispositional factors for pancreatic cancer like tobacco or alcohol have been identified, the pathomechanism of carcinogenesis is still unclear [4–7]. However, the structure of the basement membrane is supposed to play an important role in liver metastasis.
ARTICLE IN PRESS 430
M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
Thus matrix metalloproteinases (MMP) [8], a group of calcium-depending secreted and membrane-bound zincendopeptidases (gelatinases, collagenases, stromelysins and membrane-bound matrix metalloproteinases), have a central position in the degradation of the extra cellular matrix (ECM) including tissue remodelling, inflammatory cell invasion, organogenesis, angiogenesis, wound healing, bone resorption, trophoblast invasion and adipogenesis [9]. Accordingly MMP are also involved in the process of carcinogenesis due to a destruction of tissue integrity, metastatic spread and entry to blood and lymphatic systems. However, the balance of degrading processes and inactivity of MMP is controlled by four tissue inhibitors of metalloproteinases (TIMP). Moreover secreted pro-MMP have to be converted into active enzymes by proteolyses. Thus Gelatinase A (MMP 2) and B (MMP 9) degrade type-IV collagen, the major component of the basement membrane. Since MMP-2 and MMP-9 are overexpressed in ductal pancreatic cancer [10–13], we evaluated the impact of synthetic MMPI RO 28-2653 with a high specifity towards MMP-2 and MMP-9 on the incidence of liver metastasis in a model of ductal pancreatic carcinoma in Syrian hamster [14–17].
2. Materials and methods 2.1. Animals, tumor induction, diets and therapy One hundred and thirty eight-weeks-old male Syrian hamsters (Fa. Harlan-Winkelmann, Soest, Germany) were kept in single cages with corncob-bedding (Bed-OCobs, Anderson Cob, Maumee, USA) under standardized climatic conditions with constant temperature (21+5 1C), humidity (70+10%) and air exchanges as well as a controlled 12-h-dark/12-h-light cycle for 30 weeks. The animals had free access to water at libitum. The hamsters were randomized into eight groups: Group (gr.) 1–3: n ¼ 15 and gr. 4–8: n ¼ 17, respectively. The animals of gr. 4–8 received 10 mg Nnitrosobis-2-oxopropylamine (BOP, Ash Stevens Chem., USA)/kg body weight by weekly subcutaneous injection over a period of 10 weeks in order to induce a ductal pancreatic adenocarcinoma, while gr. 1–3 got 0.5 ml of 0.9% sodium chloride solution instead. Gr. 1 and 4 had free access to a standard fat diet (SF). In contrast all other animals received a special high fat diet (HF) (ssniff Soest GmbH, Soest, Germany), which causes an increased incidence of liver metastases up to 90% in this pancreatic cancer model [18–21]. Therapy started after 6 weeks after tumor induction lasting for 14 weeks. Therefore hamsters of gr. 3 (HF, Eudragit) and 6 (HF, BOP, Eudragit) received 60 mg Eudragit/kg body weight/day (Roche, Pharmaceuticals, Germany), while gr. 7 (HF, BOP, ld-MMPI) was treated
with 40 mg and gr. 8 (HF, BOP, hd-MMPI) with 120 mg RO 28-2653/kg body weight/day (Roche, Pharmaceuticals, Germany) by oral application. The experiment was approved by the public animal welfare committee and carried out according to the interdisciplinary principles and guidelines for the use of animals in research, marketing and education and the UKCCCR-Guidelines for the welfare of animals in experimental neoplasia [22,23]. 2.2. Histology After 30 weeks all hamsters were sacrificed and final body weights were estimated. Accordingly pancreas and liver were removed and weighed separately. Slices of pancreatic carcinomas and the whole pancreas of animals without macroscopic pancreatic carcinomas were embedded in 10% buffered formaldehyde. Hepatic slices of macroscopically tumor free areas were taken from left lateral lobe and frozen immediately at 80 1C for biochemical analysis. Liver of BOP-induced animals was sectioned in 1 mm steps and liver metastases were counted per animal. Furthermore, 2-dimensional size of liver metastases was measured. Slices of liver metastases were embedded in 10% buffered formaldehyde for histological examination. Specimen from formaldehyde solution were cut at 3 mm and stained with hematoxylin and eosin for histological examination. Preneoplastic lesions, borderline lesions, carcinoma in situ and ductal adenocarcinomas were classified according to the criteria of Meijers et al. [24]. 2.3. Sample preparation for biochemical analysis Ultraturrax homogenizer was used for sample preparation. For analysis of MMP, tissue was homogenized in 50 mM Tris-HCL-Buffer (pH 7.4) containing 1 mM Monothioglycerol. Preceding measurement of MMP concentrations, aliquots were centrifuged at 1000 rpm for 10 min. 2.4. Measurement of MMP concentration MMP-2 and MMP-9 concentration were measured using a commercial Biotrak Activity Assay System (Amersham Biosciences Europe GmbH, Freiburg, Germany). All samples and standards were incubated in microtitre wells precoated with anti-MMP-2/-MMP-9 antibody. All present MMP-2/MMP-9 was bound to the wells and other components of the samples were removed by washing and aspiration. MMP-2/MMP-9 was detected by activation of a modified pro detection enzyme and the subsequent cleavage of its chromogenic peptide substrate. The results were compared to a standard of activated MMP-2/MMP-9 in parallel samples. Therefore spectrophotometry (MRX TC
ARTICLE IN PRESS M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
Revelation, Dynex Technologies, USA) was performed at 405 nm. 2.5. Statistical analysis Continuous data were tested for normal distribution continuous by Shapiro–Francia test, accordingly no normal distribution was found. Data were given as median and range. Groups were compared using Kruskal–Wallis test for global hypothesis and Mann–Whitney test with Bonferroni correcture for single comparisons. We compared paired data in each group (non-metastic liver tissue vs. malignant liver tissue) by Wilcoxon rank test. If appropriate, Fisher’s exact test was carried out for categorical data. P values less than 0.05 were regarded as significant, statistical analysis was carried out using the PC program SPSS 11.0s on Windows XPs.
3. Results 3.1. General characteristics and lethality According to final body and liver weight, no significant differences were observed between the groups (data not shown). Lethality did not differ significantly between all groups, it was 0% in gr. 1, 2, 3, and 7, amounted 5.9% in gr. 5 and 8 (n ¼ 1). Highest mortality (17.6%) was found in gr. 4 and 6 (n ¼ 3) (data not shown).
431
As expected we observed no liver metastasis in healthy control gr.1–3. Lowest incidence of liver metastases was detected in tumour control gr.4 (42.9%), which was fed by standard fat diet. However, incidence of liver metastases was significantly increased by alimentation with special high fat diet in gr.5 (93.8%) and gr.6 (92.9%) (po0,05 vs. gr.4), while it was decreased by therapy with MMPI in gr.7 (53.3%) and gr.8 (68.8%) (po0.05 vs. gr.5–6) (Fig. 1). 3.3. MMP-2 concentration MMP-2 concentration in non-metastatic liver did not differ between gr.1–8 (gr.1: 3.76 [2.54–7.56] ng/ml; gr.2: 2.76 [1.88–5.35] ng/ml; gr.3: 3.77 [2.45–5.10] ng/ml; gr.4: 4.22 [2.74–10.03] ng/ml; gr.5: 5.43 [3.23-7.45] ng/ml; gr.6: 4.34 [3.23–6.65] ng/ml; gr.7: 2.54 [1.65–7.96] ng/ml; gr.8: 2.88 [1.76–7.00] ng/ml) (Fig. 2). While intrametastatic hepatic MMP-2 concentration did not differ between gr.4–6 (gr.4: 17.04 [10.55– 21.54] ng/ml; gr.5: 18.54 [10.43–21.56] ng/ml; gr.6: 18.54 [14.65–22.12] ng/ml), it was decreased in low and high dose therapy gr.7–8 (gr.7: 7.6 [4.43–13.62] ng/ml; gr.8: 5.43 [3.87–7.43] ng/ml; Po0.05 vs. gr.4–6, respectively) (Fig. 2). In all tumour groups (gr.4-8) intrametastic MMP-2 concentration was higher than in non-metastatic hepatic tissue (Po0.05) (Fig. 2). 3.4. MMP-9 concentration
3.2. Incidence of liver metastases All metastases were classified as metastases of ductal adenocarcinoma. No primary liver neoplasms were detected.
MMP-9 concentration in non-metastatic liver did not differ between gr.1–8 (gr.1: 0.66 [0.35–0.96] ng/ml; gr.2: 0.74 [0.35–0.90] ng/ml; gr.3: 0.74 [0.35–0.86] ng/ml; gr.4: 0.56 [0.31–0.75] ng/ml; gr.5: 0.56 [0.13–0.89] ng/ml; gr.6:
Incidence of Liver Metastases [%] 120
100
93.8
92.9
80 68.8 * 53.3
60 * 42.9 40
20
0 SF,BOP
HF,BOP
HF,BOP,Eudragit HF,BOP,ld-Inh. HF,BOP,hd-Inh.
Fig. 1. Incidence of liver metastases [%]. *Po0.05 vs. HF/BOP and HF/BOP/Eudragit, respectively.
ARTICLE IN PRESS 432
M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
MMP-2 in liver [ng/ml] 25 NML LiMe 20
15
10
* *
5
0 SF
HF,Eudr. HF
HF,BOP SF,BOP
HF,BOP,ld-Inh.
HF,BOP,Eudr.
HF,BOP,hd-Inh.
Fig. 2. MMP-2 concentration in non-metastatic liver (NML) and liver metastases (LiMe) [ng/ml]. *Po0,05 vs. LiMe of SF/BOP, HF/BOP and HF/ BOP/Eudragit, respectively.
0.37 [0.13–0.79] ng/ml; gr.7: 0.57 [0.25–0.80] ng/ml; gr.8: 0.48 [0.35–0.89] ng/ml) (Fig. 3). Furthermore intrametastatic hepatic MMP-9 concentration showed no difference between gr. 4–7 (gr.4: 5.29 [4.23–6.46] ng/ml; gr.5: 6.44 [3.12–8.55] ng/ml; gr.6: 4.24 [3.23–6.56] ng/ml. gr.7: 3.42 [3.23–5.45] ng/ml), while it was decreased in gr.8 (2.31 [1.34–3.23] ng/ml; Po0.05 vs. gr. 4–7) (Fig. 3). Comparison between intrametastatic and non-metastatic hepatic MMP-9 concentration revealed a higher concentration in liver metastases in all groups (Po0,05) (Fig. 3).
4. Discussion Pancreatic cancer is the fourth to fifth leading cause of death from malignancies in Europe and the USA [1,2]. Despite of improvement in the treatment of cancer in the last decades, there are only disappointing survival rates in pancreatic cancer. Furthermore at the time of diagnosis the majority of patients suffers from advanced cancer, thus only less than 30% of patients with pancreatic carcinoma can undergo curative resection [3,6,7]. Accordingly the development of new therapeutic options seems necessary. In the last years several pathways in cellular processes have been examined in order to detect the mechanisms
of carcinogenesis. MMP seem to play an important role in tumor growth and metastasis, although not all details of their activation and action are known. In various trials it has been shown that especially MMP-2 and MMP-9 are increased in pancreatic cancer and that MMP-production is tumor-derived and not only in response to the tumor [10,11,25]. These results are supported by our present study as well. Thus concentrations of intrametastatic hepatic MMP-2 and MMP-9 in tumour groups 4 and 5 showed the highest levels in liver metastases. Additionally in all tumor groups MMP-2 and MMP-9 concentration was higher in liver metastases than in non-metastatic hepatic tissue. The present study demonstrates the efficacy of oral treatment with the highly selective MMPI RO 28-2653 by decreasing the incidence of liver metastases due to reduced MMP 2 and MMP 9 concentration in the widely accepted model of BOP-induced pancreatic cancer in Syrian hamsters. This solid animal model is especially favoured for such investigations because of the outstanding histopathological similarities to human disease and the high incidence of pancreatic cancer [18,26,27]. Another reason is the increased liver metastasis up to 90% by a dietary elevation of raw fat content as described in a number of studies [18–21]. In this study we observed an elevation of liver metastasis from 42.9% in gr. 4 (SF, BOP) up to 93.8% in gr. 5 (HF, BOP). The
ARTICLE IN PRESS M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
433
MMP-9 in liver [ng/ml] 10 NML LiMe 8
6
4
* 2
0 SF
HF,Eudr. HF
HF,BOP SF,BOP
HF,BOP,ld-Inh.
HF,BOP,Eudr.
HF,BOP,hd-Inh.
Fig. 3. MMP-9 concentration in non-metastatic liver (NML) and liver metastases (LiMe) [ng/ml]. *Po0,05 vs. LiMe of SF/BOP, HF/BOP and HF/ BOP/Eudragit, respectively.
only difference between these groups was the raw fat content of the diet. Some experimental studies have shown an inhibiting effect of MMPI treatment on tumor growth and metastasis in different cancers [9,28–30]. One explanation for this action is the inhibition of the degradation of the ECM. Cancer cell infiltration and entry to blood, lymphatic and distant organ systems is reached by the MMP-induced degradation of the basement membrane [31]. Our results confirm this assumption because the incidence of liver metastases was decreased in both therapy groups. Additionally the measured concentrations of MMP 2 and 9 in liver metastases were reduced in these—with different doses of RO 28-2653 treated— groups. In conclusion we were able to show increased concentrations of MMP 2 and 9 in liver metastases in this animal model. Therapy with RO 28-2653 decreased this elevated concentrations and reduced the incidence of liver metastases in BOP induced pancreatic cancer in Syrian Hamster. Further investigations, particularly clinical trials, should be performed to observe whether the promising advantage of RO 28-2653, its highly selective inhibition of MMP 2 and MMP 9 concentration, shows comparable results in patients with pancreatic cancer. Perhaps the results of such studies with a high selective MMPI will not be as disappointing as earlier clinical trials with broad-spectrum MMPI like
Batimastat or Marimastat [32–35]. Additionally, the combination of Ro 28-2653 with other established therapies should be evaluated.
References [1] R.H. Hawes, Q. Xiong, I. Waxman, et al., A multispeciality approach to the diagnosis and management of pancreatic cancer, Am. J. Gastroenterol. 95 (2000) 17–31. [2] Al. Warshaw, C. Fernandez-del Castillo, Pancreatic carcinoma, N. Engl. J. Med. 326 (1992) 455–465. [3] Zalatnai, Pancreatic Cancer—a Continuing Challenge in Oncology, Pathol. Oncol. Res. 9 (4) (2003) 252–263. [4] N. Alexakis, C. Halloran, M. Raraty, P. Ghaneh, R. Sutton, J.P. Neoptolemos, Current standards of surgery for pancreatic cancer, Br. J. Surg. 91 (2004) 1410–1427. [5] K. Amikura, M. Kobari, The time of occurrence of liver metastasis in carcinoma of the pancreas, Int. J. Pancreatol. 17 (1995) 139–146. [6] S. Shore, M.G.T. Raraty, P. Ghaneh, J.P. Neoptolemos, Review article: chemotherapy for pancreatic cancer, Aliment Pharmacol. Ther. 18 (2003) 1049–1069. [7] J.P. Neoptolemos, D. Cunningham, H. Friess, C. Bassi, D.D. Stocken, et al., Adjuvant therapy in pancreatic cancer: historical and current perspectives, Ann. Oncol. 14 (2003) 675–692. [8] J. Gross, C.M. Lapiere, Collagenolytic activity in amphibian tissues: a tissue culture assay, Proc. Natl. Acad. Sci. 48 (1962) 1014–1022. [9] E. Maquoi, N.E. Sounni, L. Devy, F. Oliver, F. Frankenne, et al., Anti-invasive, anti-tumoral, and anti-angiogenic efficacy of a Pyrimidine-2,4,6-trione derivative, an orally active and selective
ARTICLE IN PRESS 434
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17] [18]
[19]
[20]
[21]
M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434 matrix metalloproteinase inhibitor, Clin. Cancer Res. 10 (2004) 4038–4047. X. Qian, V.L. Rothmann, R.F. Nicosia, G.P. Tuszynski, Expression of Thrombospondin-1 in human pancreatic adenocarcinomas: role in matrix metalloproteinase-9 production, Pathol. Oncol. Res. 7 (4) (2001) 251–259. E.E. Zervos, A.E. Shafil, A.S. Rosemurgy, Matrix metalloproteinase (MMP) inhibition selectively decreases type II MMP activity in a murine model of pancreatic cancer, J. Surg. Res. 81 (1999) 65–68. T. Koshiba, R. Hosotani, M. Wada, Y. Miyamoto, K. Fujimoto, et al., Involvement of matrix metalloproteinase-2 activity in invasion and metastasis of pancreatic carcinoma, Am. cancer soc. 642-650 (1998). E.E. Zervos, A.E. Shafil, M. Haq, A.S. Rosemurgy, Matrix metalloproteinase inhibition suppresses MMP-2 activity and activation of Panc-1 cells in vitro, J. Surg. Res. 84 (1999) 162–167. L.M. Coussens, B. Fingleton, L.M. Matrisian, Matrix metalloproteinase inhibitors and cancer: trials and tribulations, Science 295 (2002) 2387–2392. C.M. Overall, C. Lopez-Otin, Strategies for MMP inhibition in cancer: innovations for the post-trial era, Nat. Rev. 2 (2002) 657– 672. D. Yip, A. Ahmad, C.S. Karapetis, C.A. Hawkins, P.G. Harper, Matrix metalloproteinase inhibitors: applications in oncology 17 (1999) 387–399. S.M. Wojtowicz-Praga, R.B. Dickson, M.J. Hawkins, Matrix metalloproteinase inhibitors, Invest. New Drugs 15 (1997) 61–75. F.A. Wenger, C.A. Jacobi, M. Kilian, et al., The impact of laparoscopic biopsy of pancreatic lymph nodes with helium and carbon dioxide on port site and liver metastasis in BOP-induced pancreatic cancer in hamster, Clin. Exp. Metastasis 18 (2000) 11–14. F.A. Wenger, C.A. Jacobi, M. Kilian, et al., Does dietary alphalinoleic acid promote liver metastasis pancreatic carcinoma initiated by BOP in Syrian hamsters?, Ann. Nutr. Metabol. 43 (1999) 121–126. M.K. Herrington, J. Permert, K.R. Kazakoff, et al., Effects of high fat diet and cholecystokinin receptor blockade on pancreatic growth and tumour initiation in the hamster, Carcinogenesis 14 (5) (1993) 1021–1026. D.F. Birt, A.D. Julius, L.T. White, P.M. Pour, Enhancement of pancreatic carcinogenesis in hamsters fed a high fat diet ad libitum and at controlled calorie intake, Cancer Res. 49 (21) (1989) 5848–5851.
[22] UKCCCR-Guidelines for the welfare of animals in experimental neoplasia. Cancer Metastasis Rev. 8: (1989); 82–88. [23] New York Academy of Sciences, Ad Hoc committee on animal research. Interdisciplinary principles and guidelines for the use of animals in research, marketing and education. [24] M. Meijers, A. van Garderen-Hoetmer, C.B. Lamers, et al., Role of cholecystokinin in the development of BOP-induced pancreatic lesions in hamsters?, Carcinogenesis 11 (1990) 2223–2226. [25] M. Bloomstom, A. Shafii, E.E. Zervos, A. Rojiani, A.S. Rosemurgy, MMP-2 and TIMP-1 are derived from, not in response to, pancreatic cancer, J. Surg. Res. 102 (2002) 35–38. [26] D.S. Longnecker, Experimental cancer of the pancreas, Curr. Opin. Gastroeterol 7 (1991) 731–738. [27] P.M. Pour, R.G. Runge, D. Birt, et al., Current knowledge of pancreatic carcinogenesis in the hamster and its relevance to the human disease, Cancer 47 (1981) 1573–1587. [28] E.E. Zervos, J.G. Norman, W.R. Gower, M.G. Franz, A.S. Rosemurgy, Matrix metalloproteinase inhibition attenuates human pancreatic cancer growth in vitro and decreases mortality and tumorgenesis in vivo, J. Surg. Res. 69 (1997) 367–371. [29] A. Matsushita, M. Onda, E. Uchida, R. Maekawa, T. Yoshioka, Antitumor effect of a new selective matrix metalloproteinase inhibitor, MMI-166, on experimental pancreatic cancer, Int. J. Cancer 92 (2001) 434–440. [30] R. Jimenenz, W. Hartwig, B.A. Antoniu, C.C. Compton, A.L. Warshaw, et al., Effect of matrix metalloproteinase inhibition on pancreatic cancer invasion and metastasis, Ann. Surg. 231 (5) (2000) 644–654. [31] E. Deruygina, M. Bourdon, R. Reisfeld, A. Strongin, Remodelling of collagen matrix by human tumor cells requires activation and cell surface association of matrix metalloproteinase-2, Cancer Res. 58 (1998) 3743–3750. [32] S.R. Bramhall, M.T. Hallissey, J. Whiting, J. Scholefield, G. Tierney, et al., Marimastat as maintenance therapy for patients with advanced gastric cancer: a randomised trial, Br. J. Cancer 86 (2002) 1864–1870. [33] S. Zucker, J. Cao, W.T. Chen, Critical appraisal of the use of matrix metalloproteinase inhibitors in cancer treatment, Oncogene 19 (2000) 6642–6650. [34] L.M. Coussens, B. Fingleton, L.M. Matrisian, Matrix metalloproteinase inhibitors and cancer: trials and tribulations, Science (Wash DC) 295 (2002) 2387–2392. [35] B. Fingleton, Matrix metalloproteinase inhibitors for cancer therapy: the current situation and future prospects, Expert Opin. Ther. Targets 7 (2003) 385–397.
Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434 www.elsevier.com/locate/plefa
Matrix metalloproteinase inhibitor RO 28-2653 decreases liver metastasis by reduction of MMP-2 and MMP-9 concentration in BOP-induced ductal pancreatic cancer in Syrian Hamsters: Inhibition of matrix metalloproteinases in pancreatic cancer M. Kiliana,, J.I. Gregora, I. Heukampb, M. Hanela, M. Ahlgrimma, I. Schimkec, G. Kristiansend, A. Ommere, M.K. Walze, C.A. Jacobia, F.A. Wengere a
Clinic of General, Visceral, Vascular and Thoracic Surgery, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany b Clinic of Internal Medicine IV, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany c Clinic of Internal Medicine I, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany d Institute of Pathology, Charite´ Campus Mitte, Universita¨tsmedizin in Berlin, Schumannstr. 20/21, 10117 Berlin, Germany e Clinic of Surgery and Center of Minimal-Invasive Surgery, Kliniken Essen-Mitte, Henricistr. 92, University Duisburg-Essen, 45136 Essen, Germany Received 13 March 2006; accepted 25 August 2006
Abstract Background: Matrix metalloproteinases (MMP) are proteolytic enzymes which degrade the extracellular matrix and therefore play an important role in metastasis. However, the impact of MMP inhibitors (MMPI) on pancreatic cancer is still unclear. Thus we evaluated the influence of selective MMPI Ro 28-2653 on the incidence of liver metastases and the concentration of MMP-2 and MMP-9 in ductal pancreatic adenocarcinoma in Syrian hamster. Material and methods: One hundred and thirty male Syrian hamsters were randomised into 8 groups (Gr.1–3: n ¼ 15, Gr.4–8: n ¼ 17). Pancreatic cancer was induced by weekly subcutaneous injection of 10 mg N-nitrosobis-2-oxopropylamin (BOP)/kg body weight (Gr.4–8) while healthy control Gr. 1–3 received 0.5 ml sodium chloride 0.9%. Gr.1 and 4 had free access to a standard diet, Gr. 2, 3 and 5–8 received a diet rich in polyunsaturated fatty acids, which increases liver metastasis in this model. In week 17 oral therapy started: Gr.3 and 6: 60 mg Eudragit/kg body weight/d (vehicle of MMPI), Gr.7 and 8: 40 mg, respectively, 120 mg RO 282653/kg body weight/d; Gr.1, 2, 4, 5: no therapy. After 30 weeks all hamsters were sacrificed and histopathologically examined. Additionally concentrations of MMP-2 and MMP-9 were measured in non-metastatic liver and liver metastases. Results: Concentrations of MMP-2 and MMP-9 in liver metastases were decreased by high- and low-dose therapy with MMPI. Furthermore, the incidence of liver metastases was significantly reduced by low-dose therapy with Ro 28-2653. Conclusion: Low-dose therapy with Ro 28-2653 decreased liver metastasis due to an inhibition of MMP-2 and MMP-9 concentration in ductal pancreatic cancer. r 2006 Elsevier Ltd. All rights reserved.
1. Introduction Pancreatic cancer is the fifth leading cause for death of malignancies in Europe and the USA [1,2] with a Corresponding author. Tel.: +030 450622216; fax: +030 450522905. E-mail address: [email protected] (M. Kilian).
0952-3278/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.plefa.2006.08.004
median survival time less than 6 months [3–6]. Even after resection survival rates are low because of local recurrence and hepatic metastasis. Although several predispositional factors for pancreatic cancer like tobacco or alcohol have been identified, the pathomechanism of carcinogenesis is still unclear [4–7]. However, the structure of the basement membrane is supposed to play an important role in liver metastasis.
ARTICLE IN PRESS 430
M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
Thus matrix metalloproteinases (MMP) [8], a group of calcium-depending secreted and membrane-bound zincendopeptidases (gelatinases, collagenases, stromelysins and membrane-bound matrix metalloproteinases), have a central position in the degradation of the extra cellular matrix (ECM) including tissue remodelling, inflammatory cell invasion, organogenesis, angiogenesis, wound healing, bone resorption, trophoblast invasion and adipogenesis [9]. Accordingly MMP are also involved in the process of carcinogenesis due to a destruction of tissue integrity, metastatic spread and entry to blood and lymphatic systems. However, the balance of degrading processes and inactivity of MMP is controlled by four tissue inhibitors of metalloproteinases (TIMP). Moreover secreted pro-MMP have to be converted into active enzymes by proteolyses. Thus Gelatinase A (MMP 2) and B (MMP 9) degrade type-IV collagen, the major component of the basement membrane. Since MMP-2 and MMP-9 are overexpressed in ductal pancreatic cancer [10–13], we evaluated the impact of synthetic MMPI RO 28-2653 with a high specifity towards MMP-2 and MMP-9 on the incidence of liver metastasis in a model of ductal pancreatic carcinoma in Syrian hamster [14–17].
2. Materials and methods 2.1. Animals, tumor induction, diets and therapy One hundred and thirty eight-weeks-old male Syrian hamsters (Fa. Harlan-Winkelmann, Soest, Germany) were kept in single cages with corncob-bedding (Bed-OCobs, Anderson Cob, Maumee, USA) under standardized climatic conditions with constant temperature (21+5 1C), humidity (70+10%) and air exchanges as well as a controlled 12-h-dark/12-h-light cycle for 30 weeks. The animals had free access to water at libitum. The hamsters were randomized into eight groups: Group (gr.) 1–3: n ¼ 15 and gr. 4–8: n ¼ 17, respectively. The animals of gr. 4–8 received 10 mg Nnitrosobis-2-oxopropylamine (BOP, Ash Stevens Chem., USA)/kg body weight by weekly subcutaneous injection over a period of 10 weeks in order to induce a ductal pancreatic adenocarcinoma, while gr. 1–3 got 0.5 ml of 0.9% sodium chloride solution instead. Gr. 1 and 4 had free access to a standard fat diet (SF). In contrast all other animals received a special high fat diet (HF) (ssniff Soest GmbH, Soest, Germany), which causes an increased incidence of liver metastases up to 90% in this pancreatic cancer model [18–21]. Therapy started after 6 weeks after tumor induction lasting for 14 weeks. Therefore hamsters of gr. 3 (HF, Eudragit) and 6 (HF, BOP, Eudragit) received 60 mg Eudragit/kg body weight/day (Roche, Pharmaceuticals, Germany), while gr. 7 (HF, BOP, ld-MMPI) was treated
with 40 mg and gr. 8 (HF, BOP, hd-MMPI) with 120 mg RO 28-2653/kg body weight/day (Roche, Pharmaceuticals, Germany) by oral application. The experiment was approved by the public animal welfare committee and carried out according to the interdisciplinary principles and guidelines for the use of animals in research, marketing and education and the UKCCCR-Guidelines for the welfare of animals in experimental neoplasia [22,23]. 2.2. Histology After 30 weeks all hamsters were sacrificed and final body weights were estimated. Accordingly pancreas and liver were removed and weighed separately. Slices of pancreatic carcinomas and the whole pancreas of animals without macroscopic pancreatic carcinomas were embedded in 10% buffered formaldehyde. Hepatic slices of macroscopically tumor free areas were taken from left lateral lobe and frozen immediately at 80 1C for biochemical analysis. Liver of BOP-induced animals was sectioned in 1 mm steps and liver metastases were counted per animal. Furthermore, 2-dimensional size of liver metastases was measured. Slices of liver metastases were embedded in 10% buffered formaldehyde for histological examination. Specimen from formaldehyde solution were cut at 3 mm and stained with hematoxylin and eosin for histological examination. Preneoplastic lesions, borderline lesions, carcinoma in situ and ductal adenocarcinomas were classified according to the criteria of Meijers et al. [24]. 2.3. Sample preparation for biochemical analysis Ultraturrax homogenizer was used for sample preparation. For analysis of MMP, tissue was homogenized in 50 mM Tris-HCL-Buffer (pH 7.4) containing 1 mM Monothioglycerol. Preceding measurement of MMP concentrations, aliquots were centrifuged at 1000 rpm for 10 min. 2.4. Measurement of MMP concentration MMP-2 and MMP-9 concentration were measured using a commercial Biotrak Activity Assay System (Amersham Biosciences Europe GmbH, Freiburg, Germany). All samples and standards were incubated in microtitre wells precoated with anti-MMP-2/-MMP-9 antibody. All present MMP-2/MMP-9 was bound to the wells and other components of the samples were removed by washing and aspiration. MMP-2/MMP-9 was detected by activation of a modified pro detection enzyme and the subsequent cleavage of its chromogenic peptide substrate. The results were compared to a standard of activated MMP-2/MMP-9 in parallel samples. Therefore spectrophotometry (MRX TC
ARTICLE IN PRESS M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
Revelation, Dynex Technologies, USA) was performed at 405 nm. 2.5. Statistical analysis Continuous data were tested for normal distribution continuous by Shapiro–Francia test, accordingly no normal distribution was found. Data were given as median and range. Groups were compared using Kruskal–Wallis test for global hypothesis and Mann–Whitney test with Bonferroni correcture for single comparisons. We compared paired data in each group (non-metastic liver tissue vs. malignant liver tissue) by Wilcoxon rank test. If appropriate, Fisher’s exact test was carried out for categorical data. P values less than 0.05 were regarded as significant, statistical analysis was carried out using the PC program SPSS 11.0s on Windows XPs.
3. Results 3.1. General characteristics and lethality According to final body and liver weight, no significant differences were observed between the groups (data not shown). Lethality did not differ significantly between all groups, it was 0% in gr. 1, 2, 3, and 7, amounted 5.9% in gr. 5 and 8 (n ¼ 1). Highest mortality (17.6%) was found in gr. 4 and 6 (n ¼ 3) (data not shown).
431
As expected we observed no liver metastasis in healthy control gr.1–3. Lowest incidence of liver metastases was detected in tumour control gr.4 (42.9%), which was fed by standard fat diet. However, incidence of liver metastases was significantly increased by alimentation with special high fat diet in gr.5 (93.8%) and gr.6 (92.9%) (po0,05 vs. gr.4), while it was decreased by therapy with MMPI in gr.7 (53.3%) and gr.8 (68.8%) (po0.05 vs. gr.5–6) (Fig. 1). 3.3. MMP-2 concentration MMP-2 concentration in non-metastatic liver did not differ between gr.1–8 (gr.1: 3.76 [2.54–7.56] ng/ml; gr.2: 2.76 [1.88–5.35] ng/ml; gr.3: 3.77 [2.45–5.10] ng/ml; gr.4: 4.22 [2.74–10.03] ng/ml; gr.5: 5.43 [3.23-7.45] ng/ml; gr.6: 4.34 [3.23–6.65] ng/ml; gr.7: 2.54 [1.65–7.96] ng/ml; gr.8: 2.88 [1.76–7.00] ng/ml) (Fig. 2). While intrametastatic hepatic MMP-2 concentration did not differ between gr.4–6 (gr.4: 17.04 [10.55– 21.54] ng/ml; gr.5: 18.54 [10.43–21.56] ng/ml; gr.6: 18.54 [14.65–22.12] ng/ml), it was decreased in low and high dose therapy gr.7–8 (gr.7: 7.6 [4.43–13.62] ng/ml; gr.8: 5.43 [3.87–7.43] ng/ml; Po0.05 vs. gr.4–6, respectively) (Fig. 2). In all tumour groups (gr.4-8) intrametastic MMP-2 concentration was higher than in non-metastatic hepatic tissue (Po0.05) (Fig. 2). 3.4. MMP-9 concentration
3.2. Incidence of liver metastases All metastases were classified as metastases of ductal adenocarcinoma. No primary liver neoplasms were detected.
MMP-9 concentration in non-metastatic liver did not differ between gr.1–8 (gr.1: 0.66 [0.35–0.96] ng/ml; gr.2: 0.74 [0.35–0.90] ng/ml; gr.3: 0.74 [0.35–0.86] ng/ml; gr.4: 0.56 [0.31–0.75] ng/ml; gr.5: 0.56 [0.13–0.89] ng/ml; gr.6:
Incidence of Liver Metastases [%] 120
100
93.8
92.9
80 68.8 * 53.3
60 * 42.9 40
20
0 SF,BOP
HF,BOP
HF,BOP,Eudragit HF,BOP,ld-Inh. HF,BOP,hd-Inh.
Fig. 1. Incidence of liver metastases [%]. *Po0.05 vs. HF/BOP and HF/BOP/Eudragit, respectively.
ARTICLE IN PRESS 432
M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
MMP-2 in liver [ng/ml] 25 NML LiMe 20
15
10
* *
5
0 SF
HF,Eudr. HF
HF,BOP SF,BOP
HF,BOP,ld-Inh.
HF,BOP,Eudr.
HF,BOP,hd-Inh.
Fig. 2. MMP-2 concentration in non-metastatic liver (NML) and liver metastases (LiMe) [ng/ml]. *Po0,05 vs. LiMe of SF/BOP, HF/BOP and HF/ BOP/Eudragit, respectively.
0.37 [0.13–0.79] ng/ml; gr.7: 0.57 [0.25–0.80] ng/ml; gr.8: 0.48 [0.35–0.89] ng/ml) (Fig. 3). Furthermore intrametastatic hepatic MMP-9 concentration showed no difference between gr. 4–7 (gr.4: 5.29 [4.23–6.46] ng/ml; gr.5: 6.44 [3.12–8.55] ng/ml; gr.6: 4.24 [3.23–6.56] ng/ml. gr.7: 3.42 [3.23–5.45] ng/ml), while it was decreased in gr.8 (2.31 [1.34–3.23] ng/ml; Po0.05 vs. gr. 4–7) (Fig. 3). Comparison between intrametastatic and non-metastatic hepatic MMP-9 concentration revealed a higher concentration in liver metastases in all groups (Po0,05) (Fig. 3).
4. Discussion Pancreatic cancer is the fourth to fifth leading cause of death from malignancies in Europe and the USA [1,2]. Despite of improvement in the treatment of cancer in the last decades, there are only disappointing survival rates in pancreatic cancer. Furthermore at the time of diagnosis the majority of patients suffers from advanced cancer, thus only less than 30% of patients with pancreatic carcinoma can undergo curative resection [3,6,7]. Accordingly the development of new therapeutic options seems necessary. In the last years several pathways in cellular processes have been examined in order to detect the mechanisms
of carcinogenesis. MMP seem to play an important role in tumor growth and metastasis, although not all details of their activation and action are known. In various trials it has been shown that especially MMP-2 and MMP-9 are increased in pancreatic cancer and that MMP-production is tumor-derived and not only in response to the tumor [10,11,25]. These results are supported by our present study as well. Thus concentrations of intrametastatic hepatic MMP-2 and MMP-9 in tumour groups 4 and 5 showed the highest levels in liver metastases. Additionally in all tumor groups MMP-2 and MMP-9 concentration was higher in liver metastases than in non-metastatic hepatic tissue. The present study demonstrates the efficacy of oral treatment with the highly selective MMPI RO 28-2653 by decreasing the incidence of liver metastases due to reduced MMP 2 and MMP 9 concentration in the widely accepted model of BOP-induced pancreatic cancer in Syrian hamsters. This solid animal model is especially favoured for such investigations because of the outstanding histopathological similarities to human disease and the high incidence of pancreatic cancer [18,26,27]. Another reason is the increased liver metastasis up to 90% by a dietary elevation of raw fat content as described in a number of studies [18–21]. In this study we observed an elevation of liver metastasis from 42.9% in gr. 4 (SF, BOP) up to 93.8% in gr. 5 (HF, BOP). The
ARTICLE IN PRESS M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434
433
MMP-9 in liver [ng/ml] 10 NML LiMe 8
6
4
* 2
0 SF
HF,Eudr. HF
HF,BOP SF,BOP
HF,BOP,ld-Inh.
HF,BOP,Eudr.
HF,BOP,hd-Inh.
Fig. 3. MMP-9 concentration in non-metastatic liver (NML) and liver metastases (LiMe) [ng/ml]. *Po0,05 vs. LiMe of SF/BOP, HF/BOP and HF/ BOP/Eudragit, respectively.
only difference between these groups was the raw fat content of the diet. Some experimental studies have shown an inhibiting effect of MMPI treatment on tumor growth and metastasis in different cancers [9,28–30]. One explanation for this action is the inhibition of the degradation of the ECM. Cancer cell infiltration and entry to blood, lymphatic and distant organ systems is reached by the MMP-induced degradation of the basement membrane [31]. Our results confirm this assumption because the incidence of liver metastases was decreased in both therapy groups. Additionally the measured concentrations of MMP 2 and 9 in liver metastases were reduced in these—with different doses of RO 28-2653 treated— groups. In conclusion we were able to show increased concentrations of MMP 2 and 9 in liver metastases in this animal model. Therapy with RO 28-2653 decreased this elevated concentrations and reduced the incidence of liver metastases in BOP induced pancreatic cancer in Syrian Hamster. Further investigations, particularly clinical trials, should be performed to observe whether the promising advantage of RO 28-2653, its highly selective inhibition of MMP 2 and MMP 9 concentration, shows comparable results in patients with pancreatic cancer. Perhaps the results of such studies with a high selective MMPI will not be as disappointing as earlier clinical trials with broad-spectrum MMPI like
Batimastat or Marimastat [32–35]. Additionally, the combination of Ro 28-2653 with other established therapies should be evaluated.
References [1] R.H. Hawes, Q. Xiong, I. Waxman, et al., A multispeciality approach to the diagnosis and management of pancreatic cancer, Am. J. Gastroenterol. 95 (2000) 17–31. [2] Al. Warshaw, C. Fernandez-del Castillo, Pancreatic carcinoma, N. Engl. J. Med. 326 (1992) 455–465. [3] Zalatnai, Pancreatic Cancer—a Continuing Challenge in Oncology, Pathol. Oncol. Res. 9 (4) (2003) 252–263. [4] N. Alexakis, C. Halloran, M. Raraty, P. Ghaneh, R. Sutton, J.P. Neoptolemos, Current standards of surgery for pancreatic cancer, Br. J. Surg. 91 (2004) 1410–1427. [5] K. Amikura, M. Kobari, The time of occurrence of liver metastasis in carcinoma of the pancreas, Int. J. Pancreatol. 17 (1995) 139–146. [6] S. Shore, M.G.T. Raraty, P. Ghaneh, J.P. Neoptolemos, Review article: chemotherapy for pancreatic cancer, Aliment Pharmacol. Ther. 18 (2003) 1049–1069. [7] J.P. Neoptolemos, D. Cunningham, H. Friess, C. Bassi, D.D. Stocken, et al., Adjuvant therapy in pancreatic cancer: historical and current perspectives, Ann. Oncol. 14 (2003) 675–692. [8] J. Gross, C.M. Lapiere, Collagenolytic activity in amphibian tissues: a tissue culture assay, Proc. Natl. Acad. Sci. 48 (1962) 1014–1022. [9] E. Maquoi, N.E. Sounni, L. Devy, F. Oliver, F. Frankenne, et al., Anti-invasive, anti-tumoral, and anti-angiogenic efficacy of a Pyrimidine-2,4,6-trione derivative, an orally active and selective
ARTICLE IN PRESS 434
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17] [18]
[19]
[20]
[21]
M. Kilian et al. / Prostaglandins, Leukotrienes and Essential Fatty Acids 75 (2006) 429–434 matrix metalloproteinase inhibitor, Clin. Cancer Res. 10 (2004) 4038–4047. X. Qian, V.L. Rothmann, R.F. Nicosia, G.P. Tuszynski, Expression of Thrombospondin-1 in human pancreatic adenocarcinomas: role in matrix metalloproteinase-9 production, Pathol. Oncol. Res. 7 (4) (2001) 251–259. E.E. Zervos, A.E. Shafil, A.S. Rosemurgy, Matrix metalloproteinase (MMP) inhibition selectively decreases type II MMP activity in a murine model of pancreatic cancer, J. Surg. Res. 81 (1999) 65–68. T. Koshiba, R. Hosotani, M. Wada, Y. Miyamoto, K. Fujimoto, et al., Involvement of matrix metalloproteinase-2 activity in invasion and metastasis of pancreatic carcinoma, Am. cancer soc. 642-650 (1998). E.E. Zervos, A.E. Shafil, M. Haq, A.S. Rosemurgy, Matrix metalloproteinase inhibition suppresses MMP-2 activity and activation of Panc-1 cells in vitro, J. Surg. Res. 84 (1999) 162–167. L.M. Coussens, B. Fingleton, L.M. Matrisian, Matrix metalloproteinase inhibitors and cancer: trials and tribulations, Science 295 (2002) 2387–2392. C.M. Overall, C. Lopez-Otin, Strategies for MMP inhibition in cancer: innovations for the post-trial era, Nat. Rev. 2 (2002) 657– 672. D. Yip, A. Ahmad, C.S. Karapetis, C.A. Hawkins, P.G. Harper, Matrix metalloproteinase inhibitors: applications in oncology 17 (1999) 387–399. S.M. Wojtowicz-Praga, R.B. Dickson, M.J. Hawkins, Matrix metalloproteinase inhibitors, Invest. New Drugs 15 (1997) 61–75. F.A. Wenger, C.A. Jacobi, M. Kilian, et al., The impact of laparoscopic biopsy of pancreatic lymph nodes with helium and carbon dioxide on port site and liver metastasis in BOP-induced pancreatic cancer in hamster, Clin. Exp. Metastasis 18 (2000) 11–14. F.A. Wenger, C.A. Jacobi, M. Kilian, et al., Does dietary alphalinoleic acid promote liver metastasis pancreatic carcinoma initiated by BOP in Syrian hamsters?, Ann. Nutr. Metabol. 43 (1999) 121–126. M.K. Herrington, J. Permert, K.R. Kazakoff, et al., Effects of high fat diet and cholecystokinin receptor blockade on pancreatic growth and tumour initiation in the hamster, Carcinogenesis 14 (5) (1993) 1021–1026. D.F. Birt, A.D. Julius, L.T. White, P.M. Pour, Enhancement of pancreatic carcinogenesis in hamsters fed a high fat diet ad libitum and at controlled calorie intake, Cancer Res. 49 (21) (1989) 5848–5851.
[22] UKCCCR-Guidelines for the welfare of animals in experimental neoplasia. Cancer Metastasis Rev. 8: (1989); 82–88. [23] New York Academy of Sciences, Ad Hoc committee on animal research. Interdisciplinary principles and guidelines for the use of animals in research, marketing and education. [24] M. Meijers, A. van Garderen-Hoetmer, C.B. Lamers, et al., Role of cholecystokinin in the development of BOP-induced pancreatic lesions in hamsters?, Carcinogenesis 11 (1990) 2223–2226. [25] M. Bloomstom, A. Shafii, E.E. Zervos, A. Rojiani, A.S. Rosemurgy, MMP-2 and TIMP-1 are derived from, not in response to, pancreatic cancer, J. Surg. Res. 102 (2002) 35–38. [26] D.S. Longnecker, Experimental cancer of the pancreas, Curr. Opin. Gastroeterol 7 (1991) 731–738. [27] P.M. Pour, R.G. Runge, D. Birt, et al., Current knowledge of pancreatic carcinogenesis in the hamster and its relevance to the human disease, Cancer 47 (1981) 1573–1587. [28] E.E. Zervos, J.G. Norman, W.R. Gower, M.G. Franz, A.S. Rosemurgy, Matrix metalloproteinase inhibition attenuates human pancreatic cancer growth in vitro and decreases mortality and tumorgenesis in vivo, J. Surg. Res. 69 (1997) 367–371. [29] A. Matsushita, M. Onda, E. Uchida, R. Maekawa, T. Yoshioka, Antitumor effect of a new selective matrix metalloproteinase inhibitor, MMI-166, on experimental pancreatic cancer, Int. J. Cancer 92 (2001) 434–440. [30] R. Jimenenz, W. Hartwig, B.A. Antoniu, C.C. Compton, A.L. Warshaw, et al., Effect of matrix metalloproteinase inhibition on pancreatic cancer invasion and metastasis, Ann. Surg. 231 (5) (2000) 644–654. [31] E. Deruygina, M. Bourdon, R. Reisfeld, A. Strongin, Remodelling of collagen matrix by human tumor cells requires activation and cell surface association of matrix metalloproteinase-2, Cancer Res. 58 (1998) 3743–3750. [32] S.R. Bramhall, M.T. Hallissey, J. Whiting, J. Scholefield, G. Tierney, et al., Marimastat as maintenance therapy for patients with advanced gastric cancer: a randomised trial, Br. J. Cancer 86 (2002) 1864–1870. [33] S. Zucker, J. Cao, W.T. Chen, Critical appraisal of the use of matrix metalloproteinase inhibitors in cancer treatment, Oncogene 19 (2000) 6642–6650. [34] L.M. Coussens, B. Fingleton, L.M. Matrisian, Matrix metalloproteinase inhibitors and cancer: trials and tribulations, Science (Wash DC) 295 (2002) 2387–2392. [35] B. Fingleton, Matrix metalloproteinase inhibitors for cancer therapy: the current situation and future prospects, Expert Opin. Ther. Targets 7 (2003) 385–397.