Synthetic Inhibitor of Matrix Metalloproteinases(Batimastat) Reduces Prostate Cancer Growth

in an Orthotopic Rat Model

Michael Lein,1* Klaus Jung,1 Dinh K. Le,1 Tayyaba Hasan,2 Bernhard Ortel,2Dietmar Borchert,1 Bjoern Winkelmann,1 Dietmar Schnorr,1 and

Stefan A. Loenings1

1Department of Urology, Charite University Hospital, Humboldt University, Berlin, Germany2Wellman Laboratories of Photomedicine, Massachusetts General Hospital, Harvard

Medical School, Boston, Massachusetts

BACKGROUND. Increased concentrations of metalloproteinases are associated with the in-vasive and metastatic behavior of several human malignant tumors. Normally, enzymaticactivity is tightly regulated by nonspecific mechanisms and specific inhibitors. The aim of thestudy was to determine the potential of a synthetic metalloproteinase inhibitor, batimastat, toshow its in vitro effect on MatLyLu cancer cells and its in vivo effect on tumor growth inorthotopic cancer (R3327 Dunning tumor) in rats.METHODS. In vitro, a dose response curve of batimastat was generated over 4 days using theMTT assay. Prostate cancer was injected in vivo in male Copenhagen rats by inoculatingR3327 Dunning tumor cells (MatLyLu) into the ventral prostatic lobe of 30 rats. Each of 10 ratsreceived batimastat (30 mg/kg body weight) or vehicle administered once a day by i.p.application beginning the day of cell inoculation. Ten rats remained untreated. The effect onlocal tumor growth was evaluated by measuring tumor weights 20 days after tumor cellinoculation.RESULTS. Significant inhibition of tumor cell proliferation in vitro occurred at 400 and 4,000ng/ml batimastat. After orthotopic cell inoculation, tumors grew to mean weights of 18.9 g inthe control group without treatment, to 22.3 g in the vehicle group, and to 11.1 g in the treatedgroup. In comparison to the control group and to the vehicle group, tumor weights increasedsignificantly less under treatment with batimastat.CONCLUSIONS. Batimastat is able to reduce tumor growth in the standard prostate cancermodel. Using this model, activity against cancer progression of future inhibitory agents canbe reliably assessed. Prostate 43:77–82, 2000. © 2000 Wiley-Liss, Inc.

KEY WORDS: prostate cancer treatment; R3327 Dunning tumor; inhibition of tumorprogression

INTRODUCTION

Cancer of the prostate is the most common malig-nancy in men. In 1998, an estimated 184,500 new casesof prostate cancer (PCa) were diagnosed in the USA[1]. Many factors are involved in choosing individualtreatment strategies. Radical prostatectomy is consid-ered the method of choice in the treatment of PCawithout metastases. In advanced PCa or after tumorrecurrence, hormone therapy is an established treat-ment to inhibit tumor progression with a limited ef-

fectiveness. However, these patients are not curableusing hormone therapy and will die of tumor progres-sion. This fact stimulates an enormous interest in ad-ditional methods of PCa therapy. One new promisingstrategy could be the use of synthetic inhibitors ofmatrix metalloproteinase (MMP) based on the signifi-

*Correspondence to: Michael Lein, M.D., Department of Urology,University Hospital Charite, Schumannstrasse 20/21, 10117 Berlin,Germany. E-mail: Michael.Lein@charite.deReceived 2 April 1999; Accepted 27 December 1999

The Prostate 43:77–82 (2000)

© 2000 Wiley-Liss, Inc.

cance of MMPs for tumor growth and formation ofmetastases [2].

MMPs belong to the group of endogenous prote-ases with the ability to degrade various components ofextracellular matrix [3]. These enzymes are involvedin physiological processes occurring during tissue re-modeling and repair, and play a crucial role underpathological conditions such as rheumatoid arthritis,tumor invasion, and metastasis [4,5.] A positive cor-relation between increased invasiveness of various hu-man tumors and MMP expression has been demon-strated, for instance, in breast, colon, gastric, and lungcancers [6–9].

The activity of these enzymes is regulated both bysecretion in a latent form requiring cleavage for acti-vation and by the presence of inhibitors [10]. TheMMPs are inhibited by endogenous specific and non-specific proteinase inhibitors. The specific componentsare called tissue inhibitors of metalloproteinases(TIMPs). Changes of the invasive qualities of the tu-mor cell correlate with the actual catalytic activities ofthe proteases as well as with the potential of proteaseinhibitors. Both in vitro and in vivo investigationshave shown that the balance between MMPs and theirinhibitors seems to be important for tumor invasionand progression and, in some cases, has shown asso-ciations with prognosis [3,11]. The relative importanceof MMPs and TIMPs has been elucidated in PCa [12–16]. In our studies, MMP and TIMP levels were foundto be significantly reduced in cancerous tissue of theprostate compared to the noncancerous part; however,there was an imbalance in favor of MMP [17]. Theseobservations suggest a complex interplay betweenMMPs and their inhibitors.

Recently, inhibitors were synthesized and a numberof studies have demonstrated that synthetic inhibitorsof metalloproteinases (e.g., batimastat, marimastat)limit tumor growth and metastases [18]. However,little is known about the efficacy of these new drugs inurological malignancies [19,20].

The present study examined the ability of a syn-thetic inhibitor, batimastat, to reduce tumor growth inthe hormone-independent PCa model. The DunningR3327 rat prostatic adenocarcinoma and its sublineshave been developed as a model system to study PCaprogression [21]. The Dunning tumor was chosen be-cause this tumor shows close similarities with humanPCa with regard to its behavior under hormonetherapy, chemotherapy, and/or radiation therapy.Therefore, the Dunning tumor has become the stan-dard animal model for human PCa [22].

There are sparse data on the effects of syntheticinhibitors in PCa [19,23]. This study was designed toevaluate the potential usefulness for treatment of PCain a rat model. Another aim was to find out whether

the Dunning tumor is a suitable model for testing fur-ther inhibitors of MMPs.

MATERIALS AND METHODS

Tumor Cell Line

The R3327-MatLyLu Dunning tumor is an andro-gen-independent, highly metastatic, and anaplastic ratprostate tumor, which spreads to the lymph nodesand to the lungs. Tumor cells were cultivated in RPMI-1640 (Bio Whitaker, Verviers, Belgium) with L-glutamine supplemented with 10% fetal calf serum,and were maintained at 37°C in an atmosphere of 5%CO2. Cells were passaged by trypsinization every 8–10days. The cell line was a generous gift of Dr. Heston(Memorial Sloan-Kettering Cancer Center, New York,NY).

In Vitro Cytotoxicity of Batimastat

Dunning cells were harvested in log phase growthin 60–70% confluence. The trypan blue exclusion assaywas performed to determine the real proportion ofliving and dead cells. Tumor cells were counted usinga hemocytometer. Afterwards, cells were seeded into96-well microtiter plates (Greiner, Alphen a/d Rijn,The Netherlands) and allowed to adhere during over-night incubation. Twenty-four hours after seeding, themedium was exchanged for 40, 400, and 4000 ng/mlbatimastat in culture medium, prepared as describedbelow. Cells were grown for a 5-day period. Mediumwith batimastat was renewed every 48 hr. The wellswere rinsed and the medium was changed. An MTT(3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazoliumbromide; thiazolyl blue) assay was used, which mea-sures the formation of a purple formazan from MTTcatalyzed by mitochondrial dehydrogenases. MTT so-lution dissolved in RPMI-1640 medium containing10% fetal calf serum was added and incubated at 37°Cfor 45 min. MatLyLu cells were tightly attached to thedishes, the MTT solution was carefully removed, and100 ml DMSO were added. Dishes were shaken atmoderate speed for 20 min to completely dissolve theformazan. The absorbance was read on a MicroplateReader (Anthos HT3, Anthos Labtec Instruments,Salzburg, Austria), using 540-nm or 620-nm band-passfilters.

Animals

Thirty male Copenhagen rats, 3 months old (180–200 g), were obtained from Charles River (Sulzfeld,Germany). Animal care was in accordance with insti-tutional guidelines. The animals received food and

78 Lein et al.

water ad libitum. During the tumor cell injection theanimals were anesthetized with an intraperitoneal in-jection of pentobarbital and were restrained in plasticholders allowing for exposure of the prostate.

Tumor Model and Treatment

This tumor model has been described in detail [24].In short, tumor cells were freshly harvested by tryp-sinization and resuspended in fresh complete mediumat 106/ml. PCa was orthotopically injected by implan-tation of 100,000 tumor cells (0.1 ml) into the ventralprostatic lobe of the rat, using a suprapubic transverseincision. To avoid any possible leakage of cell materialin the orthotopic model, a cotton-tipped applicatorwith iodine solution was placed over the injection sitebefore withdrawal of the needle and held in place forapproximately 1 min afterward. Dunning tumorswere injected in a total of 30 animals.

Ten rats were used in each group (control group,vehicle group, and treatment group). In group 1, allanimals remained untreated (control group). Vehiclealone (group 2) or batimastat (30 mg/kg; group 3) wasintraperitoneally administered daily, beginning on theday of tumor cell injection. Batimastat was kindly pro-vided by British Biotech Pharmaceuticals (Oxford,UK). A uniform solution of batimastat was preparedin PBS/Tween-80 (0.01%) solution by sonication ac-cording to the recommendation of the producers. Dur-ing the experiment the solution was kept at 4°C andwas shortly resonicated before use. All animals weresacrificed 20 days after tumor cell inoculation, andtumors were removed and examined (for volume andweight). Tumor growth was assessed by caliper mea-surement. Tumor volume was calculated on the as-sumption that the tumors were hemiellipsoids: V=length × width × height × 0.5236 [25].

Statistical Analysis

Statistical calculations were performed by the sta-tistical program GraphPad Prism (GraphPad, SanDiego, CA). The one-way analysis of variance(ANOVA) and Tukey’s multiple comparison test wereused. The cell proliferation curves were comparedusing ANOVA analysis with Dunnett’s multiple com-parison test for repeated measures. Statistical differ-ences of P < 0.05 were considered statistically signifi-cant.

RESULTS

In Vitro Effects of Batimastat

The proliferation curves are shown in Figure 1.There were small but significant inhibition effects

(about 10–20%) on tumor cell proliferation at the con-centration of 40 ng/ml, observed on day 4, and at theconcentration of 400 ng/ml, observed on days 3 and 4,whereas a highly reduced significant proliferation ratewas observed at 4,000 ng/ml after 1 day.

Using 4,000 ng/ml, the viability was less than 60%of vehicle control on days 2 and 3 and less than 70% onday 4.

In Vivo Effects of Batimastat

In each rat, tumors grew up after tumor cell injec-tion of the prostate. The surgical procedures werewell-tolerated.

The average rat weights are shown in Figure 2.Within 20 days there was a reduction in body weightsin the untreated groups (by 8.1% in the control group,and by 11.1% in the vehicle group) and by 9.0% in thetreatment group. Differences between the groups re-garding the body weights were not observed; thus, anadditional influence of batimastat on body weightcould not be proved.

In the untreated control group, 3 rats died, prob-ably because of the cancer, between day 19–20 (tumorweights: 19.3 g, 20.4 g, and 24.9 g). Figure 3 summa-rizes tumor growth in treated and untreated rats. Tu-mor weights were significantly different betweengroups. The orthotopic mean tumor weight 20 daysafter tumor cell inoculation was significantly smallerwithin the treated group than in the control group andthe vehicle group (11.1 g vs. 18.9 g vs. 22.3 g, respec-tively; P < 0.001). There were no significant differencesbetween the control group and the vehicle group (P >

Fig. 1. Effect of batimastat concentration on Dunning tumor cellproliferation as measured by MTT assay. Data are given as arith-metic means and standard deviation. The effect of batimastat wasrelated to the vehicle control, shown as 100%. Significant differ-ences from the control: *P < 0.05, **P < 0.001.

Batimastat Reduces Prostate Cancer Growth 79

0.05). Batimastat significantly inhibited local tumorgrowth (P < 0.001). In comparison to the untreatedgroups, tumor weights were about 50% lower. Usingtumor volumes instead of tumor weights for evalua-tion of the batimastat effect, equal results were ob-tained.

Macroscopically, the smaller tumors in the treatedgroup showed a complete capsule without infiltrationin the surrounding area. In contrast, in the control andvehicle groups, tumors were enlarged and invasive

growth was observed. The colon and the seminalvesicles showed evidence of tumor infiltration.

Some tiny deposits of batimastat were observed atthe injection sites, which demonstrated the low solu-bility of the compound.

DISCUSSION

In our study, we observed that the Dunning R3327adenocarcinoma of the Copenhagen rat is a reliablemodel for evaluating the effects of synthetic inhibitorsof MMPs on PCa. We used a dosing regimen of bati-mastat (30 mg/kg, intraperitoneal application, daily)which generates serum concentrations between 12–30ng/ml over 24 hr in nude mice [26] and about 100ng/ml in human beings (information from British Bio-tech), despite the low solubility of this compound asshown by some tiny deposits of batimastat at the in-jection sites in our experiments.

After inoculation of 100,000 MatLyLu cells, tumorsformed in 100% of rats. A tumor weight of about 20 gseems to be the limiting factor of life in this rat model,because 3 rats died with the respective tumor volume.In vitro, batimastat had little effect on the proliferationof MatLyLu cells at concentrations of 40 ng/ml and400 ng/ml. These concentrations are equal to or 10times higher than the concentrations in the animalsshowing the in vivo effect of batimastat. These datasupport the view already demonstrated in other ex-periments that batimastat only minimally influencesthe proliferation rate in cell culture but has an effect intissue by means of the inhibition of MMPs [27]. Apossible reason for this observation is the complex in-terplay between tumor progression and the surround-ing cellular matrix components within the body.

Little is known about the role of MMPs in prostatecancer [13–17,19,23,28]. In a previous study usingWestern blot technique, a correlation between tumorvolume and MMP-9 expression was observed in thesame prostate carcinoma model (Lein et al., unpub-lished results). Changed activity of MMPs in tissuemay also manifest itself in body fluids such as blood[29]. Baker at al. [19], measuring MMP-1, MMP-3,TIMP-1, and TIMP-2 in the blood of PCa patients,found increased TIMP-1 and MMP-1, unchangedMMP-3, and decreased TIMP-2 concentrations. Wewere able to show that the mean MMP-3 and TIMP-1concentrations were higher in PCa patients with me-tastases in comparison to controls and patients with-out metastases [30]. In order to measure the MMPsand TIMPs of PCa tissue in comparison to the normalprostate counterpart, prostate tissue samples obtainedfrom radical prostatectomies were analyzed [17]. Incancerous tissue samples, the median values of MMP

Fig. 2. Body weights of the treatment group (batimastat), ve-hicle group, and control group of rats. Significant differences be-tween weight curves were not observed. At the beginning of theexperiments, the mean weights were 208.9 g in the treatmentgroup, 219.3 g in the vehicle group, and 212.5 g in the controlgroup. Values are given as percentage of arithmetic means of theinitial values, indicated as 100%.

Fig. 3. Tumor weights measured 20 days after tumor cell injec-tion. Values are expressed as individual values and as arithmeticmeans. Shown are the results for the control group (18.9 ± 5.4 g)in comparison to the vehicle group (22.3 ± 4.3 g) and to thetreatment group (11.1 ± 2.6 g). Treated animals had significantlylower tumor weights than those in the vehicle group (P < 0.001)and in the control group (P < 0.001).

80 Lein et al.

and TIMP-1 were significantly lower, but the ratio ofMMP/TIMP-1 was significantly higher than in normaltissue. The decreased ratio was interpreted as an in-dicator of the imbalance between MMP and TIMP-1and as a phenomenon characteristic of prostate cancer.These results support the experimental attempt to ap-ply synthetic inhibitors in a PCa model.

Batimastat is a broad-spectrum MMP inhibitor anddisplays great inhibitory activity against many mem-bers of the MMP family, such as MMP-2 and MMP-9[18,31]. In our experimental model, we observed aclear inhibitory effect on PCa growth. This result war-rants further studies regarding metastases. The effectof batimastat using Duc-26 cells in an in vitro invasionassay and a diaphragm invasion model was evaluatedby Knox et al. [23]. These cells were generated bytransfecting the PCa DU-145 cell line derived from acentral nervous system metastasis. It was shown thatthe inhibitor is able to inhibit local tumor cell invasion,as evidenced by the PCa Duc-26 cells.

In patients with hormone-refractory PCa, prelimi-nary data indicate that the MMP inhibitor marimastatproduced an effect on the PSA level in 55% of patients[32]. Our results suggest that MMP inhibitors such asmarimastat are able to limit tumor growth and pro-gression, and thus may be effective therapeutic agents.It is possible that the development of MMP inhibitorswill yield new agents that will be particularly effectiveagainst hormone-independent PCa. These metastatictumors do not respond to conventional therapies.

In other tumor models, the systemic application ofsynthetic inhibitor was demonstrated to be effective inreducing local tumor growth. Watson et al. [33] evalu-ated the effect of batimastat on a human colorectalcancer ascites model, which was initiated by injectionof C170HM2 cells into the peritoneal cavity of severecombined immunodeficient (SCID) mice. The inocula-tion of ascites cells resulted in solid tumor depositsand ascites formation. Batimastat inhibited ascites for-mation in 47% of the animals and reduced the solidperitoneal tumor growth. In a similar study by Davieset al. [34], human ovarian carcinoma cells were im-planted into the peritoneal cavity of nude mice. Ad-ministration of batimastat resulted in a significant in-crease in survival. In a xenograft breast cancer model,tumor cells were implanted in the mammary fat padof nude mice [8]. Following batimastat treatment, asignificant reduction in local tumor regrowth and inlung metastases was described. Some studies haveshown batimastat to possess antimetastatic activityagainst a number of different tumor types. In a B16melanoma tumor model, batimastat was highly effec-tive in reducing hematogenous spread [35]. There arefew other reports of attempts to inhibit distant metas-

tases with batimastat. Eccles et al. [36] were able toobtain therapeutic effects against HOSP.1P lung colo-nies following intravenous inoculation of the cells.This is a rat mammary carcinoma which arose spon-taneously in a female CBH/cbi rat. The inhibitor re-duced the total tumor burden in all animals treated.Taken together, these investigations have shown thatthe inhibitor of MMPs, batimastat, can limit tumorprogression in animal models, and such investigationsfurther suggest that batimastat may be effective as anadjunctive cancer therapy. Because of the almost com-plete insolubility of batimastat, direct injection intobody cavities was necessary. Therefore, more solubleagents have been developed. Oral administration ofmarimastat generates reasonable blood levels of thedrug (information from British Biotech).

CONCLUSIONS

In conclusion, this study shows the ability of thebroad-spectrum MMP inhibitor batimastat to influ-ence local tumor growth in prostate cancer. The resultsmay offer an experimental basis for the evaluation ofnew and more specific MMP inhibitors with betterbioavailability (solubility) properties in the treatmentof prostate cancer.

REFERENCES

1. Landis SH, Murray T, Bolden S. Cancer statistics, 1998. CA Can-cer J Clin 1998;48:6–29.

2. Brown PD, Giavazzi R. Matrix metalloproteinase inhibition: areview of anti-tumor activity. Ann Oncol 1995;6:967–974.

3. Liotta LA, Stetler-Stevenson WG. Tumor invasion and metasta-sis: an imbalance of positive and negative regulation. CancerRes 1991;51:5054–5059.

4. Crawford HC, Matrisian LM. Tumor and stromal expression ofmatrix metalloproteinases and their role in tumor progression.Invasion Metastasis 1995;14:234–245.

5. Gibbons RP, Cole BS, Richardson RG, Correa RJ, Brannen GE,Mason JT, Taylor WJ, Twining SS. Regulation of proteolytic ac-tivity in tissues. Crit Rev Biochem Mol Biol 1994;29:315–383.

6. Anderson IC, Sugarbaker DJ, Ganju RK, Tsarwhas DG, RichardsWG, Sunday M, Kobzik L, Shipp MA. Stromelysin-3 is overex-pressed by stromal elements in primary non-small cell lungcancers and regulated by retinoic acid in pulmonary fibroblasts.Cancer Res 1995;55:4120–4126.

7. Murray GI, Duncan ME, O’Neil P, Melvin WT, Fothergill JE.Matrix metalloproteinase-1 is associated with poor prognosis incolorectal cancer. Nat Med 1996;2:461–462.

8. Davies B, Miles DW, Happerfield LC, Naylor MS, Bobrow LG,Rubens RD, Balkwill FR. Activity of type IV collagenases inbenign and malignant breast disease. Br J Cancer 1993;67:1126–1131.

9. Yamagata S, Yoshii Y, Suh JG, Tanaka R, Shimizu S. Occurrenceof an active form of gelatinase in human gastric and colorectalcarcinoma tissues. Cancer Lett 1991;59:51–55.

10. Stetler-Stevenson WG, Krutzch HC, Liotta LA. Tissue inhibitor

Batimastat Reduces Prostate Cancer Growth 81

of metalloproteinase (TIMP-2). A new member of the metallo-proteinase inhibitor family. J Biol Chem 1989;264:17374–17378.

11. Zeng ZS, Huang Y, Cohen AM, Guillem JG. Prediction of colo-rectal cancer relapse and survival via tissue RNA levels of ma-trix metalloproteinase 9. J Clin Oncol 1996;14:3133–3140.

12. Powell WC, Knox JD, Navre M, Grogan TM, Kittelson J, NagleRB, Bowden GT. Expression of the metalloproteinase matrilysinin DU 145 cells increases their invasive potential in severe com-bined immunodeficient mice. Cancer Res 1993;3:417–422.

13. Hamdy FC, Fadlon EJ, Cottam D, Lawry J, Thurrell W, SilcoksPB, Anderson JB, Williams JL, Rees RC. Matrix metalloprotein-ase 9 expression in primary prostatic adenocarcinoma and be-nign prostatic hyperplasia. Br J Cancer 1994;69:177–182.

14. Stearns M, Stearns ME. Evidence for increased activated metal-loproteinase 2 (MMP-2a) expression associated with humanprostate cancer progression. Oncol Res 1996;8:69–75.

15. Knox JD, Wolf C, McDaniel K, Clark V, Loriot M, Bowden GT,Nagle RB. Matrilysin expression in human prostate carcinoma.Mol Carcinog 1996;15:57–63.

16. Lokeshwar BL, Selzer MG, Block NL, Gunja-Smith Z. Secretionof matrix metalloproteinases and their inhibitors (tissue inhibi-tors of metalloproteinases) by human prostate explant cultures:reduced tissue inhibitor of metalloproteinase secretion by ma-lignant tissues. Cancer Res 1993;53:4493–4498.

17. Jung K, Lein M, Ulbrich N, Rudolph B, Henke W, Schnorr D,Loening SA. Quantification of matrix metalloproteinases in andtissue inhibitors of metalloproteinase in prostatic tissue: analyti-cal aspects. Prostate 1998;34:130–136.

18. Rasmussen HS, McCann PP. Matrix metalloproteinase inhibi-tion as a novel anticancer strategy: a review with special focuson batimastat and marimastat. Pharmacol Ther 1997;75:69–75.

19. Baker T, Tickle S, Wasan H, Docherty A, Isenberg D, Waxman J.Serum metalloproteinases and their inhibitors: markers for ma-lignant potential. Br J Cancer 1994;70:506–512.

20. Gohji K, Fujimoto N, Fujii A, Komiyama T, Okawa J, NakajimaM. Prognostic significance of circulating MMP2 to TIMP2 ratioin recurrence of urothelial cancer after complete resection. Can-cer Res 1996;56:3196–3198.

21. Dunning WF. Prostate cancer in the rat. NCI Monogr 1963;12:351–369.

22. Lucia MS, Bostwick DG, Bosland M, Cockett ATK, Knapp DW,Leav I, Pollard M, Rinker-Schaeffer C, Shirai T, Watkins BA.Workgroup I: rodent models of prostate cancer. Prostate 1998;36:49–55.

23. Knox JD, Bretton L, Lynch T, Bowden T, Nagle R. Syntheticmatrix metalloproteinase inhibitor, BB-94, inhibits the invasionof neoplastic human prostate cells in a mouse model. Prostate1998;35:248–254.

24. Kager M, Spruss T, Schneider MR, von Angerer E. DunningR3327-G prostate carcinoma of the rat: an appropriate model fordrug evaluation. J Cancer Res Clin Oncol 1992;118:334–338.

25. Dethlefsen LA, Prewitt JMS, Mendelsohn ML. Analysis of tu-mor growth curves. J Natl Cancer Inst 1968;2:389–406.

26. Wang X, Brown PD, Crimmin MJ, Hoffmann RM. Matrix me-talloproteinase inhibitor BB-94 (batimastat) inhibits human co-lon tumor growth and spread in a patient-like orthotopic modelin nude mice. Cancer Res 1994;54:4726–4728.

27. Zervos EE, Norman JG, Gower WR, Franz MG, Rosemurgy AS.Matrix metalloproteinase inhibition attenuates human pancre-atic cancer growth in vitro and decreases mortality and tumori-genesis in vivo. J Surg Res 1997;69:367–371.

28. Wilson MJ, Kapoor S, Vogel MM, Sinha AA. Characterization ofgelatin degrading metalloproteinase activities of the Dunningrat prostate tumor grown in nude mice. Prostate 1991;19:237–250.

29. Zucker S, Lysik RM, DiMassimo BI, Zarrabi HM, Moll UM,Grimson R, Tickle SP, Docherty AJ. Plasma assay of gelatinaseB: tissue inhibitor of metalloproteinase complexes in cancer.Cancer 1995;76:700–708.

30. Jung K, Nowak L, Lein M, Priem F, Schnorr D, Loening SA.Matrix metalloproteinases 1 and 3, tissue inhibitor of metallo-proteinase 1 and the complex of metalloproteinase 1/tissue in-hibitor in plasma of patients with prostate cancer. Int J Cancer1997;74:220–223.

31. Botos I, Scapozza L, Zhang D, Liotta LA, Meyer EF. Batimastat,a potent matrix metalloproteinase inhibitor, exhibits an unex-pected mode of binding. Proc Natl Acad Sci USA 1996;93:2749–2754.

32. Boasberg P, Harbaugh B, Roth B, Eisenberger M, Langleben A,Allen R, Rasmussen H. Marimastat, a novel matrix metallopro-teinase inhibitor in patients with hormone-refractory prostatecancer. Proc ASCO Thirty-Second Annu Meet 1996;15:258.

33. Watson SA, Morris TM, Parson SL, Steele RJC, Brown PD.Therapeutic effect of the matrix metalloproteinase inhibitor, ba-timastat, in a human colorectal cancer ascites model. Br J Urol1996;74:1354–1358.

34. Davies B, Brown P, East N, Crimmin MJ, Balkwill FR. A syn-thetic matrix metalloproteinase inhibitor decreases tumor bur-den and prolongs survival of mice bearing human ovarian car-cinoma xenografts. Cancer Res 1993;53:2087–2091.

35. Chirivi RGS, Garofalo A, Crimmin MJ, Bawden LJ, Stoppac-ciano A, Brown PD, Giavazzi R. Inhibition of the metastaticspread and growth of B-16-BL6 murine melanoma by a syn-thetic matrix metalloproteinase inhibitor. Int J Cancer 1994;58:460–464.

36. Eccles SA, Box GM, Court WJ, Bone EA, Thomas W, Brown PD.Control of lymphatic and hematogenous metastasis of a ratmammary carcinoma by the matrix metalloproteinase inhibitorbatimastat (BB-94). Cancer Res 1996;56:2815–2822.

82 Lein et al.