[CANCER RESEARCH 60, 1761–1765, March 15, 2000]

Continuous Administration of Angiostatin Inhibits Accelerated Growth ofColorectal Liver Metastases after Partial Hepatectomy1

Tom A. Drixler, Inne H. M. Borel Rinkes, Ewan D. Ritchie, Theo J. M. V. van Vroonhoven,Martijn F. B. G. Gebbink, and Emile E. Voest2

Departments of Surgery [T. A. D., I. H. M. B. R., E. D. R., T. J. M. V. v. V.] and Internal Medicine [T. A. D., M. F. B. G. G., E. E. V.], Laboratory of Medical Oncology, UniversityMedical Center Utrecht, 3584 CX Utrecht, the Netherlands

ABSTRACT

Human plasminogen-derived angiostatin is one of the most potentantiangiogenic agents currently known. However, it is unclear whetherangiostatin is also effective against accelerated tumor growth induced bylocal up-regulation of growth factors, including angiogenesis stimulators,such as in regenerating liver. Prior to addressing this question, we tested,in mice, whether continuous administration of angiostatin could improveits biological effects. This assumption was based on the relatively shorthalf-life of angiostatin in mice, as well as on the theoretical necessity tosuppress tumor-induced angiogenesis continually. The findings presentedhere clearly indicate continuous administration to be superior to theconventional twice-daily bolus injections. Using the maximally effectiveregimen of 100 mg/kg/day via s.c. pump infusion, we found angiostatin tonot only suppress s.c. primary tumors but also to significantly inhibit theoutgrowth of colorectal hepatic metastases in resting liver and even toinhibit accelerated tumor growth in regenerating liver after 70% partialhepatectomy. In conclusion, angiostatin could play an important role inpatients subjected to partial hepatectomy to prevent outgrowth of residualmicrometastases, provided it is administered continuously to obtain max-imal biological effects.

INTRODUCTION

Antiangiogenic treatment has become one of the most exciting newdevelopments in anticancer therapy. Among the most potent antian-giogenic substances is angiostatin, a proteolytic cleavage product ofplasminogen (1). Angiostatin was discovered in mice bearing Lewislung carcinoma as an endogenous peptide, generated by the primarytumor, and capable of suppressing outgrowth of metastatic tumor cellsbeyond microscopic dimensions. In contrast, endogenously formedangiostatin was unable to inhibit propagation of the primary malig-nancy. This phenomenon was attributed to local production of proan-giogenic factors by the tumor, which outweighed the antiangiogenicregulators such as angiostatin, shifting the balance between the twoforces toward angiogenesis (i.e.,the angiogenic switch; Ref. 2). Incontrast, exogenous human plasminogen-derived angiostatin has beenshown to effectively suppress growth of s.c. primary tumor lesions inmice, regardless of tumor cell type (3).

In analogy herewith, liver regeneration after extensive hepaticresection is accompanied by a similar peak in local growth factorproduction, including proangiogenic factors, eliciting proliferativeresponses along a variety of auto- and paracrine pathways (4–9). Asa consequence, it has been suggested that residual “dormant” micro-metastases in the liver remnant might also display stimulated growthafter major hepatic resection. Recent experimental data appear to

support this concept (10–12). The role of angiogenesis under thesecircumstances is unknown. One could speculate that antiangiogenictreatment to prevent outgrowth of liver metastases might be problem-atic because of the overwhelming number of secondary tumor celldeposits. These deposits each produce their own proangiogenic mi-croenvironment in a host organ where angiogenesis dependency hasbeen demonstrated to differ from that in s.c. deposits (13). Moreover,the local secretion of large amounts of growth factors after partialhepatectomy may present additional loss of effectiveness of angio-genesis inhibitors. On the other hand, there is sufficient experimentalevidence that antiangiogenic treatment using exogenously adminis-tered, plasminogen-derived human angiostatin effectively inhibits out-growth of pulmonary metastases (3). In addition, angiostatin causesregression of existing macroscopic solid tumors, supporting our hy-pothesis that angiostatin does inhibit accelerated metastatic outgrowthin regenerating liver (3, 14, 15).

To test this hypothesis, optimal administration of angiostatin is aprerequisite. Hitherto, most animal studies on angiostatin have used atwice-daily s.c. treatment schedule. The half-life of human angiostatinin mice is around 4–6 h (1). This implies that twice-daily injectionsdo not allow for continuous suppression of angiogenesis. Accord-ingly, we have previously demonstrated that twice-daily bolus injec-tions with IFN-g at a sublethal concentration only partially suppressedangiogenesis, whereas continuous infusion of IFN-g at half the dailydose of the bolus injections completely inhibited angiogenesis (16).The idea that antitumoral effects could be improved by continuousadministration of an antiangiogenic peptide (TNP-470), instead ofintermittent administration, was also supported by Yamaokoet al.(17).

The present study was undertaken to define the optimal route ofadministration of angiostatin and to evaluate the antitumoral efficacyof such treatment in experimental colorectal hepatic metastases inresting liver, as well as in liver regeneration after major partialhepatectomy. We show that continuous administration of angiostatinsignificantly improves its antiangiogenic and antitumoral effects whencompared with twice-daily s.c. bolus injections. Furthermore, contin-uous administration of angiostatin was able to suppress acceleratedgrowth of colorectal liver metastases in a regenerating liver.

MATERIALS AND METHODS

Cell Lines and Cell Culture

The mouse colon adenocarcinoma cell line C-26 was maintained as amonolayer culture in DMEM supplemented with 10% heat-inactivated FCS,100 units/ml penicillin, and 100mg/ml streptomycin in a 10% CO2 environ-ment. Early passage cells (,15 times) were used. Confluent cultures wereharvested by brief trypsinization (0.05 trypsin in 0.02% EDTA), washed threetimes with PBS, and resuspended to a final concentration of 106 cells/ml (s.c.tumor) and 107 cells/ml (liver metastases) in PBS. The presence of single-cellsuspension was confirmed by phase contrast microscopy, and cell viability wasdetermined by trypan blue staining.

Received 9/8/99; accepted 1/18/00.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby markedadvertisementin accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by the Wijnand M. Pon Foundation (to T. A. D.), theNetherlands Digestive Diseases Foundation Grant WS 96-29 (to T. A. D.), and the DutchCancer Society Grant 99-2114 (to M. F. B. G. G.).

2 To whom requests for reprints should be addressed, at Department of InternalMedicine, Division of Medical Oncology, Laboratory of Medical Oncology, UniversityMedical Center, P. O. Box 85500, 3508 GA Utrecht, the Netherlands. Phone: 31-30-2506265; Fax: 31-30-2523741; E-mail: E.E.Voest@digd.azu.nl.

1761

Research. on September 7, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

Angiostatin Production

Human angiostatin was generated as described by O’Reillyet al. (1) usingminor modifications. Briefly, recovered outdated human plasma was diluted2:1 with PBS, supplemented with 3 mM EDTA, warmed up (37°C), and filtered(0.1 mm). The plasma was then applied to a lysine-Sepharose column (Phar-macia, Uppsala, Sweden) at room temperature. After washing the column with0.5M phosphate buffer, plasminogen was eluted with 0.2M prewarmed (37°C)e-aminocaproic acid at pH 7.4. SDS-PAGE of the eluant revealed one band ofapparentMr 92,000 corresponding to plasminogen. The eluant was dialyzedagainst demi-water (MWCO: 6–8000 Spectra/Por;.4 3 107 dilution; 4°C),followed by proteolytic digestion (12 h at 37°C; 120 rpm) with porcinepancreatic elastase (Calbiochem, San Diego, CA) in a concentration of 0.8unit/mg plasminogen using a shaker (24 h at 37°C; 120 rpm). Next, thesolution was applied to a lysine-Sepharose column again that had been equil-ibrated with a salt solution pH 7.4 (0.5M NaCl, 0.2 M e-ACA, 0.03 M

NaH2PO4, 0.02 NaN3, and 0.1% Triton). Furthermore, the column was re-equilibrated with a 30 mM phosphate buffer at pH 7.4. Finally, angiostatin waseluted with 0.2 mM e-ACA and dialyzed against demi-water (MWCO: 6–8000Spectra/Por;.4 3107 dilution). SDS-PAGE revealed three distinct bands ofapproximatelyMr 40,000,Mr 42,000, andMr 45,000, resembling the tripletfirst described by O’Reillyet al. (1). After freeze-drying, angiostatin wasdissolved in PBS (175 mg/ml) and stored at280°C.

Animals

Male Balb/c mice, 8–10 weeks of age, were used in all experiments andwere purchased from the General Animal Laboratory, University MedicalCenter Utrecht. Animals were maintained under specific pathogen-free condi-tions, food and waterad libitum, and kept on a 12-h light/12-h dark cycle.Experiments were performed according to the guidelines of the Utrecht AnimalExperimental Committee, University Medical Center Utrecht.

Cornea Neovascularization Assay

Antiangiogenic effects and dose dependency of angiostatin were tested incornea neovascularization assay as described previously, using minor modifi-cations (18). Briefly, mice (n5 8 eyes/group) were anesthetized using amixture of Hypnorm (0.3 mg/mouse i.p.; Janssen-Cilag, Brussels, Belgium)and Dormicum (12.5 mg/mouse i.p.; Roche, Brussels, Belgium). In addition,both corneas were anesthetized locally with 4 mg/ml oxybuprocaine (0.4%).Corneal micropockets were created with surgical blade number 10 and a pairof microscopic tweezers, followed by implantation of a;100 ng basic fibro-blast growth factor (Life Technologies, Inc., Rockville, MD) containing su-crose aluminum sucralfate pellet coated with Hydron (IFN Sciences, NewBrunswick, NJ). To prevent corneal infection, Aknemycin (Merck-BelgolaboNV, Overijse, Belgium) was applied to the cornea. The corneas of mice wereexamined daily by microscope. The surface area of newly formed bloodvessels was determined using a formula (0.23 p 3 maximal vessellength3 clock hours) described by Kenyonet al. (19). The experiment wasended when in untreated animals the newly formed blood vessels had reachedthe pellet (6 days after implantation of pellets).

Tumor Models

s.c. Tumor. Mice (n 5 7/group) received a s.c. injection of 106 C-26 cellsin 100 ml of PBS. Tumor diameters were determined daily by caliper, andtumor volume was calculated by the formula: width2 3 length 3 0.52. Theexperiment was ended after 14 days when the tumors of control mice reacheda volume that started to affect the quality of life.

Liver Metastases. Liver metastases were induced as follows. After anaes-thetizing mice (n5 7/group), a transverse incision in the left flank wasperformed, exposing the spleen, followed by intrasplenic injection of 105 C-26tumor cells in 100ml of PBS using a 27-gauge needle (20). A visible “paling”of the spleen and the lack of bleeding were the criteria for a successfulinoculation. Five min later, the spleen was removed to prevent growth of tumorcells in the spleen. In this model, metastases are confined to the liver (data notshown). Seven and 14 days after tumor inoculation, mice were sacrificed, andwet liver weight was measured. Then, left lateral and median liver lobes werefrozen in liquid nitrogen and stored at280°C until determination of the hepaticreplacement area.

Accelerated Intrahepatic Tumor Growth. Five min after intrasplenicinjection of 105 C-26 tumor cells, the spleen was resected; 5 min later, the leftlateral and caudal liver lobes were ligated and resected, resulting in a partialhepatectomy of 70%. Tumor growth was evaluated on days 7 and 14. Micewere randomized into four groups: tumor growth in resting liver (n5 12);tumor growth in resting liver1 angiostatin treatment (n5 11); tumor growthin regenerating liver (n5 6); and tumor growth in regenerating liver1 an-giostatin treatment (n5 6).

Intrahepatic tumor angiogenesis was evaluated by sectioning 10 standard-ized cryosections (5mm) of each liver lobe and reacting with monoclonalantibodies (MEC 13.3) against CD31 and counterstained with hematoxylin(21). Furthermore, in each section the intrahepatic metastases were encircledusing a digitalized pen that was connected to a light microscope unit (Axios-kop; Zeiss, Oberkochen, Germany) and video camera (Sony CCD/RGB colorvideo camera; Hamburg, Germany). In this way, microscopic images weretransferred into a computer frame store. Calculation of the hepatic replacementarea,i.e., the ratio of tumor volume/total liver volume (%), was calculatedusing software of Videoplan 2.2 (Kontron Elektronik GmbH, Eching, Ger-many).

Treatment Schedules

All animals, except hepatectomized mice, were treated with angiostatintwice-daily (bolus group) or continuously (pump group). Bolus injections weregiven as s.c. injections (twice-daily) with 100ml of angiostatin (12.5 mg/mlPBS) or 100ml of PBS (control). Continuous administration was performed byloading a mini-osmotic Alzet pump [Alza, Palo Alto, CA; type 2001 (7 days)or type 2002 (14 days)] with 200ml angiostatin (type 2001, 87.5 mg/ml PBS;type 2002, 175 mg/ml PBS). The pump was implanted s.c. in the dorsal skinfold. In mice bearing s.c. tumors, absence of contact between tumor depositand pump was assured by implanting the pump in the contralateral side.Animals received s.c. an initial bolus injection of 2.5 mg angiostatin in 100mlof PBS. In view of the results on continuous administration, subsequentexperiments on accelerated tumor growth in regenerating liver were performedusing continues administration only.

Antiangiogenic effects and dose response were tested in the cornea neovas-cularization assay for 6 days using three daily doses (1, 10, and 100 mg/kg;n 5 8/group). The Alzet 2001 pump was used. In all tumor models, acumulative daily dose of 100 mg/kg was used. Furthermore, in all of the pumpgroups in the tumor models, the Alzet 2002 type was used.

Statistical Analysis

Data were expressed as mean6 SE unless mentioned otherwise in the text.The significance of differences in corneal angiogenesis response, s.c. tumorgrowth (volume), and liver metastases (wet liver weight and hepatic replace-ment area) among groups was determined by the unpaired Studentst test.P , 0.05 was considered to be statistically significant.

RESULTS

Dose-dependent Inhibition Affects Treatment Schedule

After 6 days, new blood vessels from the limbal plexus had reachedthe pellet in all control mice. Both treatment groups consistentlyshowed positive angiogenic responses, but these were only a fractionof the response of control mice (Fig. 1). In the bolus group, inhibitionwas dose dependent: twice-daily injection with 1, 10, and 100 mg/kg/day resulted in 14.02, 33.38, and 72.10% inhibition, respectively.Furthermore, continuous administration of angiostatin resulted in adose-dependent inhibition, which was significantly greater than bolusinjection at all dose levels: 22.71, 67.07, and 92.68% inhibition. Nocorneal edema was seen. Histological examination revealed no signsof inflammation.

Antitumoral Effects Improved by Continuous Administrationof Angiostatin

Of the three daily doses tested in the cornea neovascularizationassay, 100 mg/kg revealed maximal angiogenesis inhibition. There-

1762

ANGIOSTATIN INHIBITS ACCELERATED TUMOR GROWTH

Research. on September 7, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

fore, this daily dose was used in subsequent experiments to compareantitumoral effects.

s.c. Tumor. Fourteen days after tumor cell inoculation, tumors ofthe control group had reached a volume of 18486 84 mm3 (Fig. 2).In mice treated twice-daily with angiostatin, the mean s.c. tumorvolume was 9196 94 mm3. Near total suppression of tumor growthwas observed in mice that received angiostatin continuously (tumorvolume5 104 6 16 mm3).

Liver Metastases.Mean wet liver weight (g) of normal mice was1.356 0.07 g. Seven and 14 days after tumor inoculation in controlmice, wet liver weight had increased to 2.296 0.22 g and3.786 0.16 g, respectively. The wet liver weight in the bolus groupwas 2.166 0.03 g (POD7;

3 not significant versus control) and2.986 0.06 g (POD14; P , 0.05versuscontrol). Tumor growth wasinhibited significantly in the pump group, as judged by wet liverweight of 1.776 0.12 g on POD7 (P , 0.05 versuscontrol) and2.24 6 0.13 g on POD14 (P , 0.01 versuscontrol). The differencebetween both treatment groups was significant at both time points.Similar results were obtained using the hepatic replacement area (%)as parameter for intrahepatic tumor growth (Fig. 3). In control ani-mals, the respective hepatic replacement areas were 59.36 7.5% and97 6 2% on POD7 and POD14. In the bolus group, they were43.36 8.1% (POD7) and 816 10.1% (POD14), respectively, whereasmaximal inhibition of tumor growth was observed in the pump group,with hepatic replacement areas of 126 4.6% on POD7 and 436 5.3%on POD14.

Accelerated Tumor Growth in Regenerating Liverand Angiostatin

All mice recovered well from liver resection. When compared withintrahepatic tumor growth in nonresected, nonregenerating liver, ac-celeration of growth of colorectal hepatic metastases was observed inthe remnant liver after 70% partial hepatectomy (Fig. 4). On POD7,the hepatic replacement area of control mice carrying colorectal livermetastases was 24.66 4.1%, whereas in the hepatectomized group,the hepatic replacement area had more than doubled: 536 3.2% (Fig.

5). This discrepancy between nonhepatectomized and hepatectomizedgroups had disappeared on POD14 (64 6 7.1% and 686 3.7%,respectively). Accelerated tumor growth in regenerating liver wasinhibited significantly by angiostatin. On POD7, the hepatic replace-ment area of nontreated hepatectomized mice and angiostatin-treatedhepatectomized mice was 536 3 and 20.86 1.4%, respectively.Seven days later, on POD14, the hepatic replacement area in angiosta-tin-treated mice was 346 1%, i.e., approximately half that in non-treated control mice (686 3.7%).

DISCUSSION

In the present study, we demonstrated that continuous administra-tion significantly improves both inhibition of angiogenesis and tumorgrowth as compared with bolus injections of angiostatin. This wasshown not only for primary s.c. tumors but also for metastatic out-growth using a highly aggressive cell line, which in normal circum-stances leads to murine death within 156 2 days after inoculation.Furthermore, using the same route of administration and a daily dose

3 The abbreviations used are: POD7 or POD14, postoperative day 7 or 14, respectively.

Fig. 2. Influence of method of administration on the antitumoral effects of humanangiostatin in mice bearing s.c. tumors.l, PBS control (n5 6); Œ, pump (n 5 7;angiostatin, 100 mg/kg/day);f, twice-daily (n5 7; angiostatin 100 mg/kg/day). Signif-icantly different from control value:p, P , 0.001.Bars,SE.

Fig. 3. Influence of method of administration on the antitumoral effects of humanangiostatin in murine colorectal liver metastases (n5 7/group).f, control mice treatedwith PBS;z, bolus group treated with angiostatin (100 mg/kg body weight/day);M, pumpgroup treated with angiostatin (100 mg/kg body weight/day). Tumor growth is expressedas a percentage using the hepatic replacement area on histological slides. Significantlydifferent from control values:p, P , 0.001; pp, P , 0.05. POD7 and POD14 valuesbetween bolus and pump groups were significantly different:P , 0.001.Bars,SE.

Fig. 1. Evaluation of antiangiogenic effects of human angiostatin in the corneaneovascularization assay. Measurements were taken at 6 days after implantation. Eachgroup (column) represents eight mice.f, control group. All dose levels differ significantlyfrom the control group. At a dose level of 1 mg/kg/day, the difference between the bolusand pump group was not significant (P5 0.1778). At the dose levels of 10 and 100mg/kg/day, the difference between the bolus and pump groups was significant (P, 0.001andP 5 0.001, respectively).Bars,SE.

1763

ANGIOSTATIN INHIBITS ACCELERATED TUMOR GROWTH

Research. on September 7, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

of 100 mg/kg body weight, accelerated outgrowth of colorectal he-patic metastases after 70% partial hepatectomy was inhibited as well.

As reported by others, we confirmed that performing a partialhepatectomy leads to accelerated intrahepatic tumor growth (22, 23).Because angiostatin strongly inhibited liver metastases in resting liver,we expected that angiostatin could suppress the accelerated outgrowthof liver metastases in regenerating liver as well. However, we did notexpect that antitumoral effects of angiostatin would be as strong as theinhibition of liver metastases in a resting liver. Surely, in a regener-ating liver, both remnant liver and colorectal liver metastases contrib-ute to local expression of growth factors, whereas in a resting liver,only the tumor contributes to up-regulation of local growth factors.This is supported by others; vascular endothelial growth factor, one ofthe major proangiogenic factors, is up-regulated in colorectal hepaticmetastases in resting (24, 25) and in remnant hepatocytes in regener-ating liver (26). On POD7, no significant differences could be ob-served between inhibition of tumor growth of colorectal metastases inresting (58% inhibition) and regenerating liver (60% inhibition). Buton POD14, inhibition of tumor growth of metastases was significantlystronger in the resting liver (63% inhibition) compared with theregenerating liver (49% inhibition;P , 0.05). This might be ex-

plained by the concept that compared with tumor growth in a restingliver, in a regenerating liver the local imbalance of growth factors maybe shifted more to the side of the proangiogenic factors. One way toovercome this problem could be to further increase the total daily doseof angiostatin. Furthermore, the effect of angiostatin, or any antian-giogenic treatment, on liver regeneration after partial hepatectomydeserves attention. In preliminary experiments, we have found thatangiostatin, in addition to an antitumor effect, reduces the rate ofphysiological liver regeneration.4

It is important to note that antitumor activity of angiostatin wasmeasured differently in the s.c. and liver metastases model; livermetastases were measured by surface area (mm2), whereas sc. tumorgrowth was evaluated using tumor volume (mm3). It is a mathematicalfact that any difference in tumor growth between control and testgroups will be more strongly exhibited using volume then area.Therefore, it appears that angiostatin is less effective in the livermetastases in comparison with s.c. tumors. The sc. injection of tumorcells in the s.c. dorsal skinfold results in a localized tumor depositcontaining 106 tumor cells, whereas in the liver model, an intrasplenicinjection with 105 tumor cells results in a large number of intrahepatictumor deposits. The liver metastases may all grow to 1–2 mm2

without neovascularization. This results in a substantial tumor growthnot affected by antiangiogenesis treatment. However, all describedmodels demonstrate that continuous administration compared withtwice-daily injections is crucial for inhibiting angiogenesis and tumorgrowth.

The role of antiangiogenic treatment for colorectal hepatic micro-metastases after partial hepatectomy is unclear. In agreement to ourresults, Tanakaet al. (27) reported previously that the angiogenesisinhibitor TNP-470 inhibited growth of colorectal liver metastases inrabbits. Our findings are relevant to the future use of antiangiogenicagents in cancer patients. In this study, we focused on colorectalhepatic metastases. The optimal clinical use of angiogenesis inhibitorsmay be in the adjuvant setting (28). Patients at risk for recurrence,either after complete surgical resection of the primary tumor/hepaticmetastases or after successful chemotherapy or radiation therapy,could be candidates for antiangiogenic treatment in an attempt toimpose dormancy on residual clinically undetectable micometastases.Continuous suppression appears to be crucial in preventing the out-growth of tumors. This may be done by continuous administration of

4 T. A. Drixler, M. F. B. G. Gebbink, T. J. M. V. v. Vroonhoven, E. E. Voest, and I.H. M. Borel Rinkes. Does angiostatin affect physiological angiogenesis?, manuscript inpreparation.

Fig. 4. Histological sections of tumor-bearing livers, showing microscopic appearance of colorectal hepatic metastasis 4 days after intrasplenic injection of 105 tumor cells (C-26).In mice subjected to 70% partial hepatectomy (A), the tumor area is significantly larger in comparison with control mice (B; H&E,340).

Fig. 5. Inhibition of angiostatin on tumor growth in resting and regenerating liver onPOD7 and POD14. f, tumor growth in resting liver (n5 12);z, tumor growth in restingliver 1 angiostatin treatment (n5 11; on POD14 P , 0.01versusf); s, tumor growthin regenerating liver (n5 6); M, tumor growth in regenerating liver1 angiostatintreatment (n5 6; on POD14 P , 0.001versuss). Bars,SE.

1764

ANGIOSTATIN INHIBITS ACCELERATED TUMOR GROWTH

Research. on September 7, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

these agents by ambulatory infusion pumps, by modulating antiangio-genic drugs to increase their plasma half-lives, or by new approachessuch as gene therapy.

In conclusion, plasminogen-derived human angiostatin is a verypotent antiangiogenic and antitumoral agent. Its biological effects canbe improved remarkably by continuous administration instead of onlytwice-daily. Furthermore, angiostatin might play an important role asan adjuvant antitumoral agent in patients subjected to partial hepate-ctomy for colorectal hepatic metastases.

ACKNOWLEDGMENTS

We kindly thank Colinda Aarsman for technical support. We gratefullyacknowledge Lillem Slot (Buck Meditec, Vaerlose, Denmark) for providingsucralfate, Dr. Peled (Weizmann Institute of Science, Rehovot, Israel) for theC-26 cell line, and Dr. Vecchi (Istituto di Ricerche Farmacologicalhe MarioNegri, Milano, Italy) for monoclonal antibodies against CD31.

REFERENCES

1. O’Reilly, M. S., Holmgren, L., Shing, Y., Chen, C., Rosenthal, R. A., Moses, M.,Lane, W. S., Cao, Y., Sage, E. H., and Folkman, J. Angiostatin: a novel angiogenesisinhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell,79: 315–328, 1994.

2. Hanahan, D., and Folkman, J. Patterns and emerging mechanisms of the angiogenicswitch during tumorigenesis. Cell,86: 353–364, 1996.

3. O’Reilly, M. S., Holmgren, L., Chen, C., and Folkman, J. Angiostatin induces andsustains dormancy of human primary tumors in mice. Nat. Med.,2: 689–692, 1996.

4. Michalopoulos, G. K., and DeFrances, M. C. Liver regeneration. Science (Washing-ton DC),276: 60–66, 1997.

5. Rai, R. M., Yang, S. Q., McClain, C., Karp, C. L., Klein, A. S., and Diehl, A. M.Kupffer cell depletion by gadolinium chloride enhances liver regeneration after partialhepatectomy in rats. Am. J. Physiol.,270: G909–G918, 1996.

6. Mead, J. E., and Fausto, N. Transforming growth factora may be a physiologicalregulator of liver regeneration by means of an autocrine mechanism. Proc. Natl. Acad.Sci. USA,86: 1558–1562, 1989.

7. Jones, D. E., Jr., Tran Patterson, R., Cui, D. M., Davin, D., Estell, K. P., and Miller,D. M. Epidermal growth factor secreted from the salivary gland is necessary for liverregeneration. Am. J. Physiol.,268: G872–G878, 1995.

8. Kan, M., Huang, J. S., Mansson, P. E., Yasumitsu, H., Carr, B., and McKeehan, W. L.Heparin-binding growth factor type 1 (acidic fibroblast growth factor): a potentialbiphasic autocrine and paracrine regulator of hepatocyte regeneration. Proc. Natl.Acad. Sci. USA,86: 7432–7436, 1989.

9. Michalopoulos, G., Houck, K. A., Dolan, M. L., and Leutteke, N. C. Control ofhepatocyte replication by two serum factors. Cancer Res.,44: 4414–4419, 1984.

10. de Jong, K. P., Lont, H. E., Bijma, A. M., Brouwers, M. A., de Vries, E. G., van Veen,M. L., Marquet, R. L., Slooff, M. J., and Terpstra, O. T. The effect of partialhepatectomy on tumor growth in rats:in vivo and in vitro studies. Hepatology,22:1263–1272, 1995.

11. Pliskin, M. E. Depression of hostversusgraft immunity and stimulation of tumorgrowth following partial hepatectomy. Cancer Res.,36: 1659–1663, 1976.

12. Picardo, A., Karpoff, H. M., Ng, B., Lee, J., Brennan, M. F., and Fong, Y. Partialhepatectomy accelerates local tumor growth: potential roles of local cytokine activa-tion. Surgery (St. Louis),124: 57–64, 1998.

13. Fukumura, D., Yuan, F., Monsky, W. L., Chen, Y., and Jain, R. K. Effect of hostmicroenvironment on the microcirculation of human colon adenocarcinoma.Am. J. Pathol.,151: 679–688, 1997.

14. Kirsch, M., Strasser, J., Allende, R., Bello, L., Zhang, J., and Black, P. M. Angiostatinsuppresses malignant glioma growthin vivo. Cancer Res.,58: 4654–4659, 1998.

15. Lannutti, B. J., Gately, S. T., Quevedo, M. E., Soff, G. A., and Paller, A. S. Humanangiostatin inhibits murine hemangioendothelioma tumor growthin vivo. CancerRes.,57: 5277–5280, 1997.

16. Voest, E. E., Kenyon, B. M., O’Reilly, M. S., Truitt, G., D’Amato, R. J., andFolkman, J. Inhibition of angiogenesisin vivo by interleukin 12. J. Natl. Cancer Inst.,87: 581–586, 1995.

17. Yamaoka, M., Yamamoto, T., Masaki, T., Ikeyama, S., Sudo, K., and Fujita, T.Inhibition of tumor growth and metastasis of rodent tumors by the angiogenesisinhibitor O-(chloroacetyl-carbamoyl)fumagillol (TNP-470; AGM-1470). CancerRes.,53: 4262–4267, 1993.

18. Kenyon, B. M., Voest, E. E., Chen, C. C., Flynn, E., Folkman, J., and D’Amato, R. J.A model of angiogenesis in the mouse cornea. Invest. Ophthalmol. Visual Sci.,37:1625–1632, 1996.

19. Kenyon, B. M., Browne, F., and D’Amato, R. J. Effects of thalidomide and relatedmetabolites in a mouse corneal model of neovascularization. Exp. Eye Res.,64:971–978, 1997.

20. Giavazzi, R., Jessup, J. M., Campbell, D. E., Walker, S. M., and Fidler, I. J.Experimental nude mouse model of human colorectal cancer liver metastases. J. Natl.Cancer Inst.,77: 1303–1308, 1986.

21. Vecchi, A., Garlanda, C., Lampugnani, M. G., Resnati, M., Matteucci, C.,Stoppacciaro, A., Schnurch, H., Risau, W., Ruco, L., and Mantovani, A. Monoclonalantibodies specific for endothelial cells of mouse blood vessels. Their application inthe identification of adult and embryonic endothelium. Eur. J. Cell Biol.,63: 247–254, 1994.

22. Slooter, G. D., Marquet, R. L., Jeekel, J., and Ijzermans, J. N. Tumour growthstimulation after partial hepatectomy can be reduced by treatment with tumournecrosis factora. Br. J. Surg.,82: 129–132, 1995.

23. Panis, Y., Ribeiro, J., Chretien, Y., and Nordlinger, B. Dormant liver metastases: anexperimental study. Br. J. Surg.,79: 221–223, 1992.

24. Nanashima, A., Ito, M., Sekine, I., Naito, S., Yamaguchi, H., Nakagoe, T., and Ayabe,H. Significance of angiogenic factors in liver metastatic tumors originating fromcolorectal cancers. Dig. Dis. Sci.,43: 2634–2640, 1998.

25. Warren, R. S., Yuan, H., Matli, M. R., Gillett, N. A., and Ferrara, N. Regulation byvascular endothelial growth factor of human colon cancer tumorigenesis in a mousemodel of experimental liver metastasis. J. Clin. Investig.,95: 1789–1797, 1995.

26. Mochida, S., Ishikawa, K., Inao, M., Shibuya, M., and Fujiwara, K. Increasedexpressions of vascular endothelial growth factor and its receptors, flt-1 and KDR/flk-1, in regenerating rat liver. Biochem. Biophys. Res. Commun.,226: 176–179,1996.

27. Tanaka, H., Taniguchi, H., Mugitani, T., Koishi, Y., Masuyama, M., Koyama, H.,Hoshima, M., and Takahashi, T. Angiogenesis inhibitor TNP-470 prevents implantedliver metastases after partial hepatectomy in an experimental model without impairingwound healing. Br. J. Surg.,83: 1444–1447, 1996.

28. Drixler, T. A., Voest, E. E., van Vroonhoven, T. J. M. V., and Borel Rinkes, I. H. M.Angiogenesis and surgery: from mice to men. Eur. J. Surg., in press, 2000.

1765

ANGIOSTATIN INHIBITS ACCELERATED TUMOR GROWTH

Research. on September 7, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

2000;60:1761-1765. Cancer Res   Tom A. Drixler, Inne H. M. Borel Rinkes, Ewan D. Ritchie, et al.   HepatectomyGrowth of Colorectal Liver Metastases after Partial Continuous Administration of Angiostatin Inhibits Accelerated

  Updated version

  http://cancerres.aacrjournals.org/content/60/6/1761

Access the most recent version of this article at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/60/6/1761.full.html#ref-list-1

This article cites 25 articles, 12 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/60/6/1761.full.html#related-urls

This article has been cited by 31 HighWire-hosted articles. Access the articles at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  .permissions@aacr.orgDepartment at

To request permission to re-use all or part of this article, contact the AACR Publications

Research. on September 7, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from