ANTICANCER RESEARCH 24: 2201-2208 (2004)

TZT-1027 Elucidates Antitumor Activity through Direct Cytotoxicity and Selective Blockade of Blood Supply

NAOKO HASHIGUCHI1, TETSURO KUBOTA1, JUN-ICHI KOH2, YOSHINORI YAMADA3, YOSHIRO SAIKAWA1, YOSHIHIDE OTANI1, MASAHIKO WATANABE1, KOICHIRO KUMAI1, MASAKI KITAJIMA1, JUN-ICHI WATANABE4 and MOTOHIRO KOBAYASHI4

1Department of Surgery, School of Medicine, Keio University, Tokyo 160-8582; 2Department of Surgery, Saitama Social Insurance Hospital, Saitama 336-0002; 3Department of Surgery, Kitasato Institute Hospital, Tokyo 108-8642; 4Pharmacological Research Department, Teikoku Hormone MFG, Co., Ltd., Kanagawa 213-8522, Japan

Abstract. Background: TZT-1027 is a newly developed preclinical study indicated that TZT-1027 had antitumor antitumor agent derived from dolastatin 10. Materials and activity against several murine tumors, including P388 Methods: The in vitro activity of TZT-1027 on MCF-7 and R- leukemia, colon 26, B16 melanoma, M5076 sarcoma and 27 cells was evaluated by MTT assay. TZT-1027 1 mg/kg/week human tumor xenografts, MX-1 and LX-1 (3). Watanabe was administered i.v. for 4 weeks into nude mice bearing MCF- and colleagues investigated whether TZT-1027 inhibited the 7 and R-27. Subsequently, primary cultured cells from growth of various human cancer cells and whether the cell xenografts were also used for CD-DST. Two mg of TZT-1027 death caused by TZT-1027 was due to apoptosis in human or 40 mg docetaxel per kg were injected i.v. into nude mice leukemia HL-60 cells, human breast cancer MCF-7 cells and bearing R-27. 0.2% Evans blue was injected to assess the blood human prostate cancer DU145 cells, suggesting an flow. Results: TZT-1027 suppressed the in vitro growth of association of the growth-inhibitory effect of TZT-1027 with MCF-7 cells, while R-27 cells were resistant to TZT-1027, the induction of apoptosis and indicating that the apoptosis although its in vivo antitumor activity was remarkable. TZT- induced by TZT-1027 was followed by G2/M arrest (4). In 1027 blockaded R-27 tumor blood flow immediately after the present study, we further investigated the mode of injection; blood flow was not affected by docetaxel. Conclusion: action of TZT-1027 in terms of direct and indirect TZT-1027 exerts its antitumor activity through direct cytotoxicity cytotoxicity through vascular blockade, using the human against MCF-7 cells and through selective blockade of tumor breast cancer cell lines MCF-7 and R-27. blood flow against R-27 cells. Materials and Methods Since the introduction of colchicine derivatives to medical Drugs. TZT-1027 was synthesized in the Pharmacological Research treatment for gout, tubulin has been considered a molecular Department of Teikoku Hormone Mfg. Co. Ltd., Japan, and was target of antitumor agents. Plant alkaloids, such as dissolved in 0.05 M lactate buffer (pH 4.5), which was further vincristine and vinblastine, inhibit the polymerization of diluted before experimental use. The chemical structure of TZT- tubulin by binding microtubules, while newly developed 1027 is shown in Figure 1. Docetaxel was purchased from Aventis taxanes (paclitaxel and docetaxel) exert their antitumor Pharma. Ltd., Tokyo, Japan. activity through the stabilization of microtubules. TZT-1027 (1) is a newly synthesized derivative of dolastatin 10, a Human breast cancer cell lines. Human breast cancer cell lines MCF-7 and R-27 were used for the experiments. MCF-7 was compound isolated by Pettit and colleagues from the Indian initially established by Soule and coworkers (5) as a cultured cell Ocean sea hare, Dolabella auricularia, in 1987 (2). A line of hormone-dependent human breast cancer and we transplanted it into a nude mouse treated with estrogen and progesterone (6). R-27 was established as a tamoxifen-resistant variant of MCF-7 by Nawata and colleagues (7) and was also Correspondence to: Tetsuro Kubota, Department of Surgery, transplanted into nude mice by the same procedure as for MCF-7 School of Medicine, Keio University, 35 Shinanomachi, Shinjuku- (8). Both cell lines were maintained in RPMI 1640 (GIBCO, ku, Tokyo 160-8582, Japan. Tel: +81-3-33531211, Ext. 62334, Fax: Gaithersburg, MD, USA) supplemented with 10% fetal bovine +81-3-33554707, e-mail: [email protected] serum (FBS; CSL Limited, Australia), 100 I.U. penicillin, 100 Ìg streptomycin and 0.25 Ìg amphotericin B per ml (RPMI 1640 Key Words: TZT-1027, breast cancer, nude mouse, vascular flow. medium). The xenografts were maintained by serial subcutaneous

0250-7005/2004 $2.00+.40 2201 ANTICANCER RESEARCH 24: 2201-2208 (2004)

Figure 1. Chemical structure of TZT-1027, a derivative of dolastatin 10.

inoculation into BALB/c female nude mice (CLEA, Tokyo, Japan) Antitumor activity of TZT-1027 against human breast carcinoma that were treated with 5 mg/kg 17·-estradiol dipropionate and 250 xenografts in nude mice. Tumor tissue fragments, each measuring mg/kg 17‚-progesterone caproate, which accounts for 0.1 ml EP approximately 3 x 3 x 3 mm, were inoculated bilaterally into hormone depotì (Teikoku Hormone Mfg, Co. Ltd., Tokyo, subcutaneous tissues of the dorsum of ether-anesthetized nude Japan) per mouse. These two cell lines and xenografts are mice, using a trocar needle. EP hormone depotì was administered estrogen and progesterone receptor-positive in vitro and in vivo. once intramuscularly into the thigh of the mice on the inoculation day and repeated on Days 7 and 14 for MCF-7. Implanted tumors 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) were measured (length and width) with a sliding caliper three times assay. We evaluated the in vitro cytotoxicity of TZT-1027 against weekly by the same observer. Tumor weight was calculated MCF-7 and R-27 cells in culture using the 3-(4,5-dimethylthiazol-2- according to the method of Geran and colleagues (14) from linear yl)-2,5-diphenyl tetrazolium bromide (MTT, Sigma Chemical Co., measurements using the following formula: St. Louis, MO, USA) assay as reported by Mosmann (9), with some tumor weight (mg) = length (mm) x [width (mm)]2/2. modifications (10-13). When tumors reached 100 to 300 mg, tumor-bearing mice were After the cells had reached confluency, they were harvested, allocated randomly to test groups each consisting of 6 mice and centrifuged once, suspended in RPMI 1640 medium and diluted to treatment was initiated. One mg of TZT-1027 per kg suspended in 1 x 105 cells/ml. Aliquots (100 Ìl) plated into 96-well microplates 0.1 ml sterile water per mouse was administered intravenously once (GIBCO) resulted in approximately 104 cells per well. Drug weekly for four weeks. solutions were dissolved in RPMI 1640 medium and 100-Ìl aliquots Relative mean tumor weight (RW) was calculated as RW = were added to each well, giving final concentrations of 0.002, 0.02, Wi/Wo, where Wi is the mean tumor weight at any given time and 0.2, 2, 20, 200, 2,000, 20,000 pg of TZT-1027 per ml. Control wells Wo is the mean tumor weight at the time of initial treatment. The contained 100 Ìg cell suspension and 100 Ìl RPMI 1640 medium, antitumor activity of TZT-1027 was evaluated as the lowest T/C while 200 Ìl RPMI 1640 medium was used as a blank. Plates were value (%) during the experiment, where T is the RW of the treated incubated for 48 or 72 hours at 37ÆC in a humidified atmosphere group and C the RW of the control group at any given time. containing 95% air and 5% CO2. A mixture of 0.4% MTT (Sigma) Tumors were characterized as sensitive when the lowest T/C value and 0.1 M sodium succinate (Wako Pure Chemical Ind. Ltd., Osaka, was equal to or less than 42%, which was calculated from (0.75)3, Japan), each dissolved in 10 Ìl phosphate-buffered saline and corresponding to a 25% reduction of each dimension (15). At the filtered through a 0.45-Ìm membrane filter (Millipore, Bedford, end of the experiments, all the tumors were weighed in mg, mean MA, USA), was then added and the plates were incubated for an and standard deviation were calculated, and the differences between additional 3 hours at 37ÆC. After the final incubation, 150 Ìl control and treated groups were assessed by Student’s t-test. dimethyl sulfoxide (Nacalai Tesuque, Kyoto, Japan) was added to each well to dissolve the MTT-formazan salt and the plates were Collagen-droplet drug sensitivity test. The collagen gel droplet- shaken mechanically for 10 minutes on a mixer (Model 250, embedded culture drug sensitivity test (CD-DST) was performed Sonifier, Branson, MO, USA). The optical densities of each well as reported previously (16). Briefly, "Cellmatrix" type CD, Ham’s were determined on a model EAR 340 easy reader (SLT- F-12 medium at 10-fold concentration, and a reconstitution buffer Labinstruments, Salzburg, Austria) at 540 nm and 630 nm. (all from Nitta Gelatin Inc., Yao City, Osaka, Japan) were mixed Inhibition rates were calculated using the following formula: at a ratio of 8:1:1 to homogeneity in an ice-cold bath and the (1 - A/B) x 100 (percentage), where A and B represent the mean collagen solution was prepared. The cell suspension prepared absorbance of the treated and control wells, respectively. above was added to this solution to a final density of 1 x 105

2202 Hashiguchi et al: TZT-1027 Against MCF-7 and R-27 cells/ml, affording a collagen mixture. This collagen mixture was Results placed into 6-well multi-plates (Nalge-NUNC Inc., Rochester, NY, USA) at a volume of 30 Ìl per collagen gel droplet using a The antitumor activity of TZT-1027 against MCF-7 and R-27 micropipette, and subjected to gelation in a CO2 incubator at cells is shown in Figure 2. TZT-1027 showed a 37ÆC for 1 hour. In each well, 3 ml of DF medium (DF (10)), to concentration-dependent cytotoxicity against MCF-7 cells which 10% FBS had been added, was overlaid. After overnight incubation, TZT-1027 was added at different concentrations after 48 and 72 hours of incubation. A time-dependent effect (0.0001-20 Ìg/ml) and docetaxel was added from 0.01 to10 Ìg/ml. was not obvious in this experiment. In contrast, R-27 was After 24 hours, each well was washed with Hanks’ solution twice, resistant to TZT-1027 and no 50% inhibitory concentrations 4 ml/well of PCM-2 (with FBS from removed PCM-1) medium was were obtained for either the 48- or 72-hour incubation until overlaid and the cells were cultured for 7 days. On the 8th day of the highest concentration tested (20 ng/ml), which is culture, 40 Ìg/well of neutral red solution (5 mg/ml) was added extremely high. and incubation was continued for 2 hours. After removal of the The in vivo antitumor activity of TZT-1027 is shown in solution, the cells were fixed with 10% formalin buffered at neutral pH. The plates were immersed in a tray of water for 10 Figure 3A (MCF-7) and Figure 3B (R-27). Both strains were minutes without agitation and then air-dried and subjected to sensitive to TZT-1027. The lowest T/C value and RW of analysis. To obtain the total volume quantification, binary images MCF-7 were 25.5% and 0.312, respectively and there was a of proliferating cells in collagen gel drops were obtained with an statistically significant difference between the actual tumor image analyzing apparatus (VIDAS-plus, Carl Zeiss Inc., Tokyo, weights in the control (249.2±151.0 mg) and treated Japan), fibroblasts on the image were selectively eliminated and (96.8±53.1 mg) groups (p<0.002). R-27 was also highly the volume of live cancer cell colonies was calculated. sensitive to TZT-1027 and the lowest T/C value and RW Imaging colorimetric quantification method. Using a measuring were 28.6% and 0.8, respectively. There was also a statistically apparatus consisting of a personal computer and a video significant difference between the control (647±301.7 mg) microscope, we measured the image density. First, an image not and treated (247.4±131.8 mg) tumors (p<0.05). containing cells as a blank and an image with the light covered When the control tumors were used for CD-DST, only were obtained with the video microscope (VH-5910, KEYENCE, MCF-7 was sensitive to TZT-1027, with a 50% inhibitory Osaka), and the upper and lower limits were set. Next, an image of the cells in the air-dried collagen gel drops was obtained and a concentration of 3,400 pg/ml, as shown in Figure 4. On the binary image with a cut-off at a density level of 128 (density steps: other hand, R-27 was resistant to TZT-1027 in CD-DST, 0-255, 8 bits) was obtained. The cumulative density on the binary where no 50% inhibitory concentration was observed until image was calculated and the image optical density (A) was 20 Ìg/ml, while the control cells grew until they reached a obtained according to the following formula: density 2.3 times that of the initial seeding (Figure 5A). In A = log10{(™WC-™BC)/(™TC-™BC)} contrast, docetaxel was cytotoxic against R-27, as detected where ™WC is the cumulative density value of the blank image, by CD-DST at a concentration of 10 Ìg/ml (Figure 5B). ™BC is the cumulative density value with the light covered and ™TC is the cumulative density of the test sample image. For this Figure 6 shows the Evans blue contents of R-27 tumors processing, a personal computer (Apple Power Macintosh G-3) treated with TZT-1027 or docetaxel, representing the with grayscale image digitizer (LG-3 Image ONE, Tokyo) and vascular blood flow of the tumors. TZT-1027 blocked tumor "Primage" (Nitta Gelatin Inc.), a modification of the NIH Image blood flow at 1 hour after TZT-1027-injection, when the macroprogram, were used. Evans blue content decreased to almost half of that of the Anti-vascular activity of tzt-1027 against R-27 tumors. The anti- untreated control in both the 1- and 2-mg/kg TZT-1027 vascular activity of TZT-1027 against R-27 tumors was measured groups. This down-regulation continued until 24 hours after according to the method of Otani and colleagues (17). injection in both groups, while the 2 mg/kg TZT-1027 group Tumor fragments (2 x 2 x 2 mm in size) of R-27 were implanted had continuously suppressed blood flow until 8 mg at 6 hours subcutaneously into the right flank of female BALB/c nude mice and 9 mg at 24 hours of Evans blue per tumor weight in g supplemented with 5 mg/kg 17‚-estradiol dipropionate and 250 mg/kg after injection. In the 1-mg/kg TZT-1027 group, the Evans 17·-progesterone caproate. When the estimated tumor weight blue contents in the tumor were 14 mg/g at 6 and 24 hours reached 100-300 mg, TZT-1027 and docetaxel were administered intravenously at doses of 1 and 2 mg/kg for TZT-1027 and 40 mg/kg after TZT-1027 injection. On the other hand, the suppression for docetaxel. At 0, 1, 6 and 24 hours after drug injection, 0.2% of blood flow by docetaxel was minimal, where the Evans Evans blue was given intravenously at 0.2 ml/head and the mice blue content of the tumor was 16 and 18 mg/g tumor at 6 and were killed 2 minutes after Evans blue injection. Immediately after 24 hours after TZT-1027 injection, respectively. the mice had been sacrificed, tumors were isolated and frozen in liquid nitrogen. The tumor fraction was suspended in 1-3 ml of Discussion formamide and incubated for 48 hours at 60ÆC to extract the Evans blue (18). Then, the suspension was filtered on a 0.5-Ìm membrane filter and the absorbance at 620 nm (reference at 690 nm) Dolastatin 10 was initially isolated from the Indian Ocean was measured. The intra-tumoral amount of Evans blue was sea hare, Dolabella auricularia, and TZT-1027 was newly calculated as mg/g. synthesized as a derivative of dolastatin 10 with potent

2203 ANTICANCER RESEARCH 24: 2201-2208 (2004)

antitumor agents (8, 15, 20, 21). Figure 7 shows the chemo- sensitive pattern of MCF-7 (Figure 7A) and R-27 (Figure 7B) to several agents tested by their maximum tolerated doses. MCF-7 was sensitive to mitomycin C (MMC), doxorubicin (DXR) and UCN-01, and the effect of TZT-1027 followed from that of MMC and UCN-01, while its efficacy was more potent than that of DXR, UFT, cyclophosphamide (CPA) and tamoxifen (TAM), which are the standard drugs used to treat breast carcinoma. R-27 was more sensitive to MMC, CPA and TZT-1027, than to paclitaxel (PAC), DXR, hexamethylmelamine and UFT. This result was comparable to that in the report by Fujita and coworkers (22) who evaluated the antitumor effects of TZT-1027 against 16 human tumors xenografted in nude mice from gastric, breast, colon, lung, liver, renal cell and ovarian cancer lines. In their study, TZT-1027 was shown to be more potent in three cancer models than irinotecan, and to have markedly effective antitumor activity in two cancers in which vincristine was ineffective and in ovarian cancer in which irinotecan, cisplatin Figure 2. Antitumor activity of TZT-1027 on R-27 &MCF-7 in vitro and DXR were ineffective. From these results, they suggested detected by MTT assay. TZT-1027 showed a concentration-dependent cytotoxicity against MCF-7 that TZT-1027 was an excellent candidate for clinical trials for cells after 48 and 72 hours of incubation. In contrast, R-27 was resistant the treatment of cancer. The direct cytotoxicity of TZT-1027 to TZT-1027, and no 50% inhibitory concentrations were obtained for has been evaluated in several ways. Its main target is either the 48- or 72-hour incubation until the highest concentration tested microtubulin polymerization, as for vinca alkaloids, including (20 ng/ml). vinblastine (VLB), vincristine (VCR) and vindesine (VDS). Since TZT-1027 at 10 ÌM almost completely inhibited the assembly of porcine brain microtubules, its mechanism of antitumor action seems to be, at least in part, ascribable to the antitumor activity. Kobayashi and colleagues have evaluated inhibition of microtubule assembly (19). Natsume and its antitumor activity against a variety of murine tumors, coworkers have compared the activity of TZT-1027 with that including P388, B16 melanoma and colon 26 (19). The of dolastatin 10 as well as these vinca alkaloids (23). In their antitumor activities of TZT-1027 against these tumors were study, Scatchard analysis of the binding of 3H-VLB suggested superior or comparable to those of the reference agents, one binding site, while that for 3H-TZT-1027 suggested two dolastatin 10, cisplatin, vincristine, 5-fluorouracil (5-FU) binding sites, one of high affinity and one of low affinity. 3H- and E7010. In the experiments with drug-resistant P388 TZT-1027 was completely displaced by dolastatin 10 but only leukemia, TZT-1027 showed excellent antitumor activity incompletely by VLB. TZT-1027 affected the binding of VLB against cisplatin-resistant P388 and moderate activity to tubulin, but its binding site was not completely identical to against vincristine- and 5-fluorouracil-resistant P388, but no that of VLB. These results suggested that TZT-1027 had a activity against adriamycin-resistant P388. TZT-1027 was potent inhibitory effect on tubulin polymerization and differed also effective against human xenografts, where tumor from vinca alkaloids in its mode of action against tubulin regression was observed in MX-1 and LX-1 strains(19). polymerization. Watanabe and colleagues (23) have used In the present study, MCF-7 was sensitive to TZT-1027 in three cancer cell lines, human leukemia HL-60, human breast vitro, as detected by MTT assay and CD-DST, and in vivo in cancer MCF-7 and human prostate cancer DU145, and nude mice, while R-27 was resistant to TZT-1027 in vitro by reported that TZT-1027 induced DNA fragmentation. The both MTT assay and CD-DST. Although R-27 was flow-cytometric analysis revealed that the cells treated with established as a tamoxifen-resistant subline of MCF-7, its TZT-1027 were arrested at the G2/M-phase and subsequently sensitivity to the conventionally available agents is known to showed fragmented DNA smaller than that of G1-phase cells. be similar to that of MCF-7, compared with human breast From these results, they suggested an association of the carcinoma xenografts tested in our institute. Its different growth-inhibitory effect of TZT-1027 with the induction of characteristics in vivo are less dependent on estradiol and less apoptosis, which was followed by G2/M arrest. Natsume and sensitive to hormonal agents compared with the wild-type, colleagues (24) have investigated the sensitivity of surgical MCF-7. We have previously reported the chemosensitivity of specimens obtained from patients with non-small cell lung MCF-7 and R-27 to conventionally available and new carcinoma (NSCLC) and renal cell carcinoma (RCC) to TZT-

2204 Hashiguchi et al: TZT-1027 Against MCF-7 and R-27

Figure 3. A. Antitumor activity of TZT-1027 on MCF-7 transplanted in nude mouse. The lowest T/C value and RW of MCF-7 were 25.5% and 0.312, respectively, and there was a statistically significant difference between the actual tumor weights in the control and treated groups (p<0.002). B. Antitumor activity of TZT-1027 on R-27 transplanted in nude mouse. R-27 was sensitive to TZT-1027 and the lowest T/C value and RW were 28.6% and 0.8, respectively. There was also a statistically significant difference between the control and treated tumors (p<0.05).

1027 and other DNA-damaging agents, comparing the status when detected by CD-DST using the primary cultured R-27 of the p53 gene in those specimens. When 29 NSCLC that was sensitive to TZT-1027 in vivo. Since the control specimens and 22 RCC specimens were analyzed for their cells grew until they reached a density 2.3 times that of the sensitivity by flowcytometry and their p53 status was initial seeding, this CD-DST was successfully conducted and compared, TZT-1027 was less influenced by the p53 status of actually docetaxel was evaluated to be effective against R- specimens than the DNA-damaging agents. 27 tumor cells in this assay. On the other hand, MCF-7 was In our present study, R-27 was sensitive to TZT-1027 in also sensitive to TZT-1027 in CD-DST, which reproduced vivo with a T/C ratio of 28.6%, similar to that of MCF-7 at the cytotoxicity of this drug on MCF-7 in nude mice. This 30.0%, although R-27 was highly resistant to TZT-1027 discrepancy between the in vitro and in vivo sensitivity of

2205 ANTICANCER RESEARCH 24: 2201-2208 (2004)

Figure 4. Antitumor activity of TZT-1027 on MCF-7 detected by CD- DST. The primary cultured MCF-7 cells were obtained from xenografts in nude mouse to be assessed by the collagen-gel droplet drug sensitivity test (CD-DST). MCF-7 was sensitive to TZT-1027, with a 50% inhibitory concentration of 3.4 ng/ml. Data indicate mean±S.D.

Figure 6. Vascular blood flow of R-27 tumors treated with TZT-1027 or docetaxel detected by Evans Blue content. TZT-1027 blocked tumor blood flow at 1 hour after TZT-1027-injection, which continued until 24 hours after injection in 1 and 2 mg/kg TZT-1027 groups, while the suppression of blood flow by docetaxel was minimal. Data indicate mean±S.D.

R-27 was considered to be due to the indirect cytotoxicity of TZT-1027, which is mediated through an anti-vascular effect, the cytotoxicity on tumor vascular endothelial cells as reported by Otani and colleagues (17). They evaluated TZT-1027-induced tumor vascular collapse and tumor cell death in an advanced tumor model, murine colon 26 adenocarcinoma. Tolerable doses of TZT-1027 induced tumor-selective hemorrhage within 1 hour, which occurred mainly in the peripheral area of the tumor mass. They measured tumor hemoglobin content and dye permeation detected by Evans blue. When 2% Evans blue is injected, Evans blue molecules will bind to plasma albumin and behave like albumin (18). In their study, it was revealed that the hemorrhage occurred first and tumor blood flow stopped secondarily. They also studied the effects of TZT-1027 on cultured human umbilical vascular endothelial cells, where TZT-1027 induced marked cell contraction with membrane blebbing in 30 minutes, which was completely inhibited by a broad-spectrum inhibitor of protein kinases, K252a. As a result, they concluded that TZT-1027 quickly attacked the well-developed vascular system of advanced tumors, resulting in the blockade of tumor blood flow. These results may explain the indirect cytotoxicity of TZT-1027 against R-27 through vascular blockade detected by Evans blue Figure 5. Antitumor activity of TZT-1027 and docetaxel on R-27 detected assay. It was interesting that docetaxel did not exert this by CD-DST. The primary cultured R-27 cells were obtained from anti-vascular effect, while it was effective against R-27 as xenografts in nude mouse to be assessed by the collagen-gel droplet drug sensitivity test (CD-DST). R-27 was resistant to TZT-1027 in CD-DST, detected by CD-DST. where no 50% inhibitory concentration was observed until 20 Ìg/ml, while In conclusion, TZT-1027 exerts its antitumor activity through docetaxel was cytotoxic against R-27. Data indicate mean±S.D. direct cytotoxicity for MCF-7, as well as via selective blockade

2206 Hashiguchi et al: TZT-1027 Against MCF-7 and R-27

Figure 7. A. Sensitivity pattern of MCF-7 in vivo. MCF-7 was sensitive to mitomycin C (MMC), doxorubicin (DXR) and UCN-01 and the effect of TZT-1027 followed from that of MMC and UCN-01, while its efficacy was more potent than that of DXR, 5-FU, cyclophosphamide (CPA) and tamoxifen (TAM). B. Sensitivity pattern of R-27 in vivo. R-27 was more sensitive to MMC, CPA and TZT-1027 than to paclitaxel (PAC), DXR, hexamethylmelamine (HMM) and UFT (a mixed compound of tegafur and uracil at a molar ratio of 1:4).

2207 ANTICANCER RESEARCH 24: 2201-2208 (2004) of tumor blood flow for R-27, resulting in its remarkable 14 Geran RI, Greenberg NH, Macdonald MM, Schumacher AH and antitumor activity against both strains in vivo. TZT-1027 has Abbott BJ: Protocols for screening chemical agents and natural potent antitumor activity against human breast carcinoma in products against animal tumors and other biological systems (Third Edition). Cancer Chemother Rep (Part 3) 3: 51-61, 1972. vivo compared with conventionally available agents for breast 15 Kubota T, Nakada M, Tsuyuki K, Inada T, Asanuma F, Ishibiki cancer, including DXR, CPA, 5-FU and PAC. Further K and Abe O: Cell kinetics and chemosensitivity of human investigation of TZT-1027 is warranted, in particular its unique carcinomas serially transplanted into nude mice. Jpn J Cancer mode of action. Because of its promising preclinical activity, Res 77: 502-507, 1986. TZT-1027 has been entered into phase I clinical trials and the 16 Kobayashi H, Higashiyama M, Minamigawa K, Tanisaka K, recommended dose for phase II study was reported to be 1.8 Takano T, Yokouchi H, Kodama K and Hata T: Examination mg/m2 in a weekly regimen for 3 weeks with the dose-limiting of in vitro chemosensitivity test using collagen gel droplet toxicities being neutropenia and constipation (25). culture method with colorimetric endpoint quantification. Jpn J Cancer Res 92: 203-210, 2001. 17 Otani M, Natsume T, Watanabe JI, Kobayashi M, Murakoshi References M, Mikami T and Nakayama T: TZT-1027, an antimicrotubule agent, attacks tumor vasculature and induces tumor cell death. 1 Miyaszaki K, Kobayashi M, Natsume T et al: Synthesis and Jpn J Cancer Res 91: 837-844, 2000. antitumor activity of novel dolastatin 10 analogs. Chem Pharm 18 Matsumura Y and Maeda H: A new concept for macromolecular Bull 43: 1706-1718, 1995. therapeutics in cancer chemotherapy: mechanism of tumoritropic 2 Pettit GR, Kamano Y, Herald CL et al: The isolation and structure accumulation of proteins and the antitumor agent smancs. of a remarkable marine animal antineoplastic constituent: Cancer Res 46: 6387-6393, 1986. Dolastatin 10. J Am Chem Soc 109: 6883-6885, 1987. 19 Kobayashi M, Natsume T, Tamaoki S, Watanabe J, Asano H, 3 Kobayashi M, Natsume T, Tamaoki S et al: Antitumor activity Mikami T, Miyasaka K, Miyazaki K, Gondo M, Sakakibara K of TZT-1027, a novel dolastatin 10 derivative. Jpn J Cancer Res and Tsukagoshi S: Antitumor activity of TZT-1027, a novel 88: 316-327, 1997. dolastatin 10 derivative. Jpn J Cancer Res 88: 316-327, 1997. 4 Watanabe J, Natsume T, Fujio N, Miyasaka K and Kobayashi 20 Koh J, Kubota T, Migita T, Abe S, Hashimoto M, Hosoda Y M: Induction of apoptosis in human cancer cells by TZT-1027, and Kitajima M: UCN-01 (7-hydroxystaurosporine) inhibits the an antimicrotubule agent. Apoptosis 5: 345-353, 2000. growth of human cancer xenografts through disruption of signal 5 Soule HD, Varguez J, Long A, Albert S and Brennan M: transduction. Breast Cancer 9: 50-54, 2002. Human cell line from pleural effusion derived from a breast 21 Kubota T, Matsuzaki WS, Hoshiya Y, Watanabe M, Kitajima carcinoma. J Natl Cancer Inst 51: 1409-1416, 1973. M, Asanuma F, Yamada Y and Koh J: Antitumor activity of 6 Kubota T, Kubouchi K, Koh J, Enomoto K, Ishibiki K and Abe paclitaxel against human breast carcinoma xenografts serially O: Human breast carcinoma (MCF-7) serially transplanted into transplanted into nude mice. J Surg Oncol 64: 115-121, 1997. nude mice. Jpn J Surg 13: 381-383, 1983. 22 Fujita F, Koike M, Fujita M, Sakamoto Y and Tsukagoshi S: 7 Nawata H, Bronzert D and Lippman ME: Isolation and Antitumor effects of TZT-1027, a novel dolastatin 10 derivative, characterization of a tamoxifen-resistant cell line derived from MCF- on human tumor xenografts in nude mice. Jpn J Cancer 7 human breast cancer cells. J Biol Chem 256: 5016-5021, 1981. Chemother 27: 451-458, 2000. (in Japanese with English 8 Kubota T, Ishibiki K and Abe O: The clinical usefulness of human abstract) xenografts in nude mice. Prog Clin Biol Res 276: 213-225, 1988. 23 Natsume T, Watanabe J, Tamaoki S, Fujio N, Miyasaka K and 9 Mosmann T: Rapid colorimetric assay for cellular growth and Kobayashi M: Characterization of the interaction of TZT-1027, survival: application to proliferation and cytotoxicity assays. J a potent antitumor agent, with tubulin. Jpn J Cancer Res 91: Immunol Methods 65: 55-63, 1983. 737-747, 2000. 10 Shimoyama Y, Kubota T, Watanabe M, Ishibiki K and Abe O: 24 Natsume T, Kobayashi M and Fujimoto S: Association of p53 Predictability of in vivo chemosensitivity by in vitro MTT assay with gene mutations with sensitivity to TZT-1027 in patients with reference to the clonogenic assay. J Surg Oncol 41: 12-18, 1989. clinical lung and renal carcinoma. Cancer 92: 386-394, 2001. 11 Suto A, Kubota T, Shimoyama Y, Ishibiki K and Abe O: MTT 25 Yamamoto N, Andoh M, Kawahara M, Fukuoka M and Niitani assay with reference to the clinical effect of chemotherapy. J H: Phase I study of TZT-1027, and inhibitor of tubulin Surg Oncol 42: 28-32, 1989. polymerization, given weekly x 3 as a 1-hour intravenous 12 Fujita K, Kubota T, Matsuzaki SW, Otani Y, Watanabe M. infusion in patients with solid tumors. The 7th International Teramoto T, Kumai K and Kitajima M: Further evidence for Symposium on Cancer Chemotherapy "New Antitumor Agents the value of the chemosensitivity test in deciding appropriate under Development in US, Europe and Japan". pp. 26, 2002. chemotherapy for advanced gastric cancer. Anticancer Res 18: 1973-1978, 1998. 13 Abe S, Kubota T, Matsuzaki SW, Otani Y, Watanabe M, Teramoto T, Kumai K and Kitajima M: Chemosensitivity test is useful in evaluating the appropriate adjuvant cancer Received October 20, 2003 chemotherapy for Stage III non-scirrhous gastric cancer. Revised February 11, 2004 Anticancer Res 19: 4581-4586, 1999. Accepted April 5, 2004

2208