Activity of Mitozolomide (NSC 353451), a New Imidazotetrazine, Against Xenografts from Human Melanomas, Sarcomas, and Lung and Colon Carcinomas

(CANCER RESEARCH 45, 1778-1786, April 1985]

Activity of Mitozolomide (NSC 353451), a New Imidazotetrazine, against Xenografts from Human Melanomas, Sarcomas, and Lung and Colon Carcinomas

Oy stein Fodstad, ' Steina r Aamdal,2 Alexander Pihl, and Michael R. Boyd

Norsk Hydros Institute for Cancer Research, Montebello, Oslo 3, Norway [0. F., S. Aa., A. P.] and Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute, NIH, Bethesda, Maryland 20205 [0. F., M. R. B.Â¡

ABSTRACT In this investigation, we first tested the anticancer activity of mitozolomide against cells from different xenografted cancers in The chemosensitivity of human tumor xenografts to mitozo- a HTCF assay in vitro. When a pronounced inhibition of colony lomide, 8-carbamoyl-3-{2-chloroethyl)imidazo[5-1 -d]-1,2,3,5-te- formation was observed, we next examined in the HTCF assay trazin-4(3H)-one, was studied in 3 different assay systems. In the efficiency of the drug on a panel of tumors for each histolog concentrations of 1 to 500 Â¿/g/ml,mitozolomide completely inhib ical type. Since in vitro test systems have inherent limitations ited the colony-forming ability in soft agar of cell suspensions (23), we also measured the in vivo effect of mitozolomide on the from sarcomas, melanomas, lung and colon cancers, and a same tumors, using the 6-day subrenal capsule assay in immu- mammary carcinoma. When a panel of tumors of the different nocompetent mice (1-3), as well as s.c. growing tumors in histological types was tested for its sensitivity to mitozolomide athymic, nude mice (5-7, 10, 16, 18). In all 3 assay systems, in vitro, in the 6-day subrenal capsule assay in conventional mice, marked anticancer activity against human melanomas, sarco and, in some cases, as s.c. growing tumors in nude mice, good mas, lung cancers, and colon cancers was observed. agreement between the different assay systems was seen. In most cases, a very pronounced antitumor effect was observed. MATERIALS AND METHODS The efficacy of mitozolomide was as good or better than that of the drugs clinically used against the tumor types tested. Tumor Animals size measurements and histological examinations indicated that nude mice carrying a melanoma, a small cell lung cancer, and an Male C57BL/6 x DBA/2 F, (hereafter called B6D2F,) mice, 6 to 12 osteosarcoma were cured of their tumors. The approach here weeks old, were used for the subrenal capsule assay. BALB/c athymic, used for evaluating the effect of a new drug on human cancers nude mice were purchased from the Laboratory Breeding and Research may be useful for selecting the tumor types which primarily Center, Gl, Bomholtgaard, Ry, Denmark. NMRI nude mice were bred in our own nude mouse facility. The nude mice were housed in laminar-air should be studied in clinical trials. The results indicate that clinical flow rooms at constant temperature (24-26 Â°C)and humidity (30 to responses to mitozolomide may be anticipated in sarcoma, mel 50%). The cages and bedding for the nude mice were sterilized by anoma, small cell lung cancer, and possibly in colon cancer. autoclaving, and the food by 7-irradiation.

Tumors INTRODUCTION Altogether, 18 human tumor lines were used, all of which were In a systematic study of imidazotetrazines, a novel agent, maintained as s.c. growing xenografts in athymic, nude mice (NMRI or mitozolomide3 (previously Azolastone; NSC 353451 ), with potent BALB/c strain). Two lung cancer lines (H-249 and N-417) were kindly activity against murine tumors, has been discovered by the provided by Desmond N. Carney, National Cancer Institute-Navy Medical Cancer Research Campaign Experimental Chemotherapy Group, Oncology Branch, Bethesda, MD. One lung adenocarcinoma, T 402, was University of Aston, Birmingham, United Kingdom (8, 9, 22). The obtained from Nathan 0. Kaplan, University of California, San Diego, CA. mechanism of action of mitozolomide has not yet been eluci Of the 2 colon cancers, Co-115 was obtained from B. Sordat, Epalinges/ dated, but evidence has been presented that the formation of Lausanne, Switzerland, and one, WiDr, as well as the mammary cancer interstrand DNA cross-links may be involved (8, 9,12). line MDA-MB 231, were obtained from American Type Culture Collection, Rockville, MD. The other tumors (5 melanoma, 2 osteosarcoma, 3 soft In view of the promising effects of mitozolomide in murine tissue sarcoma, and 2 lung cancer lines) were established in this labo systems (11, 22), we initiated experiments to see whether mito ratory from tumor biopsies of patients at the Norwegian Radium Hospital. zolomide might be active also against human cancers and, if so, to establish which of the tumor types examined were most Drugs sensitive, to aid in the planning of subsequent clinical trials. Mitozolomide (Chart 1) was a gift from M. G. Stevens, Cancer Re ' To whom requests for reprints should be addressed, at Norsk Hydros Institute search Campaign, Experimental Chemotherapy Group, Department of for Cancer Research, The Norwegian Radium Hospital, Montebello, 0310 Oslo 3, Pharmacy, University of Aston in Birmingham, United Kingdom. Abrin Norway 2 Fellow of the Norwegian Cancer Society. was prepared in our laboratory (15). The other drugs were commercial 'The abbreviations used are: mitozolomide, 8-carbamoyl-3-(2-chloro- preparations obtained from the Department of Pharmacy, The Norwegian ethyl)imidazo|5.1-ci]-1.2.3,5-tetrazin-4(3H)-one (also designated NSC 353451, and Radium Hospital. MSB 39565); CCNU, 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea; CDDP, c/s-di- Mitozolomide was dissolved in a minimal amount of dimethyl sulfoxide chlorodiammineplatinum (II); ADM, doxorubicin; HTCF, human tumor colony-form and kept refrigerated in glass tubes. Immediately before use, the drug ing; SRC, subrenal capsule; FVR, fractional volume reduction; SCLC, small cell lung carcinoma. was diluted in 0.9% NaCI solution (saline) to appropriate concentrations Received 10/5/84; revised 12/26/84; accepted 12/28/84. and injected into the animals in volumes of 0.2 ml. CCNU was dissolved

CANCER RESEARCH VOL. 45 APRIL 1985 1778

CONH, Xenografts in Athymic Mice. The nude mouse experiments were carried out as described previously (5, 6). Mice with progressively grow ing s.c. xenografts were divided into groups of 8 to 10 animals each. N When the tumors had reached a mean diameter of 6 to 8 mm, the animals were treated with the drugs in the doses indicated in the legends to charts. The maximum tolerable dose of mitozolomide, given i.v. 3 to 4 times on a once-weekly schedule, was found to be 40 mg/kg/dose. The tumor size was assessed by measuring twice weekly 2 perpen Chart 1. Molecular formula of mitozolomide. dicular diameters. By subtracting, for each diameter, the value at the start of treatment, the mean tumor diameter differences (Â±SE), were in Cremophor EL (Sigma) and injected i.v. into B6D2F, mice and i.p. into calculated (6). The results are presented as growth curves where the nude mice, whereas the other drugs were used as aqueous solutions mean tumor diameter difference for each treated and untreated group is and given i.v. in volumes of 0.2 to 0.4 ml. plotted versus time. At the start of treatment this value is by definition zero. Assay of Anticancer Activity The effect of treatment was expressed numerically as the FVR (21), defined by the expression: HTCF Assay. Human tumors serially transplanted in athymic, nude mice were mechanically disaggregated by the use of a Stomacher 80 mean Vc FVR = logio Lab-Blender (Seward Surgical, London, United Kingdom). For the sar mean V, comas, an enzymatic procedure (digestion with collagenase, 1 mg/ml, and 1000 IU hyaluronidase, 37 Â°Cfor 1 h) was included before the Vc and V, are the relative tumor volumes in the control group and the mechanical disaggregation. The cell suspensions were filtered through a treated group, respectively, at a given time (usually the time when the 4- to 5-^m nylon mesh, centrifuged, and resuspended in serum-contain tumor diameter difference between control and treated tumors was ing Ham's F-10 medium. The cultivation in soft agar and the chemosen- maximal). FVR gives the number of decades by which the treated tumor sitivity testing were performed by a modification of the procedure of volume is reduced below the control value. Thus, e.g., a FVR value of Courtenay and Mills (4) as described previously (23). The cells were 1.0 implies that the mean volume of the treated tumors is reduced to incubated for 1 h with 4 different drug concentrations. Triplicate cultures one-tenth of that of the control tumors (21). The tumor volume was were prepared in culture tubes containing rat RBC, and the tubes were calculated according to the formula: incubated in an atmosphere containing 5% O2, 5% CO2, and 90% N2. After incubation for 5 days, 2 ml of serum-containing medium were added 7T/6(MDf to each tube, and the incubation was continued for another 1 to 3 weeks. Colonies of more than 30 cells were scored by means of a Zeiss stereo where MD is the mean tumor diameter (6). microscope. The plating efficiency was defined as the number of colonies formed as a percentage of the number of viable cells plated. Four different concentrations of each compound, over a 3-log range, RESULTS were used. Since the 50% lethal dose value of mitozolomide after a single i.v. dose was found to be 80 mg/kg, i.e., 5 times that of ADM, we Initial Screening. The first examination of the sensitivity of used in the in vitro experiments mitozolomide concentrations that were different xenografts to mitozolomide involved testing in soft agar. 5 times the ADM concentrations previously found to give a convenient In Chart 2A, typical results are presented for a SCLC, and in number of colonies. The colony formation in the treatment cultures was Chart 2B for a soft tissue sarcoma. In both cases, mitozolomide expressed as a percentage of colony formation in the untreated control in concentrations of 10 Â¿ig/mlwas able to completely inhibit cultures. colony formation. In the case of the SCLC cells, the dose- SRC Assay. The assay was carried out as described previously (1, 2). Tumors grown s.c. (6 to 15 mm) were removed aseptically from nude mice and immediately placed in RPMI medium at room temperature. Viable tissue was dissected out and cut by scalpels into approximately 1-cu mm cubes which were implanted below the transparent kidney Â£ 100 e N-met hyIforma mide capsule. The tumor size was measured by means of a stereoscopic u VÂ«â€”Ifosfamtde microscope, calibrated in ocular units (OMU) (10 OMU = 1 mm). 80 The animals were divided into groups of 5 to 7 animals each, and in -Mitozolomide the treatment groups the mice were given the different drugs i.v. on Days 1 and 2. The doses, which are given in the legends to charts, were O based on toxicity studies using the same schedule. The maximum doses giving no toxic deaths and a weight loss less than 15% were used. For .. ADM â€” vâ€”COOP mitozolomide, this dose was 40 mg/kg/dose. 40 Six days after implantation of the tumor tissue, the animals were Â£ sacrificed, the size of the subrenal tumor was assessed in situ by jÂ»I . measuring 2 perpendicular diameters, and the mean tumor diameter different for each group was calculated. The antitumor activity is ex o pressed according to the following: 0.1 1 10 100 0.1 1 10 IN c- AD, Drug concentration (ug/ml) Antitumor activity Chart 2. Dose-response curves in vitro of tumor cells from the SCLC H-249 (A), AD, and the hemangiopericytoma DEEX (B). treated with different chemotherapeutic agents. Single-cell suspensions were exposed for 1 h at 37 Â°Cto increasing where Dc and D, represent the mean tumor diameter in the control and concentrations of the drugs, and the residual ability to form colonies in soft agar treated groups, respectively. was measured.

CANCER RESEARCH VOL. 45 APRIL 1985 1779

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. ACTIVITY OF MITOZOLOMIDE AGAINST HUMAN TUMORS response curves showed that mitozolomide inhibited colony only be used to predict the drug response in vivo if they have growth in lower concentrations than ADM, CDDP, and etoposide been carefully calibrated against in vivo results, as we have (VP-16), drugs commonly used in the treatment of lung cancers emphasized previously (23). However, the above results were (14). Another new drug, A/-methylformamide (NSC 3051) did not found sufficiently encouraging to warrant further exploration in inhibit colony formation of the SCLC cells in the dose range in vivo models. tested (5 //g to 5 mg/ml). The effect of mitozolomide on the Malignant Melanomas. When a panel of melanomas was sarcoma seemed to be comparable to those of CDDP and studied in the soft agar assay, it was found (Chart 3A) that the 5 melanomas tested varied greatly in their sensitivity to mitozo ifosfamide (Chart 2ÃŸ).In similar experiments with a malignant lomide. The drug could completely inhibit the in vitro growth of melanoma, as well as colon and breast carcinomas, substantial all the melanomas tested, and in one case, the drug concentra inhibition of colony formation by mitozolomide was found. tion required to inhibit colony formation to 50% of the control The above data are clearly only suggestive. In vitro data can was as low as 0.5 ^g/ml. The wide range of sensitivities, over several logs, suggests that the tumors tested may constitute a representative panel of human melanomas. The effect of mitozolomide on the same melanomas was then studied in the SRC assay and compared with the effects of 2 drugs commonly used in the clinic. In some cases, the sensitivity of the tumors was tested also in the athymic, nude mouse model. The results for the melanoma LOX show that the growth of the subrenal grafts (Chart 4, left) was clearly inhibited by CDDP, slightly more by CCNU, and that the strongest effect was ob tained with mitozolomide. When LOX was tested as a s.c. tumor in nude mice (Chart 4, right), mitozolomide and CCNU were clearly more active than CDDP, with fractional volume reduction values (on Day 31) of 1.95,1.74, and 0.96, respectively. Although 0.5 5 50 500 0.5 5 50 500 the relative efficacy of the 3 drugs was somewhat different in the 2 in vivo systems (Chart 4), the ranking of the drugs with Mitozolomide concentration (ug/ml) Chart 3. Dose-responsecurves in vitro of tumor cells prepared from 5 different respect to activity was the same. melanomaxenografts (A)and 5 lung cancer xenografts (B), treated with increasing In the experiment shown in Chart 4 (right), the control mice concentrations of mitozolomide for 1 h at 37 Â°C.Themelanomaxenografts were had to be killed on Day 31 after implantation because of the size THX (D), LOX (O), FEMX-I (â€¢),EMX(x). and KFX (A). The lung cancer xenografts included 3 SCLC cancers. H-249 (x). N-417 (O), and BVX (D), and the 2 adeno- of the tumors (>25 mm) and the poor general condition of the carcinomas SELSX (A) and T 402 (A). animals. In contrast, in the mitozolomide-treated animals the

Time after implantation (days) Chart 4. Chemosensitivity profiles for the malignant melanomaxenograft LOX, studied in the 6-day SRC-assayin immunocompetentmice (left), and in the s.c. nude mouse assay (right). The mice were divided into groups of 6 to 8 animalseach, and the different drugs were administeredi.v. at times indicated (arrows). The drug doses used in the SRC assay were: mitozolomide, 40 mg; CCNU,25 mg; and CDDP, 6 mg/kg/dose. In the s.c. assay, the doses were: mitozolomide, 30 mg; CCNU, 20 mg; and CDDP, 6 mg/kg/dose. The s.c. tumors were measured twice weekly. Tumor size is expressed as mean tumor diameter difference, being zero on the first day of treatment.

CANCER RESEARCH VOL. 45 APRIL 1985 1780

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. ACTIVITY OF MITOZOLOMIDE AGAINST HUMAN TUMORS tumors eventually disappeared and did not reappear during the observation period (52 days). After 2 to 3 doses of mitozolomide, the grafts assumed a yellow tinge, and the tumors flattened considerably. This implies that the difference in tumor volume between the mitozolomide- treated and the other groups was in fact larger than what was inferred from the growth curves, which are based on measure ments of 2 perpendicular diameters. Thus, the results indicated that in the case of mitozolomide treatment, tumor diameter measurements and FVR calculations may underestimate the effect. The strong growth-inhibiting effect of mitozolomide on mela nomas is illustrated in a representative picture, taken on Day 31 after implantation (Fig. 1) of a mitozolomide-treated and an untreated tumor-bearing athymic mouse. The diameters of the respective tumors were 5 to 7 mm and 25 to 30 mm. In sections of the s.c. tumor of a mitozolomide-treated nude mouse, no viable tumor cells were found (Fig. 3b). These had all been replaced by necrotic and fibrotic tissue. Histological sections of the corresponding subrenal capsule xenografts in immunocompetent mice, taken 4 days after treatment, showed large, swollen tumor cells with pyknotic nuclei after mitozolomide treatment (Fig. 2o), whereas in the untreated animals the subrenal graft (Fig. 2a) was morphologically similar to the s.c. graft from an untreated nude mouse (Fig. 3a). The results suggest that in this instance tumor regrowth probably would not have occurred. 50 Altogether, the results indicate that mitozolomide may be at least as active against malignant melanomas as CCNU, a drug Time after implantation (days) that is being used in the treatment of this disease in the clinic. Chart 5. Chemosensitivity profile for the SCLC xenograft H-249 in the s.c. nude Lung Cancers. In Chart 33, the ability of mitozolomide to mouse assay. The mice were divided into groups of 6 to 8 animals each, and in the treatment groups the drugs were administered i.v. at the times indicated inhibit growth of clonogenic cells of a panel of lung cancers is (arrows). The drug doses given were: mitozolomide, 30 mg; ADM, 6 mg; CDDP, 6 shown. As was the case with the malignant melanomas, the mg; and VP-16, 30 mg/kg/dose. The tumors were measured twice weekly, and the mean tumor diameter difference was calculated. various tumors differed widely in response to the drug, with the 2 adenocarcinomas and one SCLC tumor (BVX) being the least sensitive. This particular xenografted SCLC line was established contrast, ADM, CDDP, and etoposide (VP-16) had only minor from a patient who was resistant to a combination regimen that effects on tumor growth (FVR 0.19,0.22, and 0.07, respectively). included ADM. However, linear regresson analysis, performed on the individual The response of the lung cancer H-249, growing s.c. in athymic values at each time point for each group, demonstrated that also mice, is shown in Chart 5. The tumors of the mitozolomide- these 3 drugs had statistically significant growth-inhibiting effects treated mice disappeared almost completely (FVR a 2.78). In (P S 0.003). Sarcomas. In Chart 6 (left), the effect of mitozolomide on different sarcomas in vitro is demonstrated. Comparison with Chart 2 shows that the individual sarcomas differed to a lesser extent in their sensitivity to mitozolomide than the melanomas and the lung cancers. In the SRC assay (Chart 7, left), mitozolomide had a much greater growth-inhibiting effect on the osteosarcoma TPX than ADM and the protein synthesis inhibitor, abrin (17), which seemed to be equally effective. The concomitant studies in nude mice (Chart 7, right) showed that, after 2 doses of mitozolomide, the s.c. grafts started to decrease in size, and at Day 48 the FVR value was >0.80. In some of the animals, the tumors finally disappeared completely without recurring, whereas in the other mice of the group, the tumors were still decreasing slowly when the experiment had to be terminated as the animals started to die, for reasons unrelated to tumor growth or drug toxicity. Abrin Fig. 1. Effect of mitozolomide on a melanoma xenograft growing s.c. in an had a statistically significant (P = 0.02) growth-inhibiting effect athymic, nude mouse. The upper mouse had received 3 weekly doses of 30 mg/ (FVR = 0.09), whereas ADM had no effect (FVR = 0). kg/dose of mitozolomide, starting on Day 13 after tumor implantation. The lower animal is an untreated control. The picture was taken 31 days after tumor implan Like the malignant melanoma grafts, the sarcoma grafts as tation. sumed a yellow tinge when the mice had received 2 to 3

DISCUSSION

The results presented in this paper demonstrate that mitozo lomide, a new imidazotetrazine, with broad-spectrum anticancer activity in murine tumors4 (11, 22), was highly active against several xenografted human tumors, as measured in 3 different assay systems. Of particular significance is the finding that of the 6 different human tumor lines so far tested as tumors growing s.c. in immunodeficient mice, a system generally assumed to reflect the sensitivity of the parent tumors (5, 7, 10, 13, 16, 18, 20, 21), 5 showed a strong response to mitozolomide. It is well known that, in tumor-bearing animals, procedures 50 500 5 50 500 causing a strong weight loss may in themselves induce tumor growth retardation. In our study, control experiments showed that weight losses of the magnitude found in the drug-treated Mitozolomide concentration (u,g/ml) animals did not affect tumor growth. Chart 6. Dose-response curves in vitro of tumor cells prepared from 5 different The effect of mitozolomide on the responsive tumors appeared sarcoma xenografts (A), 2 colon cancer xenografts, and one mammary cancer xenograft (B). treated with increasing concentrations of mitozolomide for 1 h at early. Thus, after 2 to 3 courses of mitozolomide treatment, a 37 Â°C. The sarcomas included 2 osteosarcomas. OHX (A) and TPX (O); one yellowish appearance and flattening of the treated tumors was leiomyosarcoma, MTMX (D), one hemangiopericytoma. DEEX (A); and one malig nant schwannoma (x). The 2 colon cancers were Co-115 (x) and WiDr (O), and evident, and histological examination revealed that already at the mammary cancer was MDA-MB-231 (D). this stage the tumor cells were dead or dying. Three of the lines (a melanoma, a SCLC, and a sarcoma), eventually disappeared completely upon mitozolomide treatment, and none of these reappeared during the observation time of more than 60 days. These findings, together with the absence of tumor cells by in 75 -H histological examination, suggest that the treated mice may have D SRC sc E been cured. For comparison, 3 of the drugs most commonly I used in the clinic against SCLC gave only minor growth inhibition of the SCLC tumor tested here. Also, in the case of the sarcoma, 10 the other drugs tested were far less active than mitozolomide in the doses tested. 'Â»ABRIN In this study, the antitumor activity in vivo of the different drugs was compared at the highest doses not giving toxic deaths MITOZOLOMIDE or unacceptable weight loss. Dose-effect curves were deter mined in the case of the 6-day SRC assay (1), but not in nude O mouse experiments. However, since equitoxic doses were used, 4 6 24 30 40 50 Â£ under conditions similar to those used in the clinic, our data Time after implantation (days) presumably reflect the relative antitumor activity of the drugs Chart 7. Chemosensitivity profiles for the osteosarcoma xenograft TPX, studied tested. in the 6-day SRC-assay in immunocompetent mice (left), and in the s.c. nude Activity of mitozolomide against human tumors has not been mouse assay (right). The mice were divided into groups of 6 to 8 animals and demonstrated previously, with the exception of one human lung different drugs were administered at indicated times (arrows). The drug doses used cancer line (LX-1 ; National Cancer Institute screen), found to be in the SRC assay were: mitozolomide, 40 mg; ADM, 6 mg; and abrin, 250 ng/kg/ dose. In the s.c. assay the doses were: mitozolomide, 30 mg; ADM, 6 mg; and sensitive, when tested as a subrenal xenograft (22). The pro abrin, 250 ng/kg/dose. The s.c. tumors were measured twice weekly, and the nounced activity consistently observed here against several hu mean tumor diameter difference was calculated. man tumor forms for which there is currently no satisfactory chemotherapy, is obviously of considerable clinical interest. administrations of mitozolomide. Since the mitozolomide solution Mitozolomide is chemically similar to the nitrosoureas. Like administered to the animals is yellowish, the fact that the grafts these, it probably acts by causing DNA interstrand cross-link obtained this color, may reflect accumulation of the drug in the formation (8,9,12). In our human melanomas, it showed a similar tumor, as well as cell death and necrosis. pattern of antitumor activity as CCNU. In a Phase I trial, mito zolomide caused thrombocytopenia,4 similar to what is typical Colon and Breast Cancers. The effect in vitro of mitozolomide on 2 colon cancers and on one breast cancer was also tested for CCNU. In the light of these similarities, the spectrum of (Chart 6, right). Inhibition of colony formation was observed in activity in human tumors of these 2 drugs should be compared. the same dose range as for the other tumors. The in vivo results Such studies have been initiated in our laboratory. for one of the colon cancers (Chart 8) were consistent with In spite of large efforts, rather few new, clinically effective clinical experience, insofar as 5-fluorouracil was the most effec anticancer drugs have become available in recent years. Numer tive drug tested (FVR = 0.37). However, also mitozolomide and ous drugs are being produced and screened in murine test abrin showed clear growth-inhibiting effects with FVR values of systems (24), but many drugs that show promise in experimental systems prove to be ineffectual in patients. Conversely, screen- 0.25 and 0.20, respectively. For all 3 drugs, the effects were significant (P< 0.001). 4M. F. G. Stevens, personalcommunication.

CANCER RESEARCH VOL. 45 APRIL 1985 1782

"TÃ¡r .1 i. t .1 40 50 Time after implantation (days) Chart 8. Chemosensitivity profile for the colon cancer xenograft WiDr in the 6-day SRC assay in immunocompetent mice (/eft) and in the s.c. nude mouse assay (right). The mice were divided into groups of 6 to 8 animals,and in the treatment groups the drugs were administeredi.v. at the indicated times (arrows).The drug doses used in the SRC assay were: 5-fluorouracil, 80 mg; mitozolomide, 40 mg; and abrin, 250 ng/kg/dose. In the s.c. assay, the doses were: 5-fluorouracil, 100 mg; mitozolomide, 30 mg; and abrin, 350 ng/kg/dose. The tumors were measuredtwice weekly, and the mean tumor diameter differencewas calculated.

ing of new compounds on human tumor cells may possibly detect mouse and SRC models were observed in tumors that were anticancer activity that would not be discovered in murine tumor highly sensitive in the HTCF assay. It should be noted that some models (19). of the tumors not yet tested in vivo (melanomas, sarcomas, lung The strategy here used for preclinical evaluation of new anti- cancers, and a colon cancer) appeared to have high in vitro cancer drugs may be of some general interest. By this procedure, sensitivities (Charts 3 and 6), and may be expected to respond tumor types with high probability of clinical response to the drug to mitozolomide in vivo. may be selected. The effect of mitozolomide on human cancer The encouraging results in this study should expedite clinical cells was first demonstrated in a soft agar assay, used for evaluation of the antitumor activity of mitozolomide. They sug preliminary screening. The results suggested that mitozolomide gest that sarcomas, melanomas, and possibly colon cancers may be at least as effective as several established agents against should be studied in Phase II trials of mitozolomide. Among the the tumors studied. However, since experiments on single cell lung cancers, SCLC tumors seem to be the most promising suspensions do not reflect the importance of factors such as the candidates. If the tumor panels here tested are indeed repre pharmacokinetic properties of the drugs, their ability to penetrate sentative for their histological types, significant clinical responses into the tumor, and their toxicity to normal tissues, in vivo may be anticipated. experiments are necessary before definite conclusions can be drawn, and in our opinion promising agents should be evaluated ACKNOWLEDGMENTS in human tumor models in vivo before being introduced in the clinic. We therefore measured the effect of mitozolomide in vivo The excellent technical assistance of Unni Panning, Anne Pharo, Marianne Isaksen, Vivi Ann Ftorenes,and Ingrid Risan Mathisen is gratefully acknowledged. on the same tumors transplanted under the renal capsule of We thank Olav Kaalhusfor performing the statistical studies and Thorunn Kielland immunocompetent mice as well as when they were growing s.c. and Kirsten Bugge for typing the manuscript. in athymic, nude mice. Since the validity of the nude mouse model is well established REFERENCES (5, 7, 10, 13, 16, 18, 20, 21), it may be used as a reference 1. Aamdal, S., Fodstad, 0., and Pihl, A. The 6-day subrenal capsule assay for system. It is of interest from a methodological point of view that testing the response of human tumors to anticancer agents. Validity and the results obtained with the 2 in vivo systems were by and large usefulness in cancer researchand treatment. Ann. Chir. Gynecol.,(Suppl ). in in good agreement. Thus, when the effects of different drugs press, 1985. 2. Aamdal, S., Fodstad, 0, and Pihl, A. Human tumor xenografts transplanted were compared, almost without exception, the same ranking underthe renalcapsuleof conventionalmice. Growth rates and host response. order was obtained with the 2 models. The results provide strong Int. J. Cancer, 34: 725-730, 1984. evidence that the 6-day SRC assay in immunocompetent mice, 3. Bogden, A. E., and Von Hoff, D. D. Comparison of the human tumor cloning and subrenal capsule assays. Cancer Res., 44: 1087-1090, 1984. carried out on xenografted human tumors, can be used in 4. Courtenay, V. D., and Mills, J. An in vitro colony assay for human tumours evaluation of anticancer activity of new drugs. Moreover, good grown in immune-suppressedmice and treated in vivo with cytostatic agents. Br. J. Cancer, 37: 261-268, 1978. correlation between the in vitro and in vivo Chemosensitivity data 5. Fodstad. 0., Aass, N., and Pihl, A. Response to chemotherapy of human, was also seen. Thus, the pronounced responses in the nude malignant melanoma xenografts in athymic, nude mice. Int. J. Cancer, 25:

453-458. 1980. trial. Eur. J. Cancer Clin. Oncol., 20: 347-351. 1984. 6 Fodstad, 0., Aass. N , and Pihl, A. Assessment of tumour growth and of 15. Olsnes, S. Toxic and nontoxic lectins from Abrus precatorius. Methods En- response to chemotherapy of human melanomas in athymic, nude mice. Br. zymol., 50: 323-330, 1978. J Cancer. 4i(Suppl. IVy 146-149,1980. 16. Ovejera, A. A., Houchens, D. P., and Barker, A. D. Chemotherapy of human 7 Fodstad, 0., Olsnes. S.. and Pihl. A. Inhibitory effect of abrin and ricin on the tumor xenografts in genetically athymic mice. Am. Clin. Lab. Sci., 8: 50-56, growth of transplantable murine tumors and of abrin on human cancers in 1978. nude mice. Cancer Res., 37: 4559-4567, 1977. 17. Pihl, A., Fodstad, 0., and Olsnes, S. Cytostatic properties of the toxic lectins 8. Gibson. N. W., Erickson, L C., and Hickman. J. A. Effects of the antitumor abrin and ricin. In: K. Letnansky (ed.), Biology of the Cancer Cell, pp. 261- agent 8-carbamoyl-3-(2-chloroethyl)imidazo[5,1 -d]-1,2,3,5-tetrazin-4(3H)-one 267. Amsterdam: Kugler Medical Publications, 1980. on the DMA of mouse L1210 cells. Cancer Res.. 44: 1767-1771,1984. 18. Povlsen, C. D., and Jacobsen, G. K. Chemotherapy of a human malignant 9. Gibson. N. W., Hickman. J. A., and Erickson. L C. DNA cross-linking and melanoma in the nude mouse. Cancer Res., 35: 2790-2796, 1975. cytotoxicity in normal and transformed human cells treated in vitro with 8- 19. Shoemaker, R. H., Wolpert-Defilippes, M. K., Melnick, N. R., Venditti, J. M., carbamoyl-3-(2-chloroethyl)imidazo[5,1 -d]-1,2,3,5-tetrazin-4(3H)-one. Cancer Simon, R. M., Kern, D. H., Lieber, M. M., Miller, W. T., Salmon, S. E., and Von Res., 44: 1772-1775, 1984. Hoff, D. D. Recent results of new drug screening trials with a human tumor 10. Giovanella, B. C., Stehlin, J. S.. and Williams, L. J. Heterotransplantation of colony forming assay. In: S. E. Salmon and J. M. Trent (eds.). Human Tumor human malignant tumors in "nude" thymusless mice. II. Malignant tumor Cloning, pp. 345-357. Orlando. FL: GruÃ±eand Stratton, Inc., 1984. induced by injection of cell cultures derived from human solid tumors J. Nati. 20. Shorthouse, A. J., Smythe, J. F., Steel. G. G., Ellison, M., Mills, J., and Cancer Inst., 52. 921-930,1974. Peckham, M. J. The human tumor xenograftâ€”a valid model in experimental 11. Hickman. J. A., Gibson, N. W., Stone, R., Stevens, M. F. G., Lavelle, F.. and chemosensitivity? Br. J. Surg., 67: 715-722,1980. Fizames. C. M & B 39565; a novel heterocycle with potent antitumour activity 21. Steel, G. G., Courtenay, V. D., and Peckham, M. J. The response to chemo in mice. In: Proceedings of the Thirteenth International Cancer Congress, p. therapy of a variety of human tumour xenografts. Br. J. Cancer, 47: 1-13, 551. Geneva: Union International Contre Cancer, 1982. 1983. 12 Horgan. C.. Stevens. M. F. G.. and Tisdale. M. J. Preliminary investigations on 22. Stevens, M. F. G., Hickman, J. A., Stone, R., Gibson, N. W., and Baig, G. U. the mode of action of CCRG 81010 (M & B 39565). Br. J. Cancer, 48: 132. Antitumor imidazotetrazines. 1. Synthesis and chemistry of 8-carbamoyl-3-<2- 1983 chloroethyl)imidazo-[5,1-d]-1,2.3,5-tetrazin-4(3H)-one. a novel broad-spectrum 13. Kopper, L., and Steel. G. G. The therapeutic response of three human tumor antitumor agent. J. Med. Chem., 27: 196-201,1984. lines maintained in immune-suppressed mice. Cancer Res., 35: 2704-2713, 23. Tveit, K. M., Fodstad, 0., Olsnes, S., and Pihl, A. In vitro sensitivity of human 1975. melanoma xenografts to cytotoxic drugs. Correlation with in vivo chemosen 14. Mitrou. P. S., Graubner. M.. Berdel, W. E.. Mende, S., Gropp, C., Diehl, W., sitivity. Int. J. Cancer, 26. 717-722, 1980. and Khppstein, T. H. c/s-Platinum (DDP) and VP 16-213 (etoposide) combina 24. Venditti, J. M. Preclinical drug development: rationale and methods. Semin. tion chemotherapy for advanced non-small cell lung cancer. A phase II clinical Oncd., 8. 349-361,1981.

CANCER RESEARCH VOL. 45 APRIL 1985 1784 Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. Fig. 2. Effect of mitozolomide treatment on the histology of a malignant melanomaxenograft (LOX)growing under the renal capsule of conventional mice. Untreated controls (a) and after mitozolomide treatment (b). x 620.

CANCER RESEARCH VOL. 45 APRIL 1985 1785 Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. Fig. 3. Effect of mitozolomide treatment on the histology of a malignant melanoma xenograft (LOX) growing s.c. in athymic, nude mice. a. untreated controls, x 620. D, after mitozolomide treatment, x 155. Inset, x 620.

CANCER RESEARCH VOL. 45 APRIL 1985 1786

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1985 American Association for Cancer Research. Activity of Mitozolomide (NSC 353451), a New Imidazotetrazine, against Xenografts from Human Melanomas, Sarcomas, and Lung and Colon Carcinomas

Øystein Fodstad, Steinar Aamdal, Alexander Pihl, et al.

Cancer Res 1985;45:1778-1786.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/45/4/1778

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/45/4/1778. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.