[CANCER RESEARCH 44, 1688-1692, April 1984] Experimental and Clinical Activity of Mitomycin C and c/s- Diamminedichloroplatinum in Malignant Mesothelioma1 A. Philippe Chahinian,2 Larry Norton, James F. Holland, Lenina Szrajer, and Ronald D. Hart Department of Neoplastia Diseases, Mount Sinai School of Medicine of the City University of New York, New York, New York 10029 [A. P. C., L. N., J. F. H., L S.], and Department of Medicine, St. Luke's Hospital, Milwaukee, Wisconsin 53215 [R. D. H.] ABSTRACT taining Roswell Park Memorial Institute Medium 1640 (Grand Island Biological Co., Grand Island, NY) and trimmed of cutaneous, connective, Little information is available regarding the effectiveness of and adipose tissues under a fiberglass tissue culture hood (Fisher chemotherapeutic agents in malignant mesothelioma. Human Scientific Co., Pittsburgh, PA). Small tumor cubes measuring 1 to 2 mm malignant mesothelioma specimens from two patients were suc were inserted in a trocar and transplanted s.c. into 4 sites/mouse in the cessfully transplanted and maintained in homozygous nude mice. right and left groin and axilla, or into one site/mouse in the right axilla. Screening of chemotherapeutic agents revealed that cisplatin Since single-cell suspensions were not used, the number of cells was the most active single agent in one line, and mitomycin C in transplanted in these specimens can only be determined on histological the other. The combination of mitomycin C and cisplatin, how sections. For that purpose, tumor specimens were fixed in 10% neutral ever, was the most effective regimen for both lines of xeno- formalin and embedded in paraffin after dehydration in alcohol. Five- to e-^m sections were made and stained with hematoxylin and eosin. grafted human mesothelioma. Based on these results in nude Histological sections were examined with a Nikon light microscope mice, a clinical trial of mitomycin C and cisplatin was undertaken. (Nikon, Inc., Garden City, NY) equipped with an ocular micrometer Four of 12 patients showed objective response (one complete calibrated with a hemacytometer (American Optical Scientific Instru and three partial). These clinical results support the usefulness ments, Buffalo, NY). The number of tumor cells per cu mm was deter of the nude mouse model for malignant mesothelioma. mined by numerical analysis in 20 random fields at x400 for each line BG and ES (9), after correction for section thickness, fixation shrinkage (2), and proportional tumor cell content. Under these conditions, the INTRODUCTION number of tumor cells per cu mm was [4.8 ±0.18 (S.E.)] x 10s for BG xenografts, and [3.3 ±0.32] x 105 for ES xenografts. Malignant mesothelioma is a disease of growing importance. As predicted (25), its incidence is increasing (17) as a result of Mice were observed every weekday and weighed once weekly. Both lines BG and ES have retained their original histological, histochemical, the widespread use of asbestos, its major etiological agent. and ultrastructural features despite serial transplantation (6, 30). Kary- Results of various therapies for malignant mesothelioma have otypes have confirmed their human nature (6). Line BG is an epithelial been generally poor (5, 8). Neither surgery nor radiotherapy is malignant mesothelioma, and line ES is a mixed or biphasic, but predom curative. Very little information is available regarding the activity inantly epithelial malignant mesothelioma (6). After the growth of s.c. of chemotherapeutic agents (4, 8). implants was established, cohorts of mice transplanted with the same The availability in our laboratory of 2 distinct cell lines of human tumor specimen (one or 4 s.c. implants/mouse) were randomized be malignant mesothelioma serially transplanted into nude mice (6) tween chemotherapy (2 to 6 animals/group), and control with 0.9% NaCI allowed us to study systematically the activity of chemothera solution (2 animals/group). All injections were given i.p. peutic agents alone and in combination in this model. One Quantification of Antineoplastic Activity. Tumor volume was mea sured with calipers once or twice weekly, using the formula for a prolate combination, mitomycin C and cisplatin, exhibited efficacy ellipsoid against both lines in the nude mouse system (7). Based on this research, a clinical trial of this combination was initiated in 12 consecutive patients with malignant mesothelioma. where L is the greatest length and W is the perpendicular width of the MATERIALS AND METHODS tumor corrected for skin thickness (12, 19) (Fig. 1). Overall treatment effect was measured using the mean ratio T/C (expressed as a percent Nude Mouse Xenografts. Two lines of human malignant mesothe age) of the tumor volumes between Days 35 to 49 after tumor implan lioma (designated BG and ES) have been serially transplanted s.c. since tations for treated animals (T) over the tumor volumes of corresponding 1978 in nude mice, as previously described (6). For these experiments, controls (C) (12). A T/C ratio of 42% or less is considered to be an 8- to 10-week-old homozygous female nude (nu/nu) mice were obtained indication of activity (12). Life span was measured from the time of tumor from the Animal Genetics and Production Branch of the National Cancer implantation to death. Statistical comparison of means between treated animals and controls was carried out using Student's f test, with modi Institute, Bethesda, MD. Filtered air and sterilized bedding, cages, food, and water were used. Serial transplantation of tumors was carried out fication for small numbers. by excision of tumors under light anesthesia by inhalation with methox- In addition, a method of quantification of antineoplastic effect was yflurane (Penthrane; Abbott Laboratories, North Chicago, IL). The tumor designed which did not assume a preconceived pattern of growth but specimens were dissected under sterile conditions in a Petri dish con- which examined the efficacy of therapy as reflected in the shrinkage of tumor mass in each individual experiment. For individual tumors, re garded as independent, each tumor volume N(t) as a function of time f 1Supported in part by Cancer Center Grant CA-23102 from the National Cancer was fit by a polynomial spline technique (10, 23), and integrated analyti Institute, and by the T. J. Marteli Memorial Foundation for Leukemia and Cancer cally from the time of initiation of therapy at f„to the time td of last Research. Presented in part at the 74th Annual Meeting of the American Association for Cancer Research, San Diego, May 25 to 28, 1983 (7). volume measurement prior to death. The parameter A, which can be 2 To whom requests for reprints should be addressed. conceptualized as an average tumor volume between f„and td was Received September 13, 1983; accepted January 6, 1984. therefore calculated as 1688 CANCER RESEARCH VOL. 44 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1984 American Association for Cancer Research. Malignant Mesothelioma i.p. weekly for 3 weeks resulted in 4 of 22 acute deaths (18%), = (ta- N(t)dt r which is about equitoxic to each drug given as a single agent at these doses. In the experiments detailed below, treatment was Parameter A ¡sfunctionally related to the ratio of growth rates in the started between Days 13 and 20 after tumor transplantation. perturbed (treated) versus unperturbed (control) situation, as previously The overall results of these experiments, summarized in Table described (18). Values for A for each tumor line and each therapeutic situation were transformed logarithmically to stabilize variances, com 1, revealed that this combination of mitomycin C and cisplatin pared by an analysis of variance, and then compared pairwise by was the most active regimen for both lines ES and BG in nude Student's f tests, one-tailed. mice. No difference was seen between mice bearing one or 4 Patients. Twelve consecutive patients with malignant mesothelioma xenografts each. were treated between the end of 1981 and the beginning of 1983 with Quantification of antineoplastic effect in a representative ex the best regimen found in the nude mouse system. Pretreatment evalu periment shown in Chart 1 and Table 2 (BG, 29th generation; ation procedures, performance status categories, and criteria of re ES, 31st generation) revealed that, for tumor line ES, there was sponse to chemotherapy using clinically palpable tumors, roentgeno- no significant difference between control tumors and those grams, and/or computed tomography, have been previously described treated wtih cisplatin. Mitomycin C caused tumors smaller than (8). All patients had advanced disease at the initiation of chemotherapy (Stages III and IV from a TNM (tumors-nodes-metastases) classification those of controls, but this did not achieve statistical significance (4) for patients with pleural mesothelioma). Ten patients had a history of in this particular experiment, where the number of tumors was small (p = 0.1). The combination of the ineffective drug cisplatin prior asbestos exposure. Cell type was epithelial, except for Patient 1, who had a biphasic or mixed type. All patients gave informed consent. with mitomycin C, however, produced a significant difference All patients had normal hematological and renal functions. The initial 5 between tumors so treated and controls (p < 0.005). In addition, patients had all been treated previously with doxorubicin-containing tumors treated with the combination were significantly different regimens and were in progression. In Patients 1 to 4, these regimens from those treated with mitomycin C alone (p < 0.01). Tumor included doxorubicin, 25 mg/sq m/day i.v. on Days 1 to 3, and 5- azacytidine, 120 mg/sq m/day by continuous i.v. infusion on Days 1 to Table1 5, cycles being repeated every 4 weeks (8). After 7 cycles, doxorubicin Overall results of the effect of mitomycin C and cisplatin alone and in combination was discontinued and replaced by cyclophosphamide, 200 mg/sq m/day in nude mice with human malignant mesothelioma xenografts i.v.