George W. Woolley

HOST CONDITIONING IN EXPERIMENTAL CHEMOTHERAPY

The screening of potential anticancer agents against transplantable human tumors in mammals would be virtually nonexistent without procedures for altering the species resistance of animals to heterologous tissues. Whatever information of clinical value results from the study of transplantable human tumors will probably stem directly from the fact that host conditioning has made possible the extensive study of such tumors. I t is sometimes necessary to emphasize this thought in order that the questions raised by discussing variations in conditioning procedures will not obscure the real reason for using a heterologous tumor-host system for experimental chemotherapy, that is, that this tissue was derived from human patients, has retained human characteristics,” and may be more like the human nebplasm in its response to chemical compounds than animal tumors. The development of a test system using transplantable human tumors involves the same principles and procedures important in the development of any tumor-host test ~ y s t e m . ~ Moreover, as with any tumor-host system, confidence in the reliability of test results depends on an awareness and consideration of the influence on these results of variables operating within that system. The atypical variable encountered in assaying human tumors for response to potential anticancer agents is the resistance of mammalian species to grafts of heterologous tissue and the effect on both tumor and host of the agents used to lower that resistance. The objective in conditioning for experimental chemotherapy of transplantable human tumors is to use the least amount of conditioning consistent with maximum growth of the transplant and minimum effect on the host. The agents primarily used for host conditioning are X irradiation-generally 150 r--and/or either cortisone or hydrocortisone. These corticoids possess many metabolic activities in addition to their ability to depress the lymphoid organs and cells important to the defenses of the host to foreign tissue. I n rats receiving doses of cortisone of a magnitude needed for effective conditioning for human tumor growth, interference with metabolism is evidenced by the inability of the animals to gain weight normally and by a reduced tolerance to various chemicals (P. C. Merker, N ! . N. Teller and J. E. Palm, unpublished data; see also Teller et al., “The Human Tumor in Cancer Chemotherapy in the Conditioned Rat,” elsewhere in this monograph). The extent to which the corticoids either directly, by acting with test drugs, or indirectly, by side effects on metabolism, influence experimental chemotherapy results is not

The human tumor in the mouse.

The successful transplantation of heterologous tumors was achieved by Murphy Beginning in 1938, Greene3-j clearly demonstrated that human tumors can be heterotransplanted successfully to the anterior chamber of the eye of the guinea pig, rabbit, and mouse, although with great variation in success. In general, the growth of the tumors was poor. Hurst et d7 in 1939 cultivated a variety of human neoplasms on the chorioallantois of chick and duck embryos, but the growth was disappointingly small and insufficient to allow serial transfer. Then, in 1951, an important advance in the technique of human tumor transplantation was achieved by Toolan? On the basis of Murphys results she irradiated rats and mice before implantation and obtained positive evidence of tumor growth. Even more important, she succeeded in transferring 2 to 3 per cent of these tumors for many generations in X-irradiated and/or cortisone-treated rats and hamster^.^ Many reports have now appeared concerning the transfer of human tumors, primarily in rats and harn~ters. ~~ The mouse, however, appears to have been almost neglected as a possible host. In our experiments, attempts were made to transfer serially the human epidermoid carcinoma HEp 3 into treated or conditioned mice. These studies were conducted in the hope of establishing mouse-grown tumors that could be used for the evaluation of anticancer drugs, chemicals, and regimes. Experiments in this direction are now in progress. Since many data showed the importance of the site of implantation to the growth of both the homotran~plant~-~8 and the heterotran~pIant, ~* 2o the tumor was transplanted to three sites. In addition, an attempt was made to block the immunological response of the host by methods other than the usual combination of X ray and cortisone. In a third seriesof experiments, which was preliminaryin nature, we compared the transplantability of HEp 3 to mice of different inbred strains possessing different histocompatibility (H) genes in the H-2

Strain specificities of leukemia agent.

Strain specificity evaluations have been secured from three groups of experiments: (1) cortisone studies using highand low-incidence leukemia (2) experiments on cell-free transmission of mouse le~kernia;~, and (3) studies with a new leukemia of viral origin.6 Strain-specificity data have, for the most part, been incidental in these problems. It seems probable that all three areas are concerned with the operation of subcellular particles in the neoplastic process.

DISCUSSION OF PART VI

Historically, we have heard that attempts to transplant human cancer to experimental hosts have been recorded for a century or more: Peyrilhe, 1775; Langenbeck, 1840; Zahn, 1884; and Klebs, 1890.l More recently, heterotransplantation has been forwarded by Murphy (in 1912)2 through transplantation of tissue to the chorioallantois of the chick embryo. The use of X irradiation to reduce resistance to foreign tissue also was explored: More recent work includes that of Greene (1938)4 and others. In 1951 and later, Toolan,5 re-explored the use of X irradiation and instituted the use of cortisone in host-conditioning procedures. She achieved successful and continuous implantation of human tumors in conditioned rats and hamsters.6 Elsewhere in these pages she reviews some of her basic and exciting new findings, placing emphasis on the growth and morphologic potentialities of these tissues in their unusual environment, their capabilities for long-term preservation by freezing, and their stability in retaining their species specificity. The chemotherapy studies reported from our laboratory are dependent on the methodology of heterotransplantation developed by Toolan. We believe that use of human tissue enhances the probability of picking up, indicating, and evaluating compounds useful to man. Use of the rat as a host for human tumors has been presented from two laboratories. The basic methodology has been similar. Several hundred agents now have been evaluated, and workable systems of operation are in more-orless routine use. In our studies, as presented by Teller and discussed by Merker, two human tumors originally introduced by Toolan were the basis for the chemotherapy studies. In this system, young female Wistar rats approximately 4 weeks old (50 to 65 gm.) were X irradiated with a single total body dose of 150 r 1 to 4 days prior to implantation of tissue. Two 180 kvp. tubes were used, one above and one below the animals (not parabolic) a t a TSD of 50 cm. (50 r/min.). For the preparation of inoculum for transplantation, 9t o 1 1-day-old HEp 3 and 10to 14-day-old HS 1 tumors were used. Tumors grown subcutaneously in the X-irradiated and cortisone-treated rats were excised, minced with scalpels, and prepared as a suspension in sterile Ringer’s salt solution. Each 100 ml. of Ringer’s solution contained, in addition, 600 mg. glucose, 20,000 u. penicillin, 60 mg. streptomycin, and 1 ml. of McIlvane’s phosphate-citrate buffer (pH 7.35 to 7.40). A sample of minced tumor tissue was removed for a test for sterility prior to addition of the diluent.

Effects of Hadacidin on Human Tumors Grown in Eggs and Rats.

Summary Four transplantable human tumors (H. Ep. 3, H.S. 1, . Ad. 1 and A-42) were gorwn on the CAM of chick embryos and used to determine the anti-tumor potential of hadacidin. In testing samples of hadacidin, there was consistent anti-tumor activity. When a comparison among the tumor types of their response to hadacidin was made, A-42 appeared to be consistently more sensitive to hadacidin than were the other types of tumors. Using two different routes of administration, intraperitoneal and per os, hadacidin inhibited the growth of H.S. 1 in the x-irradiated and cortisone treated rat. The positive results obtained in the mammalian host are further indications of the antitumor potentials of hadacidin.

SUCCESSFUL INTRAMUSCULAR GROWTH OF HUMAN TUMOR IN CONDITIONED MICE

The classic experiment of heterologous transplantation of human tumor, which was an attempt to transfer cancer from man to dog, was recorded by Peyrilhe in 1775. In 1840 Langenbeck injected the vein of a dog with the juice from a human medullary carcinoma of the humerus. Two months later several round nodules were found in the lung. However, Rudolf Virchow, the pathologist, found the structure more similar to a spontaneous cancer in the dog than to that in the human. In 1884 Zahn2 transplanted a human enchondroma to the anterior chamber of the eye of a rabbit, but failed to find evidence of its growth. Klebs in 1890 examined fragments of human carcinoma removed from the peritoneal cavities of white rats at varying intervals after introduction. He found that epithelial constituents of the graft had vanished by the third day. reported that the Jensen rat sarcoma grew serially in the chick embryo; however, he did not succeed in propagating the tumor in chickens. In 1938 Greenej- reported that human uterine adenoma and adenocarcinoma could be maintained in the rabbit. This tumor was serially transferred for several generations. In 1951 a new approach to the problem of heterologous transplantation of human cancer was described by Toolan, who later reported the survival and proliferation of 90 per cent of 101 human tumors in rats and hamsters treated with cortisone and/or X irradiation.8-0 Recently many experiments dealing with heterologous transplantation of human tumor have been reported by Sommers et aZ.,I Hoch-Ligeti and Hsue,12 and Patterson et all3 The experiments reported in the first part of this paper were set up in the hope of developing human tumors transplantable to mice. In this work the mice were treated with cortisone, X irradiation, and ThorotrastJ either singly or in various combinations. The second part of the paper follows histologically the development of a human tumor from one day to one week after transplantation. * The research for this paper was supported in part by Grant CY-3784 from the National Cancer Institute, Public Health Service, Bethesda, Md., and in part by Grant T-47 from the American Cancer Society, Inc., New York, N. Y. t Visiting Research Fellow, Sloan-Kettering Institute for Cancer Research, and SloanKettering Division, Cornell University Medical College, New York, N. Y.

Testing of hormones and related substances with human tumors growing on the chick embryo.

A proprietary colloidal suspension of thorium dioxide In 1912 Murphy3,

Experiments on cell-free transmission of mouse leukemia.

The use of the chorioallantoic membrane (CAM) of the chick embryo for heterologous tumor growth began with the classic studies of Murphy (1912). Later the chick embryo was used to grow tumors from a variety of animals, including man. Stevenson (1918) reported the successful growth of a variety of mouse tumors. The chick embryo was used by Schrek and Avery (1937) to grow rat and rabbit tumors, and the serial passage in the egg of established transplantable human tumors was reported in 1954 by Dagg et al. Human tumors grown on the CAM have been shown to respond to nitrogen mustard (Dagg, et al., 1954), triazene (Dagg et al., 1955), triethylenemelamine (Harris, 1956), and a number of compounds of microbial origin (Harris, 1957, and in his article elsewhere in this monograph). These studies are concerned primarily with experimental chemotherapy utilizing hormones and related substances as test agents against human tumors growing on the CAM of chick embryos. Preliminary chemotherapy studies using hormones and related substances as test agents have indicated that such tumors may be useful in the evaluation of potential antitumor agents in these categories. Some preliminary results of our tests are presented in this report. Cornell University Medical College, New York, N . Y.