Document 4aZX64Bb7gZvORRG5ywyxrKv1

NATIONAL CANCER INSTITUTE MONOGRAPH 46 (December) 1 977 i Modern Concepts in Brain Tumor Therapy: Laboratory and Clinical Investigations DHEW Publication No. (NIH) 77-1236 U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE Public Health Service National Institutes of Health NATIONAL CANCER INSTITUTE. BETHESDA, MARYLAND 20014 OLI 6784 Chemical- and Virus-Induced Bruin Tumors1 James 4. Swenber** 1 ABSTRACT--Experimental animal models resembling most hu man brain tumor types can be induced by exposure to oncogenic viruses or chemical carcinogens: Astrocytomas and glioblastoma multiforme can be produced experimentally by Intracerebral injection of oncornaviruses, whereas medulloblastomas, choroid plexus papillomas, and ependymomas can be induced by the papovaviruses. Adenoviruses have been utilized to cause medulloepitheiiomas, neuroblastomas, and retinoblastomas. All three groups of viruses can result in sarcoma production. Gliomas represent the primary tumor type induced in the brain by chemical carcinogens. These autochthonous tumor systems are reviewed, with emphasis on methods, tumor type, latency period, advan tages, and disadvantages, in addition, recent investigations ot molecular events involved in neoplastic transformation by chem ical carcinogens are summarized__Natl Cancer Inst Monogr 46: 3-10, 1977. A variety ot experimental brain tumor models resembling most human tumor types have been developed during the past decade. Biain tumors can be induced with intracerebral injection of viruses or chemicals, or by transplacental, paren teral. oral, or even topical exposure to chemical carcinogens. Depending on the model, such tu tors arise weeks to years after exposure. The ailability of suitable animal models has stimu lated increased research on brain tumor kinetics, immunology, chemotherapy, and histogenesis. Re cently, these models have been used in the study of molecular events responsible for neoplastic transformation. This review is intended to sum marize the experimental brain tumor systems available todav and to highlight some of the exciting research being done with these systems. VIRUS-INDUCED BRAIN TUMORS Work on experimental brain tumor induction with viruses was reviewed comprehensively in 1 Presented .it the Symposium on Modern Concepts in Brain Tumor i'hcrjpy: Laboratory and Clitm.il Investiga tions, held in Atlanta. Ga., February 26-2H, 1976. and sponsored bv the Clinical Investigation Branch and the Cancer Clinical Investigations Review Committee. Division of Cancer Treatment. National Cancer Institute, Hethesda, Md. 2001 4. * Pathologv and Toxicology Research Unit. The Upjohn Company, Kalamazoo, Mich, 4901)1. 1 Dr. Dareli Bigner provided many helpful suggestions during the preparation ol this manuscript. TABLE 1__Specificity of experimental brain tumor induction with oncogenic vimaee Experimental brain tumor type Inducing viruses Group Type* Anaplastic astrocytoma or glioblastoma multiforme Medulloblastoma Neuroblastomas or retinoblastomas Ependymomas or choroid plexus papillomas Sarcomas or meningeal tumors RNA-Oneomaviruses ASV, MuSV. SSV DNA-Papova DNA-Adenovirus DNA-Papova DNA-Adenovirus RNA-Oncomavirus Human papova JC Human adenovirus type 12 SSV40. human papova PML-1, JC, and BK SA7and CELO ASV and MuSV DNA-Papova DNA-Adenovirus Bovine papilloma, murine polyoma, human papova JC SA7 *ASV - *vin sxrcomx virus; MuSV 3 murine sarcoma virus; SSV = simian sarcoma virus; BK - human papovavirus BK; SA7 - simian adenovirus 7. several publications (1-6). Intracranial tumors have been induced by representatives of most oncogenic tumor virus groups, including the ade noviruses, papovaviruses, and oncornaviruses. To date, techniques for induction and morphologic descriptions were emphasized in most studies on these models. Future investigations will probably de-emphasize morphologic classification and in stead turn toward utilizing such models in basic and applied research. Morphologic studies have led to several generalizations that are summarized in table 1. For example, only oncornaviruses induce astrocytomas and glioblastomas, whereas the human papovavirus JC is the only virus known to cause the development of medulloblas tomas. Choroid plexus papillomas are induced by simian virus 40 (SV40), the SV-PML papovavi ruses isolated from patients with progressive mul tifocal leukoencephalopathy (PML) and the avian adenovirus (CELO), whereas medulloepitheliomas and retinoblastomas are only elicited with human adenoviruses. In contradistinction, sarcomas were induced by members of most tumor virus groups. OLI 6785 SVVISBERG RNA Tumor Viruses Avian Sarcoma Virus Mammalian neuro-oncogenicity of ASV was first described bv Rabotti and Raine (7). Subsequemlv, gliomas and intracranial sarcomas were induced m cats, dogs, guinea pigs, mice, rabhits, rats, and subhuman pumates. The relative pro portion of glial and mesenchymal elements in these biain tumors \aiies considerably among species. Homogeneous glial tumors have clearly been pioduced in the rat and dog with ASV. Bigner et al. (<V) induced gliomas in 27 of 30 dogs. 19 of which had gliomas only, with no evidence of sarcoma formation. They later dem onstrated that the site of ASV inoculation influ enced tumor type: i.e., injection of ASV deep in periventricular rones resulted exclusively in astro cytomas, whereas only sarcomas were induced if the virus was injected superficially over the cere bellar vermis (9). Well-differentiated astrocytomas also developed in F344 (JO) and Sprague-Dawley (II) rats after neonatal intracerebral injections of ASV. In addition, Copeland et al. (12) demon strated that rats remained susceptible to the neuro-oncogenic effects of ASV at 100 days of age. Although the incidence of brain tumors in rats inoculated with ASV as adults was only 50%, nearly all were astrocytic. In contrast to the rat and dog, a large percent age of the hamster, guinea pig, mouse, cat, and subhuman primate tumors induced with intracere bral ASV have a mesenchymal nature. Whereas neoplastic astrocvtic elements can be observed, prominent reticulin fibers are also present. Tu mors composed of large balloon-like cells (13) and "giant cell glioblastomas" (14, 15) usually contain such reticulin. This cell type probably represents a histiocytic element in a mesenchymal or mixed glial-mesenchymal tumor. Several other features of ASV-induced brain tumors deserve comment: 1) There is generally no replication of infectious or noninfectious virus in mammalian tumor systems. However, virus structural antigens and virus-related transplanta tion antigens are usually produced in transformed mammalian cells. One can "rescue" ASV from such cells by growing the "nonpermissive" trans formed cells with normal avian cells and allowing cell fusion to occur. 2) The dose of virus required to elicit intracranial tumors is lower than that for extracranial tumor induction. This is probably due to a greater immunologic response against the latter. 3) Improvements in the techniques for growing and concentrating ASV have allowed large pooh of stand,udi/ed vims to !>e produced (lb). Using this virus, lesearchers have obtained repioducihle survival curves for rats bearing a 100% incidence of astrocytomas. This, system pro vides the first primary glioma model that is well suited for chemotherapy trials (//I. Preliminarv results of these trials have been reported (5, 1618). Other Oncornaviruses MuSV has been reported to cause brain tumors in mice and rats following intracerebral inocula tion (5). Gliomas, meningiomas, and hemangioen dotheliomas represent the most common tumor types induced with MuSV, In contrast to ASVinduced mammalian tumors, replicating virions were easily detectable in these mouse tumors bv electron microscopy (19). Recently, a 100% inci dence of extremely vascular brain tumors contain ing neoplastic astrocytes and extensive endothelial proliferation was induced in Wistar-Furth rats with the Kirsten MuSV (20). No C-type virions were detected by electron microscopy. Intracerebral inoculation of neonatal marmosets with SSV induced solitary tumors in 6 of 10 animals (21). The tumors were composed of pleomorphic neuroglial cells, areas of necrosis, hemorrhage, palisading, and prominent endothe lial proliferation. These SSV-induced marmoset tumors closely resemble the human glioblastoma multiforme. Infectious virus was isolated from brain and cerebral spinal fluid of tumor-bearing animals. DNA Tumor VlruoM Papovaviruses Several members of the papovavirus group cause tumors when inoculated intracerebrally into animals (5). Bovine papilloma virus induced men ingiomas, fibromas, and Ftbrosarcomas in ham sters and calves following latency periods ranging from 20 days to nearly a year. No virus particles were detected in the tumors by electron micros copy. Fibrosarcomas were also detected in the leptomeninges after polyoma virus was injected intracerebrally into newborn rats, hamsters, and rabbits. Intracerebral inoculations of SV40 into newborn hamsters induced Fibrosarcomas in the leptomeninges and choroid plexus papillomas in the ventricles. Of greatest interest, however, are the tumors that developed in hamsters after intracerebral NATIONAL CANCER INSTITUTE MONOGRAPH NO. 46 OLI 6786 CHEMICAL- and VIRl s-INDLCED BILALS IT MORS injections ot JC virus, ,i hum.in p.ipmuvirus isol.ited from a patient with PMl, (22-24). Brain tumors developed m .id of 63 hamsters inoculated vith JC virus, with medullohlastomas representing .he most common tumor tspe. t hese appear to arise from the internal granule tell laser of the cerebellum and trequeiulv are multifocal. Several other tumor types have been induced with JC virus, including primitive gliomas, papillary epen dymomas, meningiomas, and pineocytomas. Virus was recovered from 5 ot 7 tumors tested. The JC virus is morphologicallv indistinguishable from SV40 and SV40-PML strain 1, but differs from them in its cell culture host range and antigenic properties (5, 25). Adtnovinjses The experimental induction of neurogenic tu mors in rats, hamsters, and mice after the animals received intracerebral, intraocular, or ip inocula tions of human adenovirus 12 provides an excel lent model for future research on neuroblastomas, medulloepitheliomas, and retinoblastomas (2631). Medulloepitheliomas were induced in 88% of the rats given intracerebral injections of human denovirus 12. These appeared to develop from ae subependymal plate and were characterized by the formation ol rosettes, high levels of cholin esterase following in vitro cultivation, and the presence of cilia containing 9+0 tubules (a hall mark of normal sensory neuronal cells). Similar 9+0 tubules were demonstrated ultrastructurally in the cilia of retinoblastomas induced in rats and hamsters following neonatal intraocular injections ol human adenovirus 12. Other members ol the adenovirus group that elicited intracranial tumors (5) include the simian anti avian adenoviruses |SA7, SV20, and CELO). ,3.47 induced poorly ditlereniiated tumors of the choroid plexus, trigeminal nerve, and dura mater. Injection ol SV20 into hamsters caused extremely undillerentiated tumors, which had the same morphology whether induced sc or intracerehratly. Intracerebral tumors produced with CELO had a morphology distinctly different from neo plasms induced with either the human or the simian adenoviruses; they were composed of pap illary arrangements ot cuboidal and columnar cells that tilled the ventricles. Although originally thought to be ependymal tumors, the CELO tumors probably represent choroid plexus papil- tas. CHEMICALLY INDUCED TUMORS OF THE CENTRAL NERVOUS SYSTEM (CNS) The susceptibility of neuroectodermal tissue to the oncogenic effects of chemical carcinogens was first demonstrated in 1939 (32). Since then, brain tumors have been induced bv local implantation of several polycyclic hydrocarbons (PCH). More recently, additional classes of carcinogens were shown to be oncogenic for the nervous system following systemic exposure. These ranged from agents such as 2-acetylaminofluorene. which caused a few tumors of the nervous system along with many tumors elsewhere, to the nitrosoureas that caused neuroectodermal tumors in 100% of the animals. Several reviews (33-36) and a mono graph (5) have been published on the chemical induction of brain tumors. In addition to reports by some scientists on the induction and character ization of various brain tumor models, other investigators used animal models for neuro-onco genesis to study the molecular basis for neoplastic transformation. Whether similar mechanisms are involved in human brain tumors is unknown. It has been estimated that 80-90% of all human cancer is caused by chemicals; however, no chem ical carcinogen has been clearly linked to human brain tumors. In the section to follow, various brain tumor models have been arranged by the class of carcin ogen used to induce them. Polycyclic Hydrocarbon* Over 200 years ago, Sir Percival Pott first theorized the potential carcinogenic hazard of PCH. Therefore, it was fitting that the first ex perimental tumors of the nervous system were also induced with PCH (32). Since that time, brain tumors have been produced with several PCH, including 3-methvlcholjiithreiie, benzofajpvrene, dibenzanthraceue, dimeth vlbenzfci]anthracene (DMBA), and trimethyl benzanthra cene. Such tumors have been induced in frogs, toads, mice, rats, hamsters, and dogs. However, birds, guinea pigs, rabbits, cats, and monkeys were resistant to the neuro-oncogenic effect ol PCH. In general, brain tumors occur only when the PCH are in direct contact with the CNS. To accomplish this, the carcinogens are directly im planted in the brain. The incidence and type of tumor produced by PCH are greatly influenced by the positioning of the pellet (37). Few gliomas but many sarcomas and meningiomas result from superficial placement of the pellet near the dura. Ependymomas were primarily induced when the MODERN U1NC.EPIS IN HRAIN IL'MOR [IILRAPY OLI 6787 XWFSBFRO pellet was placed in the ventricles, wheieas oligo dendrogliomas were associated with implantation in the frontal white matter. Astrocytomas arose near pellets placed in the subcortical regions of the panetal lobe, and medulloblastomas lormed alter pellet implantation into the cerebellum. Although several investigators have successfullv induced brain tumors in rats with FCH. mice appear to be the most susceptible animal species (J-/). Tumor incidence varies from less than 10 to nearly 100$, with an average of 40-60$ of the animals developing tumors. Most animals die be tween 200 and 300 days; however, latency periods have varied from as short as 68 to as long as 750 days. The primary advantage of this model is that the site of tumor formation can be localized. Disadvantages of the system include the variable incidence and latency period of the experimental tumors. The model is poorly suited for quantita tive dose-response studies and investigations of early biochemical processes, since the time of exposure can vary from minutes to months be cause of the extremely long half-life of the pellet. The importance of unique doughnut-shaped vi rions detected in macrophages surrounding pel lets of carcinogens prior to tumor formation requires further elucidation (38). Whether these virions represent activated viruses that play an important role in tumor induction remains un known. However, they have never been detected in experimental gliomas and have induced fibro sarcomas rather than gliomas when they were injected intracerebrally into mice. When pregnant rats were exposed to DMBA iv on day 21 (39) or orally on days 14, 17, and 20 of gestation (JO), the most frequently detected neoplasms induced in the offspring were tumors of the nervous system. Transplacental exposure to additional PCH will be necessary to determine whether this is a general phenomenon or a spe cific effect of DMBA, In any event, it should now be possible to design pulse exposure experiments and to demonstrate dose-response relationships for DMBA. Nltroso Compoirvdi A landmark in experimental neuro-oncology occurred in 1964 when Druckrey and co-workers (41) selectively induced tumors of the rat nervous system after iv injections of methylnitrosourea (MNU). This provided the first reproducible brain tumor model that did not traumatize the brain. Since then, several additional carcinogens have been discovered that primarily induce neuroecto dermal tumors following systemic administration. I hese animal models represent some ol the iimim powerful tools for modern rcscaiih in cliemit.d neuro-oncogenesis. i Methylnitrosourea This acyl-alkvl-nitrosamide is among the most potent of all the chemical carcinogens known todav. When administered iv at repeated low doses, the nervous svstem ile.ulv icpievents the target organ. The highest incidence of neuiogemc tumors is obtained with repeated iv injections of MNU at a dose schedule of 5 mg/kg/week for 32-36 weeks (42). This dosage has consistenilv produced a 90-100$ incidence of grosslv detect able brain tumors in Spraguc-Dawlev rats. Welldiflerentiated neurinomas are also induced: how ever, the incidence of these is much lower. This low incidence of tumors of the peripheral nervous system (PNS) can be reversed if F344 rats are used instead of Sprague-Dawley (43). The rat and the rabbit appear most susceptible to brain tumor induction with MNU (5). Dogs have been more variable; some brain and peripheral nervous system tumors have been induced in various breeds. Mice were quite resistant to neuro-onco genesis with MNU. Denlinger et al. (44) induced 2 gliomas in 19 C3H mice exposed iv to 25 mg/ kg every 4 weeks (total dose, 175 mg/kg). In spite of observation periods exceeding 6 years, attempts to induce brain tumors in rhesus monkeys have been unsuccessful (5). When cats were exposed to iv or oral doses of 25 or 20 mg/kg, respectively, all animals died within 10 days from severe bone marrow toxicity and septicemia (35). Additional resistant species include guinea pigs, sheep, and swine. Anaplastic gliomas, mixed gliomas, and oligo dendrogliomas represent the most common brain tumors induced with MNU. These are frequently located near periventricular regions, in subcortical white matter, and in the hippocampus. If the dose of MNU is raised from 5 mg/kg/week to 10 or 20 mg/kg twice a week, a shift from glial to mesenchymal cell types becomes apparent, with most tumors being sarcomas or gliosarcomas {45). MNU is also a potent carcinogen when adminis tered locally, as shown clearly in experiments utilizing the sc, ip, or oral routes. Tumors develop at the site of administration and to a lesser extent in the nervous system. The lower incidence of tumors detected in the nervous system can best be explained by the decreased levels of circulating MNU. In addition to tumors at the site of injec tion and in the nervous system, a high incidence NATIONAL CANCER INSTITUTE MONOGRAPH NO. 46 OLI 6788 CHEMICAL- AND VIRL'S-INDfCED BRAIN TL'MORS d1 ihvmic lvmphomas can be induced after oral or sc administration of 20 mg/kg twice a week for 9 weeks. These types of experiments have clearly shown that MNU is not '`neurospecific.1' Rather, MNU is a potent carcinogen for many tissues. Potential mechanisms involved in determining which target site is affected by this and related carcinogens will be discussed later. Ethytutrosourea (ENU) The exquisite sensitivity of the fetal rat's nerv ous system to the oncogenic effects of ENU was first demonstrated in the experiments of Ivankovic and Druckrey (96). A 100# incidence of neurogenic tumors developed in offspring ex posed to a single dose of 20 mg/kg or more during the last trimester of pregnancy. Clear dose-response relationships exist with single doses of I, 5, 20, or 50 mg/kg inducing a 12, 79, 100, and 100% incidence of neurogenic tumors, re spectively (97). While the tumor incidence in-, creased, the latency period for tumor induction decreased from a mean of 655 days for 1 mg/kg to 211 days for 50 mg/kg. Age of the individual greatly influences the susceptibility to the neurooncogenic effects of ENU. The rat fetus is resist ant to the carcinogenic effects of ENU before the 12th day of gestation; however, it is susceptible to teratogenic effects before this stage of develop ment. Susceptibility to neuro-oncogenesis in creases from the 12th day of gestation to birth, after which it begins decreasing [33). By 30 days of age, the rat has a susceptibility comparable to the adult; i.e., the nervous system no longer represents the target organ and much higher doses of ENU are required to induce a similar incidence of neoplasia. ENU has usuallv been administered iv to preg nant rats, although it can be given orally, ip, or sc. Neurogenic tumors were even reported in 38T of the offspring whose mothers had 50 mg/ kg ol ENU applied topically to the skin (9#), It has aho been possible to induce tumors by admin istration of precursors of ENU in the food and water of pregnant rats. Under the mildly acidic conditions of the rat's stomach, ethylurea and sodium nitrite undergo nitrosation and form ENU in vivo (99). Rats are the species most sensitive to the neurooncogenic effects of ENU. Hamsters, mice, and opossums have also developed neurogenic tumors. In hamsters, tumors have been confined to the PN5. whereas in mice a low incidence of CNS and PNS tumors develops. Mixed gliomas and oligodendrogliomas represent the most common brain tumors induced with ENU. wherea- anaplas tic neurinomas are the tvpe occurring most fre quently in the PNS. Anaplastic gliomas and sarco mas are rare. Animals bearing anaplastic neurinomas die significantly earlier than those bearing gliomas. When the sequential develop ment of these tumors was investigated, evidence of neoplastic proliferation was detected in trigem inal nerves as early as 3 weeks after exposure to ENU [50). Comparable changes were not evi dent in the brain until 4 months after ENU exposure. Mechanisms involved in this differential triggering of neoplastic proliferation between glial and Schwann cells remain unknown. Omar Nitroso Compounds Additional nitroso compounds possessing neurooncogenic properties have been described since the original discovery of MNU (5, 33), including dimethylnitrosourea, trimethylnitrosourea, propylnitrosourea, butylnitrosourea. dinitrosopiperazine, nitrosopiperidine, nitrosomorpholine, ethylnitrosobiuret, methylnitrosobiuret, ethylnitrosourethane, and methylnitrosourethane. In general, these offer no major advantages over MNU and ENU and will not be discussed in detail. Mur Carcinogana Tnazants Several of the dialkyl-arvl-triazenes have been shown to be carcinogenic for the nervous system as well as for other tissues [51). Neuro-oncQgenesis has been successful both in adult rats and transplacentally. As noted for MNU and ENU. the methvl compounds were most effective in adults, whereas the corresponding ethvl com pounds were highly oncogenic to the fetal i.uN nervous system. i.2-Dieitiyihydra/int. Azoethana, and Azoxyathana Transplacental neuro-oncogenesis has been in duced in the rat with these three carcinogens [33). In contrast to the triazenes, adult rats devel oped multiple intestinal tumors instead of neuro genic neoplasms when the methyl counterpart was administered. INVESTIGATIONS ON THE MECHANISMS OF CHEMICAL NEURO-ONCOGENESIS Scientists conducting studies at several I.iImujtor es have suggested a possible mecham-m lor MODERN CONCEPTS IN BRAIN TUMOR THERAPY OLI 6789 SWENBERC, ihe nt'iirospccilicitv ol some themic.il carcinogens. I hat chemical carcinogens are electrophilic re agents has been known tor several years (52). Those carcinogens that are not electrophilic must he metahoh/ed to form electrophiles, which inter act with matromoiecules such as DNA, RN'A, and protein. Neoplastic transformation represents a heritable change at the cellular level, with the most likely site for this change being the DNA of the original!' transformed cell (55). Investigations of several carcinogens and tissues demonstrated alkylation of DNA bases. The great est amount of this alkylation occurred at the N-7 position on guanine, with lesser amounts being detected at the N-3 position of adenine and the Opposition of guanine. When the extent of N-7 ulkvlguanine formation was determined in brains and livers of adult and 10-day-old rats given injections of ENU. no correlation was found between DNA alkylation and carcinogenicity (5-/), Subsequent studies determined the degree of purin? alkylation at various times after ENU exposure. Goth and Rajewsky (55, 56) demon strated surprising dif ferences between the repair rates of 0*-ethylguanine and those of N-7 ethylguanine or iV-3-ethyladenine. 0*-Ethylguanine persisted for long periods in the target organ (brain) but was repaired much more rapidly in the liver. N-7-Ethylguanine was removed rapidly in both tissues. If one determines the braindiver ratio for each of the major sites of purine alkyla tion. a high degree of correlation is evident between alkylation at the Opposition of guanine and carcinogenicity. Further support for this hypothesis has been reported by Kleihues and co-workers (57-59), who demonstrated persistence of 0*-methylguanine in brains of rats treated with MNU and methylmethanesulfonate (MMS). Both methylat ing agents caused brain tumors in rats; however, MMS has a much weaker neuro-oncogenic effect. When equimolar doses of MNU and MMS were administered to rats, only 0.05 as much 0*methylguanine was present after exposure to MMS. Thus the extent of initial OPalkylguanine formation and its persistence over time correlates with carcinogenicity. A possible mechanism for this correlation was suggested by Loveless (60). Alkylation at the Opposition of guanine forces the base into the enol form, which causes anoma lous base paning during DNA replication. Instead of pairing with evtosine, 0*-alkylguanine pairs with thymine. Fixation of anomalous base pairing requires that at least one cell replication take place before repair of the damaged DNA. The lack of this-replication in entl-stage neurons mas explain why no neuronal tumors have lieen in duced with MNU, even though DNA alkylation occurs to a greater extent in neurons (than in glia (6/). It also helps to explain the predilection sites for ENU- and MNU-induced gliomas, since these areas contain the glia with the greatest propensity for cell replication. Enzymic repair systems for 0P alkylguanine have recently been demonstrated (62, 63). If these systems are deficient in rat brain, the exquisite susceptibility of the fetal rat brain to carcinogens may thus be explained. Over half the 35 or more compounds that caused transplacental carcinogenesis induced tumors of the rat nervous system. In summary, pathogenetic mechanisms involved in experimental neuio-oncogenesis seem to be as well or better known than those of other neo plasms. Delayed repair of OPalkylguanine ap pears to be a major factor in sensitizing the rat nervous system to carcinogenesis. Whether this mechanism is responsible for human brain tumors remains to be investigated. Sensitivity to neuro oncogenesis may be a general phenomenon for many carcinogens if effective tissue concentrations reach the nervous system. At present, the somatic mutation theory of cancer induction best explains chemical neuro-oncogenesis. REFERENCES (/) Yohn DS: Oncogenic viruses: Expectations and applica tions in neuropathology. Prog Exp Tumor Res 17:74-- 92, 1972 (2) Ikuta F, Kumanishi T: Experimental virus-induced brain tumors. Prog Neuropathol 253-334, 1973 (5) Johnson RT. Naravan O: Experimental neurological diseases of animals caused by viruses. In Models of Human Neurological Diseases (Klawans HL. ed). Am sterdam: Excerpta Medica, 1974. pp 43-82 (4) BIGNER DD, Pegram C: A review of virus-induced experimental brain tumors and of the putative associ ations of viruses with human brain tumors. Adv Neurol 13:57-83, 1976 (5) Janisch W, Schreiber D: Experimental Tumors ot the Central Nervous System. [Bigner D, Swenberg J. eds (first English edition)]. Kalamazoo: The Upjohn Co.. 1977 (6) Rabotti CF: Experimental intracranial tumors of viral etiology. In The Experimental Biology of Brain Tu mors (Kirsch WM, Grossi-Paoletti E, Paoletti P, eds>. Springfield, III.: Charles C Thomas. 1972, pp. MSISO (7) Rabotti CF, Raine WA: Brain tumors induced in hamsters inoculated intracerebrally at birth with Rous sarcoma virus. Nature 204:898-899. 1964 (8) Bigner DD, Odom GL. Mahaley MS Jr. et al: Brain tumors induced in dogs by the Schmidt-Ruppin strain of Rous sarcoma virus: Neuropathological and im munological observations. J Neuropathol 28:648-680. 1969 NATIONAL CANCER INSTrTVTE MONOGRAPH NO. 46 OLI 6790 CHEMICAL- AM) VIKl S-I.M)L(.El) BHAIN I tMOHS 'I hic.SEi DD. Kvedar JP. Shaffer TC, cl al Fuctorv intlucncing the cell type ol brain tumors induced in dot's !>v Sihmidt-Riippin-Rous sarcoma virus. J Neurop.ithol Exp Neurol 31:5H3--595. 1972 i/(/i Wilfonc Rf, Bicner DD. Self DJ, et al Hr,on tumor opes induced by the Schmidt-Ruppin strain ol Rous sarcoma virus in inbred Fischer rats. Acta Neuropathol (Berlin) 25 196-206, 1975 ill) Swenberg JA. Bicner DD, Hall TL. Experimental (hemotheraps of viral-induced bi.tin tumors Proc \in Assoc Canter Res 15 -Cl, 197-4 (12' C opeland DD, Vogel FS, Bicner DD The induction ol initacramal neoplasms b\ the inoculation ol avian -atconia virus in perinatal and adult rats. J Neuropathol Exp Neurol 34.340-358. 1975 i/I) Birger PC. Bicner DD, Self DJ: Morphologic obser vations ol brain tumors in PD -4 hamsters induced hv tour strains ol avian sarcoma virus, Acta Neuropathol Berltni 26-1-21. 1975 (14) hiMAMSHi T: Biain tumors induced vsith Rous sarcoma inis, Schnudi-Ruppin strain. I. Induction ol brain tumors in adult mice with Rous chicken sarcoma cells. JpnJ Exp Med 37:461-474, 1967 (I5\ Rlmanishi T, Ikuta F, Nishida K, et al: Brain tumors induced in adult monkeys by Schmidt-Ruppin strain of Rous sarcoma virus. Gann 64:641-643, 1973 (16) Bicner DD. Self DJ, Frey J, et al: Refinement of the avian oncornavirus-induced primary rat brain tumor model and its potential lor therapeutic screening. Recent Results Cancer Res 51-20-34. 1975 (17) Crafts D, Wilson CB: Animal models of brain tumors. Natl Cancer Inst Monogr 46:11-17, 1977 (18) Svvenberc JA. Bicner DD, Hall TL: Experimental chemotherapv ol viral-induced brain tumor*. Proc 5m Assoc Cancer Res 15:43, 1974 (19) Ditty PE: Virus-induced cerebral sarcoma. J Neuropathol Exp Neurol 29:370-391, 1970 (20) Yung WK, Blank NK. Vick NA: Glioblastoma. Induc tion of a reproducible autochthonous tumor in rats with murine sarcoma virus. Neurology 26:76-83, 1976 121) Johnson L, Wolfe LG, Wfiisler W. et al: Induction of gliomas in marmosets with simian sarcoma virus, type 1 (SSY-I). Proc Am Assoc Cancer Res 16:119, 1975 (22) Padgett BL, Walker DL, Zu Rhein GM, et al: Cultiva tion of papova-like virus from human brain with progressive multifocal leukoencephalopathy. Lancet I 1257-1260. 1971 (23) Walker DL, Padcett BL, Zu Rhein GM, et al: Human papovavirus <JC) induction of brain tumors in ham sters. Science 181:674-676. 1973 (24) Zu Rhein GM, Varaus J: Morphology of brain tumors induced in Syrian hamsters after inoculation with JC virus, a new human papovavirus. In Proceedings Vll International Congress on Neuropathology (Kornyey S. Tariska S. Gosztonvi G. eds). vol I. Amsterdam: Excerpta Medica: Budapest: Akadc'miai Kiado. 1975, pp 479-481 (25) Becker LE, Narayan O, Johnson RT: Demonstration of papovavirus tumor antigen in brain tumors. J Neuropathol Exp Neurol 34:95, 1975 (261 Ogawa K, Hamaya K, Fujh Y, et al: Tumor induction bv adenovirus type 12 and its target cells in the central nervous svstem. Gann 60:383-392, 1969 (27) Mlkai \. Kobayashi S: Human adenovirus-induced medulloepitheliomatous neoplasms in Sprague-Dawicv rats. AmJ Pathol 73:671-690. 1973 (28) -------- . Pritn.try hrjin and spm.il tord tumuts imimed bv human adenovirus "I vpe 12 in haimteis ) \euiopathol Exp Neurol 32:523-541, 1973 (29) Ml'kai N. Kobayashi S. Ogl'RI M: L'li'rasmictutal vtud* ies ol human adenoviruv-produccd reimolilasloma like neoplasm-, in Sprague-Dawlev rats. Invesi Opthalmol 13:59:4-601, 1974 (30) Mukai N, Mlrao T: Retinal tumor induction hv ocular inoculation ol human adenovirus in 3-dav-old lals, J Neuropathol Kxp Neurol 34 28-37, 1975 (J/i Kobayashi S, Mlkai V Retinohlasinmu-like tumors induced In human adenovirus tvpe 12 in rais, Cancer Res 34:1646-1651. 1974 (32) Seugman AM. Shear MJ: Studies in carcinogenesis VI11. F.xperiniental production ol brain tumors in mice with meihvlcholanthrene. Am J Cjncer 37 364395. 1939 (33) Druckrey H, Ivankovic S. Preussmann R. et al Selec tive induction ol malignant tumors ol the nervous svstem bv resorptive carcinogens. In The Experimen tal Biologs of Brain Tumors (Kirsch WM, GrovsiPaoletti E. Paolctti P, eds). Springfield. 111.: Charles C Thomas, 1972. pp 85-147 (34) Kirsch WM. Schulz D: The chemical induction ol brain tumors with carcinogenic hydrocarbons and miscellaneous agents. In The Experimental Biology of Brain Tumors (Kirsch WM. Grossi-Paoletti E. Paoletti P. eds). Springfield, HI.: Charles C Thomas, 1972, pp 1 HI-193 (35) Swenberg JA: Chemical induction of brain tumors. In Neoplasia in the Central Nervous System (Thompson RA, Green JR, eds). New York: Raven Press, 1976. pp 85-99 (36) KlEihues P, Iantos PL. Magee PN: Chemical carcino genesis in the nervous system. Int Rev Exp Pathol 15:153-232, 1976 (37) Zimmerman HM: The nature of experimental gliomas. Clin Neurosttrg 7:247-260, 1959 (38) Ikuta F. Zimmerman HM: Virus particles in reactive cells induced by intracerebral implantation of dibenzanthracene. J Neuropathol Exp Neurol 24:225-243. 1965 (39) Alexandrov VA. Napalxov NP: Carcinogenic effect of DMBA in rats exposed transplacentallv. Vopr Onkol 20:50-54, 1974 (Russ) (40) Joshi SR, Shenefelt RE, Rice JN: Transplacental carcinogenicitv of DMBA in the rat. Proc Am Assoc Cancer Res 15:27. 1974 (41) Druckrey H. I vankovic S. Preussmann R: Selective induction of brain tumors in rats with MNL\ Natur* wissenschaften 51:144. 1964 (Ger) (42) Svvenberc JA. Koestner A, Wechsler W: The induc tion of tumors of the nervous system with intravenous methylnitrosourea. Lab Invest 26:74-85, 1972 (43) Denuncer RH. Koestner A, Swenberg JA. An experi mental model for selective production of neoplasms of the PNS. Acta Neuropathol (Berlin) 23:219-228, 1973 (44) Denuncer RH. Koestner A. Wechsler W: Induction of neurogenic tumors in C3HeB/FeJ mice by nitro sourea derivatives: Observations by light microscopv, tissue culture, and electron microscopv. Int J Cancer 13:559-571. 1974 (45) Swenberg JA. Koestner A. Wechsler W. et al Differ ential oncogenic effects of methylnitrosourea. J Natl Cancer Inst 54:89-96, 1975 MODERN concepts in brain tumor therapy OLI 6791 SWLMSLKt. {!<> h asm Alt S. Dhukkly H I taiisplucc-ntate Er/cugutg maligner I uinoren dcs Nencrissstems. I Atlivlmirosoliarnsloil (ANH) dn BD Xl-Rutien. / Krebslorsch 7I-320-360. 1968 '47 i 5wF\'RERCi I A, Koestner A. Wechsler W. et al: (Juanti- i.nm aspects ol transplacental tumor induction with cihslnitrosnurea in rats. Cancer Res 32:2656-2660, 1972 (/i Craw J. Zeller WJ, Ivankovic S Entstehung son neurogenen Malignotnen bei den Nachkommen son Nprague-Dawlev Ratten nach perkutaner Applikatmn son AthsInitrosoharnstolt (ANH) wahrend der rr.ig/cit / Krebslorsch 81.169-172. 1974 t4^i Koestner A. Wechsler W, Denuncer RH: Induction <>l nersous svstem and Ivmphoid neoplasms bv feed ing thiesbold lesels ot methylnitrosourea and ethslmtrosoiirea precursors to aduli rats. Food Cosmet foxicol 13:605-609. 1975 (5bi Sssenberc. JA. Clenoenon N, Denlinger R, et al Se quential deselopment of ethvlmtrosourea-induced neurinomas: Morphology, biochemistry, and transplanubiliis. J Natl Cancer Inst 55 147-152, 1975 15/1 Prelssmasn R, Ivankovic 5, LandschL'TZ C, et al;, Carcinogene wirkung von 13 Aryldialkyltriazenen an BD-Ratten. Z Krebslorsch 81:285-310, 1974 <321 .Miller J A: Carcinogenesis by chemicals: An overview. (The G. H. A. Clowes Memorial Lecture). Cancer Res 30.559-576, 1970 (J3) Irving CC: Interaction ol' chemical carcinogens with DNA. In Methods tn Cancer Research VII (Busch H, ed). N'esv York: Academic Press, 1973, pp 189-244 (54) Goth R, Rajewskv MF: Ethylation of nucleic acids by ethylmtrosourea-l-l<C in the fetal and adult rat. Cancer Res 32:1501-1505, 1972 (55 I _______ Moles ill,ir and sellul.u mechanisms assos i.itcil s, nb pill's-s ait moitenesis in the rat ncisous 'ssicin bv EM ftbslalion ol nucleic acids and elimination late' ol silisl.itccl bases from the DN \ ol ditletent tissues / Ktehslorsch 82:37-64. 1974 iSai---------- . I'cMsisiente ol O-6-cthvlgujmne in rat-brain l)NA, (.miel.uion with nercous ssstein-specilic cuiunogenesis bs etbslniirosourea. Proc Natl Acad Sci USA 71:639-613. 1974 (57) KleihleS I'. Mar<,ison GP: Garcinogemcits ol .V-methvl,V-mtrosom ea Possible role of excision repair ol 0`mei li v lu iianim horn l)N A. J Natl Cancel Inst 53. 1839-I M|. 1974 (5#) Kleihue.s P. Mac.tE P\ Alkslation ol rut brain nucleic acids bv .V-niet hs l-.V-mtrosoiirea and methsl uieihunc'iilphniiute. J Neurochem 20:595-606. 1973 (59) Kleihue.s I'. Patzschke K. Marcison GP. et al Reaction of tnelhcl tnelbanesulphonate with nucleic acids ot lelul and newliorn rats in vivo. /. Krebslorsch x 1,273283. 1971 (60) Loveless A: Possible relevance ol 06 alkvlation ol deoxsgu.mosine lo the mutagenicits and carcinogenicitv ol niiisisamines and mtrosamides. Nature 223:206-207, 1969 <61) Kleihues I*. Magee PN. Austokxr J, et al: Reaction of N-methvl-.V-nitrosourea with DNA of neuronal and glial cells in vicci. FEBS Lett 32:105-108. 1973 (62) Kirtikar DM, Goldthwait DA: The enzymic release of 0*-methvlguanme and 3-methyladenine from DNA reacted with the carcinogen N-methyl-.V-nitrosourea. Proc Natl Acad Sci USA 71:2022-2026, 1974 <63) Lindahl T: New class of enzymes acting on damaged DNA. Nature 259:64-66, 1976 OLI 6792