Tobacco Documents Online

T!09271624

.Lntroc~uct~on There can be no doubt that viruses are a cause of cancer in certain animals.TM While the total process of oncogenesis may be a complex of many known factors such as chronic inflammation, hormonal influences, radiant energy, chemicals, x-irradiation, and the proper genetic constitutlon,a2"'~* the known virus-induced malignancies are the sole natu- rally occurring neoplasms of known and spe- cific cause.° Facts relating viruses to tumor formation have accumulated rapidly during the last 25 years, largely through improved tissue culture techniques?~'a~'~° electron microscopy?1'a7"~1 immunofluorescent microscopy, and ferrltin antibody procedures in electron microscopy.~2 Improved immunologic methods, the great ad- vances in genetics and protein chemistry, and the demonstration of infectious nucleic acids ~'~x are all responsible for provoking intense interest in the potential relationship between viruses and cancer in man. The purpose of this review is to trace the development of viral oncology which has led to the present ferment and to dehneate the evidence for viral carcinogenesis. Historical Survey A chronology of developments in viral on- co.iogy is given in Table 1. The initial accept- ance of viruses as biological entities came through the work of Iwanowski in 1892 and its confirmation in 1899 by Beijcrinck.~.~a After unsuccessful attempts to discover a bacterial etiology of cancer, Borrel, in 1903, first suggested that viruses might be a cause of neoplasia.~'~ The first demonstration of a tumor-induc- ing virus was given in 1908 by Ellermann and Bang with 2 varieties of the avian leuko- sis complexF Attempts to transmit solid tu- mors with cell-free filtrates failed until Rous in 1911 transmitted a breast sarcoma in Plymouth Rock hens.~ Soon myxosarcoma, Received for publication Aug. 13, 1962; accepted Sept. 11. From the ~£edidne Branch and the Laboratory of Viral 0neology, National Cancer Institute. C[ C¢ v! V f f t~ r T!09271625

T^a~ I.---.4 Chrondo.~ of Viral Oncoto~y 1033 ' 1~1 19~2 Avian leuk~Is agent Fowfsa~coma virus Human wen vkus Flht'ome v~us Papillom~ v~ua Fowl l~m~homa~ ~}llom~relnoma Mouse ma~nm~ ~rctnoma ~en¢ dls~ ~d ~l~on Purlfi~tfon o~ Shopo papilloma v~s Frog renal ~dnoma v~ Mudno leukemia vJ~ Avian ery~hroblost~ ~d my~ob~ ~ a~di~ ]'aro~ld tumor ~lr~ Natural ~nsm~lon of avon v~tl l~p~ms~Is v~ quantitative rolatlon~ip o[r~verabl~ ~m~ ~ to infecting d~o Chlorolvukcmla mutiny MuHne loukeml~ Marine leukemia v~us In vl~o ~lru~indu~d eytomall~a~y: d~¢~ anR qu~tl~t~n Dlsco,ary of mauso mammary carcinoma vk'ua Virus-like partle]~ In human [ouk~la ~nlmu~mont of tumor lnduct~n by aonon~genlc Cul~vatiou of ~rot{d tumor ~ and ~pt of~yoma Natu~l ~ansml~lon of ~l~ma M~lne leukom~ ~latlon of lnfc~lous DNA ~m ~lyom~f~ t~ue ~ukemf¢ ~tlvlty of fll~at~ from r~htfon.~d~ ~ukemla Addltlonal mm~e leuk~ osteosarcoma, angiosarcoma, and other con- nective tissue malignancies had also been transferred3°,~° It was observed with great interest that inoculation of Rous sarcoma virus into youn9 chickens produced, not typi- cal sarcomas, but a curious hemorrhagic dis- ease.mm Thus the importance of age in conditioning the host response to oncogenic viruses was demonstrated early. The Rous virus differs appreciably from many other tumor viruses later described in its ability to induce distinctive cancers at the site of inocu- lation. In 1938 Balduin Luck~ demonstrated a viral etioiogy fora renal-ceil carcinoma in the leopard frog.es.u The virus has been care- fully characterized by electron mlcroscopy?$ While producing only renal cancer in adult frogs, the agent can be adapted, after passage through tadpoles and newts, to ~roduce le- sions of the iris and cartilaginous tumors.~,¢* Whether or not lymphocystlc disease o~ fish represents a virus-induced new.growth re- mains in doubt.~.e~ Porter Although Sanarelli in I898 had shown that rriyxomatosis could be transferred among rab- bits, he did not use cell-free filtratesY° That mammalian neoplasms could be transferred with cell-free filtrates was demonstrated when Shape discovered the viruses responsible for rabbit fibroma and papitlomaYam In domes- tic and wild cottontail rabbits, cross-immunlty was shown between fibroma and myxoma in- fection. Later it was possible by inoculating benign cutaneous papilloma fragments into visceral organs to produce an invasive, metastasizing, and fatal new growth.¢aa~ Fur- thermore, when domestic rabbits with virus- induced benign cutaneous papillomas were observed for 4 to 7 months, malignant trans- formation occurred in a high proportion?* In doraeaic rabbits, papillomas developed after injection with cell-free filtrates, but the tumors could not be transmitted from one domestic rabbit to another. Virus was finally found on direct examination, and neutraliz- ing antibody could be detected which quanti- tatively paralleled the extent of papillomatous : T!09271626

involvement. Virus xvas not found in the squamous carcino.mas that developed from the benign lesions. That virus or its influence remained in the carcinoma was clearly shown by passing through 14 generations a squa- mous-cell malignancy from a lymph node me- tastasis, originating in a papilloma, and by showing antibody against the Shope virus in the serum of all cancerous animals,r* Occasionally a squamous carcinoma devel- ops from a papilloma in the ~ild cottontail rabbit.¢~'. These cancers also lack virus par- ticles, whi!e the preceding papillomas contain virus in abundance. Demonstration of a ma- lignant change associated with disappearance of visible virus was significant and led to the presumption that cancerous change might be associated with somatic mutation, Later, virus protein was purified5~'~° Research on virus-induced tumors in fowl was renewed in 1933 with the demonstration that avian lymphomatosis could be transmit- ted by cell-free filtrates,st The detailed pre- sent knowledge concerning the avain leul~oses is the labor of many workers.°'~,~,s~'~z One of the 3 principal diseases in this group is visceral lymphomatosis, a condition in which few malignant cells reach the peripheral cir- culation. Erythroblastosis and myeloblasto- sis, described as transmissible with cell-~ree filtrates much earlier,~¢ are the remaining 2 major members. A close similarity has been shown immunologically among the chicken tumor viruses,sz,o~ Detailed classifications of the member entities in this complex of dis- eases are available.~ Study of virus-determined mouse cancer was inaugurated by Bittner's discovery, in 1936, of an agent in milk responsible for mouse mammary carcinoma.~ Subsequent studies, largely the .wprk of Bittner and of Andervont,¢'os'~°s illus~rated the necessity f~ w a critical study of th~wirus' exact role. Al- ~though virus is usually required for mou ;c mammary carcinoma'to develop with high frequency, the recognition of the ~i~.~.ultane- ously indispensable ~actors o~ .g.~.~ic con- stitution and proper hormonal sta-:'t~has led to debate as to the primacy o~ vlrus2 "Bittner had been inclined to believe that hormdnal MRCHIVES OF I.'gTER;'<.dL MEI~ICIA'I£ factors are uppermost, the virus' role being only permissive. Others argue that the hered- itary and hormonal factors simpl.v augmea~t the action of virus,s Mouse mammary carcinoma virus was dis- covered while studies of the genetic and hor- monal factors were in progress and only fl~e importance of hormonal stlmnlation was known and inbred murine strains had been developed. Andervont has emphasized that the virus is not absolutel£ essential in all mouse mammary carcinomas2or This system clearly indicates the intricate compl~ity of the separate factors which probably operate to induce cancer.*°,ut A great impetus to research in viral on- cology occurred in 1951 when Ludwik Gross demonstrated that celi-Iree extracts ft~m AI( and C58 mice with spontaneously occuring leukemia produced leukemia when inoculated into newborn mice.*°s.*°~ In 1953, Gross ~o. *~ and Stewart **~.~u described parotid, sub- cutaneous, and adrenal malignancies induced by cell-free filtrates from AK leukemic mouse tissues. Subsequent work showed that a virus different from that which induced leukemia was responsible for these other neoplasms and that 2 rinses were present in the AK strain,as4*~-*a° Techniques of preparing an active virus ino~lum progressed to ~e extent that certain strains, 1-14 days old, developed leu~mia in over ~% of ~ses after a latent peri~ of 2~ to 4 months,xtv Ia t955 S~wartz and co-workers ~escribed acceleration of leukemia production in a strain with a high leukemia incidence, by using cell- free filtrates of brains from patients d~d leukemia,tu°'*~ Comparable findings have ~en reported by De~ng.*"s Schwartz and co-workers have also isolated a viral agent from the brains o~ Swiss and ~HeB mice with spontaneous leuk~ia that on serial pas- ~ge through mice induced lympho~rcoma in C3~ and Swiss strains.*za.:u* .Bergol'ts has re~rted the production of murine leukemia using cell-free human materials.~ur,*u* Gra~, working wi~ seve~l spontaneously occurring murine sa~omas of ~e reticulum-cdl type and the Ehfli~ ~rcinoma, isolated a factor whi~, u~n inoculation into new~rn mice, ~oL llI, Ma~, 19~ 86 V fil ag in in S~ R ot in o! di A S C S C T!09271627

:d- .iS- :as all o~ ~te ,K ng ed ~b- ed ~se us an :nt Y lte U.q i'~--C.'LV C F- R produccd myclogcnous leukemia in a high percentage of animals,t=.x~ Many of fl~ese neoplasms ~ere d~loroleukcmias. In 1956, Yriend, working ~th cell-free filtrat~ of Ehflich ~rcinoma, produced reticulum-cell neoplasm with ewthroblastosis in Swiss and DBA/2 mlee.~.ln° in contr~t to the ~H rccipient of the Gross leukemia virus, thymectomy does not protect tim Swiss strain from cancer induction by the Friend ageutJ~ The modifying effects of thymecto- my, splenectomy, and gonadectomy on the induction of routine virus leukemia have pro- voked much interest.~a~'t~o Moloney, in 1959, isolated a virus Sarcoma 37 which produced a l~nphoid leu- kemia in BALB/c and other mice.~a~4a~ Lieberman and Kaplan, using x-irradiation in mice, produced lymphoid tumors, cell-free ~tracts of which ~used leukemia in iso- logous strain mice?as Thus the concept that an ~ternal carcinogen might activate a latent oncogenic virus ~s advanced. Recently, additional murine leukemia vi- ruses have been reported by Rauscher,ta~ Stansly~~° and Bather.t~t A f~ture of the Raus~er vies leukemia, shared with some o~er murine vies-induced leukemi~, is intense viremla, making blood the best source of vies. So potent is ~is agent ~at a 10-¢ dilution still permits pr~uction of the dis- ease. T~ 2.--The Mu~nt Leukem~ ed Of :' " AK, 0~8 (spontaneous) (0ro~a vlru~) 883* Swiss, C3HeB (s~n~n~s) 101.7 VC ~ (8ehw~Sch~lman) 3d ~ variant of 8~~1~ 101.~. ~t *~ua (Manak~) S~m= I and II, 80V ce [ ~rclnom~ ~m~ a7 tS- ~ ~hrllch ~aoma (Friend v~) ~.~" in aS [ (Kaplsn-Ll~b~an) x ~ ~hrlkh ~o~ or Table 2 categorizes the known nmrine leu- kemia viruses. The isolation o£ 3 agents oi vaD:ing biological activity ~rom Sarcoma 37 is noteworthy. As these murine leukemia viruses are generally poor antigens and are similar morphologically, their fundamental distinctiveness must be established or rc/uted on other bases.~a~n43 With the discover7 fl~at C3H mice, inocu- lated with tissue filt~tes from AKR leukemic mice, oRen developed parotid, adrenal, and other dissimilar neoplasms, there arose the concept of a polyoma vi rus.~°4~a'~'t~a While potyoma virus and parotid carcinoma virus arc synonymous,t° the agent is frequently re- ferred to as the S.E. Polyoma virus after the exhaustive studies of Stewart and Eddy and their co-workers,aoa~o'~t The virus can be grown in emb~o and monkey kidney cell ~ltures.a~a~o.x~'~ An ~tremdy large va- rie~ of tumors has been produced, including carcinomas of the breast and tumors of kidney, thymus, m~othdium, subcu~neous connective tissues, thyroid, adrenal, buc~l mucosa, sweat glands, stomach, and various epide~oid ~rcinomas. The agent is remark- able in its relative lack of species and strain specificity requirements among rodents. Particles resembling "immature" vies but without demonstrable biolo~c activity have ~en descried in detail from a tmnsplan~ble mouse plasma-cell tumor?~aa6~ Nothing to Pargde Origin 81=e (m~,) Host(s) Histology Beferenco Ti09271628

suggest the presence of virus has been seen in normal plasma cells or in plasma ceils from 4 patien~ with'multiple myeloma.~° Grace, .-X'lirand, and their associates prepared cell-free filtrates from a number of human tumors and inoculated mice and tissue culture media. Iq'o tumors resulted in animals subseRuently receiving the tissue culture teriais, but new growths, particularly breast carcinomas, occurred in newborn mice in- jected directly with cell-free preparations of human cancers. Immunologic studies made it unlikely that the mouse tumor polyoma virus was implicated. It has been suggested a~8 that if there was nonviral oncogenic material in the human tumors, the dilution in tissue cul- ture might explain the failure to produce malignancies with this method. Since 1957, virus-like particles have been demonstrated in human leukemic tissues.7. 1,~-~7 While their causal relationship to leu- kemia remains to be determined, it is of interest that the~e qli'us-like particles are mor- phologically similar .to those seen in fowl and murine leukemia. Some of the evidence that cell-flee extracts of human leukemic tissue enhanced the development of leukemia in cer- taln mice has already been c, ted.~-°'*~2 De- Long was able to produce leul~eahta m an untreated strain of mice with a Iowineidence of leukemia by using cell-free materials from patients with acute lymphocytic, monocytlc, and myelocytic leukemias,t"s Bone marrow cultures were used as the inocula. No disease was produced with identically prepared fil- trates from normal human controls. While it is not possible to ~traw any conclusions about the exact pathogenesis of these leukemias, the observation that a filterable agent from human - leukemia can provoke murine leukemia cannot be ignored.~ Similar experiments on a larger scale appear indicated, partidularly us- ing primates as recipients for cell-free human leukemia filtrates. Epldemiologic investigations of ease clus- ters of human leukemia, are being expanded in an effort to gain ecologic evidence for a possible infectious etiology of this disease.~0. m Studies to provide corroborative or con- trary data concerning the association of virus .dRCH1P'ES OF 1NTERN/IL MEDICI;v'E particles with human leukemia arc needed in hrge numbers, for patients both in relapse and in clinical remission. Desoxyribonucleic-acid-rich inclusion bod- ies have been discovered in human rectal polyps x¢~ but not thus far in maliga~ant le- sions of the rectum. Whether thls phenome- non has anything in common with the Shope rabbit papilloma, in which virus is plentiful before but absent after malignant transfor- mation, remains to be seen. More electron microscopic studies on human benign neo- plasms with a propensity for malignant cl~ange are necessary. Although cancer has been reported to de- velop on the base of a virus-associated human condyloma accuminatum,0.~ the only estab- lished human neoplastic virus, is that respon- sible for common skin warts,x¢~ Although transmission to humans using cell-free ex- tracts has been demonstrated,x¢*'x¢o attempts to propagate the virus in tissue culture have met with difflculty.1~ There is evidence for virus presence in human laryngeal papil- lomas,~s°.ts~ and Dahmann has established a close relationship antigenically between these laryngeal papillomas and cutaneous warts,is° In addition, Haguenau has observed virus- like particles in a small number of human breast carcinomas,xs~ Stewart and Irwin studied concentrates from cell-free filtrates of human tongue papil- lomas, an embryonal hepatic tumor, urine from 3 children with acute lymphocytic leu- kemia, renal neuroblastoma, and Hodgkin's dlsease.~sa With these filtrates, focal prolif- erative changes were observed in tissue cul- ture with human amnion and embryonic cells. The effect was lost when cellular elements were entirely eliminated through careful fil- tration and centrifugation of the tissue culture fluids. However, perhaps it should be restated that cell fragments hrc necessary for the transfer of certain viruses, such as herpes zostcr, in tissue culture,tS~,ts~ The Tumor Viruses • The oncogenic viruses are not fundamen- tally different from other animal viruses.~, t~."~ The requirement for living cells, the Vol. 111, May, I96~ 88 T!0927"1629

d in ~pse lod- :ctal : |co ope lot- ton leo- ,ant de- ab- on- pts for ta ese tes ul- : .~l- ire ed he P'IRU.YF_.S--CANCER species or cell specificity with certain excep- tions, ability, for long latency, the importance of numerous conditioning factors, and the ability to exist in free or occult form are characteristic of all. Electron Microscopy.--Virus particles have been demonstrated now by electron mi- croscopy in the majority of neoplasms for which there is biological evidence of a viral etiology.°'aT"~x.t~2,1ssq~'~ Although the initial detection of characteristic virus-like particles was exceedingly laborious in some tumor sys- tems, the demonstration of Shope rabbit papilloma virus, some strains of mouse mam- mary carcinoma agent, and avian myeloblas- tosis virus were exceptions. Although the ~Rous sarcoma virus particles were demon- strated early, subsequent efforts at identifica- tion were tedious,tga.xca Thus it has been with many of the known tumor viruses. Among the smallest virus particles are those of polyoma. These are very dense bodies roughly 30 my in diameter and pos- sessing central nucleold and a single closely adherent surrounding membrane. They are most often located in the host cell nucleus. The Shope papilloma virus is morphologically similar to polyoma virus. The spherical virus particles seem in human warts have an aver- age diameter of 52 rap. when tightly packed and arranged in characteristic crystalline hex- agons,ma~ The Rous sarcoma virus has an external menabrane, a centrally-located nude- old, and an average diameter of 75-80 Two distinct morphologic types of virus have been described in mouse mammary car- cinoma?ss Termed Types A and B by Bernhard, the particles have average dia- meters, respectively, of 70 and 105 m~,. A- type particles are doughnut-shaped with a double membrane system, no central nucleoid, and are predominantly cytoplasmic. B-type particles possess an eccentric nudeoid and are more often located extracellularly. The majority of the avian and murine leu- "kemia viruses are anatomically similar. A centrally placed nucleoid and a single mem- brane are commonly described £eatures. The viruses of avian myeIoblastosis and erythro- blastosis have average diameters of 120 and i02 m/~, respectively, while the visceral lym- phomatosis virus is considerably smaller with a diameter of 72 In a separate size-range category altogether must be placed rabbit fibroma and myxoma viruses. Approximately 220 m~. in diameter, they are among the largest oncogeuic viruses known,a~ Detailed studies of the morphologi¢ characteristics of the known tumor viruses are now available,as,~.xss,~.x°~ Dalton and colleagues have made anatomi- cal studies of the routine leukemia viruses.4z Particles abound within megakaryocytes. While most partldes average from 60 to 150 m~. in diameter, smaller particles are occa- sionally identified as in the Moloney virus routine leukemia. Twenty per cent o[ the megakaryocytes from peripheral blood of rats with Moloney virus-induced leukemia tained particles, and virus was observed free in the plasma. This association of central nudeoid and circumferential membrane sys- tem resembles no native cellular structure. While the interpretation of foreign intracel- lular particles as viruses must be made very cautiously, there is as yet no other satisfy- ing or permissible hypothesis. In a number of tumors for which no biological evidence of viral origin exists, structures morphologically similar to those observed in the known virus-induced neo- plasms have been described3z In those situa- tions in which particles are seen with the electron microscope but show no biological activity, their significance remains uncertain. There is no justification, however, for postu- lating a coincident or fortuitous presence of such particles when biological activity has been proved as in the murine leukemias. Con- sidering the lack of any cytologic lesion in tissue culture and the poor antigenicity of the murine leukemia viruses, the continuing importance of electron microscopy and bioas- say of cell-free filtrates is obvious. The pres- ence of particles and the high biologic activity of these filtrates in inducing murlne leukemia constitutes the best "proof" of cause and effect presently available. Mode of Vim Replication.~A$.though the exact mode of intracellular replication of 8~ i 89 Porter T109271630

---

A'E virus remains a central problem in virology, electron microscopic studies which clearly show a process of budding from the cellular membrances represcnt a great achievement. zn0.a~ This plmnomenon of buddinghas been demonstrated for a nutnber of the known on- ¢r~genie vh'uses, Including those responsible for certain murine leukemias.4x In the case of the latter, virus particles are abundant in megakaryocytes where they can be seen to bud from membranes lining cytoplasmic vac- uoles.~ Budding has also been demonstrated in the particles resembling viruses that have been seen in transplantable plasmacytomas, but these bodies are as yet ~ithout known biological activity.~°~,~ It has been pointed out that partldes have not been demonstrated at the exact time of entry into or release from human cancer cells. Regardless of how entrance is achieved, the combination of cytoplasmic filaments and the cell border effects a "new" particle through budding. This process is first recognized as a dense thickening and protrusion on the cell. surface. Edwards' studies reveal that, while still attached to the cell surface, the particle exhibits reorganization of internal material, suda that a distinct nucleo~dal area becomes visible. This central density is encompassed by filaments which create an outer border with as many as 3 or 4 membranes,z~a Final- ly the particle is detached from the cell sur- face to become free of direct connection with the cellular cytoplasmic membrane of the host. Stages in this process of virus particle bud- ding are illustrated in the Figure. Whether the process of budding will prove a general phenomenon throughout the entire spectrum of the tumor viruses is uncer- tain.~r,~s At present, evidence for similarity in structure and in mode of emergence from the host cell is increasing.~,~'x°~'a~'a~'~'~°°'~°z Recognition of the ability of virus partldes to emerge directly from normal-appearlng cells has added a new dimension to the study of vlrus-host rel~tionshlps. Chemical Compodt~on.--Purification of the tumor viruses has not reached as yet the level of achievement known for plant viruses and certain other animal viruses. 90 i 91 Porter ~Vhile Rous virus preparations are difficult to puri~y,7n highly purified yiehls of rabbit papilloma and avian nayeloblastosis viruses have been obtaincd.~'s~ Avian myeloblastosis virus coutains appreciable lipid and is un- expectedly loxv in nucleic acid content. A high adenoshm trlphosphatase (A'i'l'a~) activity distinguishes this virus from the erythroblastosls agent.~°'~'-"°~ All of the umr- ine leukemia and fowl viruses thus far studied are apparently of tim ribonucleic acid (R~A) variety. Characterization of the physical and chemical properties of tbe oucogetfic viruses has far to go, and much fundamental work remains to be done. Conditioning Factors ~or Viral Oncogenesis It is plain that even in those conditions in which an obvious relationship between virus and tumor exists, the determinants of genetic constitution, hormonal status, age, immunity, dose, and presence of other viruses are cer- tainly multiple and probably interrelated. Genetic and hormonal factors strongly influ- ence susceptibility to the mouse mammary carcinoma virus,~r'~°'~ and chickens vary con- siderably in individual resistance to the avian lymphomatosis vlrus2z Breeding for in- creased resistance to avian lymphomatosis has been achieved2~ Also there is evidence among fowls for a heritable factor responsi- "ble for nonsusceptibility to the Rous sar- • coma virt~s.2°8 Although the absolute time limits can be modified, the susceptibility of mice to cer- tain leukemia viruses is markedly influenced by age. It has been mentioned that depend- ing on the age of birds, either typlcal tu- mors or an atypical hemorrhagic condition results on exposure to the Rous virus.~ Younger animals are generally more sus- ceptible to infection by an oncogenic virus. However, there are exceptions; erythroblas- tosis virus is more pathogenle for older chickens.~ In general, the estimated virus dose and the percentage of animals developing tumors are usually proportional. Since there is evl- denee that many animal strains carry virus T109271632

i ~80 without displaying neoplasia in that partic- ular generation, it has been postulated that the quantity o7 virus is only greater in those strains exhibiting malignancy,t°.lss Even strains with a low tumor incidence can be induced to show mouse mammary carcinoma if a high infective virus close is used.~° Evi- dence is abundant that such tumor systems as mouse mammary carcinoma, the avian leukoses, and polyoma provide examples of animals carrying virus throughout life and not outwardly showing the lesions in ques- tion.~'~°a.:°r This is particularly true if the inoculating doses are small,is° Such obser- vations support the concept that the circum- stances of tumor formation are so highly specific that only a fraction of animals carry- ing virus may develop cancer. The impor- tant application of quantitative techniques which related the amount of virus recover- able to the infecting dose showed that the absence of demonstrable virus in tumor tis- sue extracts was no justification for assum- ing the lesion to be of nonviral origin2~a Riley has recently demonstrated an inter- esting ~iru.~-tumor .¢ynergi~m.~°~ In mice inoculated with both a filterable virus and virus-free tumor, synergism was demon- strated by accelerated tumor growth and by elevation of host plasma lactic a.cid dehydro- genase concentration. This elevation i~ zyme concentration was significantly gre;::er than that produced by tumor or virus This ability of virus-like agents to influence the biological and biochemical character of a host or a tumor may prove extremely important. Also, it is well known that virus- induced tumor cells support the multiplica- tion of other viruses. Newcastle disease virus and vaccinia virus in Rous sarcoma cells are exam.pl.e~s.~° The extent to ~'hich the presence of mul- tiple carcinogens :i~fluences the induction of malignancy is als~ of importance. The co- carcinogenic effect of viral infection has been shown.~ In one sense, virus sb~..uld be con- sidered a cocarcinogen for.tb, e p$6.duction of mouse mammary ¢arcmor~,',~:lthough Rs presence is not absolutely necessary in all cases.~°~ ~Intradermal injections o[ vaccinia ~qRCHIVF~ OF INTERNAL MEDICLVE virus into mouse skin painted with methyl- cholanthrene, combined with injections of corticosteroids, produce cancer at the site of injection.~ Chemicals, irradiation, end other mutagenic stimuli are able to produce asymptomatic changes that become manifest only when a second proliferative stimulus is applied.~o.:a Naturally, the possibility is real that chemical and physical agents may acti- vate latent neoplastic viruses as may be the case with the irradlatlon-induced routine leukemia.~as Virus-like particles have also been discovered recently in cells from can- cers originally induced by methylcholan- threne.:au In connection with the "masking" 'of tumor viruses within their hosts, one re- calls that Rous sarcoma virus is not recov- erable from the sarcomas when the infective virus dose is small,ta°,u*~ A critical review of the entire concept of virus "masking" has been furnished by Beard.~* The evidence suggesting that tumor growth is augmented by the presence of virus re- gardless o~ whether or not virus caused the tumor has already been cited.:~ It should also be mentioned that some animal viruses have a predilection for destroying tumor ceils. Certain neurotropie viruses, such as the West Nile and Bunyamwera, exhibit an oncolytic action on transplanted routine neo- plasms.:~s.:~ Although previous attempts to treat human cancer with viruses have not met with any regular success or enthusi- asm,zt¢'2t8 YIowie and Crosby have recently reported encouraging results in human lyre- photon patients using attenuated Venezuelan equine encephalitis virus preparations.~-~s~ It is not clear to what extent natural or artificially acquired immunity influences the response to oncogenie viruses. Immunologic study of a number of the tumor viruses has been undertaken nonetheless. The Rous agent possesses an excellent antigen, and very detailed studies are available.~to Mouse mammary carcinoma virus ~s apparently not antigenic, and the group of murlne leukemia viruses are also poor antigens• Rowe, with Huebner and associates, has examined the serologic characteristics o~ polyoma virus- induced int~ectlon~¢'~°.~n A viral hemag- Vol. III, May, 196] 92 T!09271633