ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 13, No. 3 Copyright © 1983, Institute for Clinical Science, Inc.

Viruses and Human Cancer JOHN P. MANOS, M .D. and ERNEST M. WALKER, JR., M .D., P h .D. Department of Laboratory Medicine, Medical University of South Carolina, Charleston, SC 29425 ABSTRACT The cause of human cancer is probably multifactorial and the role of viruses is unclear. The study of retroviruses has led to the identification of oncogenes responsible for transformation and tumor induction. Human viruses associated with malignancies include the JC virus (associated with progressive multifocal leukoencephalopathy) and some adenoviruses. No human malignancies have been associated with the latter group. A number of herpes viruses of lower animals have been associated with malignancies and herpes simplex virus type 2 has been associated with carcinoma of the cervix and vulva. The Epstein-Barr virus has been associated with Burkitt’s lymphoma and nasopharyngeal carcinoma. Some circumstantial evidence suggests that cytomegalovirus may be associated with Kaposi’s sarcoma among homosexuals. The hepatitis B virus (HBV) has been associated with hepatocellular carcinoma. The fullfillment of Koch’s postulates presents ethical problems regarding man and proving the viral etiology of human malignancy. Social experiments may elucidate some of these questions. An increase in venereal herpes should be associated with an increase in carcinoma of the cervix and use of the HBV vaccine in populations with high incidences of HBV carrier states should decrease the incidence of hepatocellular carcinoma. The investigation of the retroviruses though not establishing viruses as causing human malignancies at least will improve our understanding of the malignant process. That a particular hum an cancer is caused by a virus is yet to be proven. The cause of human cancer is probably m ultifactorial in th at environm ental, physical, and chemical stimuli, with per­ haps specific genetic factors in the right combination, can bring about malignant changes.151622 If and how viruses or viral products play a role in the development of human malignancy is under intense in­ vestigation. It is well established that a num ber of viruses of lower animals are

able to cause malignancies in the natural host or in other species.615 It would be ludicrous to expect that viruses do not play a role in the development of human cancer. The causal relationship betw een vi­ ruses and human cancer is difficult to es­ tablish and the fullfillment of Koch’s pos­ tulates to prove the infectious etiology of some cancers remains difficult if not im­ possible. Intensive investigation at­ tem pting to dem onstrate the causal re­

201

0091-7370/83/0500-0201 $00.90 © Institute for Clinical Science, Inc.

202

M ANOS A N D W A LK ER

lationship betw een viruses and hum an cancer provides only circumstantial evi­ dence. The identification of virus parti­ cles, virus precursors, or virus specific nucleic acid from biopsies or cultured tu­ mor cells of human origin does not prove that a virus has caused the malignancy. A significant problem with this approach is that the virus may merely be a passen­ ger or a contam inating virus.6 The at­ tem pt to induce tumors in animals by the inoculation of human tumors or extracts of human tumors, in an effort to dem ­ onstrate a candidate human virus, is lim­ ited. Animal viruses natural to the host injected may have been introduced. Hu­ man viruses are able to cause malignant transformation in vitro and are potential suspects as hum an oncogenic viruses. Significant inferences have been ac­ quired from seroepidemiological surveys in an attem pt to correlate viral antibody with tumor incidence, examples being the Epstein-Barr virus (EBV) and its associ­ ation w ith B urkitt’s lym phom a8 24 and herpes simplex virus type 2 (HSV-2) and carcinoma of the cervix.6’91516 A num ber of anim al and hum an vi­ ruses have been associated eith er di­ rectly or indirectly with oncogenesis and malignancies. The retroviruses (oncor­ naviruses) are a large heterogenous group of small ribonucleic acid (RNA) containing viruses that infect many animal species (possibly including man) and in many in­ stances are proven to be oncogenic in their natural host or in other animal spe­ cies,1-2 Among the human viruses which have oncogenic potential at least in other animal species are the papoviruses,61314 the adenoviruses,61516 and some herpes­ viruses.691516 An important characteristic of many tumor causing viruses is the abil­ ity to cause transform ation (i.e., the transforming of a normal cell into a ma­ lignant state).6-91516 C haracteristics of transformed cells include the capability to grow indefinitely in culture, the loss of contact inhibition, the alteration of the host cell membranes and, frequently, the

capacity to induce tumors in laboratory animals. One of the most studied group of vi­ ruses w ith oncogenic potential am ong many of its m em bers are the retro v i­ ru se s.12 These viruses contain single stranded RNA and range from 80 to 110 nanometers in size. They contain deoxy­ ribonucleic acid (DNA) polym erase (a reverse transcriptase) which causes the formation of double stranded DNA which is produced from a RNA template. This is a reversal of the usual direction of genetic information which is DNA to RNA. This DNA intermediate is then in­ tegrated into the host genome which sub­ sequently may lead to neoplastic trans­ formation. This new characteristic is per­ p etu ated to the progeny of the tran s­ formed cell. The mechanism by which this occurs is poorly understood. One of the well known viruses in this group, the Rous sarcoma virus (RSV) has been studied extensively.12 The RSV virus is known to contain four genes, one of which has been term ed the src (for sarcoma) gene or oncogene which carries the in­ formation needed for transformation. Ev­ idence suggests that the src gene pro­ duces a kinase which is important in the phosphorylation of tyrosine which is in­ volved in cell glycolysis. Since tran s­ formed cells make 50 times as much viral kinase as normal cells there is an increase in glycolysis in the transform ed cell. There are some 15 oncogenes which have been described among the various retro­ viruses.223 Similar evidence of an src like gene has been found in vertebrate cells, including cells from a hum an bladder carcinoma.12’22 In vertebrates these cel­ lular type src genes are called c-src or ‘cell oncogenes’ as opposed to v-src for ‘viral oncogenes’. The c-src genes are similar in size and in chemical structure to their viral counterparts. Both catalyze phosphorylation of tyrosine and both ap­ parently are tightly bound to the plasma membrane of cells.12 One hypothesis as noted previously is

203 viruses. The case with human viruses and proven causation of cancer in man con­ tinues to be elusive; however, a num ber of human viruses have been associated indirectly with malignancies although, as of this writing, there is no definitive ev­ idence that any virus is, in fact, the cau­ sation of cancer in man. The papova group of viruses have a number of member viruses that can cause transformation and induce tumors in lab­ oratory animals.6-9 Included in this group of viruses is the human papilloma virus (wart virus) which is a tum or virus; how­ ever, it is not associated with malignan­ cies. The JC virus is associated with pro­ gressive multifocal leukoencephalopathy (PML) and occurs in patients with im­ m unosuppressive disease secondary to malignancies such as Hodgkin’s disease. This virus, which has been isolated from a num ber of patients, induces brain tu­ mors in new born ham sters, histologi­ cally resem bling m alignant glioblasto­ mas.14 It apparently is an ubiquitous vi­ rus among humans since antibody to this virus is present in up to two-thirds of children by the age of 14 years. Condy­ loma acuminata (venereal warts), caused by an antigenic variant of the human pap­ illoma virus, has been demonstrated to undergo m alignant transform ation.13 Other viruses in the papova group such as the polyoma and the SV40 viruses have been shown to produce cancers in a va­ riety of laboratory animals.6 A num ber of human adenoviruses are also known to cause a variety of malig­ nancies in new born rodents, the most oncogenic being types 12, 18 and 31. However, there is no evidence that ad­ enoviruses are oncogenic in man.6 T here are several herpes viruses of lower animals that have been definitively associated with malignancies.6-15 Marek’s disease is an infectious malignant lymphoproliferation in fowl and has recently been controlled by an attenuated viral vaccine. The Lucke frog virus is known to be the etiological agent of renal ad­

VIRUSES AND HUMAN CANCER

that the c-src gene is a normal and nec­ essary part of the cell and is involved in the normal growth and developm ent of the cell. However, under proper stimu­ lation, that is by a carcinogen, be it chemical, radiation, or possibly even a non-oncogenic virus, the activity of the c-src gene is augm ented and results in an inappropriate expression resulting in un­ controlled or cancerous growth. This the­ ory is referred to as the dosage hypoth­ esis.1-2 An alternative hypothesis is that viral oncogenes differ from their cellular progenitors in subtle but important ways as a result of m utations w hen cellular genes were copied into the retrovirus genomes possibly through viral infection early in evolution. Thus, the v-src and csrc oncogenes might have different tar­ gets in the cell and, therefore, different effects on cellular behavior.1-2 The find­ ing of oncogenes similar to those in ret­ roviruses in human cells, however, is difficult because of the enormous size of the DNA in the human genome. Recent reports have shown serological evidence of antibody to a human retrovirus as­ sociated with human T-cell lymphoma leukemia virus (HTLV) in a cluster of Japanese with adult T-cell leukem ia.17 Antibody was not found in those with non T-cell leukemia, in healthy donors in the endemic area and in random donors. Do oncogenic viruses have to have on­ cogenes? A pparently not, since some known tumor viruses lack an oncogene. They do have somewhat different char­ acteristics than those tum or viruses pos­ sessing oncogenes in that the former act slowly in animals, taking several months for a tumor to develop, and do not have the characteristic of causing transforma­ tion of normal cells in vitro J-2 Do human viruses, which have at least circumstantially been associated with hu­ man cancer, have oncogenes such as EBV and HSV-2? Both are transform ing vi­ ruses; therefore, oncogenes should be part of the genome. Many investigators are trying to identify such genes in these

2 04

M ANOS A N D W A L K E R

enocarcinoma in frogs. Herpes viruses of lower primates are also associated with carcinoma, including herpesvirus saimiri which has been isolated from healthy squirrel monkeys and herpesvirus ateles isolated from healthy spider monkeys. Neither of these viruses causes cancer in their respective natural hosts but both induce m alignant lymphomas in other monkey species. Apparently the host im­ m une response plays a crucial role in preventing the host from developing the malignancy. A common sexually trans­ mitted disease among humans, HSV-2, has been, in the past several years, as­ sociated with carcinoma of the cervix; however, most of the incriminating evi­ dence has been seroepidemiological.6'915 Genital HSV and cervical carcinoma fre­ quently have been found to occur in sim­ ilar groups, and antibody to HSV-2 has been found more often in females with cervical carcinom a than in healthy fe­ males. Antibodies to other veneral dis­ eases do not correlate with cervical car­ cinoma as do HSV-2 antibodies. The oncogenic potential of HSV-2 has also been dem onstrated by the fact that it can cause transform ation of normal animal and human cells in vitro and these cells in turn have induced tumors in new born laboratory anim als.915 There have also been recent reports demonstrating HSV2 type antigens in association with squa­ mous cell carcinoma in situ of the vulva (9 of 10 patients in one reported series).1112 The rise in squamous cell carcinoma of the vulva and genital HSV-2 suggests an association. On the other hand, this also may merely be a reflection of an oppor­ tunistic infection by an ubiquitous agent among man. The EBV is one of the most common viruses that affects man.6'81520 This virus selectively infects the B lymphocytes of man and, like HSV-2, causes in vitro transformation. Transformed cells by the EBV show oncogenicity when inoculated into monkeys. There is also significant

seroepidemiological evidence associating the EBV with malignancies, including B urkitt’s lymphoma in Africa24 and na­ sopharyngeal carcinoma in the oriental male and with infectious mononucleosis in th e U nited S tates.6-815-20 Recently, there has been an association of the EBV w ith o th er m alignancies, such as lym ­ phomas following renal transplantation7 and with acute leukemias associated with infectious mononucleosis.19 It is interest­ ing that of the herpesviruses m entioned only the EBV and HSV-2 have the ability to transform normal cells in vitro.15 All of the viruses in the Herpes group are able to induce tumors by the isolated virus, except for HSV-2, with results for the lat­ ter virus being equivocal.15 Another m em ber of the Herpes group of viruses, the cytomegalovirus (CMV) has received much attention in the re­ cent literature regarding its association with Kaposi’s sarcoma occurring in an al­ most epidem ic form among the hom o­ sexual population.5 This sarcoma tends to occur in those individuals who have what has recently been term ed an ‘acquired im m unodepressive syndrom e’ (AIDS). Although other chronic infections also occur in the hom osexual population, especially Pneumocystis carinii induced pneumonitis, CMV apparently is one of the common threads that has seroepidem iologically been associated w ith pa­ tients having Kaposi’s sarcoma. Infection with CMV has been shown to cause im­ m unosuppression and its genom e has been detected in Kaposi’s sarcoma tis­ sue.5 The virus, after ultra violet irradia­ tion, can cause transformation of normal ham ster fibroblasts into tumor cells.10 Finally, the Hepatitis B virus (HBV) has been associated with hepatocellular carcinoma (HCC).3'4'61218 In a recent study investigating 20 patients who died with alcoholic cirrhosis and HCC, all had HBV DNA integrated into the genome of the neoplastic liver cells, as demonstrated by hybridization studies, although only 9 of

205 factor. The administration of the newly developed HBV vaccine to populations with high incidences of HBV carrier states should, in time, demonstrate a decrease in the incidence of HCC in these popu­ lations if HBV is a major factor in the development of this malignancy.3 Since there appears to be an increase in ve­ nereal transm itted HSV-2 world wide and since it has been incriminated as a possible causation of cervical carcinoma, there should be an increase in cervical carcinoma in the next decade or two if this malignancy is associated with HSV-2.15 The intense investigation of the tumor producing retroviruses has revealed much in the understanding of the malignant cell in lower animals and in man. Continued investigation in this area, although not revealing a definitive causation between viruses and human malignancies, will at least continue to give us a better under­ standing of the malignant process itself.

VIRUSES AND HUMAN CANCER

16 of these patients whose blood was available were found to contain serolog­ ical markers of HBV infection.4 The oc­ currence of HCC tends to correlate somewhat with the incidence of chronic carriers of HBV. In areas such as the United States where the incidence of the carrier state is low so is the incidence of HCC. Ongoing epidemiological studies of oriental male HBV carriers over the past two to four years have shown that 50 cases of HCC have occurred during this period, and all but one occurred in chronic carriers. The investigators, from following this group to date, estimate that the relative risk of HCC is more than 250 times greater in carriers than in non-carriers.3 Again, HBV may only be one of many hepatotoxic agents that may even­ tually lead to HCC and that HBV is, in fact, a cause of H CC rem ains to be proven. There is at present no conclusive evi­ dence that viruses cause human tumors. However, with the knowledge gained from the study of oncogenic viruses of low er anim als and w ith a num ber of H erpes viruses in man being strong sus­ pect, it seems inconceivable that at least some human malignancies are not caused directly or indirectly by viruses. It is probable, as many feel, that the cause of human malignancy is multifactorial and that a variety of circumstances, such as genetic predisposition, the state of the im m une system , environm ental expo­ sure and the presence of certain viruses, all must occur at the right time and the right place in order for a malignancy to develop. The fullfillment of Koch’s pos­ tulates presents ethical problems regard­ ing man and definitive proof of the viral etiology for human cancer must come by other indirect or possibly natural mech­ anisms. The use of an attenuated virus vaccine in successfully eradicating Marek’s disease in fowl is well docum ent­ ed .15 Analogous vaccines for humans is, of course, complicated by the safety

References 1. Bishop, J. M.: The molecular biology of RNA tu­ mor viruses: A physician’s guide. New Eng. J. Med. 303:675-682, 1980. 2. Bishop, J. M.: Oncogenes. Sci. Amer. 246:8092, 1982. 3. B lumberg, B. S. and L ondon, W. T.: Hepatitis B virus and the prevention of primary hepato­ cellular carcinoma. New Eng. J. Med. 304:782784, 1981. 4. Brechot, C., N alpas, B., C ourouce, A-M, D u hamel, G., C allard, P., C arnot, F. T iollais, P., and Berthelot, P.: Evidence that hepatitis B virus has a role in liver-cell carcinoma in al­ coholic liver disease. New Eng. J. Med. 306:1384-1387, 1982. 5. D urack, D. T.: Opportunistic infections and Ka­ posi’s sarcoma in homosexual men. New Eng. J. Med. 305:1465-1467, 1981. 6. F enner, F. J. and W hite, D. O.: Oncogenic vi­ ruses. Medical Virology, 2nd ed. Fenner, F. J. and White, D. O. eds. New York, Academic Press, 1976, pp. 162-181. 7. H anto, D. W., F rizzerà, G., G ajl-P eczalska, K. J., Sakamoto, K., P urtilo, D.T., Balfour, H. H., Simmons, R. L., and N ajarían, J. S.: Epstein-Barr virus-induced B-cell lymphoma after renal transplantation. New Eng. I. Med. 306:913918, 1982. 8. H enle, W., H enle , H., and L ennette, E. T.:

206

9. 10. 11.

12. 13. 14. 15.

VIRUSES A N D H U M A N C A N C ER

The Epstein-Barr virus. Sci. Amer. 241:48—59, 1979. HlRSCH, M. S.: Herpes simplex virus. Principles and Practice of Infectious Diseases, 1st ed. Mandell, G. L., Douglas, R. G., and Bennett, J. E., eds. New York, Wiley & Sons, 1979, pp. 12831294. Ho, M.: Cytomegalovirus. Principles and Prac­ tice of Infectious Diseases, 1st ed. Mandell, G. L., Douglas, R. G., and Bennett, J. E., eds. New York, Wiley & Sons, 1979, pp. 1307-1323. Kaufman, R. H., D reesman, G. R., Burek, J., Korhonen, M. O., M atson, D. O., M elnick , J. L., P owell , K. L., P urifoy, D. J. M., C ourtney, R. J., and Adam, E.: Herpesvirus-in­ duced antigens in squamous-cell carcinoma in situ of the vulva. New Eng. J. Med. 305:483-488, 1981. Kew , M. C., H odkinson, J., P aterson, A. C., and Song, E .: Hepatitis-B virus infection in black children with hepatocellular carcinoma. J. Med. Vir. 9:201-207, 1982. Kovi, J., T illman, R. L., and L ee , S. G.: Ma­ lignant transformation of condyloma acumina­ tum. Amer. J. Clin. Pathol. 67:702-710, 1974. L E H R IC H , J. R.: Polyomavirus. Principles and Practice of Infectious Disease. 1st ed. Mandell, G. L . , Douglas, R. G., and Bennett, J. E., eds. New York, Wiley & Sons, 1979, pp. 1436-1441. Rapp, F. and W estmoreland, D.: Do viruses cause cancer in man? Cancer. 25:215-229, 1975.

16. Rapp, F.: Viruses and cancer:Update. Cancer 28:356-361, 1978. 17. Robert-G uroff, M., Nakao, Y., N otake, K., Ito, Y., Sliski, A., and G allo, R. C.: Natural anti­ bodies to human retrovirus HTLV in a cluster of Japanese patients with adult T cell leukemia. Sci­ ence 215-,975-978, 1982. 18. Robinson, W. S.: Hepatitis B virus. Principles and Practice of Infectious Diseases, 1st ed. Man­ dell, G. L., Douglas, R. G., and Bennett, J. E., eds. New York, Wiley and Sons, 1979, pp. 13881423. 19. Rosner, F. and G runwald, H. W.: Infectious mononucleosis and acute leukemia. J. Amer. Med. Assoc. 246:1783-1784, 1981. 20. Schooley, R. T. and D olin, R.: Epstein-Barr vi­ rus. Principles and Practice of Infectious Dis­ eases, 1st ed. Mandell, G. L., Douglas, R. G., and Bennett, J. E., eds. New York, Wiley and Sons, 1979, pp. 1324-1340. 21. Schwartz, P. E. and N aftolin, F .: Type 2 herpes simplex virus and vulvar carcinoma in situ. New Eng. J. Med. 305:517-518, 1981. 22. SHODELL, M.: The intim ate enemy. Science 82 3:46-51, 1982. 23. Varmus, H. E.: Form and function of retroviral proviruses. Science 2i6:812-820, 1982. 24. Z iegler, J. L.: Burkitt’s lymphoma. New Eng. J. Med. 305:735-745, 1981.