CA: A Cancer Journal for Clinicians, Vol 25, 215-229, Copyright
© 1975 by American Cancer Society
Do Viruses Cause Cancer in Man?
Fred Rapp Ph.D.1 and
Diana Westmoreland B. Phil.2
1 Professor and Chairman, Department of Microbiology, the Pennsylvania State University College of Medicine, Hershey, Pennsylvania.
2 Fellow of the International Agency for Research on Cancer, the Pennsylvania State University College of Medicine, Hershey, Pennsylvania.
There is a good deal of circumstantial evidence linking RNA- and DNA- containing viruses with human cancer. Yet, despite the expenditure of many man-hours and millions of dollars, not a single human cancer has a proven viral etiology. Why has so little success been achieved? Why is it difficult to find a human oncornavirus? If most candidate human tumor viruses are ubiquitous, why do the majority of people not get cancer?
Because it is impossible to induce tumors in man, the study of human neoplasia has been somewhat hampered. Indeed, the first priority of a clinician is to destroy the tumor as quickly and completely as possible. Prior to the development of colonies of inbred mice and the use of newborn animals for experimental inoculation, virologists were also unable to find oncogenic viruses in mammals. However, "wild" oncornaviruses have recently been isolated from outbred animals such as the cat and it is highly probable that, despite the difficulties involved, human oncornaviruses will soon be identified. But whether any such virus will ever be directly proven oncogenic in man is doubtful. Perhaps the best proof of a viral etiology of cancer would come from the development of anti-virus vaccines. The example of Marek's disease suggests that, for her-pesviruses at least, the possibility of vaccine development is not remote, although preliminary studies have not been encouraging. Once again, ethical restraints on the use of intact killed or attenuated virus vaccines should be considered.
It is probable that viruses involved in cancer are frequently defective, and virus genetic material within the malignant cell may consist of only a few "transforming" genes, representing a fragment of the whole genome. This might explain why it has often been difficult to detect complete virus or virus products in both RNA and DNA transformed or tumor cells. To circumvent this problem, many laboratories are now using more sensitive techniques for detecting small amounts of viral RNA and DNA within cells, and for examining the transforming potential of viral nucleic acid fragments.
Although exposure to most candidate human tumor viruses appears widespread, many people never develop cancer, and many more do not have the disease until late in life. In fact, evidence of human oncornavirus information has been found in normal breast tissue, and infection with EBV and herpes simplex occurs in most individuals tested. If these viruses do play a role, other factors must be operating to control oncogenic expression. Genetic predisposition to cancer has been clearly shown in mice and chickens for both leukosis and Marek's disease. Examples of genetic factors in human cancer include the high incidence of nasopharyngeal carcinoma among Chinese in different environments, and the susceptibility of patients who are found to have certain chromosome abnormalities to leukemia.
Whether or not cancer develops following viral infection probably depends on the immunologic response of the infected animal. Such is the case in Marek's disease, Burkitt's lymphoma and monkey lymphomas. The importance of the immune response is also seen by the increased incidence of cancer in people who, whether due to age, disease or immunosuppressive therapy, have reduced immunocompetence. In addition, environmental hazards, such as irradiation and exposure to carcinogens, may activate the oncogenic potential of latent viruses.
Finally, the common disease manifestation of many viruses is only one of numerous consequences of infection. Herpesviruses are particularly "multipotential." For example, herpes simplex type 1 may cause fever blisters, encephalitis, keratoconjunctivitis, or no symptoms at all. As a result of "stress," the latent virus in the trigeminal ganglia of the host can recur causing a similar variety of symptoms. During long periods of latency, the dormant virus might change into a form with increased potential. Indeed, it would not be surprising if "stress" sometimes involved host or environmental conditions which favored malignancy. Thus, cancer may be an "accident," involving the convergence of multiple factors, viruses being of prime importance.
This is, however, speculation. The only conclusions which can be drawn from extensive data are that a number of human viruses have oncogenic potential, and that more will be discovered in the near future with the identification of human oncornaviruses. In several human tumors, an oncogenic virus appears to be a necessary, but not a sufficient, condition for tumor development. The expression of the oncogenic potential of the virus is controlled by a number of factors, most of which, like all pathogenesis, are not completely understood.
