|
|
|
Viruses, Retroviruses, |
The concept that viruses cause human cancer dates back to the first decade of the 20th century, when experiments on animals showed that tumors could be induced in chickens by an agent that could pass through a filter. This pioneering work by Francis Peyton Rous in the United States was recognized more than 50 years later with the 1972 Nobel Prize that he shared with Howard Temin and David Baltimore, who characterized the molecular biology of retroviruses, first detected in Rous's experiments. Over the last decade, the new tools of molecular biology have led to profound discoveries about the role of viruses in human cancer and the mechanisms of disease causation. Viruses are a type of infectious agent that must invade living cells in order to reproduce. The two major types of viruses have either DNA or RNA as their genetic material. Viruses often invade cells by attaching to receptors on the surface of the target cell. Once inside the cell, they often integrate their genetic material into that of the host and alter the cell in ways that predispose to cancer through a variety of mechanisms. In some cases, the virus is thought to induce cancer directly; in other cases, indirect effects of the virus (e.g., immunodeficiency) predispose to malignancy. The major classes of viruses that are linked to cancer are retroviruses, herpes viruses, papilloma viruses, hepadnaviruses (hepatitis B), and flavaviruses (hepatitis C) (Evans and Mueller, 1990). In addition, genes called oncogenes, first discovered as part of the genetic material of acutely transforming retroviruses but now recognized as part of the normal genetic makeup of the cell, have been identified as critical factors in the oncogenic process. The first human retrovirus, discovered by Robert C. Gallo of the National Cancer Institute, is called human lymphotropic virus type I (HTLV-I) (Blattner, 1990). This prototype human retrovirus is strongly associated with malignant lymphomas of T-lymphocytes, first recognized in Southern Japan and called adult T-cell leukemia/lymphoma. A characteristic feature of these tumors is the monoclonal integration of the viral genome in the tumor tissue. Because the leukemia may occur years to decades after infection--which often occurs at birth--it is hypothesized that other factors play a role in pathogenesis. However, viral genes have the capacity to turn on genes of the host cell, promoting cell growth that may lead to uncontrolled cell proliferation. Recently, HTLV-I has been linked to a chronic neurologic syndrome called HTLV-associated myelopathy, which, because of demyelination, bears some resemblance to multiple sclerosis. The HTLV-I virus also causes a pediatric immunodeficiency syndrome called infective dermatitis and adult autoimmune diseases such as polymyositis and arthritis. Because of these immunologic effects of HTLV-I, it has been hypothesized that some more common cancers might be enhanced by HTLV-I infection through indirect mechanisms such as immunodeficiency. The closely related HTLV-II virus has not been definitively linked to cancer, although it was first isolated from a patient with a rare form of T-lymphocyte leukemia. The recent discovery that HTLV-II occurs naturally among some native American populations has created new opportunities to determine whether this "orphan" virus contributes in any way to cancer or other disease causation (Blattner, 1993). Since 1990, the American blood supply has been screened for HTLV-I. As a consequence, numerous donors have been identified as seropositive and have sought counseling concerning the health implications of this infection. While disease risk is not fully characterized, experts agree that the risk of leukemia and other complications is low, with an estimated lifetime risk of approximately 3 to 5 percent for an HTLV-associated disease (Blattner, 1993). The emergence of an epidemic of Kaposi's sarcoma in the United States among gay men, first recognized in 1981, was soon linked to the epidemic of acquired immunodeficiency syndrome, and an infectious agent postulated. The techniques pioneered by Robert C. Gallo were critical to the isolation of human immunodeficiency virus by Gallo and Luc Montagnier of The Institut Pasteur in Paris. This class of virus has as its major effect the induction of profound immunodeficiency through its ability to infect T-lymphocytes and cause their destruction (Blattner, 1991). While the process of HIV-associated immunodeficiency is complex, including direct killing effects of the virus on T-lymphocytes, lymphokine-mediated immune perturbations, and "autoimmune" mechanisms, the end result is a progressive depletion of CD-4 positive T-helper lymphocytes. This depletion results in heightened susceptibility of the infected individual to a variety of "opportunistic" pathogens as well as numerous cancers. Kaposi's sarcoma is a cancer of the lining of blood vessels which can occur on the skin, or be more widely disseminated in vital organs. Before the AIDS-associated epidemic, Kaposi's sarcoma was a rare tumor reported largely among older men, often of Mediterranean ancestry, and in residents of central Africa. The epidemic form is largely seen among gay men and is much more rare in other groups at risk of HIV infection. Some epidemiologic data suggest that an infectious agent may be involved and studies are under way to search for such an agent (Beral et al., 1990). Recently, a novel herpesvirus has been implicated in the pathogenesis of AIDS-related body-cavity-based lymphomas (Cesarman et al., 1995). The other major tumor type is non-Hodgkin's lymphoma, which occurs in all risk groups and appears strongly related to profound immunodeficiency (Rabkin et al., 1991). The pattern of tumor types and the high frequency of lymphomas of the central nervous system mirror the pattern of lymphomas associated with congenital and transplantation-associated immunosuppression (MacMahon et al., 1991). In one recent analysis, it was estimated that between 8 percent and 10 percent and up to 25 percent of all lymphomas in the United States will be AIDS-associated in coming years (Gail et al., 1991). Molecular studies have suggested that these lymphomas are linked to the Epstein-Barr virus (EBV), often with the pattern of oncogene translocation associated with Burkitt's lymphoma, which is a type of virally associated cancer originally described by Sir Dennis Burkitt in the 1960s. Thus, AIDS-associated cancers may represent examples of a process of immunosuppression allowing other oncogenic viruses such as EBV or the more recently discovered human herpes virus 6 (HHV-6) to be expressed as cancer (Cohen, 1991). It is likely that some other virally-associated cancers will show increases among HIV-positive immunosuppressed persons. For example, preliminary data suggest that human papilloma viruses may be increased in HIV-infected persons, with the potential for enhancing induction of associated tumors. As noted above, Epstein-Barr virus has been linked to Burkitt's lymphoma as well as other lymphomas, Hodgkin's disease (Herbst et al., 1991), and nasopharyngeal carcinoma (Litter, 1991). Intervention studies with an EBV vaccine are now under way in hopes of preventing some of these cancer types. Worldwide, hepatocellular carcinoma is a leading cause of death. A role for hepatitis B in the etiology of this tumor is well established (Blumberg and London, 1982). For example, very large studies of persons from populations where hepatitis B is frequent have shown an exceptional risk for cancer among antigen carriers who have not developed an adequate antibody response to the virus (Paterlini et al., 1990). Vaccine trials are also under way with this virus in order to prevent infections and associated hepatocellular cancer. The recently discovered hepatitis C, has also been linked to hepatocellular carcinoma (Simonetti et al., 1992). Given the very different nature of these two viruses, important clues about the role of viruses that cause liver damage and cancer will emerge as more is learned. The decade of the 1990s is exciting because of the advances in techniques for detecting and characterizing oncogenic viruses. It is likely that new agents that cause cancer will be discovered, and cancers of unknown cause linked to known and yet to be discovered agents. |
* From the Viral Epidemiology Branch, Division of Cancer Etiology, National Cancer Institute, Bethesda, Maryland
