Kaposi's Sarcoma

Kaposi's sarcoma is an opportunistic disease that occurs as a result of immunosuppression. In HIV-1-infected individuals, it can be seen at a wide range of CD4⁺ T-cell levels, most commonly in bisexual and homosexual patients. Contrary to its name, Kaposi's sarcoma is not an actual sarcoma (i.e., a malignancy of mesenchymal origin). The lesions that characterize Kaposi's sarcoma are not a result of neoplastic transformation of cells but arise from an excessive proliferation of spindle cells mixed with endothelial cells, fibroblasts, and inflammatory cells, accompanied by new blood vessel formation (neoangiogenesis) (Safai et al. 1985). The spindle cells exhibit characteristics common to endothelial and smooth muscle cells of vascular origin (Roth et al. 1992). The clinical manifestations of Kaposi's sarcoma include multiple vascular nodules in the skin, mucous membranes, and viscera. The disease may follow an indolent course, as evidenced by restricted skin or lymph node involvement, or an aggressive course with widespread cutaneous and visceral spread.

Epidemiologic studies have suggested that Kaposi's sarcoma is primarily caused by a sexually transmitted agent that may be transmitted by fecal contact (Beral et al. 1990, 1992; Peterman et al. 1993). It is infrequently observed in patients who are not HIV-infected, with clusters among men in Mediterranean countries and Africa. The disease is not equally distributed among all groups of HIV-seropositive individuals. About 20% of homosexual men with AIDS develop Kaposi's sarcoma. In other groups of AIDS patients, less than 1% of patients develop Kaposi's sarcoma. This distribution suggested an infectious etiology. Kaposi-sarcoma-de-rived cells have been screened for HIV and for a variety of known infectious agents, including CMV, HBV, HHV-6, EBV, human papillomavirus (HPV), and mycoplasma, with negative results (Jahan et al. 1989; Roth et al. 1992).

Recently, a new herpesvirus has been found in biopsy samples of Kaposi's sarcoma (Chang et al. 1994). This virus has been called Kaposi-sarcoma-associated herpesvirus (KSHV), as well as human herpesvirus type 8 (HHV-8). This unique herpesvirus is present not only in AIDS-associated Kaposi's sarcoma, but also in HIV-negative classical and African Kaposi's sarcoma (Dupin et al. 1995; Huang et al. 1995; Moore and Chang 1995; Su et al. 1995). The virus is also associated with body cavity lymphomas, a relatively rare, often EBV-negative, malignancy associated with AIDS (Cesarman et al. 1995). In a prospective study, the presence of HHV-8/HKSV sequences in blood was used to predict the subsequent appearance of Kaposi's sarcoma lesions (Whitby et al. 1995). HHV-8/KHSV is related to, but is distinct from, herpesvirus saimiri, a New World primate herpesvirus in the gamma, or lymphotropic, subfamily of herpesviruses. Gammaherpesviruses (a group that includes EBV) are capable of transforming cells in culture. This supports a causative role for HHV-8/KHSV in Kaposi's sarcoma. However, this virus is difficult to grow in cell culture, although the isolation of persistently infected cell lines from body cavity lymphomas (Renne et al. 1996b) has allowed initial studies of the properties of the virus (Renne et al. 1996a; Zhong et al. 1996).

Studies with spindle cells cultured from Kaposi's sarcoma lesions suggested the possibility that Kaposi's sarcoma involves a number of cytokines and growth factors: IL-1β, IL-6, TNF-α, basic fibroblast growth factor, and GM-CSF (Nakamura et al. 1988; Ensoli et al. 1989; Corbeil et al. 1991). Kaposi-sarcoma-derived cells secrete high levels of these factors in vitro, and conditioned media from the cultures induce normal capillary endothelial cells to proliferate. These observations suggest that Kaposi-sarcoma-derived cells might regulate their own growth by autocrine pathways. Oncostatin-M, a cytokine produced by activated lymphoid cells, is a mitogen for Kaposi-sarcoma-derived cells in vitro. Tumor cells exposed to oncostatin-M become spindle-shaped, grow more easily in soft agar, and secrete higher levels of IL-6 (Miles et al. 1992; Nair et al. 1992). Oncostatin-M can also upregulate the growth of Kaposi-sarcoma-derived cells in the absence of IL-6 (Radka et al. 1993). The HIV Tat protein has also been shown to serve as a growth factor in Kaposi-sarcoma-derived cells in vitro (Ensoli et al. 1993). Herpesvirus saimiri and other large DNA viruses are known to encode a number of cytokines, chemokines, and stimulatory and receptor-like molecules, which could account for some of the effects seen both in vitro and in vivo (Albrecht et al. 1992; McFadden and Graham 1994), and the sequence of HHV-8 also reveals a number of cytokine-like genes (Russo et al. 1996).

Other Opportunistic Neoplasms

HIV infection is also associated with a markedly increased susceptibility to the development of B-cell lymphomas, cervical carcinomas, and other neoplasms (Levine 1993). In fact, B-cell lymphomas and cervical carcinomas are now considered AIDS-defining conditions in HIV-infected individuals. Approximately 3% of AIDS cases in the United States present with lymphoma; the incidence is increasing with the prolonged survival attributable to improved treatment of opportunistic infections. B-cell lymphomas have also been observed frequently in SIV-induced immunodeficiency in macaque monkeys (Feichtinger et al. 1992; Putkonen et al. 1992). HIV infection increases the risk of developing lymphoma approximately 60–100-fold over the general population. EBV is involved as an initiating agent in approximately 60% of these lymphomas (Shibata et al. 1993). EBV is an initiating cofactor for Burkitt's lymphoma in malaria-infested regions in central Africa. EBV genomes have been detected in virtually 100% of primary central nervous system lymphomas in HIV-infected patients (MacMahon et al. 1991). Although the pathogenesis of B-cell lymphomas in immunodeficient individuals is likely to be multifaceted, increased proliferative activity of B cells, such as that caused by EBV immortalization or chronic antigenic stimulation, may predispose cells to subsequent genetic events leading to the transformation to lymphoma. In addition to B-cell lymphomas, non-B-cell and mixed immunophenotype lymphomas are more rarely seen. Monoclonal HIV-1 integration has been reported in some of these rare lymphomas (Shiramizu et al. 1994), but this provocative observation has not been reproduced or confirmed.

Cancers of the anogenital tract—particularly cervical carcinoma in women—are considered to be AIDS-defining conditions. Approximately 90% of such cancers of the anogenital tract in humans have an HPV etiology (zur Hausen 1991). Although such cancers usually occur many years after the primary infection, HPV is believed to play an initiating part in the progression to full neoplastic transformation. Cervical dysplasia and vulvar and cervical cancers appear to be greatly increased, and cervical disease appears to be more aggressive and to evolve more rapidly in HIV-infected women (Brettle and Leen 1991). Immunodeficiency with consequent lack of immunological control of HPV is likely to be a key factor in the appearance of this opportunistic cancer.

As we have seen, the malignancies associated with HIV infection are primarily, if not entirely, associated with one or another viral agent. It would thus appear that they should be considered as special examples of opportunistic infections. Whether failure of immune surveillance to protect against growth of transformed cells is also a factor in AIDS malignancy remains to be determined.