Specific form of vascular neoplasm that occurs sporadically in the general population and frequently in persons with acquired immune deficiency.

Kaposi sarcoma (KS) occurs throughout the world in sporadic, iatrogenic, endemic, and epidemic forms. Before the 1980s, KS was a rare neoplasm that occurred sporadically, mostly around the Mediterranean basin (1,2). In the United States, it is a rare disease, with an incidence of 0.02 per 100,000 tumors (3). Endemic KS is centered in equatorial Africa, where it accounts for 9% of all malignancies (4). Iatrogenic KS is related to medically induced immune suppression, and KS occurs in about 0.4% of all patients undergoing renal transplantation, which is an incidence 150 to 200 times greater than would be expected in the general population (5,6). KS cutaneous and visceral occurred in patients after organ transplantation accompanied by seroconversion of HHV8, the agent of KS (7). In a large Italy-France epidemiologic study the incidence for KS was 128 fold higher in transplant recipients and 451 fold higher in HIV-infected subjects (8). Steroid therapy was also reported to induce KS even in an HIV-negative patient (9). In patients with acquired immune deficiency syndrome (AIDS), a high frequency of KS was noted from the very first reports of cases in 1981 (3,10). Following the epidemic of human immunodeficiency virus (HIV) infection, the incidence of KS climbed as high as 15% in patients with AIDS. Thus, in the United States, KS became at least 20,000 times more common in persons with AIDS than in the general population, and 300 times more common than in other immunosuppressed groups (10,11). Marked variations in the frequency of KS between AIDS risk groups were also noted. The frequency of KS may be as high as 36% among homosexuals with AIDS, whereas its rate of occurrence is only 4.3% among intravenous drug addicts and 1% among hemophiliacs with AIDS (11). A male predominance is characteristic; the male-to-female ratio ranges from 3:1 in Western series to 15:1 in African series (1,12). Children may also be affected in areas of endemicity, where KS accounts for 5% to 10% of all malignant tumors (12,13). The epidemiologic data are consistent with a causative viral agent that is transmitted mainly by sexual contact (10).

Earlier studies, even those carried out before the present unexpectedly high incidence of KS, suggested an association with a herpesvirus. Herpes-like virus particles were seen in tissue cultures of African KS (14), in a biopsy specimen of a skin lesion in an African man (15), and in several KS lesions of internal organs (16). Cytomegalovirus (CMV) was suspected to be the herpesvirus involved in the etiology of AIDS because it is sexually transmissible (17), and antibodies to CMV are prevalent in most KS patients (18). However, the amounts are small, and CMV is sometimes detected in only a few infected cells, which suggests that CMV infection although frequently associated with KS is not involved in its pathogenesis (19).

More recently, a new herpesvirus, Kaposi sarcoma-associated herpesvirus (KSHV) or human herpesvirus type 8 (HHV-8), belonging to the subfamily of Gammaherpesvirinae, was identified in KS. By means of the molecular technique of representational difference analysis, portions of DNA foreign to the host genome were identified in KS lesions (20). Thus, a specific 210-bp genomic sequence could be detected, amplified by the polymerase chain reaction, visualized on ethidium bromide gels, and confirmed on Southern blot hybridization (21). The DNA sequence specific to HHV-8 was detected in up to 88% of cases of classic KS and in 100% of cases of KS in AIDS patients (21). HHV-8 was also present in mononuclear cells of peripheral blood in 52% of patients with KS (22). HHV-8 was not detectable in normal endothelial cells, reactive or neoplastic lesions unrelated to KS, or mononuclear cells of control patients free of HIV (22,23).

By in situ hybridization, it was shown that the HHV-8 DNA sequences are localized predominantly in the nuclei of the endothelial cells lining the vascular clefts and in the characteristic spindle-shaped cells of KS (24). These findings demonstrating a close association of HHV-8 with the lesions of KS, particularly in patients with significant immune deficiencies, are compatible with an important role of HHV-8 in the pathogenesis of KS.

The transmission of HHV-8 is achieved by two main routes, sexual, predominantly in homosexual men, and nonsexual in endemic populations (25). In both routes, saliva has been implicated by shedding of oral epithelial cells containing HHV-8 (25). Also transmission of HHV-8 by blood transfusion was demonstrated by the occurrence of HHV-8 seroconversion in a large group of transfusion recipients (26).

Also, DNA sequences of the HHV-8 virus have been identified in virtually all cases of body cavity-based lymphoma, presently classified as Primary effusion lymphoma (PEL) originating in germinal center B-cells (27,28,29) as well as in a high percentage of the multicentric form of Castleman disease of both HIV+ and HIV- individuals (30,31,32).

The organs affected and the distribution of lesions in KS have important implications regarding its pathogenesis. Because most patients who have KS present with multiple lesions, usually cutaneous, it is believed that KS, unlike other neoplasms, is multicentric from inception. It is further assumed, although no proof is available, that KS, unlike other tumors, does not disseminate by metastasis. For these reasons, it has been thought that KS is perhaps not a vascular neoplasm but rather a kind of vascular hyperplasia (33). A reason for the controversial concepts of this disease and its still obscure pathogenesis is that most studies to date have been based on cases of dermatologic KS, by far the most common form, whereas studies including substantial numbers of cases of visceral KS and autopsies have been remarkably few (12,34).

In a study of 86 cases of KS of internal organs, it was shown that all organs, including bone marrow and brain, long considered exempt, can be involved (35). In the 47 autopsies included in this study, multiple organs were usually involved, which may be an indication as much of multicentricity as of tumor dissemination in the advanced stages of immune deficiency. In 17 of the 47 cases (37%), the visceral tumors were not accompanied by cutaneous or mucosal lesions. The lymph nodes, a usual site of metastasis of other tumors, were frequently involved. Some have argued that tumor emboli were not identified in the presumed metastatic sites (36); however, a marked angiotropism of KS cells was noted, manifested by frequent growth around and inside blood vessels. It is possible that the familiar metastatic patterns of other tumors may not apply to KS, and that other mechanisms and stimulatory factors are involved in the progression of this particular neoplasm. A recent study showed a lytic gene of HHV-8 may play a role in the development of KS (37). In other studies, it was shown that exposure of the TAT protein, encoded by the TAT gene of HIV and necessary for viral replication, to a number of cytokines (interleukin 1, interferon-α) that are usually increased in HIV infection, may have modulating and mitogenic effects on KS cells (28).

About one quarter of HHV-8 genes encode unique proteins which have immune modulatory roles in cytokine production, apoptosis, and cell signaling (39).

The histogenesis of KS has also been the subject of debate. Most cellular components of the mesenchyme—endothelial cells, pericytes, fibroblasts, dendritic cells, myocytes, myofibroblasts, and mesenchymal stem cells—have been considered to be the cells of origin of KS. Presently, as a result of numerous immunohistochemical studies with a variety of monoclonal antibodies (MAbs), the histogenesis of KS has been narrowed to a progenitor cell from either blood vessel (40,41,42) or lymphatic endothelia (43,44). Determining the cell of origin of KS is not only of conceptual but also of practical importance, because this knowledge can assist with the pathologic diagnosis.

The presence of CD34 on immature spindle cells of KS may identify these cells as progenitors of more mature endothelial cells. The CD34+ hematopoietic progenitor cells are infected by HHV-8 and may be the reservoir for the virus infection providing a continuous source of virally infected cells (45). If CD34+ progenitor endothelial cells are able to circulate in the peripheral blood, like CD34+ hematopoietic progenitor cells, they may give rise to KS under special conditions, which would explain the multifocal form of neoplastic dissemination. Some experiments have indicated that CD34 may downregulate the expression of some adhesion molecules (46), suggesting another mechanism for the dissemination of KS. In vitro work on long-term cultures of KS indicates that the TAT protein released by HIV-infected CD4+ T lymphocytes and macrophages may trigger a cascade of cytokines and growth factors (47). The vascular and fibroblast growth factors identified in cultures and tissues of KS may initiate the vascular hyperplasia that ultimately evolves into KS under conditions of immune deficiency.