Kaposi’s sarcoma is the most frequent neoplastic manifestation of HIV infection and is one of the CDC criteria that define an HIV-infected individual as having AIDS. AIDS-associated KS more frequently is seen among homosexual or bisexual men with HIV than in other HIV transmission risk groups. This epidemiologic observation led to a search for a second transmissible factor, which resulted in the identification of a new member of the gamma herpesvirus family, Kaposi’s sarcoma herpesvirus (KSHV) or human herpesvirus 8 (HHV-8). This virus has now been associated not just with KS, but also with a subset of B-cell lymphomas (discussed below), Castleman’s disease and, perhaps, multiple myeloma.

The epidemiologic data indicate that exposure to this virus is more common in promiscuous homosexual men and in populations with higher frequencies of “classic” or “endemic” KS not associated with HIV-1. It is found in KS tissue regardless of the epidemiologic background, but the mechanism by which the virus participates in tumor development is unclear. There are a number of KSHV genes with human homologues that suggest possible direct effects of the virus (cyclin D, bcl-2 or activated G-protein coupled receptor homologues) or effects at a distance (interleukin-6, chemokine and Ox-2 homologues). Hypotheses for how this virus is oncogenic are wide ranging and the mechanisms involved may be quite distinct from other tumor viruses. The nature of the immunologic response to KSHV remains ill defined, but clearly plays an important role in the control of KSHV-related tumors. KS in the setting of organ transplant often regresses with reduction of immunosuppressive medication. Similarly, in AIDS, patients treated with potent antiretroviral drug combinations have both a markedly reduced incidence of KS and often have pre-existing KS regress.

Histopathologically, KS lesions are a mixture of different cell types. Endothelial cells are quite present within the KS lesions, as is a prominent spindle-cell proliferation surrounded by extravasated erythrocytes and macrophages. The cell of origin of the neoplasm is still debated, as is the clonality of the disease. There are some patients who have multiple independent clones of tumor, whereas a subset of patients appear to have metastatic lesions derived from a single clone. In general, however, aggressive treatment of the original lesion has not had substantial impact on the ultimate development of other lesions.

KS often is a cutaneous nonblanching red macule. As lesions increase in size, they often have surrounding ecchymoses and acquire more of a violet hue. At times, the lesions may become nodular and, with advanced disease, the lesions may become confluent with large plaques developing, particularly on the legs. There is no orderly pattern of tumor progression, and presentation may be with lesions at multiple sites. The rate of growth of the primary lesions, as well as the appearance of new lesions, is quite variable. The lesions may occur on any cutaneous site and on mucous membranes. Lymphatic involvement is not unusual, and KS may present as lymphadenopathy. Visceral involvement, particularly of the trachea, lungs, and gastrointestinal tract, occurs commonly and may be seen in the absence of cutaneous disease. The most striking morbidity associated with KS is that of lymph node involvement and consequent lymphedema involving the lower extremities, groin, and head and neck. Extensive parenchymal or pleural involvement of the lung may result in life-threatening respiratory compromise.

The diagnosis of KS is relatively straightforward in HIV-infected individuals presenting with an erythematous or violaceous cutaneous or mucosal lesion. However, bacillary angiomatosis caused by Rochalimaea species may result in similar lesions. This process, which is treatable with antibiotics, should be excluded by Warthin-Starry staining of biopsy material. Following diagnosis, an assessment should be made of the distribution of the lesions. The presence of visceral disease does not necessarily correlate with poor response of lesions to therapy, so that an extensive evaluation for gastrointestinal or lymphadenopathic KS is not indicated unless there are specific symptoms referable to such involvement.

In patients with symptomatic visceral disease, lesions associated with edema, or rapidly evolving extensive cutaneous disease, chemotherapy can often provide symptomatic relief (

Table 416-1) . The most active chemotherapeutic drugs appear to be paclitaxel, liposomal anthracyclines, or combinations of vincristine/bleomycin or doxorubicin/vincristine/bleomycin. Liposomal doxorubicin or liposomal daunorubicin have been shown to be active with reduced toxicity compared with standard agents and often are used as a first-line approach. For those patients who fail liposomal agents (common after 10 weeks), low-dose paclitaxel has been shown to be highly active, desirable, and well tolerated. Response to chemotherapy usually occurs within the first few weeks of treatment. Unfortunately, the lesions regrow when the chemotherapy is stopped, so treatment is generally chronic.

Patients with KS who do not have rapidly progressive disease and therefore do not require immediate intervention may be treated with several other approaches. The most important among these is aggressive antiretroviral therapy. Regression of KS is common among patients treated with combinations of antiretroviral drugs occurring 2 to 4 months after control of viremia has been achieved. For those patients whose KS persists despite antiretrovirals, options include observation, single-agent interferon-alpha, local radiation, intralesional chemotherapy, local cryotherapy, or combinations of these.

Selecting the optimal therapeutic approach involves determining the clinical status of the patient, particularly using the staging classification (

Table 416-2) , as well as lifestyle issues. Patients who are categorized as “good risk” by the TIS staging system are also good candidates for response to interferon-alpha, but toxicity is common. A number of agents currently in clinical trial appear to have potential for this patient group. Among these are thalidomide, 9- cis retinoic acid, and angiogenesis inhibitors.