Comprehensive, up-to-date information on HIV/AIDS treatment, prevention, and policy from the University of California San Francisco

This chapter focuses on local and systemic therapies for HIV-associated Kaposi sarcoma, including discussion of their indications, expected benefits, and toxicities.

Kaposi sarcoma (KS) is the most common malignancy seen in the setting of HIV infection. Although the percentage of HIV-infected patients presenting with KS as an AIDS-defining diagnosis decreased in the United States from about 30% in the early 1980s to 15% 10 years later, the absolute number of cases did not decline over this time period,(1-3) as the incidence of KS as a secondary AIDS diagnosis increased from 23% in the 1980s to 50% in the late 1990s.(4) Treatment with effective antiretroviral therapy (ART) has led to a dramatic decline in KS incidence. A prescription for ART may reduce the likelihood of KS by 50%.(5) In the current era of treatment, KS is generally identified as a late manifestation of HIV infection, occurring when immunosuppression is severe. Thus, KS-associated morbidity and mortality rates have increased, particularly in sub-Saharan Africa.(6)

KS may affect quality of life in a variety of ways. Tumor-associated edema may cause difficulty with ambulation. Plaquelike or ulcerated cutaneous lesions may be painful, and bulky oral disease may interfere with speech or swallowing. In severe cases, bronchopulmonary involvement may lead to respiratory compromise. Because cutaneous KS frequently involves the face, cosmetically disturbing lesions are common and may cause disfigurement and lead to social isolation.

KS can be indolent or a progressive debilitating disease. Although responses have been reported to a variety of treatments, KS remains an incurable tumor. Chemotherapy by itself can produce rapid relief of symptoms, but responses are often of limited duration and do not clearly improve survival.(7) Antineoplastic treatment, in combination with ART that suppresses HIV replication, may produce prolonged remission of KS.(8,9) Palliation of symptoms and improvement in quality of life remain major goals of treatment.

ART is associated with significant improvement in the survival of patients with KS, and has become an essential component of KS management. Prior to the availability of effective ART, 90% of patients with pulmonary KS progressed and died of their disease. In contrast, where effective ART is available, the proportion of patients with pulmonary KS experiencing fatal disease progression has been reduced to 47%.(9) In a multivariate analyses, the Multicenter AIDS Cohort Study demonstrated an 81% reduced risk of death for KS patients treated with combination ART.(10)

Patients with KS require frequent evaluations to determine the tempo and morbidity of their disease. Although patients with limited, stable disease may be considered for local therapy, observation alone is usually appropriate for patients with limited disease who have recently initiated ART. However, even with effective ART and well-controlled HIV infection, some patients develop progressive KS. These patients are candidates for specific antineoplastic treatments.

Uncontrolled HIV-1 viral replication may contribute to KS progression by several mechanisms. Expression of the HIV Tat protein regulates human herpesvirus (HHV)-8 viral growth.(11,12) Immune restoration consequent to ART may contribute to HHV-8 latency. In addition to the indirect effect of ART on KS growth, at least some of the protease inhibitors have specific antineoplastic effects.(13,14) Administration of either saquinavir or indinavir to nude mice blocks the development of, and induces regression of, KS-like lesions.(14) Ritonavir has been shown to inhibit activation and proliferation of primary endothelial cells in vitro, and to decrease production of tumor necrosis factor alpha, interleukin-6, interleukin-8, and vascular endothelial growth factor.(13) KS-derived cell lines underwent apoptosis after treatment with ritonavir at concentrations obtainable in the clinical setting. Furthermore, ritonavir inhibited tumor formation and progression by KS-derived cells in a mouse xenotransplantation model.(13)

Several factors may aid in the choice of local versus systemic therapy for patients with AIDS-related KS. Patients with fewer than 10 lesions and relatively stable disease (no or very few new lesions over the preceding months), associated with minor symptoms (eg, patients with isolated, cosmetically disturbing lesions or a limited number of painful lesions), are candidates for local therapy. Additionally, patients with HIV-related opportunistic diseases or advanced HIV infection may have decreased tolerance of systemic therapy and might also be candidates for local treatment. For patients with pulmonary KS, symptomatic lymphedema, rapidly progressive mucocutaneous disease (>10 new lesions in the preceding month), or symptomatic visceral disease, systemic therapy is the treatment of choice (Table 1).

The clinical course of AIDS-related KS is as varied as the course of HIV infection itself. The recognition of KS as a systemic malignant disease with increasing mortality in the HIV-infected patient has led to controversy regarding the appropriate use of local therapy. Although many patients will ultimately develop progressive disease, early in its course, KS may be indolent and associated with few symptoms. For these patients, local therapy may be an option. Advantages of local therapy, particularly in patients with advanced AIDS-related immunosuppression, include the relative absence of systemic adverse effects, the nonimmunosuppressive nature of local treatments, and the relatively circumscribed treatment course. Patients with limited cutaneous disease of highly visible sites--the face in particular--may be candidates for local treatment. Isolated nodular or painful lesions, as well as irritating conjunctival or limited mucosal disease, may be adequately palliated with local interventions. Advocates of systemic treatment note that local therapies fail to prevent the development of new lesions and provide only temporary local benefit in many cases.

For patients with stable, limited, relatively indolent KS, local therapy may be a viable option. For the majority of patients, KS requires systemic intervention. Currently available effective systemic therapies include IFN-alfa--with or without antiretroviral agent(s)--and cytotoxic chemotherapy.

Interferon

IFN-alfa was the first drug specifically approved for the treatment of KS. It is of particular interest because of its antiproliferative, antiviral (anti-HIV), antiangiogenic, and immune-modulating properties. Early investigations of IFN-alfa for the treatment of KS demonstrated an association between response and immune function, as measured by CD4 count, as well as an association between response and dose. The likelihood of response to IFN-alfa in a patient with a CD4 lymphocyte count of <100 cells/mm3 was reported to be <10%, but increased to 45-50% in patients with a CD4 lymphocyte count of >400 cells/mm3.(28-30) When used as a single agent without concurrent antiretroviral therapy, IFN-alfa appeared most effective at relatively high doses (U.S. Food and Drug Administration (FDA)-approved regimens for the two approved forms of recombinant IFN-alfa are 36 MU subcutaneously daily for IFN-alfa 2a and 30 MU subcutaneously three times per week for IFN-alfa 2b).(31,32) In selected patient populations with no history of opportunistic infections or AIDS-related symptoms and relatively preserved immune function, IFN-alfa alone results in a relatively long response duration, averaging 6-12 months for partial responders and up to 2 years for complete responders. Before the advent of viral load measurements and highly active antiretroviral therapy, studies were undertaken to attempt to increase the efficacy of IFN-alfa, decrease the required IFN-alfa dose and toxicity, and broaden the patient population that might benefit from IFN-alfa treatment. These studies evaluated IFN-alfa in combination with a variety of antiretroviral agents. Response rates of 42-46% were observed for treatment with IFN-alfa plus zidovudine.(33-35) Compared with IFN-alfa as a single agent, this combination was more likely to benefit patients with poor prognosis and high-risk KS features (CD4 count <150 cells/mm3, visceral or bulky disease) and to be effective with a lower dose of IFN-alfa. Myelosuppression was the dose-limiting toxicity of this combination, and responded to granulocyte-macrophage colony-stimulating factor, but at the cost of increased constitutional symptoms.(36) To avoid drug combinations with overlapping toxicity, the AIDS Clinical Trials Group (ACTG) evaluated didanosine, a nonmyelosuppressive antiretroviral agent, in combination with IFN-alfa, 1 MU subcutaneous daily versus 10 MU subcutaneous daily.(37) Tumor regression was observed in 40% of the patients treated in the low-dose IFN arm and 55% of those treated with 10 MU IFN-alfa. The median time to progression was >110 weeks. Treatment was well tolerated. Although treatment with single antiretroviral agents is no longer recommended because it leads to rapid selection of drug-resistant HIV, these studies suggest that response to IFN-alfa may be enhanced when the drug is used with effective antiretroviral combinations, and that these combinations should be continued during KS treatment as long as toxicities are manageable. Studies have been initiated to evaluate treatment with IFN-alfa in combination with protease inhibitor-based combination ART. The toxicity of IFN-alfa is dose related. At a daily dose of 1 MU, IFN-alfa results in little identifiable toxicity. As the dose is increased, the likelihood of developing a flulike syndrome--characterized by low-grade fever, myalgias, and malaise--increases. Tachyphylaxis to these flulike symptoms develops over the first few weeks of therapy in most patients. Long-term administration of IFN-alfa, particularly at high doses (>20 MU subcutaneously daily), may result in persistent fevers, anorexia, weight loss, fatigue, and neuropsychiatric symptoms such as depression. The time to response to IFN-alfa, either alone or in combination with antiretroviral agents, is usually about 8 weeks or more after initiation of treatment. The relatively slow onset of response seen with IFN-alfa therapy makes this intervention inappropriate for patients with advanced, rapidly progressive KS.

Cytotoxic Chemotherapy

Chemotherapy may result in rapid resolution of KS-associated symptoms and thereby improve quality of life. Cytotoxic chemotherapy is indicated for patients with extensive mucocutaneous KS, rapidly progressive cutaneous disease (more than 10 new lesions per month), symptomatic visceral disease, pulmonary disease, or extensive symptomatic lymphedema. A wide variety of chemotherapeutic agents, individually and in combination, have been evaluated for the treatment of KS. The broad range of KS response rates to single agents (21-80%) is a result of differences in the efficacy of the agents tested, variations in the patient populations treated (including level of immune function, history of prior opportunistic infections, and tumor burden), and lack of standardization of the criteria used to stage these patients or to evaluate their response to treatment.(38-54) In general, the phase III clinical trials completed since 1990 define the study population and treatment outcomes more rigorously, applying the ACTG staging and response criteria. Single Agent Chemotherapy Etoposide Etoposide (VP-16) has been evaluated as both an oral and an intravenous treatment for KS. As an oral agent, VP-16 has been evaluated primarily in patients who have undergone prior treatment with multiple cytotoxic agents. Previously treated patients treated with VP-16 (50 mg/day for 21 days every 4-5 weeks, or for 7 days out of every 2-3 weeks) have an overall response rate between 21% and 36%.(40,41,45) When VP-16 is given intravenously (150 mg/m2 on days 1 to 3 every 4 weeks), high response rates (78%) have been reported in patients without prior treatment and with good prognosis (no history of opportunistic infection and no constitutional symptoms).(53) The primary toxicity of this agent is myelosuppression, which may necessitate the use of colony-stimulating factors, particularly in patients with advanced disease receiving other myelosuppressive agents. Additionally, alopecia frequently occurs in association with etoposide therapy, and is a consideration for patients who are receiving treatment of their KS for primarily cosmetic purposes. Bleomycin Bleomycin has been evaluated both as a single agent in a variety of doses and schedules and as a component of the most frequently used combination regimens. In patients with KS of poor prognosis, bleomycin given intramuscularly (5 mg/day for 3 days) or as a 4-day continuous infusion (6 mg/m2/day) produced a 48% partial response rate.(42) Results of a small, single-institution study in which bleomycin was given as a 72-hour infusion (20 mg/m2/day) to 17 patients indicated a partial response rate of 65%.(43) Bleomycin toxicity appears to be acceptable, with neutropenia an infrequent complication. Infusional regimens are limited by patient acceptance in view of the need for central venous access and the use of infusion pumps. Doxorubicin Doxorubicin (also known by its trade name Adriamycin) is a component of the most widely used combination regimen for HIV-associated KS. In one trial, however, weekly treatment with doxorubicin (25 mg/m2) in patients with AIDS-related KS achieved a partial response rate of only 10%.(46) The primary toxicity of weekly doxorubicin is myelosuppression, even with the relatively low doses prescribed for KS treatment. Vincristine and Vinblastine The vinca alkaloids have been used as single agents or alternating on a weekly schedule.(47,55) Alternating vincristine and vinblastine weekly achieves a response rate of 33%.(55) Toxicity includes vincristine-induced neurotoxicity (which limits its usefulness as a single agent) and vinblastine-induced myelosuppression. Paclitaxel Paclitaxel, a drug with cellular targets similar to the vinca alkaloids, is also effective for the treatment of KS. A phase II trial evaluated paclitaxel (135 mg/m2) administered intravenously over 3 hours every 21 days to patients with symptomatic AIDS-related KS.(44) The dose of paclitaxel was increased by 20 mg/m2 to a maximum of 175 mg/m2 if absolute granulocyte counts remained above 1,000/mm3. No colony-stimulating factors were required for support of neutrophil counts. A partial response was reported in 13 of 20 patients (65%), with 5 of 6 patients with known pulmonary KS responding, as well as 6 previously treated patients with nonpulmonary KS achieving a partial response. Therapy was well tolerated, although the incidence of neutropenia increased with increasing doses of paclitaxel, and alopecia was seen in all patients treated. A second investigation of paclitaxel, in heavily pretreated KS patients with advanced immunosuppression (median CD4 count: 5/mm3), evaluated the efficacy and toxicity of an alternate dose and schedule: 100 mg/m2 administered intravenously over 3 hours every 2 weeks.(48) Of the 30 evaluable patients, 16 (53%) achieved a partial response. The time to response was short (median of three cycles of treatment). Dramatic improvement in symptomatic lymphedema was noted in 25 of 26 patients. Therapy was well tolerated. Grade 3 or greater neutropenia was uncommon. Vinorelbine A number of newer agents have been evaluated for the treatment of KS. Vinorelbine, a semisynthetic vinca alkaloid with a favorable therapeutic index compared to other vinca alkaloids, was evaluated in previously treated patients with progressive AIDS-related KS.(54) Thirty-five evaluable patients were treated with vinorelbine 30 mg/m2 every 2 weeks. Clinical complete and partial remissions were achieved in 43% of patients, with a median progression-free survival of 151 days. The agent was well tolerated, with neutropenia the most frequent dose-limiting toxicity. Liposomal Agents Standard chemotherapy agents have achieved a modicum of success in the treatment of KS. Overall, however, the duration of response to cytotoxic agents has been limited, particularly if HIV replication is inadequately suppressed. In an attempt to improve the therapeutic-to-toxic ratio of therapy, liposome-encapsulated anthracyclines were evaluated for the treatment of KS. Theoretically, encapsulation in liposomes prolongs the plasma half-life of a drug, increases drug accumulation in tumor tissue, decreases drug uptake by the reticuloendothelial system, and results in less drug-induced toxicity. Two liposomal agents are currently approved for the treatment of KS: liposome-encapsulated daunorubicin (DaunoXome) and liposome-encapsulated doxorubicin (Doxil). Both agents are highly active against KS and have less toxicity than the nonliposomal anthracyclines. Myelosuppression, however, remains the dose-limiting toxicity. In a phase III trial involving previously untreated patients with advanced AIDS-related KS, participants were randomly assigned to receive either liposome-encapsulated daunorubicin (40 mg/m2) intravenously over 30-60 minutes every other week or doxorubicin (10 mg/m2), bleomycin (15 units), and vincristine (1 mg) (ABV) every other week.(49) No significant difference in response rate was observed; treatment with liposome-encapsulated daunorubicin led to a 23% response rate, compared to a 30% response rate to ABV treatment. The duration of response, time to progression, and survival were similar in both treatment arms. Toxicity was significantly less for patients treated with liposome-encapsulated daunorubicin. Alopecia and neuropathy occurred less frequently in the liposome-encapsulated daunorubicin-treated group (8% and 13%, respectively), compared to those patients who received ABV (36% and 41%, respectively). Myelosuppression was the dose-limiting toxicity; grade 3 or greater neutropenia occurred in 13% of patients in each treatment arm. A phase III trial of liposome-encapsulated doxorubicin enrolled a similar patient population and randomly selected patients to receive either liposome-encapsulated doxorubicin (20 mg/m2) every other week for six treatments or doxorubicin (20 mg/m2), bleomycin (10 mg/m2), and vincristine (1 mg) (ABV) every other week for six treatments.(50) Patients receiving the liposome-encapsulated doxorubicin achieved a response rate of 46%, whereas the response rate achieved with ABV was 26% (p < .001). The duration of response was similar in the two groups. The time to response was shorter with liposome-encapsulated doxorubicin treatment (39 days), compared to treatment with ABV (50 days). Alopecia and neuropathy occurred less frequently in patients treated with the liposome-encapsulated drug (11% and 12%, respectively), compared to those receiving ABV (42% and 28%, respectively). Mucositis occurred more frequently with the liposome-encapsulated doxorubicin as compared to ABV (18% versus 8%), whereas the incidence of myelosuppression was similar in the two treatment arms. Palmar-plantar erythema was not seen as a significant toxicity in either treatment group (<1%), although prolonged treatment with liposomal doxorubicin, because of its longer half-life compared with free drug, may increase the risk of this complication.(56) In a second phase III trial of liposomal doxorubicin, 241 previously untreated patients received intravenous therapy every 3 weeks either with liposome-encapsulated doxorubicin (20 mg/m2), or with bleomycin (15 mg/m2) and vincristine (2 mg) (BV).(51) Responses were achieved by 58.7% of the liposomal doxorubicin-treated patients and 23.3% of those treated with BV. Neutropenia was the dose-limiting toxicity for patients treated with the liposomal agent. Combination Chemotherapy Regimens Combination chemotherapy regimens have been evaluated in hopes of improving both the overall response rate and duration of response to chemotherapy for patients with advanced AIDS-related KS. Response rates for combination regimens vary from 28% to 88%.(50-56) Differences in response rates are largely attributable to differences in the patient populations evaluated, the lack of strictly defined response criteria, and variations in the dosing schedules used. The combination regimen studied and used most widely is ABV given every 2 weeks. The initial report of ABV therapy in 30 patients described a response rate of 88%, whereas in more recent randomized trials, response rates were as low as 28%.(50,52) Toxicity of this regimen includes myelosuppression, mild nausea, moderate alopecia, and peripheral neuropathy. An alternative regimen, which is associated with less myelosuppression and alopecia, is BV.(57) A single-institution phase II study of BV in 18 patients reported a response rate of 72%. This result is in marked contrast to the response rate of 23.3% reported in the phase III trial of BV versus liposomal doxorubicin.(51) Early results of an ACTG phase III trial evaluating the toxicity and efficacy of liposomal doxorubicin alone versus liposomal doxorubicin in combination with BV for the treatment of advanced AIDS-related KS suggests no therapeutic advantage to the addition of BV to liposomal doxorubicin.(58)

Increasing understanding of the pathogenesis of AIDS-related KS has led to investigations into a variety of pathogenesis-based therapies. Although HHV-8 is associated with the majority of KS lesions, available data do not support the use of antiherpes agents (foscarnet, ganciclovir, acyclovir, etc) for the treatment of established KS. Whether or not this approach will be useful for prevention of KS in patients with serologic evidence of HHV-8 has yet to be determined. Over the last few years, inhibitors of both cytokines and angiogenesis have been evaluated for the treatment of KS. To date, although small numbers of responses to these biologic therapies have been reported, none of these agents are specifically approved for the treatment of KS. It is hoped that further development of antiangiogenesis agents, cytokine inhibitors, or both, will result in long-term, nontoxic treatment for patients with KS.

Myelosuppression is a frequent problem in patients with advanced HIV-related immunosuppression. In addition, a wide variety of myelosuppressive agents are used for the treatment of HIV and its complications. For example, for patients receiving zidovudine or ganciclovir, the addition of myelosuppressive cytotoxic chemotherapeutic agents often necessitates the addition of a colony-stimulating factor to support neutrophil counts. Drug interactions and the effect of combination therapies on drug concentrations must also be considered. The protease inhibitors, particularly ritonavir, which affects the hepatic metabolism of a wide variety of medications, must be used cautiously in combination with cytotoxic agents. As yet, minimal data are available regarding the pharmacokinetic interactions between the protease inhibitors and chemotherapy. It is postulated, however, that the total systemic exposure indicated by the area under the concentration curve for some drugs (eg, paclitaxel) may be increased threefold when prescribed in combination with ritonavir, resulting in a marked increase in all chemotherapy-related adverse effects.(59) Studies are underway to define appropriate chemotherapy dose modifications in the presence of protease inhibitors.

Although KS remains an incurable tumor for which no established standard therapy is defined, general principles of treatment provide a basis for rational clinical decision making (Table 1). The selection of therapy for KS must take into account the potential benefit and adverse effects of treatment, interactions with other medications, and potential impact on underlying immunosuppression.

Cosmetically disturbing lesions should be treated early and, if limited in number, may be appropriate for local therapy. Radiation therapy is the most effective local therapy.

Interferon-alfa in combination with antiretroviral agents should be considered for the treatment of patients with slowly progressive or minimally symptomatic KS. The long duration of response, the limited toxicity of low-dose IFN (1-10 MU/day subcutaneously) either alone or in combination with nonmyelosuppressive antiretroviral agents, and intrinsic antiviral activity make IFN-alfa an important therapy for KS. This combination should be considered for clinically stable patients, even those with CD4 lymphocyte counts <100 cells/mm³.

Systemic chemotherapy should be used for patients with rapidly progressive, symptomatic, or life-threatening KS. Individuals with pulmonary disease or severe lymphedema should receive chemotherapy.

For patients who require systemic chemotherapy, liposome-encapsulated anthracyclines are more effective and less toxic than ABV. Response rates to paclitaxel are high, with acceptable toxicity. This agent is effective in patients who have relapsed after prior anthracycline-based chemotherapy. Patients with pre-existing myelosuppression or individuals who require other myelosuppressive agents should receive colony-stimulating factors in combination with chemotherapy.