Bruce J Dezube
Beth Israel Deaconess Medical Center
Harvard Medical School
I n the USA, Kaposi’s sarcoma (KS) is over 20,000 times more common in people with AIDS than in the general population, and over 300 times more common than in other immunosuppressed patients, such as renal transplant recipients. KS is the most common tumour occurring in HIV-infected patients and is an AIDS-defining illness according to Centers for Disease Control (CDC) guidelines.
The pathogenesis of AIDS-related KS is multifactorial, involving the gamma-herpesvirus KSHV/HHV8, altered expression and response to cytokines, and stimulation of KS growth by the HIV-1 transactivating protein Tat.(1) Tat promotes the growth of spindle cells of endothelial origin derived from AIDS-KS lesions, but does so only in the presence of inflammatory cytokines.(2,3) KSHV/HHV8 encodes viral IL-6, which leads, in turn, to expression of vascular endothelial growth factor (VEGF).(4) The inflammatory-angiogenic state created by the virus has been proposed as the critical step in the pathogenesis of KS.(5)
The synergy between cytokines and the HIV-1 Tat protein, as well as the immunosuppression associated with AIDS, may explain why AIDS-related KS is more aggressive than the Mediterranean form of the disease in which Tat does not play a role.
AIDS-related KS has a variable clinical course, which can range from minimal disease presenting as an incidental finding to explosive growth resulting in significant morbidity and mortality. The skin lesions appear most often on the lower extremities, face (especially the nose), oral mucosa and genitalia.(6) Lymphoedema, particularly in the face, genitalia and lower extremities, may be out of proportion to the extent of cutaneous disease. KS is associated with a profound psychosocial burden comprising emotional distress, guilt, anger, ostracism and loss of employment.(7)
Extracutaneous spread of KS is common. KS in the oral cavity occurs in approximately one-third of patients and is the initial site of the disease in about 15%.(8) Gastrointestinal (GI) involvement is found in 40% of cases at initial diagnosis and up to 80% at autopsy. Involvement may occur in the absence of cutaneous disease. GI lesions may cause weight loss, abdominal pain, nausea, vomiting, upper or lower GI tract bleeding, or diarrhoea in some patients, while remaining asymptomatic in others.(9,10)
Pulmonary involvement may present as shortness of breath, fever, cough, haemoptysis or chest pain, or as an asymptomatic finding on chest X-ray. Lymph node involvement is not uncommon and may be present in patients with no evidence of mucocutaneous disease.
Exacerbation by steroids and infection
Corticosteroid therapy has been associated with the induction of KS or exacerbation of pre-existing KS in HIV-infected patients,(11) and HIV-negative patients receiving steroids for organ transplantation, autoimmune disorders or lymphoproliferative diseases.(12) This association should be emphasised, since HIV-infected patients may be treated with corticosteroids for a variety of disorders including immune thrombocytopenic purpura and Pneumocystis carinii pneumonia. Although not strictly contraindicated in KS patients, corticosteroids should be used with caution. KS lesions may regress upon reduction or withdrawal of steroids.(11,12) Opportunistic infections have also been associated with the induction of KS and with the exacerbation of pre-existing KS. High levels of proinflammatory cytokines, which have been found to accompany infections, may account for these clinical effects.
The major treatment goals for KS are palliation of symptoms; shrinkage of the tumour to alleviate oedema, organ compromise or psychological stress; and prevention of disease progression. Treatment options depend greatly on the tumour (extent of disease and rate of growth), HIV-1 viral load, and the host (CD4+ T-lymphocyte count, overall medical condition). It is important to ascertain whether there is any impairment in organ function that will increase the potential for drug-related toxicities, and to determine whether there are concomitant medications that will increase the likelihood of drug–drug interactions.
Highly active antiretroviral therapy (HAART)
KS patients should be advised to take antiretroviral drugs that will maximally decrease HIV-1 viral load. Effective antiretroviral regimens are associated with both a decreased proportion of new AIDS-defining KS cases and a reduction in the size of existing lesions. Regression and resolution of not just cutaneous lesions, but also visceral ones, have been reported in patients taking protease inhibitor-based therapy.(13–17) Highly active antiretroviral therapy (HAART) prolongs the time to treatment failure of anti-KS therapies.(18)
Alitretinoin gel 0.1% (9-cis-retinoic acid) is the only topical, patient-administered therapy approved for the treatment of KS. Alitretinoin is a naturally occurring retinoid that binds to both retinoic acid receptors and retinoid X receptors. In two phase III studies involving 305 patients, alitretinoin, when compared with vehicle gel, was associated with a shorter time to tumour response, longer duration of response, and greater time to disease progression (Table 1).(19) Dermal irritation, generally mild to moderate in severity, was not uncommon at the site of gel application. Dark-skinned patients may on occasion have undesirable skin lightening after use. In a phase II trial, oral administration of 9-cis-retinoic acid in AIDS-related Kaposi’s sarcoma led to a tumour response rate of 37%.(20)
Other local treatments include intralesional chemotherapy, radiation therapy, laser therapy and cryotherapy, all of which can be quite effective at controlling local tumour growth. Vinblastine is probably the most widely used intralesional agent and has an excellent response rate, in the range of 70%.(21) Treated lesions will usually fade and regress, although typically will not resolve completely. Radiation therapy can very effectively palliate symptomatic disease that is too extensive to be treated with intralesional chemotherapy but not extensive enough to warrant systemic therapy.(22)
In a series of 36 patients with KS of the feet, a fractionation schedule of three fractions a week at 3.5Gy/fraction to a total dose of 21.0Gy yielded an overall response rate of 91% (complete response rate, 80%). While discomfort from the radiation was frequent, it usually resolved without intervention within two weeks of completion of radiation therapy.(23)
Interferon alfa is a biological response modifier that can produce clinically significant responses in KS patients, especially those with disease limited to the skin and with relatively modest degrees of immunosuppression.(24) Although low doses of interferon (<10 million units/m(2) daily) can be used effectively when combined with nucleoside reverse transcriptase inhibitors (eg, zidovudine, didanosine), interferon therapy at higher doses is often associated with significant systemic toxicity consisting of fever, chills, neutropenia, hepatotoxicity and cognitive impairment. Poor tumour response and drug-related toxicity are particularly striking in patients with CD4+ T-lymphocyte counts <200 cells/ml. For some patients with asymptomatic but relatively disseminated cutaneous disease, interferon in combination with antiretroviral agents may be a reasonable option before deciding to use chemotherapy.
Current systemic treatments for KS include the newer liposomal anthracyclines and paclitaxel. The liposomal formulation of the anthracyclines provides the theoretical advantages of longer plasma half-life, higher tumour concentrations of drug and less toxicity in nontarget organs. Liposomal doxorubicin and daunorubicin have become the firstline chemotherapeutic treatment for patients with disseminated, symptomatic KS. In randomised multicentre trials, each of these drugs has been found to be superior to conventional chemotherapy.(25–27) The recommended dosage of liposomal doxorubicin is 20mg/m(2) every three weeks, while that of liposomal daunorubicin is 40mg/m(2) every two weeks. Each can reliably shrink tumours, lessen oedema and reduce the colour of lesions in 25–59% of patients, depending upon the specific definition of clinical response and other characteristics of the various trials. Moreover, liposomal doxorubicin is effective in 32% of patients whose tumours have progressed while receiving conventional doxorubicin, suggesting that the liposomal encapsulation increases the therapeutic effect of doxorubicin.(28) Side-effects are, in general, quite mild. Alopecia and neuropathies, in particular, are unusual with liposomal preparations, when compared with conventional forms. At higher cumulative doses, these agents are not associated with the cardiomyopathies that have often limited the use of nonliposomal anthracyclines.
Paclitaxel, the newest systemic chemotherapeutic agent approved for KS, has shown striking efficacy, even for patients with anthracycline-resistant disease. The response rates in two phase II trials were 71% and 59% (Table 2).(29,30) The median duration of response, approximately 10 months, is among the longest observed for any regimen reported for this disease.(29) Paclitaxel is well tolerated, but the higher prevalences of alopecia, myalgia/arthralgia and bone marrow suppression, and the need for a three-hour infusion, make paclitaxel less attractive than the liposomal anthracyclines as initial therapy of disseminated disease. Recommended dosing schedules for paclitaxel are 100mg/m(2) over three hours every two weeks or 135mg/m(2) over three hours every three weeks. Usual premedication regimens include diphenhydramine and ranitidine (or equivalent). Concern over corticosteroid use in these patients means that dexamethasone is usually restricted to 20mg intravenously given just before the paclitaxel, or 10mg administered orally 12 and 6 hours before.
Drug metabolism of many of the approved antiretroviral agents, particularly protease inhibitors and non-nucleoside reverse transcriptase inhibitors, involves cytochrome P450 metabolic pathways. Taxanes are oxidised to less active metabolites by hepatic cytochrome P450 enzymes. Profound toxicity observed in at least two patients receiving paclitaxel has been ascribed to a toxic interaction between this agent and antiretroviral drugs.(31) In a pharmacokinetic study in a single male patient, however, the metabolism of paclitaxel given concomitantly with indinavir, ritonavir, saquinavir or nevirapine was not statistically significantly different from that of paclitaxel as a single agent in historical control subjects.(32) Caution is advised when co-administering agents that use the same metabolic pathways.
Given the highly significant role that angiogenesis plays in the pathogenesis of KS, it is not surprising that many of the angiogenesis inhibitors in development are being tested in patients with AIDS-related KS. Angiogenesis inhibitors, which have led to durable clinical responses in patients accrued to early trials, include fumagillin,(33) thalidomide,(34) metastat(35) and interleukin-12.(36) Many of these agents are being developed clinically as both angiogenesis inhibitors and KS-inhibitors. Of note, in a phase I trial of metastat in AIDS-related KS, tumour response rate was 44% and was associated with decreases in MMP-2 levels.(35) Other potential targets for KS therapies include KSHV/HHV8 and the process of cellular differentiation.
- Martin JN, Osmond DH. Curr Opin Oncol 1999;11:508-15.
- Ensoli B, Barillari G, Salahuddin SZ, et al. Nature 1990;345:84-6.
- Ensoli B, Gendelman R, Markham P, et al. Nature 1994;371:674-80.
- Aoki Y, Jaffe ES, Chang Y, et al. Blood 1999;93:4034-43.
- Mesri EA. Blood 1999;93:4031-3.
- Dezube BJ. Semin Oncol 2000;27:424-30.
- Holland JC, Tross S. Semin Oncol 1987;14(2 Suppl 3):48-53.
- Nichols CM, Flaitz CM, Hicks MJ. J Am Dent Assoc 1993;124:78-84.
- Danzig JB, Brandt LJ, Reinus JF, Klein RS. Am J Gastroenterol 1991;86:715-18.
- Laine L, Amerian J, Rarick M, et al. Am J Gastroenterol 1990;85:959-61.
- Gill PS, Loureiro C, Bernstein-Singer M, et al. Ann Intern Med 1989;110:937-40.
- Trattner A, Hodak E, David M, Sandbank M. Cancer 1993;72:1779-83.
- De Milito A, Catucci M, Venturi G, et al. J Med Virol 1999;57:140-4.
- Murphy M, Armstrong D, Sepkowitz KA, et al. AIDS 1997;11:261-2.
- Conant MA, Opp KM, Poretz D, Mills RG. AIDS 1997;11:1300-1.
- Aboulafia DM. Mayo Clin Proc 1998;73:439-43.
- Niehues T, Horneff G, Megahed M, et al. AIDS 1999;13:1148-9.
- Bower M, Fox P, Fife K, et al. AIDS 1999;13:2105-11.
- Walmsley S, Northfelt DW, Melosky B, et al. J AIDS 1999;22:235-46.
- Miles SA, Dezube BJ, Lee JY, et al. AIDS 2002;16:421-9.
- Boudreaux AA, Smith LL, Cosby CD, et al. J Am Acad Dermatol 1993;28:61-5.
- Swift PS. Hematol Oncol Clin North Am 1996;10:1069-80.
- Gressen EL, Rosenstock JG, Xie Y, Corn BW. Am J Clin Oncol 1999;22:286-90.
- Shepherd FA, Beaulieu R, Gelmon K, et al. J Clin Oncol 1998;16:1736-42.
- Gill PS, Wernz J, Scadden DT, et al. J Clin Oncol 1996;14:2353-64.
- Northfelt DW, Dezube BJ, Thommes JA, et al. J Clin Oncol 1998;16:2445-51.
- Stewart S, Jablonowski H, Goebel FD, et al. J Clin Oncol 1998;16:683-91.
- Northfelt DW, Dezube BJ, Thommes JA, et al. J Clin Oncol 1997;15:653-9.
- Gill PS, Tulpule A, Espina BM, et al. J Clin Oncol 1999;17:1876-83.
- Welles L, Saville MW, Lietzau J, et al. J Clin Oncol 1998;16:1112-21.
- Schwartz JD, Howard W, Scadden DT. AIDS 1999;13:283-4.
- Nannan Panday VR, Hoetelmans RM, van Heeswijk RP, et al. Cancer Chemother Pharmacol 1999;43:516-9.
- Dezube BJ, Von Roenn JH, Holden-Wiltse J, et al. J Clin Oncol 1998;16:1444-9.
- Little RF, Wyvill KM, Pluda JM, et al. J Clin Oncol 2000;18:2593-602.
- Cianfrocca M, Cooley TP, Lee JY, et al. J Clin Oncol 2002;20:153-9.
- Pluda JM, Wyvill K, Little R. J AIDS Hum Retrovirol 1999;21:A29.