Cidofovir for the treatment of Kaposi's sarcoma in an HIV-negative homosexual man

Le résumé fourni par la source

S ir, Human herpesvirus 8 (HHV‐8) has been detected in the lesions of all forms of Kaposi’s sarcoma (KS) 1 and has been implicated as a probable causative agent. 2 This has led to interest in the use of antiviral drugs in the treatment of KS. In vitro studies have shown HHV‐8 to be sensitive to the antiherpesviral drugs cidofovir, foscarnet and ganciclovir. 3, 4 Cidofovir had the strongest inhibitory effect on HHV‐8 DNA synthesis. There have been case reports of responses of AIDS‐related KS to cidofovir 5 and foscarnet. 6 In addition, there is evidence of KS prevention in patients receiving antiherpesviral drugs. A retrospective study of human immunodeficiency virus (HIV)‐seropositive patients found those treated with foscarnet or ganciclovir for cytomegalovirus (CMV) infection to be protected against the development of KS. 7 In a randomized trial of AIDS patients treated with ganciclovir for CMV retinitis, those receiving systemic ganciclovir had a much lower risk of developing KS than those treated with an ocular ganciclovir implant. 8 KS has been reported in HIV‐negative homosexual men, in whom it follows the indolent course found in classical KS. 9 We describe the use of cidofovir for treating KS in an HIV‐negative homosexual man. HHV‐8 activity was monitored with serological and polymerase chain reaction (PCR) methods. A 40‐year‐old caucasian homosexual man presented with a 1‐year history of brown macular pruritic lesions on his shins and a red–brown papular lesion on his right wrist. He had had several HIV antibody tests between 1992 and 1998, all of which were negative; PCR and Western blots for HIV were also repeatedly negative. He reported HIV‐seropositive partners, but none had KS. KS was diagnosed in three skin biopsies over 1 year. Computed tomographic scans of the chest and abdomen were normal. His CD4 lymphocyte count was normal at 706 cells/mm3. PCR of peripheral blood mononuclear cells for HHV‐8 was strongly positive. His shin and wrist lesions responded well to local radiotherapy. However, over the next 6 months, KS developed more rapidly with, on average, two new lesions per week. Systemic therapy was indicated but, as he did not wish to have chemotherapy, it was decided to use the antiherpesviral drug cidofovir. An induction dose of cidofovir 5 mg/kg weekly for 2 weeks was given according to the standard schedule for CMV infection, along with probenecid and hydration. 10 Symptoms of toxicity comprised mild nausea and a neutrophil nadir of 1·8 × 109/L. Cidofovir 5 mg/kg was continued at 2‐weekly intervals for 4 months. Response was measured according to AIDS Clinical Trials Group criteria established for AIDS‐related KS. 11 Rapid progression of KS was halted, with no new lesions for 10 weeks. Of seven lesions measured, two responded completely and four raised lesions became flat, indicating a partial response. These responses were maintained until the end of cidofovir therapy. However, two new lesions developed 10 weeks into therapy. After stopping cidofovir, the patient again had rapid disease progression, with enlargement of current lesions, and developed 12 new lesions within 10 weeks. He was therefore given liposomal daunorubicin chemotherapy (40 mg/m2 every 2 weeks) with partial response, but has since relapsed. Serum samples were taken prior to cidofovir therapy and at seven further time‐points during treatment. HHV‐8 seropositivity was determined using an immunofluorescent assay. 12 HHV‐8 titres were very high throughout (1 : 406,600) and, although there was a decrease in HHV‐8 titre 3 months after the end of cidofovir therapy to 1 : 51,200, this was only 1 log10 change, and of questionable significance. Semiquantitative determination of HHV‐8 DNA was done using a nested PCR method using genomic DNA extracted from whole blood and PCR amplification of viral and cellular (control) DNA. A 213‐base pair HHV‐8 fragment was amplified in all the patient’s samples and the concentration estimated by spectrophotometry. There was no significant change in HHV‐8 DNA levels at any time‐point. This is the first report of a clinical response of KS to cidofovir alone in an immunocompetent individual, although no serological response was detected. Previous reports have been in patients who may have had confounding factors owing to immunosuppression either from AIDS 5, 13 or corticosteroids. 13 Intralesional cidofovir in classical KS has been shown to be ineffective. 14 Cidofovir is a nucleotide analogue and acts as a viral DNA polymerase inhibitor during lytic replication. However, most spindle cells are latently infected with HHV‐8. 15 In endothelial cell culture, a subset (1–6%) of cells was infected with HHV‐8 and lytic viral antigens could be demonstrated. 2 HHV‐8‐infected endothelial cells supported growth by upregulation of the vascular endothelial growth factor receptor, KDR. Cidofovir may also have a direct antiangiogenic effect: in a neonatal rat model of murine polyoma virus‐induced haemangiomas, haemangioma formation was potently inhibited by cidofovir. 16 This was not an antiviral effect, as the polyoma virus does not replicate in this model. The possible mechanism of action of cidofovir in KS could therefore be a combined effect of direct antiangiogenesis and inhibition of the small percentage of lytic HHV‐8 virus in KS lesions, blocking paracrine growth promotion of other endothelial and spindle cells in the lesion. The role of antiherpesvirus therapy may be in the prevention rather than the treatment of KS. 7, 8 HHV‐8 seropositivity in HIV‐seropositive men is predictive for the development of KS, 17 and the identification of this subgroup at high risk of developing KS may provide a useful cohort in which to study cidofovir.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.