ANTI-HERPESVIRUS TREATMENT AND RISK OF KAPOSIS-SARCOMA IN HIV-INFECTION

Objective: With the recent identification of a new herpesvirus in pati ents with Kaposi's sarcoma (human herpesvirus-8 or Kaposi's sarcoma-as sociated herpesvirus), there have been several reports on the use of a nti-herpesvirus therapy (foscarnet, ganciclovir and aciclovir) and ris k of developing Kaposi's sarcoma. We therefore investigated the associ ation between use of anti-herpesvirus drugs and Kaposi's sarcoma in a large unselected group of patients with AIDS. Patients and methods: We studied a group of HIV-positive patients at the Chelsea and Westminst er Hospital, for whom details on all AIDS-defining diagnoses made duri ng follow-up, treatment and regular CD4 counts were available. Cox pro portional hazards models with time-dependant covariates were used to a ssess the association between treatment with aciclovir, foscarnet and ganciclovir and risk of Kaposi's sarcoma. Results: A total of 3688 pat ients have been followed up for a median period of 4.2 years, during w hich time 598 patients (16.2%) developed Kaposi's sarcoma. After adjus tments for sex, exposure category, age, treatment with antiretrovirals or Pneumocystis carinii pneumonia prophylaxis, the development of AID S-defining conditions (including separate adjustment for the developme nt of cytomegalovirus and herpes simplex virus) and CD4 count, there w as a decreased risk of developing Kaposi's sarcoma with foscarnet [rel ative hazard (RH), 0.38; 95% confidence interval (CI); 0.15-0.95; P = 0.038] and with ganciclovir (RH, 0.39; 95% CI, 0.19-0.84; P = 0.015), but not with aciclovir (RH, 1.10; 95% CI, 0.88-1.38; P = 0.40). Conclu sions: These results suggest that both foscarnet and ganciclovir may h ave some activity in preventing the occurrence of Kaposi's sarcoma, bu t that aciclovir has no benefit. Further studies of the effect of thes e drugs on the risk of Kaposi's sarcoma is warranted.