HHV-8 Infection in Homosexual Men From New York City (NYC) and Washington (DC): Frequency, Risk Factors, and Relationship to Kaposi's Sarcoma (KS)

Abstract

57 Background: The observations that HHV-8 is etiologically linked to KS and that KS is much commoner in homosexual men than in other HIV-1 risk groups suggest that HHV-8 is a non-ubiquitous, sexually transmitted infection. To clarify the epidemiology of HHV-8 and its relationship to KS, we measured antibodies to HHV-8 longitudinally in homosexual men from two U.S. cities. Methods: In 1982, 245 homosexual men without evidence of AIDS were enrolled in a prospective study of AIDS incidence. We collected blood specimens and sexual behavior information anually; follow-up of HIV-1-uninfected men ceased in 1991. Antibodies to latent HHV-8 antigens were measured in all serum specimens of sufficient volume by an immunofluorescence assay. We used X2 tests for statistical testing and Kaplan-Meier methods to examine the risk of KS after infection with both HIV-1 and HHV-8. Results: The prevalence of HHV-8 at study entry was 40% among NYC subjects and 14.5% among DC subjects (RR, 2.8; p=0.001). HHV-8 prevalence was also associated with HIV-1 infection (RR 3.4; p=0.001) and number of sex partners (p=0.001, X2 for trend), but the higher prevalence in NYC men persisted after considering these variables. DC men with one or more sex partners from NYC were 3.1-fold more likely to be infected with HHV-8 at study entry (p=0.01) than were DC men with no NYC partners. The incidence of HHV-8 infection subsequent to study entry was similar regardless of residence (NYC, 0.04/person-year; DC, 0.03/person-year; p=0.5), but was strongly associated with the number of sex partners (p=0.001, X2 for trend). Among 60 men infected with both HHV-8 and HIV-1, 15 have developed KS at a median of 45 months (range, 12 to 154 months); the ten-year KS-free survival among these dually-infected men was 62%. Conclusions: 1) HHV-8 infection was common among these homosexual men in the early 1980's, especially those residing in or having sex partners from NYC. This geographic difference in HHV-8 prevalence is consistent with differences in the KS risk reported among homosexual men from these cities. 2) HHV-8 prevalence and incidence were strongly associated with the number of sex partners in this population. 3) Persons infected with both HHV-8 and HIV-1 are at high risk of developing KS; the time from dual infection with these viruses to the diagnosis of KS varies widely and may exceed a decade.