BRIEF COMMUNICATION Prevalence of Human Herpesvirus 8 Infection Before the Acquired Immuno- deficiency Disease Syndrome- Related Epidemic of Kaposi’s Sarcoma in East Africa

The etiologic role of the recently discovered Kaposi’s sarcoma (KS)associated herpesvirus (KSHV), also called human herpesvirus 8 (HHV8), in the development of all types of KS is now generally accepted (1,2). Long before the epidemic of the acquired immunodeficiency disease syndrome (AIDS) began, East Africa was known as a highrisk area for KS. For example, during the quinquennium 1964 through 1968, the annual incidence of KS in Uganda was 14.6 per million in males, representing approximately 4.6% of the pathologically examined tumors (3). By comparison, Wabinga et al. (4) observed that the annual incidence of KS in Uganda in 1993 was approximately 300 per million in males, representing 48.6% of all cancers diagnosed histologically or clinically among males at Makarere University Hospital in Uganda. To see whether the increase in the incidence of KS was associated with an increase in KSHV/HHV8 seroprevalence, we tested samples of sera obtained from the International Agency for Research on Cancer (IARC), Lyon, France. These serum samples had been collected in the early 1970s in the framework of the Ugandan prospective Burkitt’s lymphoma study (5). This allowed us to assess the KSHV/HHV8 seroprevalence in the rural populations of the West Nile District in Uganda and the North Mara District in Tanzania during the period 1972 through 1975, i.e., the period prior to the onset of the AIDSrelated epidemic of KS. Serum samples from a total of 300 children and related adults were tested: sera from 118 individuals from the West Nile District and sera from 182 individuals from the North Mara District, collected, respectively, in the period 1972 through 1975 (5) and in the period 1977 through 1978 (6). These series included sera from 95 children aged 5–9 years (46 with Burkitt’s tumor and 49 control subjects) and sera from 112 adults aged 20–39 years representing family members of these children, 64 newborns, as well as 15 endemic adult case subjects with KS from Uganda and 14 adult control subjects from Uganda (7). The serum samples were tested at a 1 : 20 dilution for the presence of antibodies to KSHV/HHV8 by an immunofluorescence assay (employing an HHV8 immunofluorescence test kit from Advanced Biotechnologies, Inc., Columbia, MD) by use of the KS-1 cell line as the source of both lytic antigen(s) and latency-associated nuclear antigen(s) of KSHV/HHV8 (8). As seen in Table 1, sera from 15 of 15 endemic adult Ugandan KS subjects were positive for KSHV/HHV8 antibodies serving as positive reference sera, whereas sera from seven (35%) of 20 children with Burkitt’s lymphoma from Tanzania and 11 (42%) of 26 children with Burkitt’s lymphoma from Uganda exhibited KSHV/HHV8 positivity, with similar rates in control children, i.e., 15 (44%) of 34 children from Tanzania and five (33%) of 15 children from Uganda. As seen in Table 2, there was an agerelated increase in KSHV/HHV8 seroprevalence, ranging from 33%–35% seropositivity in the age group 4–10 years old to 57% in the age group 11–13 years old and reaching 84% in adults. As expected, most (88%) newborns from KSHV/HHV8-seropositive mothers exhibited KSHV/HHV8 antibodies in their sera (Table 1). These results indicate that, during the 1970s, inhabitants of rural areas of Uganda and Tanzania exhibited an overall KSHV/HHV8 seroprevalence similar to that observed in recent years by Mayama et al. (9), who studied 215 outpatients aged 0–24 years in Nsambya Hospital in Kampala, Uganda, and observed the prevalence of KSHV/HHV8 antibodies to both latency-associated nuclear antigen(s) and lytic antigen(s) in 37% of the children younger than 5 years and in 49% of young adults. Our data support those published by Olsen et al. (10) that indicated a 65% KSHV/ HHV8-seropositive rate in serum samples collected from individuals in Gambia in 1985. This result suggests that HHV8 infection has been prevalent in Africa for a long time. It is worth noting that the mean age at which KS occurred in children was