Thirty-one infants living in Accra, Ghana, an area endemic for Burkitt's lymphoma, were visited a t monthly intervals for the f i r s t 15 months of life and once again a t 21 months. Sera obtained a t each visit were tested for antibodies t o EpsteinBarr virus (EBV) capsid antigen (VCA), EBVinduced early antigens (EA) and EBV-associated nuclear antigen (EBNA) by immunofluorescence techniques as well as for EBV-neutralizing antibodies and antibody-dependent cell-mediated cytolysis. A l l infants had maternal antibodies t o VCA at 1 month of age and the majority also antibodies detected by the other tests, except anti-EA. The maternal anti-VCA titers declined in subsequent months according t o a half-life of nearly 5 weeks.Depending on the initial titers, all infants became seronegative for all antibodies by 2-8 months of age, with anti-VCA persisting longest. The earliest primary EBV infection occurred i n the third month after maternal antibodies t o VCA had become non-detectable. By 12 months of age, 44% of the infants had seroconverted, by 15 months, 62% and by 21 months, 81%. There was no seasonal variation i n the incidence of infections. Infants from families of the lowest socio-economic level tended t o become infected somewhat earlier than those from better situated families (p = 0.22) but no other factors, such as household size o r crowding, seemed t o be associated w i t h enhanced r i s k of early infection.The Epstein-Barr virus (EBV) is now generally accepted as the cause of infectious mononucleosis (IM) (Henle and Henle, 1972;Miller, 1975) and it is also intimately associated with Burkitt's lymphoma (BL) and nasopharyngeal carcinoma Klein, 1975;zur Hausen, 1975). Seroepidemiologic surveys have shown that antibodies to EBV are acquired generally earlier in life under conditions of poor hygiene and/or crowding than under high living standards (Henle et al., 1969; Henle and Henle, 1970; Tischendorf et al., 1970). It has been suggested that primary EBV infections in early infancy set the stage for subsequent development of BL (de-The, 1977) but no pertinent longitudinal studies to corroborate this assumption have been reported.The present study was initiated in an area of endemic BL to provide basic epidemiologic, clinical and serologic data on primary EBV infections at an early age. The results will be reported in two sections. This report describes the decline of maternal antibodies in relation to the timing of primary infections and the influence of epidemiologic factors. In the other report (following paper) the clinical, serologic and virologic features of primary infections in these infants will be recorded.

Of 27 Ghanaian infants examined monthly for the first 15 months of life and once more a t 21 months, 12 acquired primary Epstein-Barr virus (EBV) infections by the age of 1 year and 9 others seroconverted during the subsequent period of observation. The seroconversions were not accompanied by significant clinical o r physical signs of illness and in none of the infants was a diagnosis of infectious mononucleosis (IM) suspected on the basis of hematologic observations. The 12 early seroconverters, providing the most extensive follow-up data, showed transient IgM antibody responses t o EB viral capsid antigen (VCA) which i n some cases initially exceeded the corresponding IgG antibodies in titer. Peak VCA-specific IgG titers were noted 2 months after seroconversion and were comparable t o those seen i n IM. Most of the infants developed antibodies t o the EBVinduced early antigen complex which were directed, however, against the R (restricted) component, as observed i n Burkitt's lymphoma, and not against the D (diffuse) component, as noted i n IM. Low titers of VCA-specific IgA antibodies emerged i n five of the cases. Usually, EBV-neutralizing antibodies were already present i n the first anti-VCApositive serum but antibodies t o the EBV-associated nuclear antigen (EBNA) and antibody-dependent cell-mediated cytolysis (ADCC) became detectable only months after seroconversion. N o heterophil antibody responses, o r at most barely significant ones, were recorded. Attempts t o demonstrate EBV i n throat swab specimens met with limited success, suggesting a low degree of viral excretion during silent seroconversions. Possible explanations for the differences between clinical, hematologic and serologic responses t o primary EBV infections i n infancy and later i n life are discussed.

No abstract