To serotype or not to serotype: that is still the question.

Infectious Diseases, National Institutes of Health (NIH). The NIH has patents, pending patent applications, and a license for rhesus and rhesus-human rotavirus reassortant vaccines; patents, pending patent applications, and licenses for human-bovine (UK) rotavirus reassortant vaccines; a patent for cold-adapted human rotavirus vaccines; and a pending patent application for human-porcine rotavirus reassortant vaccines. In this issue of the Journal, the report by Banerjee et al. [1] on the postneonatal protective effect of a prior neonatal infection with G10P[11] rotavirus has important implications for rotavirus vaccinology. The protective efficacy of a neonatal ro-tavirus infection has been addressed previously by various investigators. In a 1983 Australian study, neonates with asymp-tomatic rotavirus shedding during the first 14 days of life developed fewer and less-severe rotavirus-positive diarrheal illnesses postneonatally than did control children [2]. The available strains causing neonatal infection had identical electrophoretic patterns (in addition, 1 strain was identified as an " M-like " [G3] strain [3, 4]), whereas the strains causing postneonatal infection had varying patterns. Although serotypic characterization was not readily available at the time, evidence now indicates that asymptomatic neonatal G3 infections protected against community rotavirus strains, including heterotypic serotype G1 and G2 strains, during the 3-year period of surveillance [5, 6]. Two groups set out to confirm this concept by comparing the incidence of post-neonatal rotavirus illnesses in a cohort of infants infected during the neonatal period with a G9[P11] rotavirus strain 116E or with a G10P[11] strain I321 [7, 8], the latter strain being similar to the neonatal strain described by Banerjee et al. in this issue of the Journal [1]. These studies also found that a neonatal infection induced protection against postneonatal rotavirus illnesses. Data on protection against specific serotypes were not available; however, in the study on strain I321, the postneo-natal strains had varying electrophoretic patterns, suggesting heterotypic protection. A variety of other longitudinal studies of children, beginning at birth and extending up to 3 years of age, have examined the protective efficacy of a primary rotavirus infection against subsequent ro-tavirus exposure [9–13]. However, the degree of protection was found to be inconsistent , ranging from none to substantial. Substantial protection was exemplified by a Mexican study in which a single rota-virus infection induced 87% protection against moderate-to-severe rotavirus-positive diarrhea, and 2 infections induced 100% protection. When a second infection occurred, the serotype responsible tended to be different from the original infecting strain [13]. Moreover, high …