Long-term immunity against yellow fever in children vaccinated during infancy: a longitudinal cohort study

Abstract

Background: A single dose of vaccine against yellow fever is routinely administered to infants aged 9-12 months under the Expanded Programme on Immunization, but the long-term outcome of vaccination in this age group is unknown. We aimed to evaluate the long-term persistence of neutralising antibodies to yellow fever virus following routine vaccination in infancy.

Methods: We did a longitudinal cohort study, using a microneutralisation assay to measure protective antibodies against yellow fever in Malian and Ghanaian children vaccinated around age 9 months and followed up for 4·5 years (Mali), or 2·3 and 6·0 years (Ghana). Healthy children with available day-0 sera, a complete follow-up history, and no record of yellow fever revaccination were included; children seropositive for yellow fever at baseline were excluded. We standardised antibody concentrations with reference to the yellow fever WHO International Standard.

Findings: We included 587 Malian and 436 Ghanaian children vaccinated between June 5, 2009, and Dec 26, 2012. In the Malian group, 296 (50·4%, 95% CI 46·4-54·5) were seropositive (antibody concentration ≥0·5 IU/mL) 4·5 years after vaccination. Among the Ghanaian children, 121 (27·8%, 23·5-32·0) were seropositive after 2·3 years. These results show a large decrease from the proportions of seropositive infants 28 days after vaccination, 96·7% in Mali and 72·7% in Ghana, reported by a previous study of both study populations. The number of seropositive children increased to 188 (43·1%, 95% CI 38·5-47·8) in the Ghanaian group 6·0 years after vaccination, but this result might be confounded by unrecorded revaccination or natural infection with wild yellow fever virus during a 2011-12 outbreak in northern Ghana.

Interpretation: Rapid waning of immunity during the early years after vaccination of 9-month-old infants argues for a revision of the single-dose recommendation for this target population in endemic countries. The short duration of immunity in many vaccinees suggests that booster vaccination is necessary to meet the 80% population immunity threshold for prevention of yellow fever outbreaks.

Funding: Wellcome Trust.

Figure 1

Forest plots of the prevalence of seropositive children Seropositive participants at the antibody concentration threshold of 0·5 IU/mL or more are shown at the top of the figure, and seropositive and borderline study participants are presented at the bottom as a merged category including all participants with a measurable titre. p values test the differences of proportions between groups.

Figure 2

Reverse cumulative distribution plots of neutralising antibody values (A) Raw titres in the plaque reduction seroneutralisation assay for the full study groups. (B) Standardised antibody concentrations for broadly seropositive children (ie, seropositive and borderline study participants). The dotted line denotes the threshold for strict seropositivity at 0·5 IU/ml.

Figure 3

Evolution of the Ghanaian cohort Flow of study participants from a year-2·3 to a year-6·0 antibody concentration stratum. The thickness of the nodes (initial and final) and links between them are proportional to the number of participants as shown on the figure.