Zika and Dengue Interactions in the Context of a Large Dengue Vaccine Clinical Trial in Latin America

Abstract

A phase III dengue vaccine trial including 9- to 16-year-olds in Latin America (NCT01374516) was ongoing at the time of a Zika outbreak. We explored interactions between dengue and Zika, in the context of dengue vaccination. Symptomatic virologically confirmed Zika (VCZ) was evaluated using acute-phase sera from febrile participants (January 2013-March 2018). Neutralizing antibody geometric mean titers (GMTs) were evaluated pre- and post-Zika outbreak (months 25 and 72) in 2,000 randomly selected participants. Baseline dengue serostatus was determined using the plaque reduction neutralization test or inferred post hoc using nonstructural protein 1 IgG ELISA at M13 (case-cohort analysis). Vaccine efficacy against VCZ and serologically suspected Zika (SSZ) was estimated. Overall, 239/10,157 (2.4%) acute-phase samples were VCZ positive during the study. Dengue vaccine efficacy against VCZ was 27.8% (95% CI: 0.3; 47.7) among baseline dengue-seropositive participants. No vaccine effect was evident against SSZ. Zika antibody GMTs increased from pre- to post-Zika epidemic, with smaller increases observed for participants who were dengue seropositive at baseline than for those who were dengue seronegative: post-/pre-Zika GMT ratios for baseline dengue-seropositive participants were 21.5 (vaccine group) and 30.8 (placebo); and for dengue seronegatives, 88.1 and 89.5, respectively. Dengue antibody GMTs post-Zika were higher in dengue vaccine and placebo recipients with SSZ than those without SSZ in both dengue seropositives and seronegatives. Dengue vaccine did not enhance symptomatic Zika illness in dengue-seropositive individuals, rather it reduced the risk of VCZ. Zika infection boosted preexisting vaccine-induced or naturally occurring dengue-neutralizing antibodies.

Figure 1.

Virologically confirmed Zika (VCZ) cases per quarter, per calendar year in the live, attenuated, tetravalent dengue vaccine (CYD-TDV) and placebo groups. Participants were randomized in a 2:1 ratio to CYD-TDV and placebo groups. The colored bars below the graph depict when the different phases of the trial were established and the period of peak Zika incidence for each country. During the hospital phase, only samples from hospitalized cases with suspected dengue were tested; no VCZ cases were detected during this phase of the trial. This figure appears in color at www.ajtmh.org.

Figure 2.

Efficacy of the live, attenuated, tetravalent dengue vaccine (CYD-TDV) against virologically confirmed Zika (VCZ) by dengue baseline serostatus (as per nonstructural protein 1 [NS1] titers at M13) throughout the study period, from 2013 – case–cohort analysis. Participants with virologically confirmed dengue before month 13 or with undetermined dengue serostatus at M13 were excluded from analysis. Baseline dengue status was determined by NS1 at month 13 using a threshold of 9 EU/mL. VE was inferred from the proportional hazard ratio: VE = (1 − HR) × 100. N = the total number of participants included in the subcohort with serostatus and treatment group as specified; n = number of participants with virologically confirmed Zika (cases), with available data for the relevant endpoint; VCZ = virologically confirmed Zika.

Figure 3.

Clinical signs and symptoms reported for virologically confirmed Zika (VCZ) cases occurring throughout the study, by baseline dengue serostatus (safety analysis set). The y-axis shows the percentage of virologically confirmed Zika cases with available data for the relevant end point, who fulfilled criteria for the specified sign or symptom. Baseline dengue serostatus was inferred from dengue nonstructural protein 1 assay at M13. N = the number of virologically-confirmed Zika cases with baseline dengue serostatus as indicated; VCZ = virologically-confirmed Zika.

Figure 4.

Efficacy of the live, attenuated, tetravalent dengue vaccine (CYD-TDV) against serologically suspected Zika (SSZ) from the start of 2013 to the study end by baseline dengue serostatus (50% plaque reduction neutralization test [PRNT₅₀] at M0)—immunogenicity subset. N, number of person-years followed from the start of 2013; n: number of participants with Zika titers (MN) at M72 ≥ 100 1/dil; SSZ, serologically suspected Zika. Baseline dengue status by PRNT₅₀ at M0.

Figure 5.

Zika geometric mean titers (GMTs) before and after the Zika epidemic, by baseline dengue serostatus and by treatment group—immunogenicity subset. Zika antibody GMTs, measured by Zika microneutralization assay, with 95% CI, are plotted on the graphs, with data labels showing point estimates. Only participants with Zika titers (MN) at M72 ≥ 100 1/dil were included in this analysis. GMT, geometric mean titer; Pre-Zika, study month 25, before the first serologically confirmed Zika cases reported by national surveillance systems; Post-Zika, study month 72, after the peak incidence of observed Zika. Baseline dengue seronegative participants are defined as those with dengue 50% plaque reduction neutralization test (PRNT₅₀) titers < 10 (1/dil) against all four serotypes at baseline; baseline dengue-seropositive participants are defined as those with dengue PRNT₅₀ titers ≥ 10 (1/dil) against at least one dengue serotype at baseline.

Figure 6.

Dengue serotype–specific antibody responses before and after peak Zika incidence in those with (A) and without (B) serological evidence of Zika by baseline dengue serostatus in the immunogenicity subset. Pre-Zika represents study month 25, before the first serologically confirmed Zika cases reported by national surveillance systems. Post-Zika represents study month 72, after the peak incidence of observed Zika. Participants with serological evidence of Zika had Zika titers (microneutralization assay) at M72 ≥ 100 1/dil; participants without serological evidence of Zika had M72 titers < 100 1/dil. Dengue baseline serostatus was determined by 50% plaque reduction neutralization test at M0; dengue baseline seronegative (dengue negative) was defined as titers < 10 (l/dil) against all four serotypes at baseline and dengue baseline seropositive (dengue positive), titers ≥ 10 (l/dil) against at least one dengue serotype at baseline. CI = confidence interval; GMT = geometric mean titer; M = month.