Safety and immunogenicity of a tetravalent dengue vaccine in children aged 2–17 years: a randomised, placebo-controlled, phase 2 trial

Summary Background An unmet clinical need remains for an effective tetravalent dengue vaccine suitable for all age groups, regardless of serostatus. We assessed the immunogenicity and safety of three different dose schedules of a tetravalent dengue vaccine (TAK-003) over a 48-month period in children living in dengue-endemic countries. Methods We did a large, phase 2, double-blind, placebo-controlled trial at three sites in the Dominican Republic, Panama, and the Philippines. Healthy participants aged 2–17 years were randomly assigned 1:2:5:1 using an interactive web response system with stratification by age to receive either a two-dose primary series (days 1 and 91), one primary dose (day 1), one primary dose plus booster (days 1 and 365), or placebo. Participants and relevant study personnel were masked to the random assignment until completion of the study at month 48. To maintain masking, TAK-003 recipients were administered placebo doses when appropriate. The primary objective was assessment of neutralising geometric mean titres for each serotype to month 48 assessed in the per-protocol immunogenicity subset. Secondary safety endpoints included proportions of participants with serious adverse events and symptomatic virologically confirmed dengue. This study is registered with ClinicalTrials.gov, NCT02302066. Findings Between Dec 5, 2014, and Feb 13, 2015, 1800 children were randomly assigned to the following groups: two-dose primary series (n=201), one primary dose (n=398), one primary dose plus 1-year booster (n=1002), and placebo (n=199). Of them, 1479 (82%) participants completed the 48-month study. Immunogenicity endpoints were assessed in 562 participants enrolled in the immunogenicity subset, of whom 509 were included in the per-protocol subset. At month 48, antibody titres remained elevated in all TAK-003 groups compared with placebo, irrespective of baseline serostatus. At month 48, geometric mean titres were 378 (95% CI 226–632) in two-dose, 421 (285–622) in one-dose, 719 (538–960) in one-dose plus 1-year booster, and 100 (50–201) in placebo recipients against DENV 1; 1052 (732–1511), 1319 (970–1794), 1200 (927–1553), and 208 (99–437) against DENV 2; 183 (113–298), 201 (135–298), 288 (211–392), and 71 (37–139) against DENV 3; and 152 (97–239), 164 (114–236), 219 (165–290), and 46 (26–82) against DENV 4; and tetravalent seropositivity rate was 89% (79–96), 86% (80–92), 97% (93–99), and 60% (47–72), respectively. Virologically confirmed dengue was recorded in 37 (2%) TAK-003 and 13 (7%) placebo participants, with a relative risk of 0·35 (0·19–0·65). No vaccine-related serious adverse events or severe dengue virus disease were reported. Interpretation TAK-003 elicited antibody responses against all four serotypes, which persisted to 48 months post-vaccination, regardless of baseline serostatus. No important safety risks were identified. We observed a long-term reduction in risk of symptomatic dengue virus disease in vaccinees. Results from this study provide a long-term safety database and support assessment of the vaccine in the ongoing phase 3 efficacy study. Funding Takeda Vaccines.