Long-term efficacy and safety of a tetravalent dengue vaccine (TAK-003): 4·5-year results from a phase 3, randomised, double-blind, placebo-controlled trial.

BACKGROUND About half of the world's population lives in dengue-endemic areas. We aimed to evaluate the long-term efficacy and safety of two doses of the tetravalent dengue vaccine TAK-003 in preventing symptomatic dengue disease of any severity and due to any dengue virus (DENV) serotypes in children and adolescents. METHODS In this ongoing double-blind, randomised, placebo-controlled trial, we enrolled healthy participants aged 4-16 years at 26 medical and research centres across eight dengue-endemic countries (Brazil, Colombia, Dominican Republic, Nicaragua, Panama, Philippines, Sri Lanka, and Thailand). The main exclusion criteria were febrile illness (body temperature ≥38°C) at the time of randomisation, hypersensitivity or allergy to any of the vaccine components, pregnancy or breastfeeding, serious chronic or progressive disease, impaired or altered immune function, and previous receipt of a dengue vaccine. Participants were randomly assigned 2:1 (stratified by age and region) using an interactive web response system and dynamic block assignment to receive two subcutaneous doses of TAK-003 or placebo 3 months apart. Investigators, participants, and their parents or legal guardians were blinded to group assignments. Active febrile illness surveillance and RT-PCR testing of febrile illness episodes were performed for identification of virologically confirmed dengue. Efficacy outcomes were assessed in the safety analysis set (all randomly assigned participants who received ≥1 dose) and the per protocol set (all participants who had no major protocol violations), and included cumulative vaccine efficacy from first vaccination to approximately 4·5 years after the second vaccination. Serious adverse events were monitored throughout. This study is registered with ClinicalTrials.gov, NCT02747927. FINDINGS Between Sept 7, 2016, and March 31, 2017, 20 099 participants were randomly assigned (TAK-003, n=13 401; placebo, n=6698). 20 071 participants (10 142 [50·5%] males; 9929 [49·5%] females; safety set) received TAK-003 or placebo, with 18 257 (91·0%) completing approximately 4·5 years of follow-up after the second vaccination (TAK-003, 12 177/13 380; placebo, 6080/6687). Overall, 1007 (placebo: 560; TAK-003: 447) of 27 684 febrile illnesses reported were virologically confirmed dengue, with 188 cases (placebo: 142; TAK-003: 46) requiring hospitalisation. Cumulative vaccine efficacy was 61·2% (95% CI 56·0-65·8) against virologically confirmed dengue and 84·1% (77·8-88·6) against hospitalised virologically confirmed dengue; corresponding efficacies were 53·5% (41·6-62·9) and 79·3% (63·5-88·2) in baseline seronegative participants (safety set). In an exploratory analysis, vaccine efficacy was shown against all four serotypes in baseline seropositive participants. In baseline seronegative participants, vaccine efficacy was shown against DENV-1 and DENV-2 but was not observed against DENV-3 and low incidence precluded evaluation against DENV-4. During part 3 of the trial (approximately 22-57 months after the first vaccination), serious adverse events were reported for 664 (5·0%) of 13 380 TAK-003 recipients and 396 (5·9%) of 6687 placebo recipients; 17 deaths (6 in the placebo group and 11 in the TAK-003 group) were reported, none were considered study-vaccine related. INTERPRETATION TAK-003 demonstrated long-term efficacy and safety against all four DENV serotypes in previously exposed individuals and against DENV-1 and DENV-2 in dengue-naive individuals. FUNDING Takeda Vaccines. TRANSLATIONS For the Portuguese, Spanish translations and plain language summary of the abstract see Supplementary Materials section.