Study design

Both studies were phase I, randomized, placebo-controlled, and observer blind. The design, inclusion/exclusion criteria and primary objectives for both trials have been previously described in detail12,13. Briefly, the study in the predominantly dengue-naive population (NCT01666652) was conducted at the Clinical Trials Center at the Walter Reed Army Institute of Research (WRAIR), Silver Spring, Maryland, United States and enrolled healthy male and female adults between 18 and 39 years of age. The second trial (NCT01702857) was conducted at the University of Puerto Rico Medical Sciences Campus, Puerto Rico Clinical and Translational Research Consortium Center, San Juan, Puerto Rico, United States in predominantly dengue-primed healthy male and female adults aged 20–39 years, who had lived in the Caribbean for more than 10 years. Immunogencitiy analyses were restricted to dengue-naive and dengue-primed individuals in the first and second trial, respectively. In both trials, participants were randomized 1:1:1:1:1 to receive two doses of different formulations of DPIV, administered four weeks apart (on days 0 and 28): 1 μg/serotype/dose adjuvanted with either alum (Group 1 μg + Alum), AS01E (Group 1 μg + AS01E), or AS03B (Group 1 μg + AS03B), 4 μg/serotype/dose adjuvanted with alum (Group 4 μg + Alum), or placebo. AS01E is an adjuvant system containing 3-O-desacyl-4’-monophosphoryl lipid A (MPL; GSK), QS-21 (Quillaja saponaria Molina, fraction 21) (licensed by GSK from Antigenics LLC, a wholly owned subsidiary of Agenus, Inc., a Delaware, United States corporation), and liposome (25 μg MPL and 25 μg QS-21). AS03B is an adjuvant system containing DL-α-tocopherol and squalene in an o/w emulsion (5.93 mg DL-α-tocopherol). A subset of nine participants in the NCT01666652 trial also received a third (booster) dose of DPIV, at 15–21 months after the second vaccine dose12. The vaccine composition was previously described in detail12.

Exploratory objectives, presented here, evaluated the CMI responses induced by the four DPIV formulations at days 0 (pre-vaccination), 7 (7 days post-dose 1), 28 (28 days post-dose 1/pre-dose 2), 35 (7 days post-dose 2), and 56 (28 days post-dose 2) and months 7 (6 months post-dose 2) and 13 (12 months post-dose 2) for B-cell responses and at days 0 and 56 for T-cell responses. The proliferative and cytokine-producing capacity of DPIV-elicited T cells were evaluated at all time points in a subset of dengue-naive adults. B-cell responses prior to and following a DPIV booster dose (pre-booster and at 7 and 28 days post-booster) and T-cell responses (pre-booster and at 28 days post-booster) were also evaluated in a subset of participants. At each time point, ~40 mL of blood were collected from participants in each subset.

Both studies were conducted in agreement with the Declaration of Helsinki, International Conference on Harmonization and Good Clinical Practice Principles, and their protocols and associated documents were approved by institutional review boards (NCT01666652: WRAIR Institutional Review Board, the Office of Research Protections, Human Research Protection Office, the U.S. Army Medical Materiel Development Activity, and GSK; NCT01702857: U.S. Army Human Subjects Research Review Board, Office of the Surgeon General, the U.S. Army Medical Materiel Development Activity, GSK, and the Western Institutional Review Board on behalf of the University of Puerto Rico). Written and signed informed consent to participate were obtained from eligible individuals or their legally authorized representative.

Immunogenicity assessmentsDENV serotype-specific memory B-cell responses

The frequency of DENV serotype-specific antibody-secreting cells (ASCs) per million ASCs was determined using an adapted ELISPOT method developed by Crotty et al.27. Briefly, the method consisted of a first incubation of PBMCs during 5 days at 37 °C and 5% CO2 in presence of cytosine phosphate guanine (CpG 2006 – 1 µg/mL; prepared in RPMI-1640 medium supplemented with additives and 10% heat-inactivated fetal bovine serum) for differentiation into memory B cells. In parallel, polyvinylidene difluoride plates were coated with either the antigens of interest (inactivated DENV-1–4 virions; coated at 5 µg/ml; prepared in Dulbecco’s phosphate-buffered saline [DPBS]), host-cell protein (control condition: cells lysate coated at 5 µg/mL, prepared in DPBS) or anti-human immunoglobulin G (IgG) (Affinipure Goat anti-human IgG ‘H + L’ from Jackson Laboratories; coated at 50 µg/mL; prepared in DPBS). Coated plates were stored at 4 °C. At day 5, washed and counted cells were loaded at concentration of 200,000 cells per specific antigens and control condition well (in triplicates) or at concentration of 2000 cells per IgG well (in triplicates). After overnight incubation (37 °C - 5% CO2), cell removal and washing steps, spot forming cells (SFCs) are revealed by the addition of a secondary antibody (biotynilated IgG; two hours at room temperature at concentration of 2 µg/mL; prepared in saturated buffer: DPBS + bovine serum albumin 10%+ fetal bovine serum 4%), addition of a conjugate complex (extravidin peroxydase; one hour at room temperature at concentration of 4 µg/mL; prepared in saturated buffer; after washing steps), and addition of 3 amino-9 ethyl-carbazole (10 minutes at room temperature, protected from light; after washing steps). After last washing steps and drying of plates overnight at room temperature, SFCs were counted by the ELISPOT plate reader. The results were expressed as the frequencies of antigen-specific memory B cells within the total memory B-cell population (proportion of antigen-specific memory B cells per million total memory B cells).

Peptides for T-cell stimulation assays

Pools of 12–20-mer-peptides overlapping by 10–12 amino acids and corresponding to the full-length E proteins of DENV-1 Nauru/West Pac/1974, DENV-2 New Guinea C, DENV-3 Philippines/H87/1956, and DENV-4 Singapore/8976/1995 were obtained from BEI Resources. Peptide pools (16 mers overlapping by 11 amino acids) covering both the C and M proteins of DENV-1–4 were purchased from JPT Peptide Technologies. Peptide pool stocks were reconstituted in 100% dimethyl sulfoxide (DMSO) at a concentration of 200 μg/mL/peptide and stored at –80 °C.

ICS assay

Cryopreserved PBMCs were thawed and 1 × 106 PBMC were plated per well of a 96-well plate in a total volume of 200 μL of R10 medium (RPMI-1640 medium supplemented with 10% fetal bovine serum, L-glutamine, penicillin, and streptomycin) along with anti-CD28, anti-CD49d, and fluorescein isothiocyanate-conjugated anti-CD107a antibodies (BD Biosciences) and 1 μg/mL/peptide of the relevant peptide pool. R10 containing 0.5% DMSO was used as a negative control, and a mixture of 50 ng/mL phorbol 12-myristate 13-acetate and 1 μg/mL ionomycin was used as a positive control. Cells were incubated at 37 °C for one hour prior to addition of brefeldin A and monensin (BD Biosciences) and then left to continue incubating overnight. The next day, cells were washed and stained with LIVE/DEAD Aqua (Invitrogen, Life Technologies, Thermo Fisher Scientific) followed by the surface antibodies Brilliant Violet 785-conjugated anti-CD3 (BV785-CD3), BV605-CD4, BV650-CD8 (BioLegend), Alexa700-CD14, and Alexa700-CD19 (BD Biosciences). After fixation in 4% formaldehyde, cells were permeabilized and stained with the antibodies eFluor450-IFN-γ, PE-Cy7-TNFα, PE-MIP-1β, APC-IL-2, and PE-Cy5-CD154 (BD Biosciences). Data were collected using a BD LSRFortessa flow cytometer (BD Biosciences).

Short-term in vitro culture

Cryopreserved PBMCs were thawed and plated at ~1 × 106 cells per well in R10 medium with 0.5 μg/mL/peptide of the respective DENV-1–4 CME peptide pool. R10 containing 0.5% DMSO was used as a negative control. Cells were cultured at 37 °C for 6–7 days, and the supernatants were stored at –80 °C for the cytokine analysis. The cells were washed and stained with LIVE/DEAD Aqua as well as BV785-CD3, BV605-CD4, Alexa700-CD8, BV510-CD14, BV510-CD19, APC-Cy7-CD16, PE-Cy7-CD56, PE-CD38, and PE-Dazzle594-HLA-DR (BD Biosciences). After fixation in 4% formaldehyde, cells were permeabilized and stained with Alexa488-Ki67 (BD Biosciences). Data were collected using a BD LSRFortessa flow cytometer (BD Biosciences).

Cytokine analysis

Short-term cell culture supernatants were thawed and tested (in duplicate, undiluted) using the Bio-Plex Pro™ Human Cytokine 27-plex kit (M500KCAF0Y, Bio-Rad) following the manufacturer’s instructions. Protein standards were provided in the kit, and standard curves were generated with eight standard dilutions (undiluted, 1:3, 1:9, 1:27, 1:81, 1:243, 1:729, 1:2187) using a five-parameter logistic curve fit and 1/y2 weighted function. Data were acquired on a MAGPIX instrument (Bio-Rad).

Statistical analysis

Considerations regarding the sample size and power calculations were previously reported in detail for each study12,13.

At each time point, the evaluation of CMI was performed in the adapted ATP cohort for immunogenicity, including all evaluable participants who had data available for the evaluated immunogenicity endpoint at each time point. In dengue-naive adults, the ATP immunogenicity cohort at day 56 was used for all time points up to day 56, and the ATP immunogenicity cohort at month 7/13 was used for month 7/13. In dengue-primed adults, the adapted ATP cohort for immunogenicity included all participants from the day 56 analysis who had blood samples available for immunogenicity testing.

The frequencies of DENV serotype-specific CD4+ or CD8+ T cells were calculated as the difference between the frequency of CD4+ or CD8+ T cells expressing immune markers upon in vitro stimulation with DENV antigen minus the frequency of CD4+ or CD8+ T cells expressing these immune markers upon in vitro stimulation with the negative control (medium plus 0.5% DMSO), called background. Differences less than or equal to zero were imputed to one antigen-specific immune marker-expressing CD4+ or CD8+ T cell per million CD4+ or CD8+ T cells.

Data were analyzed using software packages: FlowJo (Becton, Dickinson & Company); Microsoft Excel (Microsoft Corp.); and GraphPad Prism (GraphPad Software).

All analyses were descriptive.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.