A Phase 2a Randomized, Double-Blind, Dose-Optimizing Study to Evaluate the Immunogenicity and Safety of a Bivalent DNA Vaccine for Hemorrhagic Fever with Renal Syndrome Delivered by Intramuscular Electroporation

Abstract

Hantaan virus (HTNV) and Puumala virus (PUUV) are pathogenic hantaviruses found in Asia and Europe, respectively. DNA vaccines targeting the envelope glycoproteins of these viruses have been constructed and found to elicit neutralizing antibodies when delivered to humans by various technologies including intramuscular electroporation. Here, we report findings from a Phase 2a clinical trial of a combined HTNV/PUUV DNA vaccine delivered at varying doses and administration schedules using the Ichor Medical Systems TriGrid intramuscular electroporation delivery technology. The study was designed to characterize the effects of DNA vaccine dose and number of administrations on the frequency and magnitude of immunological response. Subjects (n = 120) were divided into four cohorts. Cohorts 1 and 2 received a dose of 2 mg of DNA (1 mg per plasmid), and cohorts 3 and 4 received a dose of 1 mg of DNA (0.5 mg per plasmid) each vaccination. Each of the four cohorts received a series of four administrations (days 0, 28, 56 and 168). For cohorts 1 and 3, the DNA vaccine candidate was delivered at each of the four administrations. For cohorts 2 and 4, in order to maintain blinding, subjects received the DNA vaccine on days 0, 56 and 168, but on day 28 received only the phosphate buffered saline vehicle rather the DNA vaccine. Sera were collected on days 0, 28, 56, 84, 140, 168, 196, 252 and 365 and evaluated for the presence of neutralizing antibodies by PUUV and HTNV pseudovirion neutralization assays (PsVNAs). Day 84 was also evaluated by a plaque reduction neutralization test (PRNT). Overall the PsVNA50 geometric mean titers (GMTs) and seropositivity rates among cohorts were similar. Cohort 3 exhibited the highest frequency of subjects that became seropositive to both PUUV and HTNV after vaccination, the highest peak GMT against both viruses, and the highest median titers against both viruses.

Keywords: DNA vaccine; electroporation; hantavirus; hemorrhagic fever with renal syndrome.

Figure 1

Clinical trial.

Figure 2

Pseudovirion neutralization assays (PsVNA)50 titers. Puumala virus (PUUV) and Hantaan virus (HTNV) PsVNA50 titers for each subject at each timepoint grouped by cohort. Subjects were randomized and vaccinated with the indicated amount of DNA (either 2 mg or 1 mg) on the indicated day (0, 28, 56, 168 or 0, 56 and 168). (A) Legend. (B) PUUV and HTNV PsVNA50 titers. Note that the Y-axis for PUUV and HTNV graphs are different. The limit of quantitation was a PsVNA50 titer of 20 (grey shaded area). Dashed line indicates specimens that were not included in the efficacy evaluable population.

Figure 3

Anti-PUUV and anti-HTNV PsVNA50 seropositivity rates. (A) Legend. (B) PUUV and HTNV seropositivity rates for each cohort presented as line graphs. Percentages are based on the number of subject presenting non-missing data. (C) The same data plotted as bar graphs to highlight significant differences between groups. Percentages are based on the number of subject presenting non-missing data except Day 365. The only subjects who returned for Day 365 were the subpopulation that was still seropositive on the Day 252 visit. For Day 365 the percentage is based on the full cohort size of 30. * p-values from pairwise two-sided Fisher’s exact tests.

Figure 4

Seropositive to either HTNV or PUUV; or to both viruses. (A) Legend. (B) The percentage of subjects seropositive against either HTNV or PUUV at each timepoint were plotted as line graphs and bar graphs. (C) The percentage of subjects seropositive for both PUUV and HTNV at each timepoint was plotted. Percentages are based on the number of subject presenting non-missing data except day 365. The only subjects who returned for day 365 were the subpopulation that was still seropositive on the day 252 visit. * p-values from pairwise two-sided Fisher’s exact tests.

Figure 5

PsVNA geometric mean titers. (A) Legend. The PsVNA50 GMT for PUUV and HTNV presented as line graph (B) or bar graph (C). The limit of quantitation was 20 (grey area). Values <20 were given a value of 14.1. * p-value from post-ANOVA pairwise comparison based on the analysis of log10-transformed titers.

Figure 6

Peak PsVNA geometric mean titers (Day 196). Cohort geometric mean titers peaked on day 196. Cohorts are the same as defined in Figure 1A. Individual PsVNA50 titers were sorted lowest to highest and plotted for HTNV (top panel) and PUUV (bottom panel). GMT, median and % seropositive are plotted. Highest values are highlighted in bold text.

Figure 7

(A,B) In each graph, the black points represent the PRNT50-PsVNA50 pairs. The red line represents the ideal linear function of perfect correlation. The black line represents the observed linear function corresponding to the r value, and the blue lines represent the 95% confidence interval around the linear function. (C,D) In each graph, the blue dots represent the difference between PRNT50 and PsVNA50 for each subject plotted against each observation’s percentile value. (E,F) In each graph, the black dots represent the difference between PRNT50 and PsVNA50 plotted against the average of the PRNT50 and the PsVNA50 for each subject. Blue lines represent the mean bias and limits of agreement. (G) The bias estimates and limits of agreement from the Bland–Altman analysis are shown. (H) The results of the Deming regression model are shown. If the confidence interval for the intercept contains 0, then it can be concluded that a constant bias does not exist between the two methods. If the confidence interval for the slope contains 1, then it can be concluded that a proportional bias does not exist between the two methods.