Safety and Immunogenicity of Different Formulations of Norovirus Vaccine Candidate in Healthy Adults: A Randomized, Controlled, Double-Blind Clinical Trial

Abstract

Background: We investigated safety and immunogenicity of 1-2 doses of different bivalent virus-like particle (VLP) norovirus vaccine candidate (NoV) formulations in healthy 18- to 64-year-olds.

Methods: On days 1 and 28, participants (n = 420) randomized to 14 equal groups received intramuscular control vaccine (hepatitis A) or 1 of 11 NoV formulations containing varying dosages of GI.1 and GII.4c genotype VLP antigens with aluminum hydroxide [Al(OH)3], and 0 μg, 15 μg, or 50 μg monophosphoryl lipid A (MPL). Immunogenicity was assessed on days 1, 28, 56, 208 and 393. Solicited local and systemic reactions were recorded for 7 days, unsolicited adverse events (AEs) until day 56, and serious AEs throughout the trial.

Results: All NoV formulations induced similar increases in pan-immunoglobulin, immunoglobulin A, and histo-blood group binding antigen-blocking antibodies by day 56, mostly after 1 dose, that persisted above baseline to day 393. Higher GI.1 content interfered with GII.4c responses, and responses did not benefit from MPL. Overall reactogenicity consisted of mainly mild injection site pain, headache, and fatigue. No vaccine-related serious AEs were reported.

Conclusions: All candidate NoV formulations were well tolerated. Overall, 15 μg GI.1/50 μg GII.4c elicited the best balance of immunogenicity with no clear benefit of MPL, and is the candidate formulation being taken forward in clinical development.

Clinical trials registration: NCT02038907.

Keywords: adults; immunogenicity; norovirus; reactogenicity; vaccine.

Figure 1.

Geometric mean titers (95% confidence interval) of pan-immunoglobulin (Pan-Ig) against GI.1 and GII.4c virus-like particle (VLP) antigens. Upper panels show grouping by VLP antigen dosages, irrespective of the inclusion or dosage of monophosphoryl lipid A (MPL) (VLP 15/15 = groups 1, 4, and 7; VLP 15/50 = groups 2, 5, and 8A; VLP 50/50 = groups 3, 6, and 9). Lower panels show grouping by MPL dosage irrespective of VPL composition (0 μg MPL = groups 7, 8A, and 9; 15 μg MPL = groups 4, 5, and 6; 50 μg MPL = groups 1, 2, and 3).

Figure 2.

Geometric mean (95% confidence interval) of histo-blood group binding antigen (HBGA) antibodies (GMBT₅₀) against GI.1 and GII.4c virus-like particle (VLP) antigens. Upper panels show grouping by VLP antigen dosages, irrespective of the inclusion or dosage of monophosphoryl lipid A (MPL) (VLP 15/15 = groups 1, 4, and 7; VLP 15/50 = groups 2, 5, and 8A; VLP 50/50 = groups 3, 6, and 9). Lower panels show grouping by MPL dosage irrespective of VPL composition (0 μg MPL = groups 7, 8A, and 9; 15 μg MPL = groups 4, 5, and 6; 50 μg MPL = groups 1, 2, and 3).

Figure 3.

Geometric mean (95% confidence interval) of histo-blood group binding antigen (HBGA) antibodies (GMBT₅₀) against different GII.4 antigens: the GII.4c virus-like particle (VLP) antigen and those from GII.4 Cincinnati 2003 and GII.4 Sydney 2012 strains. Sera tested were from group 5 (n = 30) and group 8A (n = 32), which had received 15 μg monophosphoryl lipid A (MPL) and no MPL, respectively, in formulations with 15 μg GI.1 and 50 μg GII.4c VLP antigen dosages.