An influenza hemagglutinin stem nanoparticle vaccine induces cross-group 1 neutralizing antibodies in healthy adults

Abstract

Influenza vaccines could be improved by platforms inducing cross-reactive immunity. Immunodominance of the influenza hemagglutinin (HA) head in currently licensed vaccines impedes induction of cross-reactive neutralizing stem-directed antibodies. A vaccine without the variable HA head domain has the potential to focus the immune response on the conserved HA stem. This first-in-human dose-escalation open-label phase 1 clinical trial (NCT03814720) tested an HA stabilized stem ferritin nanoparticle vaccine (H1ssF) based on the H1 HA stem of A/New Caledonia/20/1999. Fifty-two healthy adults aged 18 to 70 years old enrolled to receive either 20 μg of H1ssF once (n = 5) or 60 μg of H1ssF twice (n = 47) with a prime-boost interval of 16 weeks. Thirty-five (74%) 60-μg dose participants received the boost, whereas 11 (23%) boost vaccinations were missed because of public health restrictions in the early stages of the COVID-19 pandemic. The primary objective of this trial was to evaluate the safety and tolerability of H1ssF, and the secondary objective was to evaluate antibody responses after vaccination. H1ssF was safe and well tolerated, with mild solicited local and systemic reactogenicity. The most common symptoms included pain or tenderness at the injection site (n = 10, 19%), headache (n = 10, 19%), and malaise (n = 6, 12%). We found that H1ssF elicited cross-reactive neutralizing antibodies against the conserved HA stem of group 1 influenza viruses, despite previous H1 subtype head-specific immunity. These responses were durable, with neutralizing antibodies observed more than 1 year after vaccination. Our results support this platform as a step forward in the development of a universal influenza vaccine.

Fig. 1.. Study CONSORT diagram.

Fifty-two healthy adults aged 18 to 70 years old enrolled into the trial between 1 April 2019 and 9 March 2020 to receive either 20 μg of H1ssF once (n = 5) or 60 μg of H1ssF twice (n = 47) with a prime-boost interval of 16 weeks. The COVID-19 pandemic disrupted the administration of boost doses for 11 of 12 participants who missed the boost; the remaining participant withdrew because of relocation. Participants who had altered or discontinued vaccination schedules were monitored for safety and were included in the immunogenicity analysis until their vaccination schedules were changed.

Fig. 2.. Mild solicited reactogenicity was reported after H1ssF vaccination.

Percentage of participants (x axis) reporting solicited local or systemic symptoms (y axis) in the 7 days after each H1ssF vaccination. For symptoms persisting more than 1 day, a single count per person at the maximum severity of the symptom was used for the figure. Pain/tenderness was the only local symptom reported; no swelling or redness was noted. There were also no reports of fever or joint pain. All reactogenicity was mild. N/A indicates that no boost vaccination was administered to the dose group.

Fig. 3.. H1ssF elicits broad antibody responses to group 1 influenza HA antigens.

(A) Geometric means and 95% confidence intervals (CIs) are shown for all 60-μg dose recipients’ binding antibody concentrations. (B) Fold change of binding antibody concentrations from baseline for group 1 influenza virus antigens. Binding was assessed by ECLIA. (C) Neutralizing IC₈₀ antibody titers for three viruses and (D) fold change from baseline for two viruses, as assessed by reporter-based microneutralization assay. Half circles in (A) and (B) denote stabilized stem (ss) antigens. Dotted lines in (B) and (D) indicate baseline. In (D), H5N1 fold change calculations could not be performed because of low baseline titers. Exact numbers of participant samples analyzed at each time point are listed in table S4. AU, arbitrary units; NC/99, A/New Caledonia/20/1999; Sing/57, A/Singapore/1/1957; Indo/05, A/Indonesia/5/2005.

Fig. 4.. H1ssF vaccination decreases the baseline difference in H2 binding and neutralizing antibodies between H2-exposed and H2-naïve individuals.

(A and B) Binding antibody concentrations to H2ss assessed by ECLIA (A) and neutralizing IC₈₀ antibody titers against H2N2 assessed by reporter-based microneutralization assay (B) stratified by age group; y.o., years old. (C and D) Binding antibody concentrations to H2ss (C) and neutralizing IC₈₀ antibody titers (D) stratified by expected exposure to H2 influenza; participants born in or after 1969 are considered “H2-naïve,” and those born in or before 1968 are considered “H2-exposed.” Geometric means (GMs) and 95% CI are shown. Fold changes over baseline are indicated at weeks 2, 16, and 18 for (C) and (D). Results of two-sample t tests are shown. In (A) and (B), comparisons were made at each time point for the GMs of the 60- to 70-year-old individuals to the GM of the 18- to 59-year-old individuals combined. In (C) and (D), the GMs of the H2-naïve and H2-exposed individuals were compared. Significance noted applies to each time point under the corresponding black lines. Exact numbers of participant samples analyzed at each time point are listed in table S4.