ISCOMATRIX™ adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans

G Model

ARTICLE IN PRESS

JVAC 16587 1–10

Vaccine xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Vaccine journal homepage: www.elsevier.com/locate/vaccine

ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans

1

2

3

Q1

4 5

Ka Yan Chung a , Elizabeth M. Coyle a , Dewal Jani b , Lisa R. King a , Rukmini Bhardwaj a , Louis Fries b , Gale Smith b , Gregory Glenn b , Hana Golding a , Surender Khurana a,∗ a

6

b

7

Division of Viral Products, CBER, FDA, Silver Spring, MD 20993, USA Novavax, Gaithersburg, MD 20878, USA

8

9 31

a r t i c l e

i n f o

a b s t r a c t

10 11 12 13 14 15

Article history: Received 11 March 2015 Received in revised form 29 May 2015 Accepted 5 June 2015 Available online xxx

16

30

Keywords: H7N9 Pandemic Influenza Vaccine Phage display Antibody affinity Neutralization Virus Adjuvant Epitope ISCOMATRIX (ISCO) Hemagglutinin H7N7

32

1. Introduction

17 18 19 20 21 22 23 24 25 26 27 28 29

In a previously reported phase I clinical trial, subjects vaccinated with two doses of an unadjuvanted H7N9 virus like particle (VLP) vaccine responded poorly (15.6% seroconversion rates with 45 ␮g hemagglutinin (HA) dose). In contrast, 80.6% of subjects receiving H7N9 VLP vaccine (5 ␮g HA) with ISCOMATRIXTM adjuvant developed hemagglutination-inhibition (HI) responses. To better understand the role of adjuvant, complete antibody epitope repertoires of post-vaccination sera were investigated using Whole Genome Fragment Phage Display Library (GFPDL). In addition, antibody affinity maturation following vaccination was measured against HA1 and HA2 antigenic domains using real time Surface Plasmon Resonance (SPR) based kinetic assays. Unadjuvanted H7N9-VLP vaccine generated primarily antibodies targeting the C-terminus of the HA1 domain, predicted to be mostly buried on the native HA spikes, while adjuvanted VLP vaccine generated antibodies against large epitopes in the HA1 spanning the receptor binding domain (RBD). SPR analysis using a functional H7-HA1 domain demonstrated that sera from adjuvanted H7N9-VLP vaccine induced higher total binding antibodies and significantly higher antibody affinity maturation to HA1 compared to sera from unadjuvanted vaccine. Total antibody binding and affinity to the HA1 (but not HA2) domain correlated with HI and neutralization titers. This study demonstrates that ISCOMATRIXTM adjuvanted vaccine promotes higher quality antibody immune response against avian influenza in naïve humans. © 2015 Published by Elsevier Ltd.

Q3 33 34 35 36 37 38 39 40 41 42 43 44

An outbreak of H7N9 avian-origin influenza A virus was reported in the Spring of 2013. Most human infections are result of exposure to infected poultry [1–3]. As of March 2015, the World health Organization (WHO) had reported over 650 infections and with more than 200 deaths, mostly in China (http://www.cidrap. umn.edu/infectious-disease-topics/h7n9-avian-influenza). Some H7N9 viruses isolated from infected humans possess amino acid changes known to facilitate infection of mammals, including increased binding to human-type receptors (␣2,6 sialic acid) in addition to the avian-type receptor (␣2,3 sialic acid), and mammalian-adaptive mutations E627K or D701N in the PB2 polymerase leading to transmission between non-avian hosts [4–6]

∗ Corresponding author. Tel.: +1 240 402 9632; fax: +1 301 595 1125. E-mail address: [email protected] (S. Khurana).

[1,2,7]. Therefore, avian-derived A/H7N9 influenza viruses are considered a pandemic threat due to lack of pre-existing immunity in humans. Concerted efforts are under way to generate vaccines against H7N9. Both traditional inactivated influenza vaccines (IIV) and new vaccine platforms were evaluated in clinical trials, with an emphasis on rapid response for vaccine production [8–11]. Previous vaccination with unadjuvanted IIV based on H7N7 (A/Netherlands/219/03) was poorly immunogenic in humans [12,13]. Therefore, use of adjuvant as part of the vaccination approach against novel avian H7N9 strains was considered necessary to achieve meaningful seroconversion after vaccination of naïve population. Influenza VLPs are relatively easy to develop and manufacture, and could save significant amount of time compared to the traditional egg-based inactivated influenza vaccines. There is no need for reassortment with PR8 and selection of high yield vaccine strain (add Refs 11-17, 43). Case in point, in response to initial reports of

http://dx.doi.org/10.1016/j.vaccine.2015.06.047 0264-410X/© 2015 Published by Elsevier Ltd.

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

G Model

ARTICLE IN PRESS

JVAC 16587 1–10

K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

2 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115

H7N9 virus infections, the H7N9 virus-like particle (VLP) was the first H7N9 candidate vaccine in human clinical trials. ISCOMATRIXTM adjuvant contains cage-like structures, composed of phospholipid, saponin, and cholesterol components. ISCOMATRIX was shown to promote trafficking of vaccine antigen into lymph nodes and to induce immune activation by inducing Th1 and Th2 responses, and cross-presentation of antigens to CD8 cells [14] [15] [16]. Recently, it was found that ISCOMATRIX induce IL-18 through both inflammasome-dependent and –independent mechanisms [17]. Several prophylactic and therapeutic vaccines have been formulated with ISCOMATRIXTM adjuvant and evaluated in pre-clinical and clinical trials [18]. When combined with ISCOMATRIXTM adjuvant, the H7N9 VLP vaccine elicited strong immune response in mice and ferrets and protected from lethality after challenge with homologous (A/Anhui/1/2013) and heterologous A/chicken/Jalisco/CPA1/2012 (H7N3) viruses [9,19]. Subsequently, in Phase I clinical trial, individuals vaccinated with two doses of unadjuvanted VLP responded poorly, with only 5.7% and 15.6% HI seroconversion rates (SCR) in the 15 and 45 ␮g HA dose groups, respectively. On the other hand, subjects receiving VLP vaccine (two doses of 5 ␮g HA) combined with ISCOMATRIXTM adjuvant developed H7N9 HI responses with peak SCR of 80.6% and geometric mean titers (GMT) of 64.3. Surprisingly, the adjuvanted VLP vaccine group receiving higher dose (15 ␮g HA) gave lower SCR of 64.7% with GMT of 37.1 [20]. However, several questions remained unanswered: (a) what was the adjuvant impact on antibody responses: an overall increase of B cell proliferation and differentiation into plasma cells or reshaping of the B cell repertoire? (b) was the increase in HI titers driven primarily by increase in total antibody binding to protective epitopes or to increase in their antibody net avidity (affinity maturation) following adjuvanted vaccine? (c) can we find an explanation for the inverse dose-response of the VLP HA that was observed in the clinical trial? In our earlier studies, influenza whole genome-fragmentphage-display-libraries (FLU-GFPDL) of A/Vietnam/1203/2004 (H5N1) were used for mapping of broadly neutralizing human monoclonal antibodies (MAb) and to decipher the complete epitope repertoires in polyclonal sera from individuals who recovered from H5N1 infection [21], and investigate the role of oil-in-water adjuvant (MF59) in augmenting the immune response following vaccinations against H5N1 (A/Vietnam) and the H1N1pdm09 viruses [22,23] [24]. These studies showed that GFPDL technology can identify large conformation epitopes recognized by neutralizing antibodies and can detect >97% of the epitopes recognized by influenza specific antibodies in the polyclonal sera providing an unbiased tool for in-depth understanding of antibody immune response in humans. In the current study, we evaluated pre- and post-vaccination serum samples from the H7N9 VLP clinical trial to better understand the role of ISCOMATRIXTM adjuvant in shaping the humoral immune response, especially on the antibody epitope repertoire

and antibody affinity maturation in human polyclonal sera using H7 GFPDL and surface plasmon resonance (SPR)-based real-time kinetics measurement to different antigenic domains (HA1 and HA2) of H7N9 hemagglutinin.

116 117 118 119

2. Materials and methods

120

2.1. Clinical trial samples used in the current study

121

The design and conduct of the clinical trial, including ethical review and informed consent was obtained from subjects and human experimentation guidelines of the Australian Govt. were followed in the conduct of clinical research, have been reported previously [20]. The trial was approved by the Belberry Human Research Ethics Committee in Adelaide, Australia (ClinicalTrials.gov registration number, NCT01897701). Serum pairs analyzed for this investigation were selected randomly, without regard to HI or NAI responses, from the treatment groups of interest and analyzed in blinded manner. The study at CBER was conducted with de-identified samples under Research Involving Human Subjects (RIHSC) exemption 03-118B; and all assays performed fell within the permissible usages in the original consent. ISCOMATRIXTM adjuvant (CSL Biotherapies Inc., King of Prussia, Pennsylvania, USA; ISCOMATRIX is a registered trademark of ISCOTEC Ab a CSL company; ISCO is a registered trademark of CSL). In the current study, we focused on evaluating post-vaccination sera from key groups with differential immune responses in the clinical study as listed in Table 1: placebo; unadjuvanted H7N9 VLP (15 ␮g of HA); and adjuvanted H7N9 VLP at 15 ␮g or 5 ␮g HA with 60 units of ISCOMATRIXTM adjuvant. 2.2. Measurements of HI titers against H7N7 and H7N9

123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142

143

Serum HI antibodies against A/H7N9 were measured by Focus Diagnostics (Cypress, CA), using rgA/Anhui/1/2013 NIBRG-268 (NIBSC, UK) based on the WHO Manual, using 0.5% turkey erythrocytes with 4 hemagglutination units (HAU) of homologous virus (15). Re-analysis of clinical trial sera with 1% horse erythrocytes yielded essentially identical results. A HI titer of 1:40 is considered seroprotective. 2.3. Microneutralization assay

144 145 146 147 148 149 150

151

Viral-neutralizing activity was analyzed in a microneutralization assay in MDCK cells, based on the methods of the pandemic influenza reference laboratories of the Center for Disease Control and Prevention (CDC). Low pathogenicity H7N9 (A/Shanghai/1/2013) and H7N7 (HA derived from H7N7A/Netherlands/219/03) viruses, generated by reverse genetics, were obtained from CDC and the lab of Maryna Eichelberger, CBER, respectively. The experiments were conducted with three replicates for each serum sample and performed at least twice. All plasma samples were serially diluted starting at a 1:10 dilution

Table 1 Serum neutralization (HI and MN) end-point antibody titers after the second dose of H7N9 VLP vaccinea . MN

122

HI

Group

Vaccine

Number of subjects

HA Antigen Dose (mcg)

IscomatrixTM (ISCOTM ) Units

GMT

SD

Range

GMT

SD

Range

1 2 3 4

Placebo H7N9 H7N9 H7N9

10 15 15 15

0 15 5 15

– – 60 60

10 30.67 197.3 84

0 80.04 654.89 128.05

(10) (10–320) (10–2560) (10-320)

5 15.33 105.13 44.13

0 40.02 101.39 33.63

(5) (5–160) (20–320) (20–127)

a Data are for 2 weeks post-second-vaccination sera collected on day 35 and tested in the microneutralization (MN) or hemagglutination inhibition (HI) assay. GMT; geometric mean titer; SD; standard deviation.

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

152 153 154 155 156 157 158 159 160 161

G Model JVAC 16587 1–10

ARTICLE IN PRESS K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

3

Fig. 1. Analysis of antibody repertoires elicited in adults after vaccination with unadjuvanted and ISCO-adjuvanted H7N9 VLP vaccine. (A) Distribution of phage clones after H7-GFPDL (A/Shanghai/1/2013) affinity selection with sera obtained from adults after two vaccinations with H7N9 VLP vaccine (with and without ISCO adjuvant). (B) Schematic alignment of the peptides recognized by post-second H7N9 vaccination sera in adults, identified by panning with H7-A/Shanghai/1/2013 GFPDL. The amino acid designation is based on the HA protein sequence (Fig. S1). Bars indicate identified inserts in HA1 (red bars) and HA2 (blue bars). The thickness of each bar represents the frequencies of repetitively isolated phage inserts (only clones with a frequency of two or more are shown). The HA1 receptor binding domain (RBD) is depicted as green bar. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

162 163 164 165 166

and incubated with the virus (100 TCID50 ) for 1 hr prior to addition to 96 wells containing MDCK cells. After overnight infection the cells are lysed and the content of influenza virus is detected by anti-NP MAb. A MN titer of 1:80 is considered seroprotective [20].

2.4. Protein expression, refolding and purification of H7N9 HA1 and HA2 domains E. coli BL21 cells (Novagen) were used for expression of HA1 and HA2 derivatives from the H7N9-A/Shanghai/1/2013 virus [26].

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

167 168

169 170

G Model JVAC 16587 1–10 4

ARTICLE IN PRESS K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

Fig. 2. Structural map of immunodominant epitope in the HA1 domain of the H7N9 (A/Shanghai/01/2013) virus hemagglutinin recognized by unadjuvanted compared with ISCO-adjuvanted H7N9 VLP vaccine. The immunodominant minimal C-terminal HA1 epitope amino acid residues in the H7 hemagglutinin recognized by unadjuvanted H7N9 VLP vaccine sera are depicted on the monomer of the trimeric HA spike. Amino acids in the selected epitope are depicted as green balls and the amino acid residues that are predicted to be exposed on the surface of the native trimer are represented as red surface. (A) HA sequence 206–310 (predominant epitope in the unadjuvanted vaccine group) contains less than 30% of amino acids predicted to be surface exposed (B) HA sequence 46–288 (predominant epitope recognized by the ISCO-adjuvanted H7N9 VLP vaccine sera) contains >75% surface exposed residues shown on one H7N9 monomer within the HA trimer structure [Protein Data Bank (PDB) identifier 4LN3]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

173

The H7 HA1 and HA2 proteins were produced and were properly folded that contained a high percentage of functional oligomers as previously described [26].

174

2.5. Affinity measurements by surface plasmon resonance (SPR)

171 172

sera from each cohort were used for each round of GFPDL panning All samples in each group were pooled for GFPDL analysis. GFPDL selection was carried out in solution (with protein A/G) as previously described [21,23]. 2.7. Statistical analyses

175 176 177 178 179 180 181 182 183 184 185

186 187

188 189 190

Steady-state equilibrium binding of post-H7N9 vaccination human sera was monitored at 25 ◦ C using a ProteOn surface plasmon resonance biosensor (BioRad) [23]. The recombinant HA globular domain (rHA1-His6 ) or HA stalk domain (rHA2-His6 ) for A/Shanghai/1/2013 were coupled to a GLC sensor chip. The spatial density of antigen on the chip surface was adjusted to measure only monovalent antibody binding. Antibody off-rate constants, which describe the fraction of antigen-antibody complexes that decay per sec, were determined directly from sample interactions with rHA1 or rHA2 proteins using SPR in the dissociation phase as described before [23].

Differences between groups (p-values) were examined for statistical significance by the multiple comparison adjustment using Bonferroni method. A p-value less than 0.05 was considered significant. Spearman correlations are reported for the calculation of correlations between off-rate and MN titers combined across all vaccine groups. All statistical calculations were performed using ANOVA. 3. Results

2.6. Construction of H7 gene-fragment phage display libraries (GFPDL) and panning of H7 GFPDL with polyclonal human sera

3.1. ISCOMATRIXTM adjuvant promotes diverse antibody repertoire to conformational epitopes in HA1 receptor binding domain following H7N9 VLP vaccination

The complete HA0 gene segment corresponding to the H7 strain was used for the construction of H7-GFPDL as previously described for H5N1 GFPDL[21,22]. Equal volumes of pooled polyclonal human

The clinical trial with the H7N9 VLP vaccine with or without ISCOMATRIXTM adjuvant revealed a very strong adjuvant impact on vaccine immunogenicity [20].

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

191 192 193 194

195

196 197 198 199 200 201 202

203

204 205 206

207 208 209

G Model

ARTICLE IN PRESS

JVAC 16587 1–10

K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

H7 HA1 RU

A

p = 0.0035

Resonance (RU)

p = <0.0001 p = <0.0001

5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 Pre Post Placebo

Pre Post Pre Post 15 ug 5 ug/ISCO

Pre Post 15 ug/ISCO

H7 HA1 RU vs. HI Titer

B 5000

r = 0.7327 p = <0.0001

Resonance (RU)

4500 4000 3500 3000 2500 2000 1500 1000 500 0 1

4

16

64

256

HI Titer H7 HA1 RU vs. MN Titer

C 5000

Resonance (RU)

In the current study, we investigated the quality of the antibody responses in the post-vaccination polyclonal sera from key groups and impact of adjuvant with differential immune responses in the clinical study as listed in Table 1: placebo; unadjuvanted H7N9 VLP (15 ␮g of HA); and adjuvanted H7N9 VLP at 15 ␮g or 5 ␮g HA with 60 units of ISCOMATRIXTM adjuvant (Table 1). The HI and MN titers of the samples (2 weeks after second vaccination) are shown in Table 1 and Supplementary Table 1 against both homologous H7N9 strains and the heterologous H7N7 strain. The MN GMTs were generally two-fold higher than the HI GMTs, similar to other reports [25], but the hierarchy of response rates was the same as previously reported [20]. Pooled samples from each group (pre-vaccination and 2 weeks after the second dose) were used for affinity selection using H7 GFPDL to identify the complete antibody epitope repertoire in the polyclonal sera as previously described [22]. Pooled sera from the placebo group bound to very few phages, confirming the lack of pre-existing immunity against H7 (data not shown). On the other hand, sera from all three post-H7N9 VLP vaccine groups captured similar numbers of phages ranging between 2200–2800 (Fig. 1A). The captured phages were subjected to PCRbased DNA insert sequencing, and the inserts were mapped to the H7 HA domains as shown in Fig. 1B. Sera from the group vaccinated with unadjuvanted H7N9 VLP (15 ␮g of HA) recognized both HA1 and HA2 epitopes (red and blue bars, respectively), but the majority of the HA1 fragments did not span the complete receptor binding domain (RBD) and showed a bias towards the middle and C-terminus of the HA1 domain (Fig. 1B top panel). The predominant HA1-C-terminus epitope (aa 206–310) is represented on the structure of the HA trimer in Fig. 2A, indicating that most of this sequence (>70%) is buried within the native HA trimer. In contrast, subjects in both of the adjuvanted VLP vaccine groups demonstrated “epitope spreading” with a shift to binding of large fragments spanning the entire RBD (Fig. 1A and B center and bottom panels), predicted to represent conformational epitopes known to be targeted by influenza virus neutralizing antibodies [21,22]. Indeed, as shown in Fig. 2B, the predominant HA1 epitope (aa 46–288) recognized by antibodies from post-vaccination sera of adjuvanted VLP vaccine recipients includes the entire RBD and was mostly surface exposed on the head of the HA trimer. The GFPDL analysis did not identify significant differences between the epitope profiles of the vaccine groups that received 5 or 15 ␮g of HA with ISCOMATRIXTM adjuvant (Fig. 1A and B). 3.2. Post-vaccination sera from ISCOMATRIXTM adjuvanted H7N9 VLP vaccine recipients show a significant increase in total HA1 binding antibodies

r = 0.7283 p = <0.0001

4500

5

4000 3500

Individual antibody responses to H7N9 HA was performed using SPR technology to measure total antibody binding against H7N9 recombinant HA1 and HA2 antigenic domains separately as previously described for other pandemic viruses [23,27,28].

3000 2500 2000 1500 1000 500 0 1

10

100

1000

10000

MN Titer Fig. 3. Polyclonal antibody binding to properly folded HA1 proteins from H7A/Shanghai/1/2013 and correlation between total binding and in vitro HI and neutralizing titers.(A) Steady-state equilibrium analysis of the total binding antibodies in the polyclonal human vaccine sera to properly folded functional H7-A/Shanghai/1/2013 HA1-His6 (panel A) was measured by SPR. Dots represent either pre- or post-H7N9 boost serum samples were diluted ten-fold, and injected simultaneously onto HA1 immobilized on a sensor chip through the free amine

group and onto a blank flow cell, free of peptide. Maximum resonance unit (Max RU) values for HA1 binding by serum antibodies obtained from all individuals in the placebo (black circles), unadjuvanted VLP vaccine (15 ␮g HA; blue circles), ISCOadjuvanted low dose VLP (5 ␮g HA; red circles), and ISCO-adjuvanted high dose VLP (15 ␮g HA; green circles) on Day 0 (Pre) and Day 35 (14 days post-second vaccination). Differences between groups (p-values) were examined for statistical significance by the multiple comparison adjustment using Bonferroni method. A p-value less than 0.05 was considered to be significant. (B and C) Total HA1 binding antibody (Max RU) of post-H7N9 VLP vaccinated human sera against HA1 of H7-A/Shanghai/1/2013 was correlated with the homologous virus hemagglutination inhibition (HI; r = 0.7327) (B) or microneutralization titers (MN; r = 0.7283) (C) against the H7-A/Shanghai/1/2013. Spearman correlations are reported for the calculation of correlations between total anti-HA1 antibody binding and HI (B) or MN (C) titers combined across all vaccine groups. The color scheme in panels B and C is the same as in panel A. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251

252 253 254

255 256 257 258

G Model

ARTICLE IN PRESS

JVAC 16587 1–10

K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

6

A

B

H7 HA1 kd

H7 HA2 kd

Placebo 15 ug 5 ug/ISCO 15 ug/ISCO Pre Post Pre Post Pre Post Pre Post

Placebo 15 ug 5 ug/ISCO 15 ug/ISCO Pre Post Pre Post Pre Post Pre Post

0.1

Off rate (1/sec)

Off rate (1/sec)

0.1

0.01

0.01

0.001

0.0001

0.001 p = <0.0001 p = 0.0532

p = 0.6616 p = 0.5397 p = 0.2573

p = <0.0001 Fig. 4. ISCO-adjuvanted H7N9 VLP vaccine enhances antibody affinity (slower off-rates) to H7-HA1 (but not HA2). (A and B) SPR analysis of pre- and post- second vaccination with H7N9 VLP vaccine was performed with properly folded HA1 (A) or HA2 (B) from H7N9-A/Shanghai/1/2013 virus. Off-rates of polyclonal serum antibodies from all individuals before (open symbols; Pre) or 35 days (filled symbols; Post) 14 after a second vaccine dose with placebo (black circles), unadjuvanted (15 ␮g HA; blue circles), ISCO-adjuvanted low dose VLP (5 ␮g HA; red circles), or ISCO-adjuvanted high dose VLP (15 ␮g HA; green circles) vaccine. Antibody off-rate constants that describe the fraction of antibody-antigen complexes decaying per second were determined directly from the serum sample interaction with rHA1 (1–320) protein or rHA2 (331–480) using SPR in the dissociation phase. Serum antibody off-rate constants were determined as described in Materials and Methods. Differences between groups (p-values) were examined for statistical significance by the multiple comparison adjustment using Bonferroni method. p-Value of less than 0.05 were considered significant. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282

283 284 285

286 287 288 289 290 291 292 293 294

No significant antibody binding to HA1 domain was found with individual sera from the placebo group, confirming the lack of preexisting immunity against H7 in the human population (Fig. 3A black circles). Post-vaccination serum samples from recipients of unadjuvanted vaccine showed very minimal increase in antibody binding to H7 HA1 (Fig. 3A blue circles), congruent with the observation that the predominant C-terminal epitope sequences selected in GFPDL by these sera are not exposed on the surface of HA1 trimer. In contrast, the groups that received H7N9 VLP combined with ISCOMATRIXTM adjuvant demonstrated significant increases in HA1 antibody binding in post- vs. pre-vaccination sera. Furthermore, the mean HA1 antibody binding for the group that received the lower antigen dose (5 ␮g HA) was significantly higher than for the group that received the higher dosage (15 ␮g HA) of adjuvanted H7 VLP vaccine (Fig. 3A red vs green circles, respectively p = 0.0035). Comparison of individual HI or MN titers and the total antibody binding to HA1 (Max RU) for individual post-vaccination sera from all the groups, demonstrated significant positive correlations (Fig. 3B and C). These data demonstrate that enhanced serum antibody binding to H7 HA1 domain was strong predictor of the observed adjuvant effect on VLP vaccine immunogenicity by HI or MN, and also showed the inverse HA dose-response observed during the clinical trial.

3.3. ISCOMATRIXTM adjuvant promotes affinity maturation of antibodies against H7 HA1, but not HA2, in H7N9 VLP vaccine recipients Previously, oil-in-water adjuvants as well as DNA priming prior to subunit vaccination against H5N1 avian strains were shown to significantly increase the affinity of antibodies against the HA1 globular domain [23,28]. Therefore, the dissociation rates of polyclonal serum antibodies from individuals vaccinated with unadjuvanted or adjuvanted VLP vaccine were measured using real time kinetics in the SPR assay with both the HA1 or HA2 domains from H7 virus (Fig. 4A and B). No difference in serum antibody dissociation kinetics (off-rates) between pre- and post-vaccination

samples was found for either the placebo group or the unadjuvanted vaccine group (Fig. 4A black and blue circles) indicating lack of antibody affinity maturation upon unadjvuanted VLP vaccination. In contrast, significantly slower antibody off-rates (i.e., higher antibody affinity), were observed for HA1-specific antibodies in the sera of the two adjuvanted H7-VLP vaccinated groups compared with the unadjuvanted vaccine group (p < 0.0001) (Fig. 4A red and green circles). The lower antigen dose group (5 ␮g HA) showed tighter distribution of off-rates compared with the higher (15 ␮g HA) vaccine dose (Fig. 4A red vs green circles). In contrast to the findings with the H7-HA1 domain, no statistically significant differences were observed in antibody off-rates between pre- and post-vaccination sera binding to H7-HA2 domain, and there were no clear differences among the treatment groups (Fig. 4B).

3.4. Serum antibody off-rates to H7-HA1 (but not HA2) strongly correlate with H7N9 HI and MN titers as well as H7N7 MN titers The post-vaccination HA1 and HA2 antibody off-rates were correlated with both the HI titers (Fig. 5A, B) and the microneutralization (MN) titers (Fig. 5C, D) of the individual post-vaccination serum samples. Antibody affinity (off-rates) to H7-HA1 domain demonstrated a statistically significant inverse correlation coefficient (r = −0.6564; p < 0.0001) with the homologous H7N9-HI titers; but not with the H7-HA2 domain antibody off-rates (Fig. 5B). Similarly, a strong negative correlation was observed between H7N9 MN titers and the antibody off-rates against the H7-HA1 domain (Fig. 5C) (r = − 0.6255; p < 0.0001), but not the H7-HA2 domain (Fig. 5D). We also measured cross-neutralization against the heterologous H7 from H7N7 strain (A/Netherlands/219/03). None of the post-vaccination serum samples from unadjuvanted vaccine recipients exhibit cross-neutralization. In contrast, in the ISCOMATRIXTM adjuvanted vaccine groups 9/15 and 2/15 in the 5 ␮g HA/dose and 15 ␮g HA/dose groups, respectively, displayed MN titers of >40 against the heterologous H7N7 strain (Suppl. Table 1). As shown in Fig. 6, a strong negative correlation was observed between the

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

295 296 297 298 299 300 301 302 303 304 305 306 307 308 309

310 311

312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330

G Model

ARTICLE IN PRESS

JVAC 16587 1–10

K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

A

H7 HA1 kd vs. HI Titer

B

7

H7 HA2 kd vs. HI Titer

HI Titer 1

8

HI Titer

64

512

1

1.00E-02

8

64

512

1.00E-02

Off-Rate (1/sec)

Off-Rate (1/sec)

r = -0.6564 p = <0.0001

1.00E-03

C

r = -0.2399 p = 0.0778

1.00E-03

1.00E-04

H7 HA1 kd vs. MN Titer

D

H7 HA2 kd vs. MN Titer

MN Titer 1

32

MN Titer 1024

1

1.00E-02

32

1024

1.00E-02

Off-rate (1/sec)

Off-rate (1/sec)

r = -0.6255 p = <0.0001

1.00E-03

1.00E-03

r = -0.2857 p = 0.0345

1.00E-04

Fig. 5. Serum antibody off-rates to rHA1 (but not rHA2) from H7-A/Shanghai/1/2013 following vaccination with H7-VLP (A/Shanghai/1/2013) correlate with the in-vitro neutralizing capacity against the homologous H7N9 vaccine virus. Antibody off-rate constants were determined directly from the serum sample interaction with H7 rHA1 or rHA2 proteins using SPR in the dissociation phase (Fig. 4). SPR analysis of post-boost vaccination human sera from all four vaccine groups combined was performed with rHA1 (A and C) or rHA2 (B and D) of the H7N9-A/Shanghai/1/2013 virus. Each symbol represents one individual. Serum samples on day 35 (14 days following second vaccination) with the H7N9 VLP vaccine from the placebo (black circles), unadjuvanted (15 ␮g HA; blue circles), ISCO- adjuvanted low dose VLP (5 ␮g HA; red circles), or ISCO-adjuvanted high dose VLP (15 ␮g HA; green circles) vaccination are shown. Antibody affinity of post-H7N9 VLP vaccinated human sera against HA1 (but not HA2) showed strong inverse correlation with the homologous hemagglutination inhibition (HI in A; r = −0.6564)) and virus microneutralization (MN in C; r = −0.6255) titers against the A/Shanghai/1/2013 (H7N9). Spearman correlations are reported for the calculation of correlations between off-rate and HI (A and B) or MN (C and D) titers combined across all vaccine groups. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

serum antibody dissociation off-rates against H7 HA1 domain and heterologous H7N7 MN titers (Fig. 6A; r = − 06514; p < 0.0001), but not with H7 HA2 domain (Fig. 6B) even though the HA2 domain is 333 more conserved than HA1 domain between H7N9 and H7N7 strains 334 Q4 (Suppl. The data suggests that cross-neutralization may be linked to 335 both total HA1-binding antibody titer and antibody affinity, which 336 is influenced by the adjuvanted vaccine formulation. 337 Together, these data provided in-depth understanding of the 338 impact of ISCOMATRIXTM adjuvant on the immunogenicity of the 339 H7N9 VLP vaccine. Increased total antibody binding to the H7-HA1 340 domain as well as antibody affinity maturation following adju341 vanted VLP vaccine contributed to the observed higher HI and MN 342 titers. The inverse HA vaccine dose-response was better correlated 343 with the total anti-HA1 antibody binding titers (Fig. 3) and to a 344 lesser degree with antibody affinity in the polyclonal sera. 345 331 332

346

347 348 349 350

4. Discussion Rapid vaccine production and release is an essential component of global response to a pandemic influenza threat. The process of constructing a new vaccine based on newly circulating viruses is quite lengthy, and suffers from the possibility of bottlenecks that

may pose a clear impediment to initiation of rapid mass vaccination against spreading pandemic influenza, as was evident during the 2009 H1N1 influenza pandemic [29]. Recombinant vaccine platforms offer the promise of rapid response to evolving pandemic threats. One such technology involves production of VLPs containing the HA, NA, and M1 from avian influenza virus. VLPs from multiple AIV viruses including H9N2, H5N1, H7N3, and H7N9 have been successfully produced in Sf9 cells within short time frames and were evaluated in pre-clinical and clinical trials [9,19,24,30–32]. VLP-based vaccine against the newly isolated H7N9 virus A/Anhui/1/2013 was the first H7N9 vaccine in clinical trials, and was administered alone or in combination with the saponin-based ISCOMATRIXTM adjuvant [20]. The adjuvant was required to achieve acceptable seroconversion rates, and provided dose sparing activity [20]. The current study investigated the quality of antibody immune response in depth and revealed a major shift in the antibody epitope repertoires after vaccination with adjuvanted as opposed to unadjuvanted H7N9 VLP vaccine. The unadjuvanted vaccine generated primarily antibodies targeting the HA2 domain and C-terminus of the HA1 domain, which seems to be mostly buried on the virion spikes and not overlapping with the main protective sites around the receptor binding domain (RBD). This finding was confirmed

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373

G Model

ARTICLE IN PRESS

JVAC 16587 1–10

K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

8

A

H7 HA1 kdvs. MN H7N7

B

H7 HA2 kd vs. MN H7N7

Titer

Titer

H7N3 MN Titer 1

8

H7N3 MN Titer

64

1

1.00E-02

4

16

64

256

1.00E-01

r = -0.6514

r = -0.00225 p = 0.9883

1.00E-03

Off-Rate (1/sec)

Off-Rate (1/sec)

p <0.0001 1.00E-02

1.00E-03

1.00E-04

Fig. 6. Serum antibody off-rates to rHA1 (but not rHA2) from H7-A/Shanghai/1/2013 following vaccination with H7-VLP (A/Shanghai/1/2013) correlate with the in-vitro neutralizing capacity against the heterologous H7-A/Netherlands/219/03 virus. Antibody off-rate constants were determined directly from the serum sample interaction with H7 rHA1 or rHA2 proteins using SPR in the dissociation phase. SPR analysis of post- vaccination human sera from all four vaccine groups combined was performed with rHA1 (A) or rHA2 (B) of the H7N9-A/Shanghai/1/2013 virus. Serum samples of day 35 (14 days post second vaccination with the H7N9 VLP vaccine) from the placebo (black circles), unadjuvanted (15 ␮g HA; blue circles), ISCO-adjuvanted low dose VLP (5 ␮g HA; red circles), or ISCO-adjuvanted high dose VLP (15 ␮g HA; green circles) vaccination are shown. Antibody affinity of post-H7N9 VLP vaccinated human sera against H7-HA1 showed strong inverse correlation with the heterologous virus microneutralization titers (panel A; r = −0.6514) titers against the H7N7-A/Netherlands/219/03. No correlation between antibody binding off-rates to HA2 and the MN titers was observed (B). Spearman correlations are reported for the calculation of correlations between off-rate and MN titers combined across all vaccine groups. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409

when unadjuvanted post-vaccination sera demonstrated comparatively poor binding to properly folded oligomeric H7 HA1 protein (mimicking native spikes) [26], as measured by SPR. This predominant antibody response to HA2 domain and C-terminus of the HA1 domain in the unadjuvanted vaccinees is primarily due to recall of cross reactive antibodies in these adults due to high sequence conservation with the seasonal H3 influenza strains due to prior exposure or vaccination with seasonal influenza (Supplementary Fig. 1). In contrast, the adjuvanted H7N9 VLP vaccine elicited strong HA1-RBD-targeting antibodies recognizing large conformational epitopes, with high binding to H7-HA1 by SPR. In addition to markedly increased total antibody binding, the adjuvanted H7N9 VLP vaccine elicited a significant increase in antibody affinity for the H7-HA1 (but not HA2) domain as judged by slower antigen-antibody dissociation rates. Importantly, a good correlation was observed between both antibody binding titers (Max RU) and antibody affinity with the HI and neutralization (MN) titers for individual sera against the homologous H7N9 vaccine strain and the heterologous H7-A/Netherlands/219/03 strain. GFPDL analyses revealed an “epitope spreading” in the antibody-repertoires of sera from the ISCOMATRIX-adjuvanted groups. The shift from targeting epitopes primarily in the HA2 stalk and the C-terminus of HA1, seen with unadjuvanted H7N9 immune sera, to large sequences representing conformational epitopes spanning the entire RBD in HA1, could be explained by several mechanisms. The adjuvant may have focused the HA on the VLP immunogen to increase exposure of the HA1-RBD. However, physicochemical analyses on ISCOMATRIX-adjuvanted vaccines did not reveal any direct physical interactions (personal communications). Alternatively, ISCOMATRIX, and other adjuvants, including oil-in-water MF59, could significantly increase the frequency of HA-specific T helper cells that can differentiate into Tfh and drive a sustained germinal center formation (GC). The more robust Th/Tfh responses are likely to promote expansion of naïve B cells specific to the HA1-RBD in the novel H7 HA. In contrast, the unadjuvanted vaccines are less effective in

triggering Th/Tfh responses and primarily recall pre-existing memory B cells (from prior exposure to seasonal H3N2 viruses) that cross-react with either the HA2 stem or the more conserved HA1C-terminus of H7 HA. These recall responses apparently did not target surface-exposed protective targets in RBD, resulting in poor HI and virus-neutralization titers. In the current study, most of the HA1 and HA2 binding antibodies were of IgG isotype in SPR assays and the isotype subsets were similar between unadjuvanted and adjuvanted group. It is of interest, that exactly the same shift in antibody epitopes was recently found after vaccination with inactivated H7N7 vaccine (Sanofi Pasteur) in individuals that were previously “primed” with the cold-adapted H7N7 live attenuated influenza vaccine (LAIV). The initial LAIV vaccination did not result in significant HI titers, but probably primed immune system that could drive a vigorous naïve B cell response to the inactivated vaccine boost, resulting in a epitope-spread from HA2/HA1-C terminus to the HA1-RBD conformational epitopes [33]. With regard to the “inverse” dose-response observed in the clinical trial, the total serum antibody binding to the H7-HA1 domain was significantly higher for the 5 ␮g H7-VLP-HA group compared with the 15 ␮g H7-VLP HA group with less in-group variability. ISCOMATRIX adjuvant was shown to induce both CD4 and CD8 T cell activation and improve cross-presentation by dendritic cells. It is possible that higher vaccine dose favor antigen processing by dendritic cells in the T cell areas of the lymph nodes, leading to stronger CD8 responses than CD4-B cell interactions in the B-cell follicles, required for expansion and affinity maturation of naïve B cells. Interestingly, evidence of inverse dose response was recently reported in clinical studies of MF59-adjuvanted H5N1 and H7N9 IIV vaccines, but no mechanistic explanation was provided [11,34,35]. [16–18,36–38] The requirements for generation of cross-reactive antibodies are not fully understood. However, in our previous study using recombinant H7 HA1 vaccine in ferret challenge model, we found a strong correlation between serum antibody affinity and

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445

G Model JVAC 16587 1–10

ARTICLE IN PRESS K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

455

control of nasal viral loads after both homologous and heterologous H7 virus challenge [26]. A similar trend was found for the H7N9 VLP vaccine in animal model [9,19] and to some degree in the human clinical trial [9,33]. Therefore, the combination of rapid vaccine production using the recombinant VLP technology with appropriate adjuvant could assist in early vaccination campaign against impending influenza pandemic. This vaccination approach may also overcome the minor drift in the viral HA antigen by generating high affinity crossreactive neutralizing antibodies.

456

Funding

446 447 448 449 450 451 452 453 454

All the data in the current study were generated at CBER. This Q5 work was supported by internal FDA/CBER PANFLU funds. The 458 Phase I clinical trial [14] was funded by Novavax. 459 457

460

Conflict of interest statement

469

We have the following interests: Co-authors Dewal Jani, Louis Fries, Gale Smith and Gregory Glenn are employed by Novavax and also own Novavax stock. Novavax provided support in the form of salaries for authors DJ, LF, GS and GG, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. This does not alter our adherence to Journal policies on sharing data and materials. KYC, EMC, RB, LRK, HG and SK are employees of US government and do not have any conflict of interest.

470

Acknowledgements

461 462 463 464 465 466 467 468

473

We thank Dr. Keith Peden and Dr. Carol Weiss for a thorough review of the manuscript and Dr. Maryna Eichelberger for providing the LP rgH7 containing the HA of H7N7 strain.

474

Appendix A. Supplementary data

471 472

475 476 477

478

479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vaccine.2015.06. 047 References [1] Zhang Q, Shi J, Deng G, Guo J, Zeng X, He X, et al. H7N9 influenza viruses are transmissible in ferrets by respiratory droplet. Science 2013;341:410–4. [2] Zhu H, Wang D, Kelvin DJ, Li L, Zheng Z, Yoon SW, et al. Infectivity, transmission, and pathology of human-isolated H7N9 influenza virus in ferrets and pigs. Science 2013;341:183–6. [3] Gao R, Cao B, Hu Y, Feng Z, Wang D, Hu W, et al. Human infection with a novel avian-origin influenza A (H7N9) virus. N Engl J Med 2013;368:1888–97. [4] Shi Y, Zhang W, Wang F, Qi J, Wu Y, Song H, et al. Structures and receptor binding of hemagglutinins from human-infecting H7N9 influenza viruses. Science 2013;342:243–7. [5] Xiong X, Martin SR, Haire LF, Wharton SA, Daniels RS, Bennett MS, et al. Receptor binding by an H7N9 influenza virus from humans. Nature 2013;499:496–9. [6] Watanabe T, Kiso M, Fukuyama S, Nakajima N, Imai M, Yamada S, et al. Characterization of H7N9 influenza A viruses isolated from humans. Nature 2013;501:551–5. [7] Liu Q, Zhou B, Ma W, Bawa B, Ma J, Wang W, et al. Analysis of recombinant H7N9 wild-type and mutant viruses in pigs shows that the Q226L mutation in HA is important for transmission. J Virol 2014;88:8153–65. [8] Dormitzer PR, Suphaphiphat P, Gibson DG, Wentworth DE, Stockwell TB, Algire MA, et al. Synthetic generation of influenza vaccine viruses for rapid response to pandemics. Sci Transl Med 2013;5:185ra68. [9] Smith GE, Flyer DC, Raghunandan R, Liu Y, Wei Z, Wu Y, et al. Development of influenza H7N9 virus like particle (VLP) vaccine: homologous A/Anhui/1/2013 (H7N9) protection and heterologous A/chicken/Jalisco/CPA1/2012 (H7N3) cross-protection in vaccinated mice challenged with H7N9 virus. Vaccine 2013;31:4305–13. [10] Bart SA, Hohenboken M, Della Cioppa G, Narasimhan V, Dormitzer PR, KanesaThasan N. A cell culture-derived MF59-adjuvanted pandemic A/H7N9 vaccine is immunogenic in adults. Sci Transl Med 2014;6:234ra55.

9

[11] Mulligan MJ, Bernstein DI, Winokur P, Rupp R, Anderson E, Rouphael N, et al. Serological responses to an avian influenza A/H7N9 vaccine mixed at the point-of-use with MF59 adjuvant: a randomized clinical trial. JAMA 2014;312:1409–19. [12] Couch RB, Patel SM, Wade-Bowers CL, Nino D. A randomized clinical trial of an inactivated avian influenza A (H7N7) vaccine. PLoS ONE 2012;7:e49704. [13] Treanor JJ, Campbell JD, Zangwill KM, Rowe T, Wolff M. Safety and immunogenicity of an inactivated subvirion influenza A (H5N1) vaccine. N Engl J Med 2006;354:1343–51. [14] Maraskovsky E, Schnurr M, Wilson NS, Robson NC, Boyle J, Drane D. Development of prophylactic and therapeutic vaccines using the ISCOMATRIX adjuvant. Immunol Cell Biol 2009;87:371–6. [15] Schnurr M, Orban M, Robson NC, Shin A, Braley H, Airey D, et al. ISCOMATRIX adjuvant induces efficient cross-presentation of tumor antigen by dendritic cells via rapid cytosolic antigen delivery and processing via tripeptidyl peptidase II. J Immunol 2009;182:1253–9. [16] Wilson NS, Yang B, Morelli AB, Koernig S, Yang A, Loeser S, et al. ISCOMATRIX vaccines mediate CD8 + T-cell cross-priming by a MyD88-dependent signaling pathway. Immunol Cell Biol 2012;90:540–52. [17] Wilson NS, Duewell P, Yang B, Li Y, Marsters S, Koernig S, et al. Inflammasomedependent and -independent IL-18 production mediates immunity to the ISCOMATRIX adjuvant. J Immunol 2014;192:3259–68. [18] Morelli AB, Becher D, Koernig S, Silva A, Drane D, Maraskovsky E. ISCOMATRIX: a novel adjuvant for use in prophylactic and therapeutic vaccines against infectious diseases. J Med Microbiol 2012;61:935–43. [19] Liu YV, Massare MJ, Pearce MB, Sun X, Belser JA, Maines TR, et al. Recombinant virus-like particles elicit protective immunity against avian influenza A(H7N9) virus infection in ferrets. Vaccine 2015;33:2152–8. [20] Fries LF, Smith GE, Glenn GM. A recombinant viruslike particle influenza A (H7N9) vaccine. N Engl J Med 2013;369:2564–6. [21] Khurana S, Suguitan Jr AL, Rivera Y, Simmons CP, Lanzavecchia A, Sallusto F, et al. Antigenic fingerprinting of H5N1 avian influenza using convalescent sera and monoclonal antibodies reveals potential vaccine and diagnostic targets. PLoS Med 2009;6:e1000049. [22] Khurana S, Chearwae W, Castellino F, Manischewitz J, King LR, Honorkiewicz A, et al. Vaccines with MF59 adjuvant expand the antibody repertoire to target protective sites of pandemic avian H5N1 influenza virus. Sci Transl Med 2010;2:15ra5–25ra. [23] Khurana S, Verma N, Yewdell JW, Hilbert AK, Castellino F, Lattanzi M, et al. MF59 adjuvant enhances diversity and affinity of antibody-mediated immune response to pandemic influenza vaccines. Sci Transl Med 2011;3: 85ra48. [24] Khurana S, Wu J, Verma N, Verma S, Raghunandan R, Manischewitz J, et al. H5N1 virus-like particle vaccine elicits cross-reactive neutralizing antibodies that preferentially bind to the oligomeric form of influenza virus hemagglutinin in humans. J Virol 2011;85:10945–54. [25] Rowe T, Abernathy RA, Hu-Primmer J, Thompson WW, Lu X, Lim W, et al. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol 1999;37:937–43. [26] Khurana S, Coyle EM, Verma S, King LR, Manischewitz J, Crevar CJ, et al. H5 N-terminal beta sheet promotes oligomerization of H7-HA1 that induces better antibody affinity maturation and enhanced protection against H7N7 and H7N9 viruses compared to inactivated influenza vaccine. Vaccine 2014;32: 6421–32. [27] Khurana S, Verma N, Talaat KR, Karron RA, Golding H. Immune response following H1N1pdm09 vaccination: differences in antibody repertoire and avidity in young adults and elderly populations stratified by age and gender. J Infect Dis 2011;205:610–20. [28] Khurana S, Wu J, Dimitrova M, King LR, Manischewitz J, Graham BS, et al. DNA priming prior to inactivated influenza A(H5N1) vaccination expands the antibody epitope repertoire and increases affinity maturation in a boostinterval-dependent manner in adults. J Infect Dis 2013;208:413–7. [29] Osterholm MT, Ballering KS, Kelley NS. Major challenges in providing an effective and timely pandemic vaccine for influenza A(H7N9). JAMA 2013;309: 2557–8. [30] Pushko P, Tumpey TM, Bu F, Knell J, Robinson R, Smith G. Influenza viruslike particles comprised of the HA, NA, and M1 proteins of H9N2 influenza virus induce protective immune responses in BALB/c mice. Vaccine 2005;23: 5751–9. [31] Pushko P, Tumpey TM, Van Hoeven N, Belser JA, Robinson R, Nathan M, et al. Evaluation of influenza virus-like particles and Novasome adjuvant as candidate vaccine for avian influenza. Vaccine 2007;25:4283–90. [32] Bright RA, Carter DM, Crevar CJ, Toapanta FR, Steckbeck JD, Cole KS, et al. Crossclade protective immune responses to influenza viruses with H5N1 HA and NA elicited by an influenza virus-like particle. PLoS ONE 2008;3:e1501. [33] Halliley JL, Khurana S, Krammer F, Fitzgerald T, Coyle EM, Chung KY, et al. Highaffinity H7 head and stalk domain-specific antibody responses to an inactivated influenza H7N7 vaccine after priming with live attenuated influenza vaccine. J Q6 Infect Dis 2015. [34] Nicholson KG, Colegate AE, Podda A, Stephenson I, Wood J, Ypma E, et al. Safety and antigenicity of non-adjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a randomised trial of two potential vaccines against H5N1 influenza. Lancet 2001;357:1937–43. [35] Belshe RB, Frey SE, Graham IL, Anderson EL, Jackson LA, Spearman P, et al. Immunogenicity of avian influenza A/Anhui/01/2005(H5N1) vaccine with MF59 adjuvant: a randomized clinical trial. JAMA 2014;312:1420–8.

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593

G Model JVAC 16587 1–10 10 594 595 596 597 598

ARTICLE IN PRESS K.Y. Chung et al. / Vaccine xxx (2015) xxx–xxx

[36] Davis ID, Chen W, Jackson H, Parente P, Shackleton M, Hopkins W, et al. Recombinant NY-ESO-1 protein with ISCOMATRIX adjuvant induces broad integrated antibody and CD4(+) and CD8(+) T cell responses in humans. Proc Natl Acad Sci USA 2004;101:10697–702. [37] Pearse MJ, Drane D. ISCOMATRIX adjuvant for antigen delivery. Adv Drug Delivery Rev 2005;57:465–74.

[38] Lin Y, Kwon T, Polo J, Zhu YF, Coates S, Crawford K, et al. Induction of broad CD4+ and CD8+ T-cell responses and cross-neutralizing antibodies against hepatitis C virus by vaccination with Th1-adjuvanted polypeptides followed by defective alphaviral particles expressing envelope glycoproteins gpE1 and gpE2 and nonstructural proteins 3, 4, and 5. J Virol 2008;82: 7492–503.

Please cite this article in press as: Chung KY, et al. ISCOMATRIXTM adjuvant promotes epitope spreading and antibody affinity maturation of influenza A H7N9 virus like particle vaccine that correlate with virus neutralization in humans. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.047

599 600 601 602 603 604