Immunogenicity of Oxford-AstraZeneca COVID-19 vaccine in Vietnamese healthcare workers

We studied the immunogenicity of Oxford-AstraZeneca COVID-19 vaccine in 554 Vietnamese healthcare workers who were naive to SARS-CoV-2 infection. Neutralizing antibodies increased after each dose. The sero-conversion rate reached 98.1% after dose 2. Btu at month 3, neutralizing antibodies decreased. The requirement for a third dose warrants further research.

month 1 and 3 after dose 1 from each participant alongside demographics data. We 23 measured neutralizing antibodies using a surrogate virus neutralization assay. The 554 24 study participants (136 males and 418 females) were aged between 22-71 years (median: 25 36 years). 104 and 94 out of 144 selected participants were successfully followed up at 14 26 days after dose 2 and 3 months after dose 1, respectively. Neutralizing antibodies increased 27 after each dose, with the sero-conversion rate reaching 98.1% (102/104) at 14 days after 28 dose 2. At month 3 after dose 1, neutralizing antibody levels decreased, while 94.7% 29 (89/94) of the study participants remained seropositive. Oxford-AstraZeneca COVID-19 30 vaccine is immunogenic in Vietnamese healthcare workers. The requirement for a third 31 dose warrants further research. 32 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021 Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the cause of the 34 ongoing COVID-19 pandemic [1]. Since its first detection in Wuhan, China in late 2019, 35 SARS-CoV-2 has now become an endemic virus globally. Vaccine is thus the most 36 plausible approach to return to the pre-pandemic life. As such, vaccine development has 37 ramped up globally over the last year. As of 1 st June 2021, a total of 185 and 102 vaccine 38 candidates are under the pre-clinical and clinical development phases, respectively [2]. 39 Additionally, seven vaccines have received the WHO approval for emergency use [2]. Although a vaccine must fulfill the required efficacy criteria in order to receive an approval 46 for use in humans, the rapid development and deployment of COVID-19 vaccines 47 worldwide necessitate follow up studies to better understand the development and 48 persistence of vaccine-induced immunity in different populations. Such knowledge is 49 critical to inform the global vaccination strategies and the development of next-generation 50

vaccines. 51
Despite the current surge, which has been escalating since the second week of May 2021, 52 Vietnamese people remained relatively naïve to SARS-CoV-2 infections [4,5]. As of 18 th 53 July 2021, a total of 31,391 PCR confirmed cases have been reported in Vietnam, a 54 country of over 97 million people [3]. Therefore Vietnam is an ideal setting for vaccine 55 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 28, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021 participants for subsequent follow up. The present report focused on the period from 78 baseline to month 3 after the first dose. 79

Neutralizing antibody measurement 80
Neutralizing antibodies were measured using an FDA EUA approved assay, namely 81 SARS-CoV-2 Surrogate Virus Neutralization Test (sVNT) (GenScript, USA). Prior to 82 testing, plasma samples were first diluted 1:10 and then inactivated at 56C for 30 minutes. 83 The experiments were carried out according the manufacturer's instruction. The obtained 84 results were expressed as percentage of inhibition with the 30% cut-off applied. The 85 percentage of inhibition measured by sVNT has been shown to well correlate with the 86 neutralizing antibody tiers measured by the conventional plaque reduction neutralization 87 assay [7]. 88

Neutralizing antibody data from cases of natural infection 89
To compare the development of neutralizing antibodies induced by vaccination against that 90 of natural infection, we included data from 11 Vietnamese patients who had mild or 91 asymptomatic infections. Details about these individuals and neutralizing antibody 92 measurement were detailed in our recent report [8]. 93

Statistical analysis 94
We used Fisher exact, χ 2 or Mann-Whitney U test to compare between groups (when 95 appropriate). Logistic regression was used to assess association between the probability of 96 having detectable neutralizing antibodies and age. Linear regression was used to assess the 97 association between neutralizing antibodies levels and age. The analyses were carried 98 using Prism 9.0.2 (graphpad.com). 99 Ethics 100 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. Of the 144 participants of the subgroup, 104 (72.2%) and 94 (65.3%) were successfully 111 followed-up up to 14 days after the second dose and 3 months after the first dose, 112 respectively. The age and gender distributions of these subgroups were comparable with 113 that of the whole group (Table 1). The window time between the first and the second dose 114 was six weeks. 115

Development of detectable neutralizing antibodies 116
Because HTD members of staff were naïve to SARS-CoV-2 infection [4], we first focused 117 our neutralizing antibody measurement on the baseline samples collected before the first 118 dose of the subgroup. Indeed, at baseline, none of the 104 study participants had detectable 119 neutralizing antibodies (Table 2). At day 14 and 28 after the first dose, the proportions of 120 the study participants with detectable neutralizing antibodies increased from 27.3% 121 (151/554) to 78.0% (432/554), respectively among all 554 individuals of the whole group. 122 The proportion of the study participants with detectable neutralizing antibodies reached 123 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Kinetics of neutralizing antibody levels 126
After the first dose, neutralizing antibody levels measured at day 28 were significantly 127 higher than that measured at day 14 ( Figure 1A), but comparable with that measured at 128 week 6 ( Figure 1B). At day 14 after the second dose, neutralizing antibodies significantly 129 increased, and were comparable with that obtained from Vietnamese people with 130 asymptomatic or mild infection ( Figure 1B). At month 3 after the first does neutralizing 131 antibody levels were significantly lower than that measured at 14 days after the second 132 dose ( Figure 1B) 133

Neutralizing antibodies vs. age and gender 134
At day 14 after the first dose, the development and levels of detectable neutralizing 135 antibodies among 554 study participants were negatively correlated with age. This  Figure 1B&C). 143 In terms of gender, with the exception of day 28 after the first dose, neutralizing antibody 144 levels and the proportion of study participants with detectable neutralizing antibodies were 145 comparable between males and females (Table 2 and Supplementary Figure 2). 146 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. Older individuals, especially those 80 years or above, without prior infection had lower 165 levels of neutralizing antibodies induced by the first dose than younger adults [12,13]. 166 These age-dependent responses were most profound within the first 3 weeks after 167 vaccination, but were resolved by the administration of the second dose [12]. Although 168 similar trends were observed in our study, at day 28 after the first dose, the differences in 169 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. ; https://doi.org/10.1101/2021.07.08.21260162 doi: medRxiv preprint study participant was older than 71 years, explaining why the observed differences were 171 less profound as compared to the UK population based study. 172 The results provide reassuring evidence for the effectiveness of the proposed vaccination 173 strategy aiming at prioritizing the first dose for as many people as possible in the first 174 instance [14]. However, the data also emphasize the importance of the second dose [15], 175 especially in older people, while it remains unknown whether the third dose is needed to 176 provide long-term protection. A decline in antibody titers was recorded at week 8-12 after 177 the first two doses among 75 study participants in the UK [16]. But the administration of 178 the third dose helped boost the immune response. Antibody waning is presumably more 179 profound among individuals without prior infection. Follow study is therefore critical to 180 assess the levels of antibody waning among our study participants and correlates of 181 protection, especially in the context of the rapid spread of the Delta variant globally. 182 The strength of our study includes that it was conducted in a naïve population, with no 183 prior SARS-CoV-2 infections [4]. Thus our data more naturally reflect the immunity 184 profiles induced by Oxford-AstraZeneca COVID-19 vaccine. Additionally, although the 185 correlates of protection for COVID-19 vaccine remain to be determined, neutralizing 186 antibodies are considered to be the most reliable surrogates [17]. Therefore by measuring 187 neutralizing antibodies, our findings more accurately reflect the potential of correlates of 188

protection. 189
Our study has some limitations. First, we did not study cellular immunity, especially T cell 190 response. Cellular immunity has been increasingly recognized to play a role in the 191 pathogenicity and immune response of SARS-CoV-2 infection [18]. Therefore, the 192 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. ; https://doi.org/10.1101/2021.07.08.21260162 doi: medRxiv preprint durability of both cellular and humoral immune responses should be further explored. 193 Second, due to the age and gender structure in nature of HTD staff, we did not include 194 participants older than 71 years and females were predominant among our study subjects. 195 Of note, compared with males, females seemed to better respond to Oxford-AstraZeneca 196 COVID-19 vaccine at day 28 after the first dose, which merits further research 197 In summary, Oxford-AstraZeneca COVID-19 vaccine is immunogenic in Vietnamese 198 healthcare workers who were naïve to SARS-CoV-2 infection. Neutralizing antibody 199 levels decreased at month 3 after vaccination. The requirement for a third dose warrants 200 further research. These data are critical to informing the deployment of COVID-19 vaccine 201 in Vietnam and beyond. 202 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021  . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021  is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted July 28, 2021. ; https://doi.org/10.1101/2021.07.08.21260162 doi: medRxiv preprint   Correspondence: Nguyen Van Vinh Chau, chaunvv@oucru.org, Le Van Tan,310 tanlv@oucru.org 311 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 28, 2021. ; https://doi.org/10.1101/2021.07.08.21260162 doi: medRxiv preprint Supplementary Figure 1: Probability of having detectable neutralizing antibodies among the study participants selected for assessment of the impact of the second dose. A) At 2 and 4 weeks after the first dose (n=104), B) Before the second dose (i.e.6 weeks after the first dose) and 2 weeks after the second dose (n=104) and C) at month 3 after the first dose (n=94). Shaded areas indicate 95% of confident intervals.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 28, 2021. ; https://doi.org/10.1101/2021.07.08.21260162 doi: medRxiv preprint Supplementary Figure 2: Association between neutralizing antibody levels and gender. A) At 2 and 4 weeks after the first dose of the whole group (n=554), B) From baseline to month three after the first dose of the subgroup . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted July 28, 2021. ; https://doi.org/10.1101/2021.07.08.21260162 doi: medRxiv preprint