key: cord-0716687-fo4ivew0
authors: Vania, J. F.; Bergman, M.-L.; Goncalves, L. A.; Duarte, N.; Borges, P. C.; Diwo, C.; Castro, R.; Matoso, P.; Malheiro, V.; Brennand, A.; Kosack, L.; Akpogheneta, O.; Figueira, J. M.; Casaca, A. M.; Cardoso, C.; Alves, P. M.; Penha-Goncalves, C.; Demengeot, J.
title: Population homogeneity for the antibody response to COVID-19 Comirnaty vaccine is only reached after the second dose
date: 2021-03-24
journal: nan
DOI: 10.1101/2021.03.19.21253680
sha: f585575aeb23b339e9de1c2b1b5e082401e71b58
doc_id: 716687
cord_uid: fo4ivew0

While mRNA vaccines authorised for emergency use are administrated worldwide in an effort to contain the COVID19 crisis, little is known about the heterogeneity of the immune response they induce. Here, we report the first 6 weeks of a longitudinal study that quantifies the humoral immune response to BNT162b2 mRNA COVID-19 (Pfizer/BioNTech, Comirnaty) in 1245 health care providers, the Lx1000HCW-PZF cohort. We reveal a striking inter-individual variation 3 weeks after the 1st dose administration that only in part related to age and sex. While population homogeneity in robust IgG responses was reached upon 2nd dose administration, IgM and IgA levels remain low and heterogenous. Our findings of isotypic and heterogenous antibody responses to Comirnaty highlight the need for evaluating the efficacy of COVID-19 mRNA vaccine in preventing infection aside disease, and - contrary to what has been proposed - advocate for the interval between the two doses not to be extended.

Authorization for emergency use of two mRNA COVID-19 vaccines, both encoding the most immunogenic protein of SARS-CoV-2, spike, was conceded in late 2020 by regulatory agencies such as FDA, WHO and EMA. These authorizations were based on results of phase 3 clinical trials that demonstrated high standards of safety and high levels of efficacy in preventing symptomatic SARS-CoV-2 infections [1, 2] .

While these vaccines are introduced around the world and administered to millions of people, there is a growing and acute need to evaluate their effectiveness at the population level, an endeavour that may require months of epidemiological studies.

Awkwardly, little attention has been given to whether immune responses triggered by mRNA vaccines encoding SARS-CoV-2-spike are homogenously robust. To date, immune responses have been seldom measured upon mRNA COVID-19 administration, and when this was the case, the observations concerned very small groups of participants ranging from n=8 to 20 [2] [3] [4] [5] . Immunogenicity of mRNA COVID-19 vaccines and their inter-individual variation can be easily monitored in medium to large cohorts by measuring serum reactivities to the vaccinal antigen or part of it. Notably, the receptor binding domain of spike contains the AA motifs permitting SARS-CoV-2 binding to ACE2 receptor, a prerequisite for infection, and serum reactivity to this region encompasses neutralizing activity [6] . Anti-spike immunoglobulins are also expected to mediate viral particle removal through antibody-mediated opsonization and phagocytosis, and through the recruitment of the complement system. Beyond their direct functionality, vaccine specific antibodies are markers of adaptive immunity response.

In the vast majority of cases, SARS-CoV-2 infection associates with the induction of robust IgG anti-spike responses lasting for 6 to 8 months, while strong IgM anti-spike reactivities are transient. Strong IgA anti-spike responses are frequent and may be more prevalent in symptomatic patients ( [7] , and our own observations). IgA are either dimeric or monomeric. Dimeric IgA are produced at mucosal sites and result from bonafide germinal centre reactions involving helper T cells. These have been shown to be part of the humoral response triggered by SARS-CoV-2 and to confer neutralizing capacity [8] . Monomeric IgA are abundant in blood, though partial or . CC-BY-NC 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 24, 2021. ; https://doi.org/10.1101/2021.03.19.21253680 doi: medRxiv preprint total circulating IgA deficiency is common in humans [9] . Neither the function or the origin of circulating IgA is well understood. Astonishingly, it still remains unclear whether mRNA vaccine can be expressed at mucosal sites. Overall, immune responses induced by natural SARS-CoV-2 infection seems to protect from novel infection with the same virus variant. Whether RNA vaccines provide the same broadness of Ig class has not been formally reported.

The COVID-19 vaccination campaign in Portugal was initiated in late December 2020 coinciding with a peak of disease transmission which reached 131 new daily cases per 100,000 inhabitants and caused a most intense demand for hospital care.

The vaccination roll-out started with hospital healthcare professionals at the COVID-19 response frontline, who showed during 2020 a level of exposure to the virus somewhat higher than the general population (our unpublished results). Here, we report on the humoral response to BNT162b2 mRNA COVID-19 (Pfizer/BioNTech, Comirnaty) vaccination in healthcare professionals working in a group of hospitals in Lisbon, Portugal.

The study enrolled 1245 healthcare providers working at 3 hospitals, administratively grouped in a single regional centre (CHLO), in Lisbon, Portugal. Participants were scheduled to initiate BNT162b2 mRNA (Pfizer/BioNTech, Comirnaty) vaccination in December 2020/January 2021, along the original protocol of 2 doses with a 3 weeks interval. The cohort presents a biased sex ratio (77% female, 23% male), as is common in this professional area, and encompasses a broad age range (19 to 70 years, median 41 years for females, 39 years for males). Venous blood was collected every 3 weeks, at the days of 1 st (d0) and 2 nd (d21) dose administration, and 21 days after the 2nd dose administration (d42). Drop out was of 14.8% (n=184) with 12.5% (n=156) during the first phase (Figure 1) .

Diagnosed COVID-19 prior to vaccination was an exclusion criterion in accordance with the national vaccination plan. It is estimated that only about 10-20 % of COVID- is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 24, 2021. ; https://doi.org/10.1101/2021.03.19.21253680 doi: medRxiv preprint 19 cases were diagnosed in 2020 in Europe overall, leaving most a/paucisymptomatic exposure undetected. To complete the identification of participants with prior infection, the entire cohort was tested at d0 for serum reactivity against SARS-CoV-2 nucleocapsid (N), identifying 39 such cases (3%). In addition, 8 participants were diagnosed COVID-19 during the first week of the study, and an additional 11 showed SARSCoV-2 N antigen reactivity at d21, all cases likely related to an outbreak at the participating hospitals.

All collected samples were analysed for bulk reactivity against SARSCoV-2-RBD, using a commercial ECLIA, and for isotype specific (IgG, IgM and IgA) anti-SARSCoV-2-Spike using an in-house ELISA assay. To directly determine the immunogenicity of Comirnaty, SARS-CoV-2 naïve participants, defined as negative for serum anti-SARS-CoV-2 N reactivity, were first analysed using a binary classification (Figure 2A is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 24, 2021. ; https://doi.org/10.1101/2021.03.19.21253680 doi: medRxiv preprint consistent with classical IgM responses that peak during the first week post antigen encounter, and are not significantly boosted through memory cell recall. The unchanged IgA response after the 2 nd vaccine injection strongly suggests most of the IgA anti-spike antibodies detected are of the T cell independent class and monomeric, which are unlikely to confer mucosal immunity upon subsequent SARS-CoV2 infection. In turn, this feature suggests RNA vaccine may not prevent the early stages of SARS-CoV2 infection at mucosal site.

Of note, our analysis at day 0 reveals a sizable fraction of participants presenting IgM anti-spike reactivity prior to vaccination (12,5% above threshold as compared to 0,8% when testing sera from 1000 donors collected before COVID19 pandemic). This result corroborates our unpublished longitudinal analysis of 1500 hospital healthcare workers during 2020, and the nature of these peculiar IgM reactivities will be discussed elsewhere.

As a first approach to identify the basis of the inter-individual heterogeneity in the response to Comirnaty vaccine, we performed stratified analyses for 3 parameters.

We first excluded concerns of RNA vaccine stability, by ascertaining the spread in anti-spike reactivity levels did not differ in groups of participants who received the 1 st vaccinal dose at different calendar days (not shown). We next performed stratified analysis by age groups (in 10-year bins) and revealed a cumulative negative effect of age on the level of the humoral response, more marked after the 1 st dose administration (d21) but still significant 3 weeks after the 2 nd vaccination (d42). The age effect was evident for bulk anti-RBD reactivity and for anti-spike IgG and IgM, but not IgA levels, highlighting again the unconventional nature of the IgA response to this RNA vaccine (Figure 3A and B) . Despite the overall higher immunocompetence of the youngest age strata (19-29 years), levels of specific reactivities were still strikingly spread at d21 in this age group (e.g. titre range was estimated from 2x10 2 to 2x10 4 or above for IgG). Further stratification of the cohort by age and sex revealed that in the older strata (60-70 years of age) males presented lower anti-RBD and anti-spike responses, as compared to females, a difference more marked at d21 (Figure 3C and D) . Sex effect in this older age group affected also the frequency of IgG seroconversion after 1 st dose administration, with a positivity of 82.7% for . CC-BY-NC 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Finally, and as reported by others [4, 5] , analysis of participants previously exposed to SARS-CoV-2 confirmed the 1 st vaccination dose acts as a boost. The full cohort encompassed 31 participants who tested anti-N positive at d0 and complied with the 3 collections schedules. For these participants, anti-RBD Ig and anti-spike IgG levels reached maximal values by day 21 (Figure 4A and B) . This result also serves to 

Recruitment and enrolment: This study was approved by the Ethics committee of the Centro Hospitalar Lisboa Ocidental in compliance with the Declaration of Helsinki and follows international and national guidelines for health data protection. All participants provided informed consent to take part in the study.

Blood samples processing and storage: Venous blood was collected by standard phlebotomy. Blood collection occurred at the day of the first vaccination (baseline, d0), the day of the second vaccination (d21) and 3 weeks later (day 42). Serum was prepared using standard methodology.

Immunoassays: Electro-chemiluminescence immunoassay (ECLIA) was used to quantify Ig anti-N (Elecsys® Anti-SARS-CoV2 N, Roche) and anti-RBD (Elecsys® is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Anti-SARS-CoV2 S, Roche), ran (on cobas e602) and analysed as per the manufacturer instructions, with a threshold defining positivity at index value = 0.9.

Direct ELISA was used to quantify IgG, IgM and IgA anti-full-length spike. The assay was adapted from [10] and semi-automized to measure IgM, IgG and IgA in 384-well format, according to a protocol to be detailed elsewhere. Briefly, coating was 0.5 µg/ml highly purified spike protein [11] ; sera were diluted 1/50 in duplicate, is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted March 24, 2021. ; https://doi.org/10.1101/2021.03.19.21253680 doi: medRxiv preprint

Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine

Safety and Immunogenicity of Two RNA-Based Covid-19 Vaccine Candidates

mRNA vaccine-elicited antibodies to SARS-CoV-2 and circulating variants

Antibody response to first BNT162b2 dose in previously SARS-CoV-2-infected individuals

Antibody Responses in Seropositive Persons after a Single Dose of SARS-CoV-2 mRNA Vaccine

Human neutralizing antibodies elicited by SARS-CoV-2 infection

Selective IgA deficiency

A serological assay to detect SARS-CoV-2 seroconversion in humans

Production of high-quality SARS-CoV-2 antigens: impact of bioprocess and storage on glycosylation, biophysical attributes, and ELISA serologic tests performance

Epub 2021/02/25

We thank the healthcare workers who participated to the study. We are indebted to Jorge Carneiro for help with ELISA data analysis and Tiago Paixão for guidance in data processing. We are grateful to Joao Costa and Joana Bom for providing training