key: cord-1040915-nkjji4xr
authors: Betton, Maureen; Livrozet, Marine; Planas, Delphine; Fayol, Antoine; Monel, Blandine; Védie, Benoit; Bruel, Timothée; Tartour, Eric; Robillard, Nicolas; Manuguerra, Jean-Claude; Blanchard, Anne; Ghosn, Jade; Visseaux, Benoit; Péré, Hélène; Lebeaux, David; Schwartz, Olivier; Veyer, David; Hulot, Jean-Sébastien
title: Sera neutralizing activities against SARS-CoV-2 and multiple variants six month after hospitalization for COVID-19
date: 2021-04-14
journal: Clin Infect Dis
DOI: 10.1093/cid/ciab308
sha: 677870c6ef76a8bd6cb917e9cf1dd59fd7f1d2dd
doc_id: 1040915
cord_uid: nkjji4xr

BACKGROUND: Humoral response to SARS-CoV-2 occurs within the first weeks after COVID-19. Those antibodies exert a neutralizing activity against SARS-CoV-2, whose evolution overtime after COVID-19 as well as efficiency against novel variants are however poorly characterized. METHODS: In this prospective study, sera of 107 patients hospitalized with COVID-19 were collected at 3- and 6-months post-infection. We performed quantitative neutralization experiments on top of high-throughput serological assays evaluating anti-Spike (S) and anti-Nucleocapsid (NP) IgG. FINDINGS: Levels of sero-neutralization and IgG rates against the ancestral strain decreased significantly over time. After 6 months, 2.8% of the patients had a negative serological status for both anti-S and anti-NP IgG. However, all sera had a persistent and effective neutralizing effect against SARS-CoV-2. IgG levels correlated with sero-neutralization and this correlation was stronger for anti-S than for anti-NP antibodies. The level of sero-neutralization quantified at 6 months correlated with markers of initial severity, notably admission in intensive care units and the need for mechanical invasive ventilation. In addition, sera collected at 6 months were tested against multiple SARS-CoV-2 variants and showed efficient neutralizing effects against D614G, B.1.1.7 and P.1 variants but a significantly weaker activity against B.1.351 variant. INTERPRETATION: Decrease of IgG rates and serological assays becoming negative did not imply loss of neutralizing capacity. Our results indicate a sustained humoral response against the ancestral strain and the D614G, B.1.1.7 and P.1 variants for at least 6 months in patients previously hospitalized for COVID-19. A weaker protection was however observed for the B.1.351 variant.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has been shown to induce a humoral immune response with seroconversion occurring in most patients between 7 and 21 days after diagnosis [1, 2] . This early humoral response is mostly composed of IgA, IgM and IgG directed against the viral surface glycoprotein Spike (S), the nucleocapsid protein (NP) or the spike Receptor Binding Domain (RBD) [2] . The detection of such antibodies may reflect a neutralizing activity believed to be a key point in viral clearance [3, 4] , as well as conferring a relative protection to the disease in the convalescent phase. Similarly to other coronaviruses, anti-SARS-CoV-2 antibodies decline overtime [5] , which raised questions about the extent of the protection conferred and the potential risk of reinfection. Furthermore, the recent emergence of multiple SARS-CoV-2 variants raised additional questions on cross-reactivity of the acquired antibodies after COVID-19 [6] .

Some publications have reported an association between the level of antibodies and the clinical severity, a higher level being observed in patients presenting the most critical form of the disease [4, [7] [8] [9] [10] . However, if there is consistent evidence that outpatients usually develop weaker immune response, there is fewer data relating to patients in intensive care units, a valuable population whose described antibodies response could set the upper limit of the humoral immunity against SARS-CoV-2.

While vaccination is ongoing worldwide, the insufficient supply of doses makes prioritization strategies still needed. Being able to shape levels of immunity required to protect against severe reinfection would considerably assist public health strategies in this regard, in addition to being critical information to estimate if vaccines stand the test of time and emerging variants.

A c c e p t e d M a n u s c r i p t 5 Our study explores the longitudinal evolution of antibody levels and of sera neutralizing activities in a French monocentric cohort of patients hospitalized for COVID-19 during the first wave of SARS-CoV-2 pandemic and followed-up for 6 months after hospital discharge. In addition to the ancestral viral strain, sera-neutralization activities against the emerging SARS-CoV-2 variants (B.1.1.7, B.1.351 and P.1) were evaluated.

We conducted a single-center prospective observational study on adult patients with laboratory positive SARS-CoV-2 real-time reverse-transcriptase polymerase chain reaction (RT-PCR) admitted to Hôpital Européen Georges Pompidou (APHP, Paris, France) for at least 48h. All patients were initially enrolled from March 17 th to April 29 th 2020 and were then proposed for a clinical and serological follow-up at month 3 (M3) and M6 post-infection. The study is part of The French Covid cohort (NCT04262921) sponsored by Inserm and was authorized by the French Ethics Committee CPP Ile-de-France VI (ID RCB:2020-A00256-33). This study was conducted with the understanding and the consent of each participant or its surrogate.

Demographic, clinical presentation, and comorbidity data during the index COVID-19 hospitalization were extracted from the electronic medical records collected in a standardized data collection form in the Clinical Data Warehouse (CDW) of our hospital. The dedicated medical records were stored on an i2b2 platform in a CDW together with all other hospital health records.

A c c e p t e d M a n u s c r i p t 6

Abbott SARS-CoV-2 IgG assays (Des Plaines, IL, USA) targeting SARS-CoV-2 nucleoprotein were done on Architect™ i2000SR analyzer (Abbott), according to manufacturer's instructions.

Index value threshold for positivity was 1.4 as recommended. Beckman Coulter Access SARS-CoV-2 IgG assays (Brea, CA, USA) targeting the RBD of SARS-CoV-2 spike surface protein, were done on UniCel DxI 800 Access Immunoassay System (Beckman Coulter), according to manufacturer's instructions. Index value threshold for positivity was 1 as recommended.

Qualitative results as well as index values were used for analysis for both assays.

The ancestral non-D614G SARS-CoV-2 strain (BetaCoV/France/IDF0372/2020) was isolated in [11] . Viruses were sequenced directly on nasal swabs and after one or two passages on Vero cells.

Neutralization was performed using the S-Fuse reporter system, as previously described [11] .

Briefly, U2OS-ACE2 GFP1-10 or GFP 11 cells, which become GFP + upon infection with SARS-CoV-2, were mixed (1:1 ratio) and plated at 8 × 10 3 cells per well in a μClear 96-well plate (Greiner Bio-One). SARS-CoV-2 strains were incubated with sera at the indicated dilutions for 15 min at room temperature and added to S-Fuse cells. All sera were heat inactivated 30 min at 56 °C before use. After 18 h incubation at 37°C 5%CO2, cells were fixed with 2% paraformaldehyde, washed and stained with Hoechst (1:1,000 dilution; Invitrogen). Images were acquired on an Opera Phenix high-content confocal microscope (PerkinElmer). The GFP area, the number of syncytia and nuclei were quantified using the Harmony software (PerkinElmer). The percentage of neutralization was calculated using the number of syncytia with the following formula: 100 × (1 − (value with serum − value in 'noninfected')/(value in 'no serum' − value in 'noninfected')). Neutralizing activity of each sera was expressed as the ED50, calculated using the percentage of neutralization at each different concentration. Cells were tested negative for mycoplasma. Neutralization determined with the S-Fuse reporter system correlates to pseudovirus neutralization assay [12] .

Statistics were performed using NCSS 2012 software (G Hintze, Kaysville, UT, USA). All numerical data were checked for normality and non-normal distributions were transformed (using ExpNorScore function on NCSS, which returns the expected value of the normal order statistic corresponding to X).

A c c e p t e d M a n u s c r i p t 8 Continuous variables are reported as means (SDs). Discrete variables are described as counts and percentages. Groups were compared using Two sample T-Test or Wilcoxon Rank test when necessary for continuous variables and Fisher's exact Test or χ2 for discrete variables.

We also performed a multiple regression analysis to assess variables correlated to the seroneutralization at M6. For analyses, P values <0.05 were considered significant.

The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The corresponding author (J.S.H) had full access to all the data in the study and had the final responsibility for the decision to submit for publication.

Between March 17 th and April 29 th , 2020, 354 patients were hospitalized at Hôpital Européen Georges Pompidou (Paris, France) with a confirmed SARS-CoV-2-positive pneumonia. By November 16 th , 2020, 85 deaths (24.0%) had occurred during or after hospitalization. From June 17 th to November 16 th 2020, we were able to manage a complete follow-up for 107 of these patients with two time-point visits at 3 and 6 months after hospital discharge (see Supplementary Figure 1 ). The 3-month and 6-month visits were performed with a median interval of 98 days (IQR: 91-101) and 203 days (IQR: 191-216), respectively. Among those patients with a complete follow-up, 32.7% (35/107) had required medical care in Intensive Care Unit during the acute COVID-19 phase. Fifteen (15/35, 42 .9%) of them required invasive mechanical ventilation (MV). The oxygenation maximal flow in patient in non-ICU unit, in patients not requiring MV in ICU unit and in patients before MV in ICU was 2.0 ± 1.6, 10.7± 5.2 and 13.0 ± 3.6 L/min respectively (p<0.0001). A majority of patients were male (73/107, 68.2%) with a mean age of 58.7 ± 14.0 years-old. A past history of cardiovascular risk factor (chronic cardiac disease, diabetes, obesity, hypertension, chronic kidney disease) was found in 51.4% (55/107) of them and 10.2% (11/107) had immunosuppressive diseases (cirrhosis, asplenia, sickle cell anemia, solid organ or stem cell transplantation, HIV infection, primary immune deficiency, chronic hematological disease, malignant neoplasm, autoimmune disorder). Moreover, 5.6% (6/107) of them were previously treated with an immunosuppressive therapy. All patients' characteristics are listed in Table 1 . 

We then aimed to assess if a persistent serum neutralizing activity was detected up to 6 months following COVID-19 infection, independent of anti-S or anti-NP levels. To do so, we used S-Fuse cells (specifically designed to become GFP+ when productively infected by SARS-CoV-2) to evaluate our patients' sera propensity to prevent such infection. At minimum dilution (1/30), seroneutralization was observed in all samples at M3 and M6, even when anti-S and anti-NP IgG were considered as negative regarding the commercial kit threshold.

We next quantified sero-neutralization by performing serial dilutions in order to define the ID50 neutralization (maximum dilution to maintain a 50% neutralization capacity). ID50 neutralization significantly decreased between M3 and M6 ( Figure 2) , with residual values nevertheless indicating a high neutralizing activity at M6.

As all measures were highly variable between patients, we then sought factors associated with higher levels of neutralizing activities. In a multiple regression analysis, we found that an initial management in ICU and the need for an invasive mechanical ventilation were the only two factors significantly associated with a higher rate of ID50 neutralization at M6 (Table 2 ). When considering ID50 neutralization according to ICU hospitalization and the need for mechanical ventilation, we found that M a n u s c r i p t 10 patients in ICU had significantly higher neutralizing activities as compared to non-ICU patients with the highest levels observed in ICU patients who had invasive mechanical ventilation ( Figure 3A) .

A similar trend was observed at M3, however not reaching significance ( Figure 3A ).

We then observed that anti-S modestly correlate with sera ID50 neutralization (anti-S IgG : R = 0.54, 95%CI[0.39-0.67], p value <0.0001, with a weaker association for anti-NP IgG (anti-NP IgG : R = 0.33, 95%CI[0.25-0.49], p value 0.0007).

We then analyzed anti-S and anti-NP IgG levels at M3 and M6 according to the initial management in ICU and the need for an invasive mechanical ventilation. We found higher levels of anti-NP IgG in mechanically ventilated patients in ICU vs. non mechanically ventilated patients admitted to ICU vs. patient in non-ICU medical departments at M3 (7.6±1.8 vs. 7.2±2.2 vs. 6.2±2.3, p=0.03 by Kruskall-Wallis test, Figure 3B) Figure 3B ) with a significantly higher rate in ICU patients vs. no ICU patients (4.9±2.0 vs. 3.3±2.4 p value = 0.001). In contrast, this pattern was not observed with the anti-S antibodies as there were no significant differences between the ICU vs. the no ICU groups at M3 and at M6 ( Figure 3C ).

We then used sera collected at 6 months following COVID-19 infection to assess their neutralizing activities against multiple variants, including D614G, B. 

In the present study, we described the longitudinal evolution of IgG levels and sero-neutralization at 3-and 6-months post-infection in a relatively large prospective cohort of 107 hospitalized patientswith a third of severe cases tending to be scarce in literatureand thus provides important data about the evolution of humoral immunity after a hospitalization for COVID-19 infection. We found that at least one serology assay was still positive at 6 months in 97.2% of the studied patients. Although antibodies levels decreased significantly over time, with rates dropping under the positivity threshold in a few cases, all patients' sera conserved an effective neutralizing activity at 6 months post-infection against the ancestral strain. Sero-neutralization remained higher at 3 and 6 months in patients who had required intensive care. We also used our sera collection to estimate the levels of humoral protection against the emerging SARS-CoV-2 variants. In these additional in vitro experiments, we found that sera-neutralizing activities was also effective against the B.1.1.7 and P.1 variants (also known as the UK and Brazilian variants), but was potentially weaker for the B.1.351 strain (also known as South-African variant).

Higher ID50 against the ancestral strain were observed in patients with more severe presentations, even at distance of infection. This correlation between sero-neutralization and clinical severity has been previously described [7, 10, 13, 14] , and our results now indicate that this trend might persist over time. Interestingly, we found that anti-NP IgG titers were higher according to the stage of severity, which was not observed for anti-S IgG. Early after symptoms onset, anti-NP response had already been reported as a possible marker of severity, associated to delayed viral clearance and disease severity [8] . Whether this exacerbated humoral response in severe patients is a A c c e p t e d M a n u s c r i p t 12 protective adaptation to a more intense viral load or if it plays a putative role in pathogenicity remains subject to debate [15, 16] .

At 6 months post-infection, we found that anti-S IgG titers correlated with sera ID50 neutralization, but not anti-NP IgG. This was generally in line with other works that had underlined, at different times post infection, a relatively strong correlation between neutralizing antibodies and anti-S or anti-RBD antibodies, and a usually poorer correlation to anti-NP antibodies [9, 14, 17, 18] .

The evolution of sera-neutralization overtime and in response to emerging variants is one important element to consider when questioning the extent of effective protection conferred by a priori infection and thus helps evaluating the strength of shield immunity during this pandemic. In line with first encouraging results [11, 13, 17, [19] [20] [21] , we confirmed in this study the persistence of neutralization up to 6 months post-infection against the ancestral strain, but also the existence of a broader and similarly effective neutralizing activities against novel variants including B1.1.7 and P1 variants. These results are in favor of antibodies cross-reactivity and potential protection against reinfection with these variants. As compared to other strains, we observed a weaker protection against the B.1.351 variant, however with a substantial neutralizing activity observed in most of the patients. These data suggest that these antibodies acquired during a prior COVID-19 infection might not confer a complete protection against this emerging variant firstly described in South African patients [22] . We took benefit of the available sera collection and the development of a novel assay to estimate these activities but cannot extrapolate on a higher risk of reinfection in these convalescent patients. Several publications based on pseudovirus or virus neutralizing assays A c c e p t e d M a n u s c r i p t 13 outlined that variants could partially evade humoral immunityin exposed patients as well as vaccinees [23] [24] [25] .

Our cohort of patients with the most critical forms of COVID-19 represents a valuable population to explore maximal antibodies response and set the upper limit of the humoral immunity against the SARS-CoV-2. Our results suggest that our patients should be protected at least 6 months against future re-infection. So far, there are few cases of reinfection published in the literature [26] .

Reinfections rates have been estimated as low in large recent observational studies despite waning neutralizing antibodies [27, 28] . However, it is impossible to extrapolate on future infections with novel variants. Interestingly, a recent model predicted a relationship between neutralizing levels and immune protection against the ancestral strain and novel variants, as well as a protection against severe disease [29] . Further studies are nonetheless required, especially regarding the B.1.351 variant, in order to determine if partial humoral escape can clinically lead to severe events.

Cellular immunity also appears as a major shield against SARS-CoV-2, with the development of durable T memory cells [20] , whose reactivity could be only slightly impacted by variants [30] .

Overall, patients who survived the most critical forms of COVID-19 consequently developed an intense and prolonged humoral immunity. These levels tend to correlate with the severity of the initial presentation, with patients in ICU who had invasive mechanical ventilation having the highest neutralizing activities as compared to ICU and non-ICU patients. A c c e p t e d M a n u s c r i p t 28 Figure 4 

Immunology of COVID-19: Current State of the Science

Diagnostic accuracy of serological tests for covid-19: systematic review and meta-analysis

Association between SARS-CoV-2 Neutralizing Antibodies and Commercial Serological Assays

Kinetics of viral load and antibody response in relation to COVID-19 severity

Antibody response to SARS-CoV-2 infection in humans: A systematic review

The great escape? SARS-CoV-2 variants evading neutralizing responses

Clinical and immunological assessment of asymptomatic SARS-CoV-2 infections

Antibody Responses to SARS-CoV-2 in Patients With Novel Coronavirus Disease

Anti-SARS-CoV-2 Antibody Responses in Convalescent Plasma Donors Are Increased in Hospitalized Patients; Subanalyses of a Phase 2

COVID-19-neutralizing antibodies predict disease severity and survival

Sensitivity of infectious SARS-CoV-2 B.1.1.7 and B.1.351 variants to neutralizing antibodies

IgA dominates the early neutralizing antibody response to SARS-CoV-2

A longitudinal study of SARS-CoV-2-infected patients reveals a high correlation between neutralizing antibodies and COVID-19 severity

The potential danger of suboptimal antibody responses in COVID-19

COVID-19: immunopathology and its implications for therapy

Seroprevalence of anti-SARS-CoV-2 antibodies in COVID-19 patients and healthy volunteers up to 6 months post disease onset

Evaluating the Association of Clinical Characteristics With Neutralizing Antibody Levels in Patients Who Have Recovered From Mild COVID-19 in

Sex differences in the decline of neutralizing antibodies to SARS-CoV-2

Immunological memory to SARS-CoV-2 assessed for up to 8 months after infection

Persistence and decay of human antibody responses to the receptor binding domain of SARS-CoV-2 spike protein in COVID-19 patients

Evidence of escape of SARS-CoV-2 variant B.1.351 from natural and vaccine-induced sera

Emerging SARS-CoV-2 variants reduce neutralization sensitivity to convalescent sera and monoclonal antibodies

Multiple SARS-CoV-2 variants escape neutralization by vaccine-induced humoral immunity

What we know about covid-19 reinfection so far

A 1 to 1000 SARS-CoV-2 reinfection proportion in members of a large healthcare provider in Israel: a preliminary report