Ad26.COV2.S breakthrough infections induce high titers of antibodies capable of neutralizing variants of concern

The Janssen (Johnson & Johnson) Ad26.COV2.S non-replicating viral vector vaccine, which requires only a single dose and conventional cold chain storage, is a valuable tool for COVID-19 vaccination programs in resource-limited settings. Here we show that neutralizing and binding responses to Ad26.COV2.S vaccination are stable for 6 months post-vaccination, with responses highest against the ancestral vaccine-similar D614G variant. Secondly, using longitudinal samples from individuals who experienced clinically mild breakthrough infections 3-4 months after vaccination, we show dramatically boosted binding antibodies, Fc effector function and neutralization. These responses, which are cross-reactive against diverse SARS-CoV-2 variants and SARS-CoV-1, are of similar magnitude to humoral immune responses measured in severely ill, hospitalized donors. These data highlight the significant priming capacity of Ad26.COV2.S, and have implications for population immunity in areas where the single dose Ad26.COV2.S vaccine has been deployed.


Main Text
A phase 3 clinical trial of Ad26.COV2.S in eight countries demonstrated 85% protection against severe disease 1 , including in South Africa, where the trial coincided with the emergence of the neutralization-resistant Beta (B.1.351) variant 2 . As a result, Ad26.COV2.S was made available to South African health care workers (HCWs) in early 2021 through the Sisonke open-label, phase 3b clinical trial 3 . Globally the Ad26.COV2.S vaccine has also been adopted as part of COVID-19 vaccination programs in a number of countries, including South Africa, USA, and EU member states, with 5.38, 15.68 and 16.16 million doses administered in these regions, respectively, by the beginning of November 2021.
Here, we evaluated the durability and breadth of vaccine-elicited humoral responses in nineteen HCWs vaccinated with Ad26.COV.2S in February-March 2021 (Fig. 1a). Secondly, we characterised the humoral response to breakthrough infection (BTI) in a subset of six individuals with SARS-CoV-2 PCR-confirmed infections 4-5 months (median number of months: 4.4; interquartile range: 4.1-4.8) following vaccination. Five of these participants were followed longitudinally 2-6 months post-vaccination, while for the sixth BTI participant only a single sample from 1 month post-infection was available (Table S1). These BTIs occurred during the third wave of SARS-CoV-2 infections in South Africa, which was driven by the more transmissible Delta (B.1.617.2) variant 4 . Participants, of whom 16/19 were female, had a median age of 34 (interquartile range: 30-40 years) and all presented with mild disease (Table S1). All nineteen participants were SARS-CoV-2 naive prior to vaccination, as confirmed by nucleocapsid ELISA (Fig.  1b).
We next assessed the humoral immune responses following BTI. In all participants, BTI occurred between 3-and 5-months post-vaccination. Prior to BTI, the nucleocapsid binding responses in both the BTI and non-BTI participants were negative, and only detected following BTI as expected (Fig 1b). There were also no significant differences in D614G spike binding responses between the BTI and non-BTI participants prior to 3/4-months post-vaccination (Fig. S1a). However, following infection, there was a 3.3-fold increase in spike responses, which peaked at approximately 2-weeks post infection (5-months post-vaccination) and remained constant until 1-month post-infection (6-months post-vaccination) (Fig. S1a).
Neutralization titers against D614G closely mirrored the spike binding and ADCC response, with no significant differences in titers between the BTI and non-BTI participants prior to 3/4-months post-vaccination, but with a dramatic increase in titers for all participants (407-fold increase from 102 to 41528 GMT) following infection (Fig.  1c). This increase in neutralization titers is similar in magnitude to what was previously reported for a single individual with Ad26.CoV.2S BTI 6 . Neutralization titers peaked at approximately 2 weeks post-infection (5-months post-vaccination) and declined by approximately 4.7-fold one month thereafter (Fig. 1c). Neutralization titers after BTI were also significantly higher against five SARS-CoV-2 variants relative to non-BTI participants (70-154-fold difference in GMT), and SARS-CoV-1 (9-fold difference in GMT) (Fig. 2, S2b). Thus, in contrast to vaccine-elicited responses, BTI after a single dose of Ad26.COV2.S resulted in complete coverage of SARS-CoV-2 variants at titers >1:3,000 ( Fig. 2, S2a, S2b).
Overall, our data shows durable vaccine-elicited humoral immune responses 6 months after a single dose of Ad26.COV2.S, consistent with other studies [5][6][7][8][9] . Furthermore, despite relatively modest titers after vaccination, we observe significantly boosted binding antibodies, ADCC and neutralization activity following BTI. This boost resulted in neutralization titers in BTI participants at 1-month post-infection (GMT 8,249) that were higher than those elicited by a 2 dose Pfizer-BioNTech (BNT162b2) vaccine regimen (GMT: 1,128), or those observed in acutely infected hospitalised individuals with moderate (GMT: 993) or severe disease (GMT: 3,747) (Fig. S3). Similar to BTI individuals, we have previously confirmed that ADCC, binding and neutralization are also significantly boosted following vaccination in individuals who were previously infected 10 , but not to the same levels we report here for BTI (1,372 vs 8,249 for neutralization, respectively) (Fig. S3). This confirms previous data indicating that Ad26.COV2.S is a potent priming immunogen 11 .
These data add to previous reports of BTI following mRNA vaccination which results in >30-fold increased neutralization potency, suggesting broad relevance across multiple vaccine modalities 12,13 . Taken together, these findings suggest a strongly synergistic effect of vaccination and infection which will contribute to higher levels of protective community immunity in areas with high disease burden. Furthermore, the potent priming of humoral immune responses by Ad26.COV2.S may be beneficial for homologous and 3 . CC-BY 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 November 9, 2021. ; heterologous boosting strategies 11 . Overall, this study provides insight into the magnitude and quality of humoral immune responses elicited by breakthrough infections after an adenovirus-based vaccine, with implications for public health interventions in regions with high SARS-CoV-2 transmission.  The horizontal dashed line indicates the lower limit of quantitation. Statistical analyses were performed using the Mann-Whitney test between groups, with *** denoting p<0.001, NS for non-significant and ND for no data. Each dot represents the neutralization titer of a single participant, with the BTI participants and non-BTI participants shown in red and blue, respectively. The GMT for each group against each variant is shown by a black horizontal bar. Neutralization titers in the BTI group were significantly higher than those of the non-BTI group (70 to 154-fold higher GMT against the SARS-CoV-2 variants and 9-fold higher against SARS-CoV-1). Statistical analyses were performed using the Mann-Whitney test between groups, with *** denoting p<0.001 and **** denoting p<0.0001.

Human samples
HCWs vaccinated with one dose of Ad26.CoV2.S (5x10 10 viral particles) as part of the Sisonke implementation trial were followed longitudinally and plasma sampled at 2-, 4-and 6-months post-vaccination. An additional plasma sample was collected from BTI participants at 5-months post-vaccination, which was approximately 2-weeks post-infection. Non-BTI participants were recruited from HCWs at the National Institute for Communicable Diseases (NICD) (Johannesburg), while BTI participants were recruited from HCWs at the NICD, Steve Biko Academic Hospital (Tshwane, South Africa) and Groote Schuur Hospital (Cape Town, South Africa). Ad26.CoV2.S vaccinees with prior SARS-CoV-2 infection were recruited from a longitudinal study of healthcare workers enrolled from Groote Schuur Hospital, with plasma samples collected 2-months post-vaccination. Plasma was also collected from thirteen participants that had received two doses of the Pfizer BioNTech vaccine (BNT162b2) 2-months after they had received their last dose (Johannesburg, South 5 . CC-BY 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 November 9, 2021. ; Africa). Convalescent participants were recruited as part of a hospitalised cohort at the Steve Biko Academic Hospital between May and August 2020, with plasma samples collected 10-days after the initial positive PCR test. Ethics approval was obtained from the Human Research Ethics Committees of the University of the Witwatersrand (ethics reference number: M210465), University of Pretoria (ethics reference number: 247/2020) and University of Cape Town (ethics reference numbers: 190/2020 and 209/2020). Written informed consent was obtained from all participants.

SARS-CoV-2 antigens
For ELISA, SARS-CoV-2 original variant (D614G) full spike proteins were expressed in Human Embryonic Kidney (HEK) 293F suspension cells by transfecting the cells with the respective expression plasmid. After incubating for six days at 37°C, 70% humidity and 10% CO 2 , proteins were first purified using a nickel resin, followed by size-exclusion chromatography. Relevant fractions were collected and frozen at -80°C until use. A commercial, recombinant nucleocapsid protein (BioTech Africa, Cat. no.: BA25-C) was used as the antigen in the nucleocapsid ELISAs.

SARS-CoV-2 Spike and nucleocapsid enzyme-linked immunosorbent assay (ELISA)
Spike or nucleocapsid protein (2 μg/mL) was used to coat 96-well, high-binding plates and incubated overnight at 4°C. The plates were incubated in a blocking buffer consisting of 5% skimmed milk powder, 0.05% Tween 20, 1x PBS. Plasma samples were diluted to 1:100 starting dilution in a blocking buffer and added to the plates. IgG secondary antibody was diluted to 1:3000 or 1:1000 respectively in blocking buffer and added to the plates followed by TMB substrate (Thermo Fisher Scientific). Upon stopping the reaction with 1M H 2 SO 4 , absorbance was measured at a 450 nm wavelength. In all instances, mAbs CR3022 and BD23 were used as positive controls and palivizumab was used as a negative control.

Spike plasmid and lentiviral pseudovirus production
The SARS-CoV-2 Wuhan-1 spike gene sequence, cloned into pcDNA3. 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 November 9, 2021. ; For the neutralization assay, plasma samples were heat-inactivated and clarified by centrifugation. Heat-inactivated plasma samples from vaccine recipients were incubated with the SARS-CoV-1/2 pseudotyped virus for 1 hour at 37°C, 5% CO 2 . Subsequently, 1x10 4 HEK293T cells engineered to over-express ACE-2 (293T/ACE2.MF), kindly provided by M. Farzan (The Scripps Research Institute) were added and incubated at 37°C, 5% CO 2 for 72 hours upon which the luminescence of luciferase was measured. Titers were calculated as the reciprocal plasma dilution (ID 50 ) causing 50% reduction of relative light units (RLU). Monoclonal antibodies CB6 and CA1 were used as positive controls.

Antibody-dependent cellular cytotoxicity (ADCC) assay
The ability of plasma antibodies to cross-link spike expressing cells and signal through FcγRIIIa (CD16) was measured as a proxy for ADCC. For spike assays, HEK293T cells were transfected with 5μg of SARS-CoV-2 original variant spike (D614G), Beta or Delta spike plasmids using PEI MAX 40,000 (Polysciences) and incubated for 2 days at 37°C. Expression of spike was confirmed by differential binding of CR3022 and P2B-2F6 and their detection by anti-IgG APC staining measured by flow cytometry. Spike transfected cells (1x10 5 per well) were incubated with heat inactivated plasma (1:100 final dilution) or monoclonal antibodies (final concentration of 100 μg/mL) in RPMI 1640 media supplemented with 10% FBS 1% Pen/Strep (Gibco, Gaithersburg, MD) for 1 hour at 37°C. Jurkat-Lucia™ NFAT-CD16 cells (Invivogen) (2x10 5 cells/well) were added and incubated for 24 hours at 37°C, 5% CO 2 . Twenty microliters of supernatant was then transferred to a white 96-well plate with 50 μl of reconstituted QUANTI-Luc secreted luciferase and read immediately on a Victor 3 luminometer with 1s integration time. The RLU of a no antibody control was subtracted as background. Palivizumab was used as a negative control, while CR3022 was used as a positive control, and P2B-2F6 to differentiate the Beta from the D614G variant. To induce the transgene, 1X cell stimulation cocktail (Thermofisher Scientific, Oslo, Norway) and 2 μg/mL ionomycin in R10 was added to confirm sufficient expression of the Fc receptor.

Statistical analysis
Analyses were performed in Prism (v9; GraphPad Software Inc, San Diego, CA, USA). Non-parametric tests were used for all comparisons. The Mann-Whitney test was used for unpaired comparisons between two groups, while the Kruskal-Wallis ANOVA with Dunns correction was used for multiple comparisons for unpaired groups. The Friedman test was used for multiple comparisons between paired groups. P values less than 0.05 were considered to be statistically significant. 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 November 9, 2021. ; https://doi.org/10.1101/2021.11.08.21266049 doi: medRxiv 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 November 9, 2021. ; https://doi.org/10.1101/2021.11.08.21266049 doi: medRxiv preprint