Decreased memory B cell frequencies in COVID‐19 delta variant vaccine breakthrough infection

Abstract The SARS‐CoV‐2 Delta (B.1.617.2) variant is capable of infecting vaccinated persons. An open question remains as to whether deficiencies in specific vaccine‐elicited immune responses result in susceptibility to vaccine breakthrough infection. We investigated 55 vaccine breakthrough infection cases (mostly Delta) in Singapore, comparing them against 86 vaccinated close contacts who did not contract infection. Vaccine breakthrough cases showed lower memory B cell frequencies against SARS‐CoV‐2 receptor‐binding domain (RBD). Compared to plasma antibodies, antibodies secreted by memory B cells retained a higher fraction of neutralizing properties against the Delta variant. Inflammatory cytokines including IL‐1β and TNF were lower in vaccine breakthrough infections than primary infection of similar disease severity, underscoring the usefulness of vaccination in preventing inflammation. This report highlights the importance of memory B cells against vaccine breakthrough and suggests that lower memory B cell levels may be a correlate of risk for Delta vaccine breakthrough infection.

In their manuscript, Zirui Tay et al study antibodies and memory B cells in individuals after vaccination with breakthrough infections of the SARS-CoV-2 delta variant. The topic is extremely timely and interesting as pointe out by the authors, because these infections do occur and it is currently unclear why. The research question and the technical approached are very clear and appropriate. However, my main concern is the study design. The authors have included patients with a breakthrough infection within 7 days of onset of symptoms. Hence, these are patient with an active infection. These patients are compared to uninfected housemates. Same vaccination status, but not currently experiencing an infection. Thus, the authors compare immune characteristics between infected and uninfected individuals. As a result, any differences are most likely the result of the infection status rather than anything else. I do understand why this was done, but it makes all the outcomes unclear. In fact, it would make sense if there were fewer memory B cells in infected individuals as these should all have been activated and included in the response to infection. Ideally, this study should be performed in prospectively collected samples from vaccinated individuals before any breakthrough infection and then retrospectively stratified on the basis of a subsequent infection. Then it can be addressed whether pre-existing differences underlie the risk of infection. Because of the current fault in experimental design, the results do not support the conclusions.
Referee #2 (Comments on Novelty/Model System for Author): Based on the data provided, I have serious concerns about the technical quality of the flow cytometry data and its analysis and interpretation. The supplementary figures show improper compensation of even easily distinguished surface markers, and the background subtracted stimulation data suggests that at least some samples exhibited incredibly high levels of background cytokine production (>10%) that point either to issues with the samples or the gating of the flow data.
Referee #2 (Remarks for Author): Tay et al characterise B cell, T cell and cytokine profiles of a large cohort of COVID-19 vaccine breakthrough infections and close contact controls. The cohort studied is an excellent resource and represents a large number of genuine breakthrough infections with reasonably well-matched controls. Interestingly, the authors find no differences in antibody titres between breakthrough and uninfected vaccinees, suggesting that breakthrough infection did not occur as a result of low neutralising antibody titres after vaccination.
Unfortunately, I feel the authors have missed an opportunity to incisively characterise both pre-existing and de novo T cell responses to various SARS-CoV-2 antigens in this cohort. I have serious concerns regarding the T cell flow cytometry data and its analysis. Appendix Figure S3 shows a partial gating strategy for the identification of CD4 and CD8 T cells, with apparent incorrect unmixing/compensation between CD4 and CD8. The lack of accurate unmixing and lack of representative plots showing any further T cell phenotyping or cytokine production makes it impossible to further assess the flow staining.
Furthermore, the data presented in Figure 3B are very confusing. That a median of 5% of CD8 Tcm cells would produce IL-6 in response to S peptide stimulation several months after vaccination is completely at odds with other reports of the frequency of spike-specific CD8 T cells elicited by vaccination. The presence of some data points at -10% on the graph also indicate that in several samples, the background production of IL-6 in the sample was 10% greater than the stimulated condition, which again points to either issues with the samples (such high spontaneous IL-6 production by T cells in the absence of stimulation is highly unusual) or with the gating and subsequent analysis. It is absolutely critical the authors provide FACS plots to demonstrate the cytokine staining in both stimulated and unstimulated breakthrough and control samples. For reference, what were the frequencies of IFNg+ cells in these samples? Were any positive control (mitogen) stimulations performed?
My other concern lies with the cytokine data in Figure 4. The PCA clearly shows complete overlap of the breakthrough and uninfected control cases, suggesting that the cytokine patterns analysed are not related to ongoing infection. Indeed, the vaccine breakthrough subjects in panel B do not appear to be particularly distinct from the healthy control cases, which are not even shown in the PCA analysis.
Dear Dr Durdevic, Thank you for your editorial work for our manuscript. We note the decision and would like to request for an appeal regarding our study. We believe that the referees are seriously mistaken in their understanding of the study, and on many occasions did not assess all the data presented. This is very concerning as it appears that the manuscript was not correctly reviewed.
We provide the following as support to explain the misinterpretation of the data. We sincerely hope that you will considering giving us the opportunity to resubmit with a complete point-bypoint rebuttal and revised manuscript to further improve the clarity.

Referee #1
Referee #1's major concern is that the lowered memory B cell signal could have been due to activation of the memory B cells upon infection, causing them to differentiate into plasmablasts. However, we wish to explain that our study also measured the plasmablast response in these patients, and our results show that this is not the case (these data were presented in the submitted manuscript).
If the memory B cells were "missing" because they had turned into plasmablasts, we should see higher levels of plasmablasts in the patients with low memory B cells, i.e. the two measure should be anti-correlated. However, we instead observe that memory B cells were positively correlated with plasmablast levels (Fig EV2b).
Thus, while we are in full agreement that we are here comparing infected individuals with uninfected individuals. We disagree that our main conclusions, including the surprising finding for memory B cells, are affected by this.
Notably, Referee #2 is in agreement with us here, noting that "the cohort studied in an excellent resource and represents a large number of genuine breakthrough infections with reasonably well-matched controls".

Referee #2
Referee #2's main concern was regarding the technical approach on the T cell analysis.
We wish to explain that a spectral flow cytometer was used here and standard automated unmixing analysis utilized. We are happy to provide representative plots showing further example flow cytometry plots, attached here in the Annex (Annex -Flow Cytometry Back-Gating).   Fig. 3, noting that the percentages observed by us were higher than expected.
This concern is primarily due to a difference in interpretation for Fig. 3. In Fig. 3, the Phenographs were assigned by automatic clustering, not traditional gating methods. Labels were only given afterwards to describe the cluster phenotype based on the MFI for each marker. This differs from typical gating strategies, where only a small highly distinct population is labelled positive. This was necessary to provide a qualitative description of the Phenograph clusters. This accounts for the larger percentages that we observe. To help clarify the populations in view, we have here attached flow cytometry plots for with gating for each cluster, based on the Phenograph cluster's MFI in the highlighted parameters (Annex -Flow Cytometry Back-Gating).
However, it should be noted that these gatings are approximate only, since it is impossible to completely replicate the automated multiparameter gating strategy decided by the Phenograph clustering algorithm. For each cluster, we have presented representative flow cytometry plots from one vaccine breakthrough and one close contact case respectively to showcase the differences observed.
Because the same number of cells were taken for each patient, the large negative percentages in Fig. 3c may not be due to an absolute decrease in that population, but a relative decrease due to increase in other populations. Indeed, the lowest datapoint in Fig. 3c (-12.3%) corresponds to the highest datapoint in Fig. 3b (+15.3%), and the second-lowest datapoint in Fig 3c (-4.7%) corresponds to the third highest datapoint in Fig. 3b (+7.8%). To further clarify this, we have also now modified the source data to show both the percentages of cells in each Phenograph cluster for stimulated and unstimulated conditions separately, attached in the updated Source Data for Figure 3.
Positive mitogen controls were performed for all samples (stimulating with phorbol myristate acetate (PMA)), but were not presented in the manuscript since we did not think that they were physiologically relevant. We have now added a representative set of plots showing PMA stimulation, attached in the Annex (Annex -Flow Cytometry Back-Gating).
3. Referee #2 had concerns with the cytokine data in Fig. 4, noting that "the vaccine breakthrough subjects do not appear to be particularly distinct from the healthy control cases".
We believe that the reviewer misunderstood the figure.
Actually, this is precisely our point here: vaccination is protective to such a degree that even despite vaccine breakthrough, the inflammatory cytokine profile of these individuals remains largely similar to healthy, uninfected individuals.
This was true across the comprehensive array of cytokines that were analysed in this dataset (See Figure 4 Source Data for full list).
Notably, Referee #1 is in agreement with us here, since he was pleased with the technical approach, commenting that "the technical approaches are very clear and appropriate".
We hope that these clarifications will set right the reviewers' misunderstandings and provide further substantiation for our study and its high-impact conclusions. Thank you for your response to the editorial decision on your manuscript entitled "Decreased memory B cells frequencies in COVID-19 Delta variant vaccine breakthrough infection". I have now carefully examined the arguments provided in your letter and discussed them with the other members of our editorial team. Additionally, I have sought external advice on the study from an expert in the field.
I am pleased to inform you that we decided to re-consider our initial decision and to invite major revision of your manuscript. Please provide detailed responses to the referee concerns and appropriately amend the manuscript to strengthen main message of the study.
Further consideration of a revision that addresses reviewers' concerns in full will entail a second round of review. EMBO Molecular Medicine encourages a single round of revision only and therefore, acceptance or rejection of the manuscript will depend on the completeness of your responses included in the next, final version of the manuscript. For this reason, and to save you from any frustrations in the end, I would strongly advise against returning an incomplete revision.
We would welcome the submission of a revised version within three months for further consideration. Please let us know if you require longer to complete the revision.
I look forward to receiving your revised manuscript.Please use this link to login to the manuscript system and submit your revision: https://embomolmed.msubmit.net/cgi-bin/main.plex The authors compare patients with active COVID19 with uninfected controls. Thus, the difference in memory B cells is most likely the result of the ongoing infection, rather than the underlying cause of infection (these memory B cells are low, because they have been activated and contribute to the response, likely as plasmablast in tissue). Thus, the experimental design is not appropriate and the conclusions are not supported by the outcomes of this study.

Ideally, this study should be performed in prospectively collected samples from vaccinated individuals before any breakthrough infection and then retrospectively stratified on the basis of a subsequent infection. Then it can be addressed whether pre-existing differences underlie the risk of infection.
Because of the current fault in experimental design, the results do not support the conclusions.

 "In their manuscript, Zirui Tay et al study antibodies and memory B cells in individuals after vaccination with breakthrough infections of the SARS-CoV-2 delta variant. The topic is extremely timely and interesting as pointe out by the authors, because these infections do occur and it is currently unclear why."
Response: We thank Referee #1 for his/her favourable evaluation of the importance of the paper topic.
 "The research question and the technical approached are very clear and appropriate."

Response:
We thank Referee #1 for his/her favourable evaluation of our methodological approaches and their clarity.
 "… the authors compare immune characteristics between infected and uninfected individuals." 23rd Nov 2021 1st Authors' Response to Reviewers

"Thus, the difference in memory B cells is most likely the result of the ongoing infection, rather than the underlying cause of infection (these memory B cells are low, because they have been activated and contribute to the response, likely as plasmablast in tissue)."
Response: We thank Referee #1 for the critical examination of our paper. However, we respectfully disagree with Referee #1's conclusion that the difference in memory B cells is the result of the ongoing infection, and that the lowered memory B cell signal could have been due to activation of the memory B cells upon infection, causing them to differentiate into plasmablasts. Notably, we measured the plasmablast response in these patients, and our results show that this was not the case. If the memory B cells were "missing" because they had turned into plasmablasts, we should observe higher levels of plasmablasts in the patients with low memory B cells, i.e. the two measures should be anti-correlated. However, we instead observe that memory B cells were positively correlated with plasmablast levels (Fig EV2b). Thus, while we are in full agreement that we are here comparing infected individuals with uninfected individuals, we disagree that our main conclusions, including the surprising finding for memory B cells, are affected by this. We have now inserted text into the manuscript (lines 168-171) to clarify this point.

Response:
We agree that a prospective study would provide important confirmation of the findings here. However, such a study is much more logistically challenging, especially at the scale of the number of patients analysed here. Perhaps due to this reason, previous prospective studies have focused on plasma antibody responses since plasma is much easier to process and store as compared to PBMCs. Our study is thus uniquely positioned in evaluating B and T cell responses in addition to plasma antibody responses in the context of vaccine breakthrough infection.  Figure S3 shows a partial gating strategy for the identification of CD4 and CD8 T cells, with apparent incorrect unmixing/compensation between CD4 and CD8. The lack of accurate unmixing and lack of representative plots showing any further T cell phenotyping or cytokine production makes it impossible to further assess the flow staining.
Furthermore, the data presented in Figure 3B  My other concern lies with the cytokine data in Figure 4. The PCA clearly shows complete overlap of the breakthrough and uninfected control cases, suggesting that the cytokine patterns analysed are not related to ongoing infection. Indeed, the vaccine breakthrough subjects in panel B do not appear to be particularly distinct from the healthy control cases, which are not even shown in the PCA analysis. Response: We thank Referee #2 for the favourable evaluation of the usefulness and quality of our cohort study.
 Unfortunately, I feel the authors have missed an opportunity to incisively characterise both pre-existing and de novo T cell responses to various SARS-CoV-2 antigens in this cohort.
Response: Unfortunately, we did not examine the differences in T cell responses to the different SARS-CoV-2 antigens in this cohort, since a pooled set of SARS-CoV-2 peptides (S, S1, M and N peptides) was used rather than separate antigens. In the antibody response, very few vaccine breakthrough cases (2/55) showed (presumably de novo) responses to the N protein, and it would be interesting to see if this was similar in the T cell compartment. Unfortunately, we are unable to perform this experiment since the PBMC samples from these patients are no longer available.
 I have serious concerns regarding the T cell flow cytometry data and its analysis. Appendix Figure S3 shows a partial gating strategy for the identification of CD4 and CD8 T cells, with apparent incorrect unmixing/compensation between CD4 and CD8. The lack of accurate unmixing and lack of representative plots showing any further T cell phenotyping or cytokine production makes it impossible to further assess the flow staining.

Response:
We thank the reviewer for highlighting this where in the original Appendix Fig  S3, the CD8 was undercompensated relative to CD4. We had initially accepted this since the CD4+ and CD8+ T cell populations were distinct regardless. However, we took the comments serious and we have gone over our compensations and adjusted them accordingly. We have also now revised Appendix Fig S3, such that Appendix S3a shows the full gating strategy for all T cell phenotyping, and Appendix Fig S3b shows cytokine production in unstimulated, SARS-CoV-2 peptide-stimulated, and positive control PMAstimulated conditions.

Appendix Figure S3a. Gating strategy for CD4+ and CD8+ T cells and intracellular cytokine staining.
Representative flow cytometry diagrams shown for gating CD4+ and CD8+ T cells, as well as CD27/CD45RA gating for CD4+ and CD8+ T cell differentiation status.
 It is absolutely critical the authors provide FACS plots to demonstrate the cytokine staining in both stimulated and unstimulated breakthrough and control samples.

Response:
We thank the reviewer for this suggestion and we have now included the flow cytometry diagrams for the cytokine staining for unstimulated, peptide-stimulated, and PMAstimulated conditions in Appendix Fig S3b.

Representative flow cytometry diagrams shown for gating CD4+ and CD8+ T cells for intracellular cytokine staining, example flow cytometry diagrams from unstimulated (top row), SARS-CoV-2 peptide-stimulated (middle row) and PMA-stimulated (bottom row) conditions are shown.
 Furthermore, the data presented in Figure 3B are very confusing. That a median of 5% of CD8 Tcm cells would produce IL-6 in response to S peptide stimulation several months after vaccination is completely at odds with other reports of the frequency of spike-specific CD8 T cells elicited by vaccination.
Response: We note this comment and wish to explain that this confusion was caused by the original T cell analysis. The original approach assigned clusters that were labelled based on their relative fluorescence intensity, leading to much less stringent gating cut-offs than typically used for assigning positivity. This led to elevated subpopulation frequencies. Upon further consideration, we have decided to remove the dimensionality reduction and clustering approach for the T cell analysis, and have opted for a traditional gating, highlighting common cytokine/effector molecules and identifying polyfunctionality, since these are common T We sincerely believe that with the modified figures, this now removes unnecessary complexity and better streamlines the paper.    My other concern lies with the cytokine data in Figure 4. The PCA clearly shows complete overlap of the breakthrough and uninfected control cases, suggesting that the cytokine patterns analysed are not related to ongoing infection. Indeed, the vaccine breakthrough subjects in panel B do not appear to be particularly distinct from the healthy control cases, which are not even shown in the PCA analysis.

Response:
We note Referee #2's surprise at the results, and indeed, the similarity between vaccine breakthrough subjects and uninfected controls is precisely the interesting point here: vaccination appears to be protective to such a degree that even despite vaccine breakthrough, the inflammatory cytokine profile of these individuals remains largely similar to uninfected individuals. This was the result from analysis of a comprehensive panel of 45 cytokines were included for the PCA, common infection and inflammation-related cytokines such as IL-1β, TNF, IFNγ, MIP-1α, RANTES, and CXCL10 (See Figure 4 Source Data for full list). We have now also included the unvaccinated healthy controls in the PCA as well.
Interestingly, the unvaccinated healthy controls cluster away from both the vaccinated group and primary infection group. The differences from the vaccinated group may be due to effects of the vaccine on innate immunity in addition to adaptive immunity, as has been reported by other groups (Arunachalam et al, 2021; Föhse et al, 2021). Nevertheless, we chose not to emphasize this in the paper as the comparison of vaccinated vs healthy unvaccinated groups is not the primary focus of the paper. Thank you for the submission of your revised manuscript to EMBO Molecular Medicine. We have now heard back from a referee who agreed to re-evaluate your manuscript. In addition, I have also sought external advice on the study from an expert in the field. As you will see from the report below, the referee #2 appreciates your efforts in addressing the referees' comments but also raises considerable concerns.

References
As mentioned above, I have also sought further advice on the study, and our advisor reached a conclusion that "This paper ... is solid and the technologies used is up to date. I would consider its publication following last round of reviews." Therefore, I am pleased to inform you that we will be able to accept your manuscript pending the following final amendments: 1) Please address all the points raised by the referee #2 in writing as limitations of the study. 2) In the main manuscript file, please do the following: -Correct/answer the track changes suggested by our data editors by working from the attached document.
-Make sure that all special characters display well.
-Add author contributions for Siti Naqiah Amrun and Bei Wang. -Place "Data availability" at the end of M&M section. 3) Synopsis: Please check your synopsis text and image before submission and submit their final versions with your revised manuscript. Please be aware that in the proof stage minor corrections only are allowed (e.g., typos). 4) For more information: There is space at the end of each article to list relevant web links for further consultation by our readers. Could you identify some relevant ones and provide such information as well? Some examples are patient associations, relevant databases, OMIM/proteins/genes links, author's websites, etc... 5) Press release: Please inform us as soon as possible and latest at the time of submission of the revised manuscript if you plan a press release for your article so that our publisher could coordinate publication accordingly. 6) Please be aware that we use a unique publishing workflow for COVID-19 papers: a non-typeset PDF of the accepted manuscript is published as "Just Accepted" on our website. With respect to a possible press release, we have the option to not post the "Just Accepted" version if you prefer to wait with the press release for the typeset version. Please let us know whether you agree to publication of a "Just accepted" version or you prefer to wait for the typeset version. 7) As part of the EMBO Publications transparent editorial process initiative (see our Editorial at http://embomolmed.embopress.org/content/2/9/329), EMBO Molecular Medicine will publish online a Review Process File (RPF) to accompany accepted manuscripts. This file will be published in conjunction with your paper and will include the anonymous referee reports, your point-by-point response and all pertinent correspondence relating to the manuscript. Let us know whether you agree with the publication of the RPF and as here, if you want to remove or not any figures from it prior to publication. Please note that the Authors checklist will be published at the end of the RPF. The authors have provided a revised manuscript that addresses some of the concerns of the original reviewers. In particular, the changes made to the T cell analysis in light of the issues with the original gating are appreciated, and serve to provide increased confidence in the T cell data.
With regards to the concerns raised about the B cell analysis, I am unconvinced by aspects of the revised manuscript. The authors do clearly show that the elevated plasmablast response among breakthrough cases supports the presence of active infection, and implies that reactivation of the memory B cell response has already occurred at the timepoints sampled. Under those circumstances, it is therefore confusing to conclude that differences in RBD-specific MBC reflect the immunological state prior to infection, and contribute to susceptibility to breakthrough infection. It seems equally plausible that MBC have been recruited out of the circulation due to the presence of antigen in tissues or lymph nodes.
I would note in particular that the correlation between plasmablast frequency and RBD-specific MBC frequency in EV2 is quite unconvincing -it is described in the text on line 171 as a "trend toward positive correlation", but in a sample size of n=7 with one strongly positive outlying data point, this is an overstatement. Given the number of breakthrough cases available, this analysis seems quite underpowered.

Revised Submission Referee #2
Referee #2 (Remarks for Author): The authors have provided a revised manuscript that addresses some of the concerns of the original reviewers. In particular, the changes made to the T cell analysis in light of the issues with the original gating are appreciated, and serve to provide increased confidence in the T cell data.
With regards to the concerns raised about the B cell analysis, I am unconvinced by aspects of the revised manuscript. The authors do clearly show that the elevated plasmablast response among breakthrough cases supports the presence of active infection, and implies that reactivation of the memory B cell response has already occurred at the timepoints sampled. Under those circumstances, it is therefore confusing to conclude that differences in RBD-specific MBC reflect the immunological state prior to infection, and contribute to susceptibility to breakthrough infection. It seems equally plausible that MBC have been recruited out of the circulation due to the presence of antigen in tissues or lymph nodes.
I would note in particular that the correlation between plasmablast frequency and RBDspecific MBC frequency in EV2 is quite unconvincing -it is described in the text on line 171 as a "trend toward positive correlation", but in a sample size of n=7 with one strongly positive outlying data point, this is an overstatement. Given the number of breakthrough cases available, this analysis seems quite underpowered.

Response:
We agree that the current study design does not allow us to completely rule out the possibility that MBC have been recruited out of the circulationthis would require either a prospective study design or extensive tissue sampling, both of which are out of the scope of the current study. We have previously addressed this limitation in the discussion (lines 262-266). We have now revised the manuscript to further emphasize this limitation and the possible alternative interpretation by adding the following in line 266-268: "In particular, 23rd Dec 2021 2nd Authors' Response to Reviewers reduced memory B cell levels in vaccine breakthrough cases may have been due to recruitment of memory B cells out of circulation after activation by infection."  I would note in particular that the correlation between plasmablast frequency and RBD-specific MBC frequency in EV2 is quite unconvincing -it is described in the text on line 171 as a "trend toward positive correlation", but in a sample size of n=7 with one strongly positive outlying data point, this is an overstatement. Given the number of breakthrough cases available, this analysis seems quite underpowered.

Response:
We were unfortunately unable to get paired plasmablast-memory B cell data for a number of the patients due to sample limitations as well as some technical difficulties. We agree that the current graph is strongly influenced by a strong positive data point; however, even if this point is removed, the remaining data still do not show a negative correlation between plasmablasts and memory B cells (see Figure EV3 with strong positive point removed; right). The lack of negative correlation is one piece of evidence against the alternate theory that MBCs are being lost via differentiation into plasmablasts. In light of this, we have now deleted the sentences regarding the trend toward positive correlation and its interpretation, and we have chosen to focus on the lack of inverse correlation instead (lines 166-169). Figure EV3 with strong positive point removed. common  3. Were any steps taken to minimize the effects of subjective bias when allocating animals/samples to treatment (e.g. randomization procedure)? If yes, please describe.
For animal studies, include a statement about randomization even if no randomization was used.
4.a. Were any steps taken to minimize the effects of subjective bias during group allocation or/and when assessing results (e.g. blinding of the investigator)? If yes please describe. Do the data meet the assumptions of the tests (e.g., normal distribution)? Describe any methods used to assess it.
Is there an estimate of variation within each group of data?
Is the variance similar between the groups that are being statistically compared?

Reporting Checklist For Life Sciences Articles (Rev. June 2017)
This checklist is used to ensure good reporting standards and to improve the reproducibility of published results. These guidelines are consistent with the Principles and Guidelines for Reporting Preclinical Research issued by the NIH in 2014. Please follow the journal's authorship guidelines in preparing your manuscript.

Journal Submitted to: EMBO Molecular Medicine
Corresponding Author Name: Lisa F. P. Ng

B-Statistics and general methods
the assay(s) and method(s) used to carry out the reported observations and measurements an explicit mention of the biological and chemical entity(ies) that are being measured. an explicit mention of the biological and chemical entity(ies) that are altered/varied/perturbed in a controlled manner. a statement of how many times the experiment shown was independently replicated in the laboratory.
Any descriptions too long for the figure legend should be included in the methods section and/or with the source data.
In the pink boxes below, please ensure that the answers to the following questions are reported in the manuscript itself. Every question should be answered. If the question is not relevant to your research, please write NA (non applicable). We encourage you to include a specific subsection in the methods section for statistics, reagents, animal models and human subjects.

definitions of statistical methods and measures:
a description of the sample collection allowing the reader to understand whether the samples represent technical or biological replicates (including how many animals, litters, cultures, etc.).

The data shown in figures should satisfy the following conditions:
Source Data should be included to report the data underlying graphs. Please follow the guidelines set out in the author ship guidelines on Data Presentation.
Please fill out these boxes ê (Do not worry if you cannot see all your text once you press return) a specification of the experimental system investigated (eg cell line, species name).
No statistical methods were used to calculate sample size. Sample size for vaccine breakthrough patients and close contacts was determined based on the number of patients that were enrolled and consented with the current study between Apr-Sep 2021.
graphs include clearly labeled error bars for independent experiments and sample sizes. Unless justified, error bars should not be shown for technical replicates. if n< 5, the individual data points from each experiment should be plotted and any statistical test employed should be justified the exact sample size (n) for each experimental group/condition, given as a number, not a range; Each figure caption should contain the following information, for each panel where they are relevant:

Captions
NA -no animal studies Some samples were excluded from some assays due to lack of sample availability, or assay failure. Criteria for determining assay failure were pre-established.
No animal studies. For patient samples, no randomization was required, since all available samples from eligible enrolled participants were used.
Non-parametric tests were used in all cases.
Yes -median +-interquartile range are given for all graphs.
NA -non-parametric tests used.

NA -no animal studies
Investigators were not blinded to group allocation during data collection and analysis, because samples from vaccine breakthrough patients had to be handled with additional safety precautions.
NA -no animal studies 1. Data the data were obtained and processed according to the field's best practice and are presented to reflect the results of the experiments in an accurate and unbiased manner. figure panels include only data points, measurements or observations that can be compared to each other in a scientifically meaningful way.