Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.

Gut microbiota: a new factor modulating the immunizing potential of viral and cancer vaccines

Open access

Abstract

Vaccines represent a major public health intervention against infectious diseases and potentially cancer. Surrogate markers of vaccine efficacy usually rely on neutralizing antibody titers afflicted by high interindividual variabilities. Automated multiplexed T cell assays currently allow to test the clinical relevance of T lymphocyte responses during vaccine rollout. We examined cellular and/or humoral immune responses in five independent cohorts of health care workers, young healthy individuals and patients with cancer (melanoma or lung cancer) receiving various immunizing formulations (non-replicating, viral/tumoral, mRNA/peptides/cellular/viral particles). Here we show that about 20% of vaccinees to non-replicating formulations fail to mount protective antibody and Th1/Tc1 responses while 9% receiving a live vaccine were hyperresponders. Vaccine outliers could at least in part be attributed to gut dysbiosis at baseline, evaluated by shotgun metagenomics-based machine learning or the TOPOSCORE. These findings highlight the requirement of diagnostic tools to identify intestinal dysbiosis, as well as microbiota-centered interventions to optimize the efficiency of mass vaccinations.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.21203/rs.3.rs-4294379/v1

Cites

0 citation errors · 0 not checkable

Statistics

Figures

Figure 1. Failure of COVID-19 vaccines to elicit RBD-S1-specific Th1/Tc1 immune responses. A. Graphical representation of the WELCOME patient cohort’s protocol with fecal and blood collection and timelines between vaccination and sampling (refer to Table S3). B. Bicolor map of cytokine secretion after 72h PBMC stimulation with PEPwtRBD peptide pool (positive in black, negative in grey). Patients (n=75) were ordered in columns and cytokines were ordered in rows by unsupervised hierarchical clustering using ward.D2 clustering method and binary distance calculation method. Cytokines with less than 10% frequency were excluded of this clustering. Heatmap was cut in 2 clusters of patients (orange: non-reactive (cluster 0), light blue: polyreactive Th1/Tc1 (cluster 1)). C. Volcano plot depicting the most significant cytokines after PBMC stimulation with the PEPwtRBD peptide pool between cluster 1 (light blue) and cluster 0 (orange) of panel B. Volcano plot was generated computing delta of the percentages of positive patients in cluster 1 and the percentages of positive patients in cluster 0 for each cytokine (x axis) with the −log10 of the p-values deriving from Fisher’s exact test for each cytokine (y axis). D. Barplots representing the percentage and number of positive patients (black) in each cluster of the panel B (cluster 1, n=19 and cluster 0, n=56) for prototypic cytokines GM-CSF, IFNγ, IL-13 and CXCL10. Fisher’s exact test was used to compare the number of cytokine-positive (black) versus negative (grey) patients across clusters. E. Barplots representing the percentage and number of patients that will experience a breakthrough COVID-19 infection in the future in each cluster of the Fig. 1B (cluster 0, n=19 and cluster 1, n=56). Fisher’s exact test was used to compare the number of breakthroughs COVID-19 infection-positive (red) versus negative (green) patients across clusters.
Figure 1. Failure of COVID-19 vaccines to elicit RBD-S1-specific Th1/Tc1 immune responses. A. Graphical representation of the WELCOME patient cohort’s protocol with fecal and blood collection and timelines between vaccination and sampling (refer to Table S3). B. Bicolor map of cytokine secretion after 72h PBMC stimulation with PEPwtRBD peptide pool (positive in black, negative in grey). Patients (n=75) were ordered in columns and cytokines were ordered in rows by unsupervised hierarchical clustering using ward.D2 clustering method and binary distance calculation method. Cytokines with less than 10% frequency were excluded of this clustering. Heatmap was cut in 2 clusters of patients (orange: non-reactive (cluster 0), light blue: polyreactive Th1/Tc1 (cluster 1)). C. Volcano plot depicting the most significant cytokines after PBMC stimulation with the PEPwtRBD peptide pool between cluster 1 (light blue) and cluster 0 (orange) of panel B. Volcano plot was generated computing delta of the percentages of positive patients in cluster 1 and the percentages of positive patients in cluster 0 for each cytokine (x axis) with the −log10 of the p-values deriving from Fisher’s exact test for each cytokine (y axis). D. Barplots representing the percentage and number of positive patients (black) in each cluster of the panel B (cluster 1, n=19 and cluster 0, n=56) for prototypic cytokines GM-CSF, IFNγ, IL-13 and CXCL10. Fisher’s exact test was used to compare the number of cytokine-positive (black) versus negative (grey) patients across clusters. E. Barplots representing the percentage and number of patients that will experience a breakthrough COVID-19 infection in the future in each cluster of the Fig. 1B (cluster 0, n=19 and cluster 1, n=56). Fisher’s exact test was used to compare the number of breakthroughs COVID-19 infection-positive (red) versus negative (green) patients across clusters.CC BY · PubMed Central
Figure 2. Association between gut taxonomic composition and non-reactivity to COVID-19 Health Care Workers (HCW) vaccines in the WELCOME study. A. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the WELCOME cohort microbiome samples (n=69) colored according to the immunoreactivity (orange for cluster 0 or blue for cluster 1). Clusters are defined as PEPwtRBD-specific Th1 reactivity in Fig. 1B. P-values represent Student-t tests. B. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarity matrix, colored by immune reaction cluster (orange for cluster 0 or blue for cluster 1 defined in Fig. 1B). P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. C. Intra- vs inter-individual Bray-Curtis pairwise dissimilarities including 69 baseline samples and 22 followed up-samples from the WELCOME cohort. Inter-individual pairs are colored according to the cluster of both samples (orange for cluster 0 or blue for cluster 1 defined in Fig. 1B) or grey for those pairs derived from two different clusters. D. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (orange for cluster 0 or blue for cluster 1, as defined with PEPwtRBD in Fig. 1B). E. Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a RF algorithm aiming at distinguishing cluster 1 (reactive) from cluster 0 (non-reactive) as defined with PEPwtRBD in Fig. 1B using all microbiome SGB relative abundances as features, and using the 40 most predictive species as derived by the algorithm internal feature ranking when taking into consideration training folds in order to avoid overfitting. F. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial SGB and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (cluster of association, as defined with PEPOrf in Fig. S2C). G. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster of the Fig. 1B (cluster 0, n=19 and cluster 1, n=56). Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.
Figure 2. Association between gut taxonomic composition and non-reactivity to COVID-19 Health Care Workers (HCW) vaccines in the WELCOME study. A. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the WELCOME cohort microbiome samples (n=69) colored according to the immunoreactivity (orange for cluster 0 or blue for cluster 1). Clusters are defined as PEPwtRBD-specific Th1 reactivity in Fig. 1B. P-values represent Student-t tests. B. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarity matrix, colored by immune reaction cluster (orange for cluster 0 or blue for cluster 1 defined in Fig. 1B). P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. C. Intra- vs inter-individual Bray-Curtis pairwise dissimilarities including 69 baseline samples and 22 followed up-samples from the WELCOME cohort. Inter-individual pairs are colored according to the cluster of both samples (orange for cluster 0 or blue for cluster 1 defined in Fig. 1B) or grey for those pairs derived from two different clusters. D. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (orange for cluster 0 or blue for cluster 1, as defined with PEPwtRBD in Fig. 1B). E. Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a RF algorithm aiming at distinguishing cluster 1 (reactive) from cluster 0 (non-reactive) as defined with PEPwtRBD in Fig. 1B using all microbiome SGB relative abundances as features, and using the 40 most predictive species as derived by the algorithm internal feature ranking when taking into consideration training folds in order to avoid overfitting. F. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial SGB and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (cluster of association, as defined with PEPOrf in Fig. S2C). G. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster of the Fig. 1B (cluster 0, n=19 and cluster 1, n=56). Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.CC BY · PubMed Central
Figure 3. Association between gut taxonomic composition and immunoreactivity to adjuvant treatment with dendritic cell (DC) therapy in patients with melanoma. A. Graphical representation of the MIND-DC cohort protocol with fecal and skin biopsy collection and timelines between vaccination and sampling (refer to Table S3). B. Bicolor map of cytokine secretion after delayed-type hypersensitivity (DTH) skin test (positive in black, negative in grey). Patients (n=55) were ordered in columns and cytokines were ordered in rows by unsupervised hierarchical clustering using ward.D2 clustering method and binary distance calculation method. Heatmap was cut in 2 clusters of patients (orange: non-reactive (cluster 0), blue: polyreactive (cluster 1)). C. Barplots representing the percentage and number of patients that are IL-10- and/or IL-2- and/or IFNγ- and/or IL-5-reactive/areactive in DTH responses in both autologous natural DC-vaccinated and placebo (PL) groups (orange: cluster 0, blue: cluster 1 as defined in Fig. 3B). Fisher’s exact test was used to compare the number of cytokinepositive patients across groups. D. Barplots representing the percentages and number of patients from cluster 0 (orange) and 1 (blue) as defined in Fig. 3B in patients with or without tumor relapse at 2 years. Fisher’s exact test was used to compare percentages in both clusters. E. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the MIND-DC cohort microbiome samples colored by immune reaction cluster (orange for cluster 0 or blue for cluster 1). Clusters are defined in as defined in Fig. 3B (cluster 0 versus 1). P-values are from Student-t tests. F. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarity matrix, colored by immune reaction (cluster 0 vs 1 defined in Fig. 3B). P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. G. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (orange for cluster 0 or blue for cluster 1, as defined in Fig. 3B). H. Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a RF algorithm aiming at distinguishing cluster 1 (reactive) from cluster 0 (non-reactive) from panel B using all microbiome SGB relative abundances as features, and using the 40 most predictive species as derived by the algorithm internal feature ranking when taking into consideration training folds in order to avoid overfitting. I. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster of the Fig. 3B (cluster 0, n=12 and cluster 1, n=43). Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.
Figure 3. Association between gut taxonomic composition and immunoreactivity to adjuvant treatment with dendritic cell (DC) therapy in patients with melanoma. A. Graphical representation of the MIND-DC cohort protocol with fecal and skin biopsy collection and timelines between vaccination and sampling (refer to Table S3). B. Bicolor map of cytokine secretion after delayed-type hypersensitivity (DTH) skin test (positive in black, negative in grey). Patients (n=55) were ordered in columns and cytokines were ordered in rows by unsupervised hierarchical clustering using ward.D2 clustering method and binary distance calculation method. Heatmap was cut in 2 clusters of patients (orange: non-reactive (cluster 0), blue: polyreactive (cluster 1)). C. Barplots representing the percentage and number of patients that are IL-10- and/or IL-2- and/or IFNγ- and/or IL-5-reactive/areactive in DTH responses in both autologous natural DC-vaccinated and placebo (PL) groups (orange: cluster 0, blue: cluster 1 as defined in Fig. 3B). Fisher’s exact test was used to compare the number of cytokinepositive patients across groups. D. Barplots representing the percentages and number of patients from cluster 0 (orange) and 1 (blue) as defined in Fig. 3B in patients with or without tumor relapse at 2 years. Fisher’s exact test was used to compare percentages in both clusters. E. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the MIND-DC cohort microbiome samples colored by immune reaction cluster (orange for cluster 0 or blue for cluster 1). Clusters are defined in as defined in Fig. 3B (cluster 0 versus 1). P-values are from Student-t tests. F. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarity matrix, colored by immune reaction (cluster 0 vs 1 defined in Fig. 3B). P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. G. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (orange for cluster 0 or blue for cluster 1, as defined in Fig. 3B). H. Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a RF algorithm aiming at distinguishing cluster 1 (reactive) from cluster 0 (non-reactive) from panel B using all microbiome SGB relative abundances as features, and using the 40 most predictive species as derived by the algorithm internal feature ranking when taking into consideration training folds in order to avoid overfitting. I. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster of the Fig. 3B (cluster 0, n=12 and cluster 1, n=43). Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.CC BY · PubMed Central
Figure 4. Association between gut taxonomic composition and immunoreactivity to UCPVax in non-small cell lung cancer (NSCLC) patients. A. Graphical representation of the UCPVax cohort protocol with fecal and blood collection and timelines between vaccination and sampling (refer to Table S3). TERT, telomerase reverse transcriptase. B. Boxplot representing the number of IFNγ anti-UCP spots (for 3.105 cells). Each point represents a patient. The middle line of the box represents the median and equals to 110 spots. The lower and upper hinges correspond to the first and third quartiles (the 25th and 75th percentiles). The upper whisker extends from the hinge to the largest value no further than 1.5 × IQR from the hinge (where IQR is the interquartile range, or distance between the first and third quartiles). The lower whisker extends from the hinge to the smallest value at most 1.5 × IQR of the hinge. Data beyond the end of the whiskers are called “outlying” points and are plotted individually. Colors are set depending on IFNγ-reactivity: orange stands for non-reactive patients (“cluster 0” if less than 110 spots) and blue stands for immunoreactive patients (“cluster 1” if more than 110 spots). Refer to Table S7. C. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the UCPVax cohort microbiome samples colored by immune reaction cluster (orange for cluster 0 or blue for cluster 1). Clusters are defined in Fig. 4B. P-values are from Student-t tests. D. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarity matrix, colored by immune reaction clusters defined in Fig. 4B (orange for cluster 0 or blue for cluster 1). P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. E. Intra- vs inter-individual Bray-Curtis pairwise dissimilarities considering 26 baseline samples as well as 10 followed up samples from the UCPVax cohort. Inter-individual pairs are colored according to the cluster of both samples (orange for cluster 0 or blue for cluster 1 defined in Fig. 4B) or grey for pairs derived from two different clusters. F. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (“cluster” of association, as defined in panel B). G Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a random forest (RF) algorithm aiming at distinguishing “cluster 1” (reactive) from “cluster 0” (non-reactive) using all microbiome SGB relative abundances as features, and using the 40 most predictive species as derived by the algorithm internal feature ranking when taking into consideration training folds in order to avoid overfitting. H. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster of the Fig. 4B (cluster 0, n=12 and cluster 1, n=43). Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.
Figure 4. Association between gut taxonomic composition and immunoreactivity to UCPVax in non-small cell lung cancer (NSCLC) patients. A. Graphical representation of the UCPVax cohort protocol with fecal and blood collection and timelines between vaccination and sampling (refer to Table S3). TERT, telomerase reverse transcriptase. B. Boxplot representing the number of IFNγ anti-UCP spots (for 3.105 cells). Each point represents a patient. The middle line of the box represents the median and equals to 110 spots. The lower and upper hinges correspond to the first and third quartiles (the 25th and 75th percentiles). The upper whisker extends from the hinge to the largest value no further than 1.5 × IQR from the hinge (where IQR is the interquartile range, or distance between the first and third quartiles). The lower whisker extends from the hinge to the smallest value at most 1.5 × IQR of the hinge. Data beyond the end of the whiskers are called “outlying” points and are plotted individually. Colors are set depending on IFNγ-reactivity: orange stands for non-reactive patients (“cluster 0” if less than 110 spots) and blue stands for immunoreactive patients (“cluster 1” if more than 110 spots). Refer to Table S7. C. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the UCPVax cohort microbiome samples colored by immune reaction cluster (orange for cluster 0 or blue for cluster 1). Clusters are defined in Fig. 4B. P-values are from Student-t tests. D. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarity matrix, colored by immune reaction clusters defined in Fig. 4B (orange for cluster 0 or blue for cluster 1). P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. E. Intra- vs inter-individual Bray-Curtis pairwise dissimilarities considering 26 baseline samples as well as 10 followed up samples from the UCPVax cohort. Inter-individual pairs are colored according to the cluster of both samples (orange for cluster 0 or blue for cluster 1 defined in Fig. 4B) or grey for pairs derived from two different clusters. F. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (“cluster” of association, as defined in panel B). G Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a random forest (RF) algorithm aiming at distinguishing “cluster 1” (reactive) from “cluster 0” (non-reactive) using all microbiome SGB relative abundances as features, and using the 40 most predictive species as derived by the algorithm internal feature ranking when taking into consideration training folds in order to avoid overfitting. H. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster of the Fig. 4B (cluster 0, n=12 and cluster 1, n=43). Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.CC BY · PubMed Central
Figure 5. Meta-analysis of the association between gut taxonomic composition and immunoreactivity to vaccines. A. Graphical representation of the meta-analysis encompassing three cohorts (WELCOME, MIND-DC and UCPVax, n=150). B. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the 3 cohorts’ microbiome samples colored by cohort (blue for WELCOME, green for MIND-DC and orange for UCPVax). P-values are from Student-t tests. C. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarities, colored by immune reaction clusters (orange: non-reactive (cluster 0), light blue: reactive (cluster 1)) of the pooled three cohorts; symbols indicate the three cohorts. P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. D. Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a random forest (RF) algorithm aiming at distinguishing cluster 1 (reactive) from cluster 0 (non-reactive) of the pooled three cohorts using all microbiome species-level genome bins (SGBs) relative abundances as features in each cohort independently while microbiome SGBs relative abundances of the other two cohorts are added to each training folds during the training phase. E. Condensed Forest plot on the twenty-two microbial SGBs’ showing the highest coefficient (absolute value) from a meta-analysis of log-gamma regression models and found at q<0.2 according to the same meta-analysis and in all three cohorts (sign of the coefficient corresponds to the cluster of association of the pooled three cohorts). F. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster (cluster 0 and cluster 1 as defined in Fig. 1B, 3B, 4B) in the pooled cohorts. Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster. G. Condensed Forest plot on the forty microbial enzymes showing the highest coefficient (absolute value) from a meta-analysis of negative binomial distributed regression models and found at q<0.2 according to the same meta-analysis and in all three cohorts (sign of the coefficient corresponds to the cluster of association of the pooled three cohorts).
Figure 5. Meta-analysis of the association between gut taxonomic composition and immunoreactivity to vaccines. A. Graphical representation of the meta-analysis encompassing three cohorts (WELCOME, MIND-DC and UCPVax, n=150). B. Microbial species richness, alpha-diversity (log-10-based Shannon entropy), and Inverse-Simpson of the 3 cohorts’ microbiome samples colored by cohort (blue for WELCOME, green for MIND-DC and orange for UCPVax). P-values are from Student-t tests. C. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarities, colored by immune reaction clusters (orange: non-reactive (cluster 0), light blue: reactive (cluster 1)) of the pooled three cohorts; symbols indicate the three cohorts. P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. D. Ten-times iterated, 10-fold cross-validations ROC curves and AUC of a random forest (RF) algorithm aiming at distinguishing cluster 1 (reactive) from cluster 0 (non-reactive) of the pooled three cohorts using all microbiome species-level genome bins (SGBs) relative abundances as features in each cohort independently while microbiome SGBs relative abundances of the other two cohorts are added to each training folds during the training phase. E. Condensed Forest plot on the twenty-two microbial SGBs’ showing the highest coefficient (absolute value) from a meta-analysis of log-gamma regression models and found at q<0.2 according to the same meta-analysis and in all three cohorts (sign of the coefficient corresponds to the cluster of association of the pooled three cohorts). F. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster (cluster 0 and cluster 1 as defined in Fig. 1B, 3B, 4B) in the pooled cohorts. Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster. G. Condensed Forest plot on the forty microbial enzymes showing the highest coefficient (absolute value) from a meta-analysis of negative binomial distributed regression models and found at q<0.2 according to the same meta-analysis and in all three cohorts (sign of the coefficient corresponds to the cluster of association of the pooled three cohorts).CC BY · PubMed Central
Figure 6. Gut microbiota fingerprint associated with defects in both humoral and cellular immune responses to COVID19 vaccine in the WELCOME cohort. A. RBD-S1 specific IgG titers were determined after the prime-boost COVID19 vaccine in 75 HCW from WELCOME at two timepoints between the prime boost and 200 days after (refer to Fig. 1A). The cohort was divided according to the median of the decay of antibody titers into two groups “low” (if lower to the median) versus “high” (above the median). B. Barplots representing the percentage and number of patients that will experience a breakthrough COVID-19 infection in the future in humoral immune responses’ groups of the Fig. 6A. P-value represent Chi2 test. C. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster (high and low as defined in Fig. 6A). Chi2 test was used to compare the number of SIG2+ (green) versus SIG1+ (black) patients in each humoral immune responses’ groups. D. Unsupervised hierarchical clustering of the all health care workers (HCW) according to the PEPRBDwt-specific Th1 cytokine release and RBD-S1-specific IgG titers, revealing two clusters, cluster 0 and cluster 1 for “low” and “high” combined reactivities, respectively. E. Upper panel represents microbial species alpha-diversity (log-10-based Shannon entropy) of the WELCOME cohort microbiome samples colored by cluster (orange for cluster 0 and blue for cluster 1). P-values are from Student-t tests. Lower panel represents first two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarities, colored by immune reaction clusters (orange: non-reactive (cluster 0), light blue: reactive (cluster 1)) of the WELCOME cohort. P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. F. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGBs) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (cluster of association, as defined in Fig. 1D).
Figure 6. Gut microbiota fingerprint associated with defects in both humoral and cellular immune responses to COVID19 vaccine in the WELCOME cohort. A. RBD-S1 specific IgG titers were determined after the prime-boost COVID19 vaccine in 75 HCW from WELCOME at two timepoints between the prime boost and 200 days after (refer to Fig. 1A). The cohort was divided according to the median of the decay of antibody titers into two groups “low” (if lower to the median) versus “high” (above the median). B. Barplots representing the percentage and number of patients that will experience a breakthrough COVID-19 infection in the future in humoral immune responses’ groups of the Fig. 6A. P-value represent Chi2 test. C. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1) in each cluster (high and low as defined in Fig. 6A). Chi2 test was used to compare the number of SIG2+ (green) versus SIG1+ (black) patients in each humoral immune responses’ groups. D. Unsupervised hierarchical clustering of the all health care workers (HCW) according to the PEPRBDwt-specific Th1 cytokine release and RBD-S1-specific IgG titers, revealing two clusters, cluster 0 and cluster 1 for “low” and “high” combined reactivities, respectively. E. Upper panel represents microbial species alpha-diversity (log-10-based Shannon entropy) of the WELCOME cohort microbiome samples colored by cluster (orange for cluster 0 and blue for cluster 1). P-values are from Student-t tests. Lower panel represents first two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarities, colored by immune reaction clusters (orange: non-reactive (cluster 0), light blue: reactive (cluster 1)) of the WELCOME cohort. P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. F. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGBs) and filtered for those having q<0.2 according to the log-gamma model, colored by sign of the coefficient (cluster of association, as defined in Fig. 1D).CC BY · PubMed Central
Figure 7. Elite responders to YF17D vaccine harbor an immunosuppressive intestinal microbiota. A. Graphical representation of the study comprising young healthy volunteers vaccinated with the live replicative YF17D vaccine (n=174). Serum, PBMC and stools were drawn before and 28 days after the immunization. Elite vaccine responders (n=15) were previously defined51, performing above average in all three categories neutralizing antibody titer and YFV17D specific CD4 and CD8 T cell responses. B. Microbial species alpha-diversity (log-10-based Shannon entropy) of the YF17D cohort microbiome samples colored by cluster (blue for elite and orange for others). P-values are from Student-t tests. C. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarities, colored by immune reaction groups (blue: elite; orange: others) of the YF17D cohort. P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. D. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the loggamma model, colored by sign of the coefficient (elite (n=15, blue) and non-elite (n=159, orange) vaccine responders as previously defined51. E. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1, left panel) or to the absolute numbers of prevalent SIG1 and SIG2 bacteria (right panel) in elite (n=15) and non-elite (n=159) vaccine responders as previously defined51. Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.
Figure 7. Elite responders to YF17D vaccine harbor an immunosuppressive intestinal microbiota. A. Graphical representation of the study comprising young healthy volunteers vaccinated with the live replicative YF17D vaccine (n=174). Serum, PBMC and stools were drawn before and 28 days after the immunization. Elite vaccine responders (n=15) were previously defined51, performing above average in all three categories neutralizing antibody titer and YFV17D specific CD4 and CD8 T cell responses. B. Microbial species alpha-diversity (log-10-based Shannon entropy) of the YF17D cohort microbiome samples colored by cluster (blue for elite and orange for others). P-values are from Student-t tests. C. First two principal components of a multi-dimensional scaling (MDS) based on Bray-Curtis pairwise dissimilarities, colored by immune reaction groups (blue: elite; orange: others) of the YF17D cohort. P-values are from PERmutational Multivariate ANalysis Of VAriance (PERMANOVA) test. D. Top-40 (absolute values) effect-sizes of a log-link gamma-distributed regression performed on each microbial species-level genome bins (SGB) and filtered for those having q<0.2 according to the loggamma model, colored by sign of the coefficient (elite (n=15, blue) and non-elite (n=159, orange) vaccine responders as previously defined51. E. Barplots representing the percentage and number of patients belonging to each TOPOSCORE category (SIG2, Grey and SIG1, left panel) or to the absolute numbers of prevalent SIG1 and SIG2 bacteria (right panel) in elite (n=15) and non-elite (n=159) vaccine responders as previously defined51. Fisher’s exact test was used to compare the number of SIG2 (green) versus Grey (grey) versus SIG1 (black) patients in each cluster.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

The publisher hasn’t deposited this paper’s reference list.

Research works printed but not linked: 22

  1. Carrier A. & Manghi P. & Alves Costa Silva C. & Lahmar I. et al. 6. Payton, T. et al. Immunogenicity, Safety and Tolerability of 3 Lots of 13-valent Pneumococcal Conjugate Vaccine Given With Routine Pediatric Vaccinations in the United States. Pediatr. Infect. Dis. J. 32, 871 (2013).
  2. Finan, C., Ota, M. O. C., Marchant, A. & Newport, M. J. Natural Variation in Immune Responses to Neonatal Mycobacterium bovis Bacillus Calmette-Guerin (BCG) Vaccination in a Cohort of Gambian Infants. PLOS ONE 3, e3485 (2008).
  3. Grassly, N. C., Kang, G. & Kampmann, B. Biological challenges to effective vaccines in the developing world. Philos. Trans. R. Soc. B Biol. Sci. 370, 20140138 (2015).
  4. Ferreira, R. B. R., Antunes, L. C. M. & Finlay, B. B. Should the Human Microbiome Be Considered When Developing Vaccines? PLOS Pathog. 6, e1001190 (2010).
  5. Derosa, L. et al. Custom scoring based on ecological topology of gut microbiota associated with cancer immunotherapy outcome. Cell (in press).
  6. Bol, K. F. et al. Adjuvant dendritic cell therapy in stage IIIB/C melanoma: the MIND- DC randomized phase III trial. Nat. Commun. 15, 1632 (2024).
  7. Tseng, C.-H. et al. Gastric microbiota and predicted gene functions are altered after subtotal gastrectomy in patients with gastric cancer. Sci. Rep. 6, 20701 (2016).
  8. Liu, N. et al. Supplementation with α-ketoglutarate improved the efficacy of anti-PD1 melanoma treatment through epigenetic modulation of PD-L1. Cell Death Dis. 14, 170 (2023).
  9. Santos-Peral, A. et al. Basal T cell activation predicts yellow fever vaccine response independently of cytomegalovirus infection and sex-related immune variations. Cell Rep. Med. 6, 101946 (2025).
  10. Santos-Peral, A. et al. Prior flavivirus immunity skews the yellow fever vaccine response to cross-reactive antibodies with potential to enhance dengue virus infection. Nat. Commun. 15, 1696 (2024).
  11. Cheung, K.-S. et al. Association between Recent Usage of Antibiotics and Immunogenicity within Six Months after COVID-19 Vaccination. Vaccines 10, 1122 (2022).
  12. Hong, M. et al. A comprehensive perspective on the interaction between gut microbiota and COVID-19 vaccines. Gut Microbes 15, 2233146 (2023).
  13. Jiao, J. et al. Characterization of the Intestinal Microbiome in Healthy Adults over Sars-Cov-2 Vaccination. Front. Biosci.-Landmark 27, 280 (2022).
  14. Carrier A. & Manghi P. & Alves Costa Silva C. & Lahmar I. et al. 57. Singer, J. et al. Rapamycin and inulin for third-dose vaccine response stimulation (RIVASTIM): Inulin -study protocol for a pilot, multicentre, randomised, double-blinded, controlled trial of dietary inulin to improve SARS-CoV-2 vaccine response in kidney transplant recipients. BMJ Open 12, e062747 (2022).
  15. Xu, J. et al. Boosting Vaccine-Elicited Respiratory Mucosal and Systemic COVID-19 Immunity in Mice With the Oral Lactobacillus plantarum. Front. Nutr. 8, 789242 (2021).
  16. Bonato, A. et al. 156P Microbiota-related multiomics to assess the clinical relevance of antibiotics (ATB) in immunotherapy (ICI). Ann. Oncol. 35, S277 (2024).
  17. Kim, C. G. et al. Prior antibiotic administration disrupts anti-PD-1 responses in advanced gastric cancer by altering the gut microbiome and systemic immune response. Cell Rep. Med. 4, 101251 (2023).
  18. Lamousé-Smith, E. S., Tzeng, A. & Starnbach, M. N. The Intestinal Flora Is Required to Support Antibody Responses to Systemic Immunization in Infant and Germ Free Mice. PLOS ONE 6, e27662 (2011).
  19. Bernard-Raichon, L. et al. Gut microbiome dysbiosis in antibiotic-treated COVID-19 patients is associated with microbial translocation and bacteremia. Nat. Commun. 13, 5926 (2022).
  20. Tarke, A. et al. Comprehensive analysis of T cell immunodominance and immunoprevalence of SARS-CoV-2 epitopes in COVID-19 cases. Cell Rep. Med. 2, 100204 (2021).
  21. Beghini, F. et al. Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3. eLife 10, e65088 (2021).
  22. Pasolli, E., Truong, D. T., Malik, F., Waldron, L. & Segata, N. Machine Learning Meta-analysis of Large Metagenomic Datasets: Tools and Biological Insights. PLOS Comput. Biol. 12, e1004977 (2016).

Cited by

See all 1

Metadata

Authors
Zitvogel, L., Carrier, A., Manghi, P., Silva, C. A. C., Lahmar, I., Birebent, R., Suissa, D., Laheurte, C., Schreibelt, G., Lemant, L., FAHRNER, J.-E., de Sousa, E., Berthier, F., Villemonteix, J., Chevalier, M., Piccinno, G., Hocquet, D., Lebhar, I., Maeurer, M., Caillat-Zucman, S., Kroemer, G., Derosa, L., Launay, O., Golden, E., Bol, K., De Vries, I. J. M., Adotévi, O., Segata, N., Formenti, S., Kobold, S., Pieper, D., Vital, M., Santos-Peral, A., Zaucha, M., García-Bengoa, M., Thorn-Seshold, J., Stirling, H., Rothenfusser, S., Iebba, V., Almonte, A., Ajami, N., White, M., Sahasrabhojane, P., Hoballah, Y., Losh, J., DePinho, C., Dondossola, E., & Wargo, J.
Affiliations
  • Institut Gustave Roussy (FR)
  • Fondazione Edmund Mach (IT)
  • 2. Gustave Roussy Cancer Campus (GRCC), Clinicobiome, 94800 Villejuif Cedex, France.
  • Inserm (FR)
  • Radboud University Medical Center (NL)
  • Champalimaud Foundation
  • bioMérieux (France) (FR)
  • Hôpital Saint-Louis (FR)
  • University of Trento (IT)
  • Université de Franche-Comté (FR)
  • Champalimaud Foundation (PT)
  • Assistance Publique Hôpitaux de Paris, Hôpital Saint-Louis
  • Université Paris Cité (FR)
  • Weill Cornell Medicine (US)
  • Department of Tumor Immunology, Radboud Institute for Molecular Life Sciences, Radboud University Medical Center
  • LMU Klinikum (DE)
  • Centre for Infection Research
  • Medizinische Hochschule Hannover (DE)
  • Division of Clinical Pharmacology, LMU University Hospital, Munich, LMU Munich, 80336 Munich, Germany.
  • Ludwig-Maximilians-Universität München (DE)
  • University of Trieste (IT)
  • The University of Texas MD Anderson Cancer Center (US)
Corresponding author
Zitvogel, Laurence
Venue
Research Square
Published
2025-10-28
OpenAlex
W4415623004
Cited by
1
References
0
Citations by year
2026: 1
Type
posted content
Language
en
Open access
Locations
  • Research Square (acceptedVersion, cc-by)
  • PubMed (publishedVersion)
  • Europe PMC (PubMed Central) (submittedVersion, cc-by)
Funders
  • Bristol-Myers Squibb
  • Melanoma Research Alliance (409510)
  • Regeneron Pharmaceuticals
  • Celldex Therapeutics
  • Les Laboratories Pierre Fabre
  • European Commission (825410, 101045015, HORIZON2020, 756017, 955575, 101137459)
  • Deutsche Forschungsgemeinschaft (SFB-TRR 338/1, KO5055-2-1, 369799452, 391217598, TRR 237, KO5055/3-1, SFB-TRR 338)
  • Agence Nationale de la Recherche (ANR-17-CE15-0031, ANR-17)
  • Institut National de la Santé et de la Recherche Médicale
  • ZonMw
  • Bayerische Forschungsstiftung
  • Ludwig-Maximilians-Universität München
  • Deutsche Krebshilfe
  • Assistance publique-Hôpitaux de Paris
  • Zorginstituut Nederland
  • Servier
  • Hector Stiftung
  • Institut Servier
  • National Institutes of Health (CA- 016672, 1u01ca230551-01a1, 1U01CA230551)
  • Genentech
  • Weill Cornell Medical College
  • University of Texas MD Anderson Cancer Center ((CA016672), P30-CA-016672)
  • Stand Up To Cancer
  • HORIZON EUROPE Framework Programme (101137459)
  • Elitenetzwerk Bayern (409510)
  • Helmholtz Zentrum München
  • National Cancer Institute (CA 016672, 1U01CA230551, P30CA016672)
  • HORIZON EUROPE European Research Council (101100460, 101124203)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gut microbiota and health
  • Life Sciences › Immunology and Microbiology › Immunology › Immune responses and vaccinations
  • Life Sciences › Immunology and Microbiology › Immunology › Immunotherapy and Immune Responses
Keywords
gut microbiota, vaccine immunogenicity, cancer vaccines, viral vaccines, gut dysbiosis, humoral immune response, melanoma, lung cancer, healthcare workers

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Gut microbiota: a new factor modulating the immunizing potential of viral and cancer vaccines. https://app.cochat.ai/papers/10.21203/rs.3.rs-4294379/v1
The paper
Zitvogel, L., Carrier, A., Manghi, P., Silva, C. A. C., Lahmar, I., Birebent, R., Suissa, D., Laheurte, C., Schreibelt, G., Lemant, L., FAHRNER, J.-E., de Sousa, E., Berthier, F., Villemonteix, J., Chevalier, M., Piccinno, G., Hocquet, D., Lebhar, I., Maeurer, M., . . . Wargo, J. (2025). Gut microbiota: a new factor modulating the immunizing potential of viral and cancer vaccines. Research Square. https://doi.org/10.21203/rs.3.rs-4294379/v1

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.