Induction of trained immunity by influenza vaccination - impact on COVID-19.

TLDR: In this article, a large academic Dutch hospital found that SARS-CoV-2 infection was less common among employees who had received a previous influenza vaccination: relative risk reductions of 37% and 49% were observed following influenza vaccination during the first and second COVID-19 waves, respectively.

Abstract: Non-specific protective effects of certain vaccines have been reported, and long-term boosting of innate immunity, termed trained immunity, has been proposed as one of the mechanisms mediating these effects. Several epidemiological studies suggested cross-protection between influenza vaccination and COVID-19. In a large academic Dutch hospital, we found that SARS-CoV-2 infection was less common among employees who had received a previous influenza vaccination: relative risk reductions of 37% and 49% were observed following influenza vaccination during the first and second COVID-19 waves, respectively. The quadrivalent inactivated influenza vaccine induced a trained immunity program that boosted innate immune responses against various viral stimuli and fine-tuned the anti-SARS-CoV-2 response, which may result in better protection against COVID-19. Influenza vaccination led to transcriptional reprogramming of monocytes and reduced systemic inflammation. These epidemiological and immunological data argue for potential benefits of influenza vaccination against COVID-19, and future randomized trials are warranted to test this possibility.

Figures

Figure 1: Summary of the clinical study investigating the effects of an influenza vaccine on 119 trained immunity. Blood was collected 1 week before and 6 weeks after the influenza vaccination from 120 28 participants. PBMC stimulation, quantification of plasma proteins, and single-cell RNA sequencing 121 were performed. 122

Figure 4: Ex vivo cytokine responses of the individuals before and after influenza vaccination. 210 PBMCs were stimulated with heat-killed Influenza H1N1 (California strain), heat-killed SARS-CoV-2, 211 poly I:C, and R848 for 24 hours. A. TNFα, B. IL-6, C. IL-1β, D. IL-1RA, and E. IL-12 responses were 212 quantified. SARS-CoV-2 and Influenza stimulations did not lead to detectable levels of IL-12. Wilcoxon 213 signed-rank test was performed to compare the differences in cytokine production between before and 214 after influenza vaccination. Light blue dots represent the cytokine values before vaccination, while dark 215 blue dots show after vaccination. n=28 *p<0.05, **p<0.01, ****p<0,0001. 216

Figure 2: Transcriptomic changes induced by influenza vaccination on a single cell level. A. 146 Volcano plot depicting the upregulated and downregulated genes in CD14+ monocytes 6 weeks after 147 influenza vaccination. Red dots show upregulation, while blue dots show downregulation. B. Violin plots 148 showing the expression levels of MNDA and CTSS before and after vaccination in CD14+ monocytes. 149 C. Volcano plot depicting the upregulated and downregulated genes in CD4+ naïve T cells 6 weeks after 150

Figure 5: Correlation of ex vivo anti-SARS-CoV-2 responses with circulating inflammatory 243 mediators that are significantly altered by influenza vaccination. A. Heatmap depicting the baseline 244 (before vaccination) levels of inflammatory mediators that were downregulated by influenza vaccination 245 and are significantly correlated with the induction of cytokines upon SARS-CoV-2 stimulation before vs. 246 after vaccination. Red indicates positive correlation, and blue indicates negative correlation. B. Selected 247 circulating protein and cytokine pairs that are significantly correlated. C. Correlation of circulating 248 proteins that were upregulated by influenza vaccination with anti-SARS-CoV-2 cytokine responses, and 249 dot plot depicting the association of baseline GALNT3 levels and induction IL-6 against SARS-CoV-2 250 after vaccination. r: Spearman's correlation coefficient. *p<0.05, **p<0.01, ***p<0,001. 251

Table 1: COVID-19 incidence among influenza vaccinated and unvaccinated employees of 93 Radboud University Medical Center in the first two waves of the pandemic. First wave: March - 94 June 2020, second wave: November 2020 - January 2021. Influenza vaccinations in autumn of 2019 95 and autumn of 2020 were considered for calculations regarding the first and the second COVID-19 96 waves, respectively. 97

Figure 3: Influenza vaccination downregulates circulating inflammatory proteins. 183

Full text

1
Induction of trained immunity by influenza vaccination - impact on COVID-19
1
2
Priya A. Debisarun
1,9
, Katharina L. Gössling
2,9
, Ozlem Bulut
1
, Gizem Kilic
1
, Martijn Zoodsma
3,4
,
3
Zhaoli Liu
3,4
, Marina Oldenburg
2
, Nadine Rüchel
2
, Bowen Zhang
3,4
, Cheng-Jian Xu
1,3,4
, Patrick
4
Struycken
5
, Valerie A.C.M. Koeken
1,3,4
, Jorge Domínguez-Andrés
1
, Simone J.C.F.M. Moorlag
1
,
5
Esther Taks
1
, Philipp N. Ostermann
6
, Lisa Müller
6
, Heiner Schaal
6
, Ortwin Adams
6
, Arndt
6
Borkhardt
2
,
Jaap ten Oever
1
, Reinout van Crevel
1
, Yang Li
3,4
, Mihai G. Netea
1,7,8*
7
8
1
Department of Internal Medicine, Radboud University Medical Center, 6525GA, Nijmegen,
9
Netherlands
10
2
Department for Pediatric Oncology, Hematology and Clinical Immunology, University Hospital
11
Duesseldorf, Medical Faculty, Heinrich Heine University Duesseldorf, 40225, Dusseldorf, Germany
12
3
Department of Computational Biology for Individualised Infection Medicine, Centre for Individualised
13
Infection Medicine (CiiM), a joint venture between the Helmholtz-Centre for Infection Research (HZI)
14
and the Hannover Medical School (MHH), 30625, Hannover, Germany
15
4
TWINCORE, a joint venture between the Helmholtz-Centre for Infection Research (HZI) and the
16
Hannover Medical School (MHH), 30625, Hannover, Germany
17
5
Department of Occupational Health & Safety, and Environmental Service, Radboud University
18
Medical Center, 6525EX, Nijmegen, Netherlands
19
6
Institute of Virology, University Hospital Duesseldorf, Medical Faculty, Heinrich Heine University
20
Duesseldorf, 40225, Dusseldorf, Germany
21
7
Human Genomics Laboratory, Craiova University of Medicine and Pharmacy, 200349, Craiova,
22
Romania.
23
8
Department for Immunology & Metabolism, Life and Medical Sciences Institute (LIMES), University of
24
Bonn, 53115, Bonn, Germany
25
9
These authors contributed equally.
26
27
*Correspondence: mihai.netea@radboudumc.nl
28
Mihai G. Netea, M.D., Ph.D.
29
Department of Medicine (463)
30
Radboud University Nijmegen Medical Center
31
Geert Grooteplein Zuid 8, 6525 GA Nijmegen, The Netherlands
32
Tel: +31-24-3618819; E-mail: mihai.netea@radboudumc.nl
33
. CC-BY 4.0 International licenseIt 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 thisthis version posted September 10, 2021. ; https://doi.org/10.1101/2021.09.03.21263028doi: medRxiv preprint
NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

2
ABSTRACT
34
35
Non-specific protective effects of certain vaccines have been reported, and long-term boosting of innate
36
immunity, termed trained immunity, has been proposed as one of the mechanisms mediating these
37
effects. Several epidemiological studies suggested cross-protection between influenza vaccination and
38
COVID-19. In a large academic Dutch hospital, we found that SARS-CoV-2 infection was less common
39
among employees who had received a previous influenza vaccination: relative risk reductions of 37%
40
and 49% were observed following influenza vaccination during the first and second COVID-19 waves,
41
respectively. The quadrivalent inactivated influenza vaccine induced a trained immunity program that
42
boosted innate immune responses against various viral stimuli and fine-tuned the anti-SARS-CoV-2
43
response, which may result in better protection against COVID-19. Influenza vaccination led to
44
transcriptional reprogramming of monocytes and reduced systemic inflammation. These epidemiological
45
and immunological data argue for potential benefits of influenza vaccination against COVID-19, and
46
future randomized trials are warranted to test this possibility.
47
48
Keywords: Influenza vaccine, COVID-19, SARS-CoV-2, trained immunity, cytokines, vaccination
49
. CC-BY 4.0 International licenseIt 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 thisthis version posted September 10, 2021. ; https://doi.org/10.1101/2021.09.03.21263028doi: medRxiv preprint

3
1. INTRODUCTION
50
As of May 2021, over 150 million cases and 3.1 million deaths due to the novel coronavirus disease
51
COVID-19 have been reported (1). COVID-19 is caused by severe acute respiratory syndrome
52
coronavirus 2 (SARS-CoV-2). While in the majority of cases the virus causes mild symptoms that resolve
53
spontaneously, in the elderly or patients with underlying co-morbidities, the disease is often severe and
54
potentially lethal (2). Due to the rapid spread and high clinical and socio-economic burden of COVID-
55
19, sustained efforts have been made to develop preventive and therapeutic strategies (3, 4). Several
56
effective anti-COVID-19 vaccines have been designed and successfully tested, with almost 1 billion
57
vaccine doses already administered (1). However, vaccine supply is still not able to ensure the global
58
needs, with many countries facing challenges in ensuring access to enough COVID-19 vaccines (5). An
59
additional challenge is the emergence of new SARS-CoV-2 variants which are on the one hand more
60
infective, and on the other hand can display vaccine escape (6).
61
Due to the absence of specific COVID-19 vaccines in the beginning of the pandemic, as well as the
62
current challenges posed by limited vaccine supply and emergence of new virus strains, vaccination
63
strategies using already available vaccines that can protect against a broad array of pathogens has
64
been proposed as ‘bridge vaccination’ (7). The potential interaction between vaccines and infections
65
other than their target disease has attracted a lot of attention lately. It has been demonstrated that certain
66
vaccines, such as Bacillus Calmette-Guérin (BCG), measles-containing vaccines, or oral polio vaccine,
67
have strong non-specific protective effects through long-term reprogramming of innate immunity, a
68
process called trained immunity (8). In line with this, several recent studies suggested a potential link
69
between influenza vaccination and decreased COVID-19 incidence and severity (9-12). This suggests
70
that influenza vaccination may potentially convey partial protection against COVID-19, and this potential
71
beneficial effect needs to be investigated.
72
In this study, we assessed the association between influenza vaccination and COVID-19 incidence
73
during the first two waves of the pandemic in the Netherlands, among employees of the Radboud
74
University Medical Center (Radboudumc), a large academic hospital. In addition, as possible
75
mechanisms of action, we investigated the induction of trained immunity and the impact on systemic
76
inflammation by the influenza vaccine used in the 2020 autumn season in 28 healthy adult volunteers.
77
78
. CC-BY 4.0 International licenseIt 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 thisthis version posted September 10, 2021. ; https://doi.org/10.1101/2021.09.03.21263028doi: medRxiv preprint

4
2. RESULTS
79
2.1. Quadrivalent inactivated influenza vaccination is associated with lower COVID-19 incidence
80
To investigate the effect of influenza vaccination on COVID-19 incidence, we compared the incidence
81
of COVID-19 cases, validated by SARS-CoV-2 PCR, among hospital workers who were either
82
vaccinated or not against influenza.
83
As of June 1
st,
2020, at the end of the first COVID-19 wave in the Netherlands, Radboudumc had 6856
84
employees working in the clinical departments with direct patient contact (Table 1). The total influenza
85
vaccine coverage rate (VCR) in the hospital for that season (autumn 2019) was 53% (3655/6856).
86
Among these, 184 were documented to have contracted SARS-CoV-2 PCR-positive COVID-19. 42% of
87
SARS-CoV-2 positive individuals during the first wave (77/184) had received an influenza vaccination
88
in the preceding flu season, as opposed to 54% (3578/6672) of SARS-CoV-2 negative employees:
89
3.34% of the individuals who were not vaccinated against influenza had COVID-19, compared to 2.11%
90
of the vaccinated employees (RR=0.63, 95% CI, 0.47-0.84, P=0.0016).
91
92
Table 1: COVID-19 incidence among influenza vaccinated and unvaccinated employees of
93
Radboud University Medical Center in the first two waves of the pandemic. First wave: March -
94
June 2020, second wave: November 2020 - January 2021. Influenza vaccinations in autumn of 2019
95
and autumn of 2020 were considered for calculations regarding the first and the second COVID-19
96
waves, respectively.
97
98
99
100
First wave
Second wave
SARS-
CoV-2
negative
SARS-
CoV-2
positive
Total
SARS-
CoV-2
incidence
SARS-
CoV-2
negative
SARS-
CoV-2
positive
Total
SARS-
CoV-2
incidence
Influenza
vaccination
No
3094
107
3201
3.34%
6120
250
6370
3.92%
Yes
3578
77
3655
2.11%
4438
91
4529
2.00%
Total
6672
184
6856
2.68%
10558
341
10899
3.13%
. CC-BY 4.0 International licenseIt 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 thisthis version posted September 10, 2021. ; https://doi.org/10.1101/2021.09.03.21263028doi: medRxiv preprint

5
A lower incidence of SARS-CoV-2 positivity among vaccinated individuals was also reported during the
101
second wave (Table 1), when data for the total number of 10899 Radboudumc employees became
102
available. The hospital's total influenza vaccination coverage rate during the 2020/2021 influenza
103
season was 42% (4529/10899). 91 of the 341 SARS-CoV-2 positive employees (27%) were vaccinated
104
against influenza in the autumn 2020. The COVID-19 incidence during the second wave was 3.92%
105
among unvaccinated employees and 2.00% for vaccinated employees (RR of 0.51, 95% CI 0.40-0.65,
106
P < 0.0001).
107
Our results indicate that influenza vaccination was significantly associated with lower COVID-19
108
incidence among hospital employees during the first two waves of the pandemic.
109
110
2.2. Influenza vaccination induces long-term transcriptional reprogramming
111
To assess a possible induction of trained immunity upon influenza vaccination as the underlying
112
mechanism of protection against SARS-CoV-2, a proof-of-principle study to assess the non-specific
113
immunological effects of the influenza vaccination was performed in 28 healthy individuals. Participants
114
received an influenza vaccine (Influvac Tetra), and blood was collected 1 week before and 6 weeks after
115
vaccination. The study design is summarized in Figure 1.
116
117
. CC-BY 4.0 International licenseIt 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 thisthis version posted September 10, 2021. ; https://doi.org/10.1101/2021.09.03.21263028doi: medRxiv preprint

Frequently asked questions

Q1. What are the contributions mentioned in the paper "Induction of trained immunity by influenza vaccination - impact on covid-19" ?

In this paper, the authors investigated the effect of vaccination against COVID-19 in the first two waves of the pandemic in the Netherlands, among employees of the Radboud University Medical Center. 

Q2. What was the p-value after the Benjamini-Hochberg adjustment?

R package limma 437 was used for differential expression analysis and p-values < 0.05 after Benjamini-Hochberg adjustment 438 were considered significant. 

Q3. How did the researchers determine the effect of the influenza vaccination on COVID-19?

Influenza vaccination induces long-term transcriptional reprogramming 111To assess a possible induction of trained immunity upon influenza vaccination as the underlying 112 mechanism of protection against SARS-CoV-2, a proof-of-principle study to assess the non-specific 113 immunological effects of the influenza vaccination was performed in 28 healthy individuals. 

Q4. What is the effect of influenza vaccination on the immune response to SARS-CoV-2?

influenza 266 vaccination modulated the responses against SARS-CoV-2, reducing IL-1β and IL-6 production while 267 enhancing IL-1Ra release. 

Q5. What is the role of CTSS in the induced immunity of COVID-19?

By enhancing the uptake, processing, and presentation of SARS-CoV-2 antigens, upregulation of 298 CTSS might be beneficial to induce the anti-viral immune response. 

Q6. How many cases of COVID-19 have been reported?

As of May 2021, over 150 million cases and 3.1 million deaths due to the novel coronavirus disease 51 COVID-19 have been reported (1). 

Q7. What is the role of the healthy vaccinee bias in the overestimation of the positive?

The healthy vaccinee bias could also play a role in the overestimation of the positive effect of a vaccine: 277 individuals willing to be vaccinated against influenza may be also those more likely to respect the 278 personal protection rules against COVID-19 infection. 

Q8. How many vaccines have been shown to have strong non-specific protective effects?

It has been demonstrated that certain 66 vaccines, such as Bacillus Calmette-Guérin (BCG), measles-containing vaccines, or oral polio vaccine, 67 have strong non-specific protective effects through long-term reprogramming of innate immunity, a 68 process called trained immunity (8). 

Q9. Why is the COVID-19 epidemic so prevalent in the United States?

Due to the rapid spread and high clinical and socio-economic burden of COVID-55 19, sustained efforts have been made to develop preventive and therapeutic strategies (3, 4). 

Q10. What was the upregulation of nave CD4+ T cells?

T cells exhibited 157 upregulation in translation, protein localization, and viral gene expression and downregulation in 158 lymphocyte differentiation and NFκB signaling (S2 Figure, upper panel). 

Q11. How did the authors find that Influvac Tetra 290 induce a transcriptional and functional?

the authors found that Influvac Tetra indeed 290 induces a transcriptional and functional reprogramming of innate immune cells 6 weeks after the 291 vaccination, modulating cytokine responses upon viral challenge with unrelated stimuli. 

Q12. What is the effect of the adjuvanted influenza vaccine on the immune system?

On the other hand, the adjuvanted influenza vaccine enhanced the 352 accessibility of anti-viral genes and increased the resistance of PBMCs against Dengue and Zika virus 353 infections.