key: cord-0829674-isyecifr
authors: Omersel, Jasna; Karas Kuželički, Nataša
title: Vaccinomics and Adversomics in the Era of Precision Medicine: A Review Based on HBV, MMR, HPV, and COVID-19 Vaccines
date: 2020-11-05
journal: J Clin Med
DOI: 10.3390/jcm9113561
sha: b8057c77d6923f1920895ecd295c133e41f18c75
doc_id: 829674
cord_uid: isyecifr

Precision medicine approaches based on pharmacogenomics are now being successfully implemented to enable physicians to predict more efficient treatments and prevention strategies for a given disease based on the genetic background of the patient. This approach has already been proposed for vaccines, but research is lagging behind the needs of society, and precision medicine is far from being implemented here. While vaccinomics concerns the effectiveness of vaccines, adversomics concerns their side effects. This area has great potential to address public concerns about vaccine safety and to promote increased public confidence, higher vaccination rates, and fewer serious adverse events in genetically predisposed individuals. The aim here is to explore the contemporary scientific literature related to the vaccinomic and adversomic aspects of the three most-controversial vaccines: those against hepatitis B, against measles, mumps, and rubella, and against human Papilloma virus. We provide detailed information on the genes that encode human leukocyte antigen, cytokines and their receptors, and transcription factors and regulators associated with the efficacy and safety of the Hepatitis B and Measles, Mumps and Rubella virus vaccines. We also investigate the future prospects of vaccinomics and adversomics of a COVID-19 vaccine, which might represent the fastest development of a vaccine ever.

Despite some initial skepticism, it is undeniable that the era of precision (or personalized) medicine is here, as formalized by the launch of the Precision Medicine Initiative in 2015 [1] . The paradigm that one size does not fit all is slowly taking hold in the minds of clinicians, and also of patients, and by society as a whole. What is more, we are starting to realize that personalized approaches might be a solution to the public distrust in medicines, pharmaceutical companies, and science in general that has emerged over the last few decades, and mainly in developed countries.

One of the first fields of precision medicine that developed was pharmacogenomics, which provided both the scientific basis and clinical outcomes that have together promoted the practice of personalized drug therapies; i.e., prescribing each patient with the right dose of the right drug, thus maximizing the efficiency and safety of the therapy [2] . The application of this same principle to vaccines has been named vaccinogenomics, or vaccinomics. The term vaccinomics was first used by Hoffman et al. in 1998 [3] and extensively studied by Gregory A. Poland and the Mayo Clinic Vaccine Research Group [4] . By definition, vaccinomics explores the influence of genetic (and non-genetic) factors on the heterogeneity of vaccine-induced immune responses between individuals

The current vaccination schemes do not take into account interindividual differences in vaccine efficacy. Thus, all individuals are vaccinated with the same number of doses, even though some individuals might achieve seroconversion with a lower number of vaccine doses, while others might remain unresponsive even after administration of the standard number of doses. Therefore, prediction of vaccine efficiency before the administration of the vaccine to any particular individual might be useful for health and economic reasons. Individuals deemed to be unresponsive by vaccinomic tests could then be given a further vaccine dose while monitoring for seroconversion to assure that they are protected against the disease. On the other hand, hyper-responsive individuals might be vaccinated with a smaller number of doses, which would be beneficial from the economic point of view.

Although seroconversion after three doses of the HBV vaccine is relatively high [11] , inadequate production of protective antibodies has been observed for 4% to 10% of healthy individuals [12] . The number of non-responders is even higher in patients with autoimmune diseases [13, 14] . Recently, the hypothesis has been proposed that HBV vaccine nonresponse might predict latent autoimmunity [15] .

A relatively large number of vaccinomic studies have been conducted in connection with HBV, some of which were carried out even before the term vaccinomics was coined. However, the majority of these focused on HBV vaccine efficacy, i.e., predicting good versus poor responses in terms of seroconversion on the basis of genetic markers. The immune response to the HBV vaccine shows a high level of heritability (77%), as was determined in 207 twin pairs [16] . Early studies identified individuals as HBV vaccine poor responders who expressed high levels of the human leukocyte antigen HLA-DR3 and HLA-DR7 antigens (i.e., homozygous for haplotypes HLA-B8, SC01, DR3, HLA-B44, FC31, and DR7) [17, 18] . In later studies, it was determined that the most important HLA locus for HBV vaccine response prediction was HLA-DRB1* [19] and that certain alleles at this locus are linked to nonresponse and others to good response to HBV vaccine. In several studies, alleles HLA-DRB1*0301, HLA-DRB1*0701, and HLA-DQB1*0201 were associated with no production of anti-HBV antibodies after vaccination (HBV vaccine nonresponse), while alleles HLA-DRB1*01, HLA-DRB1*11, and HLA-DRB1*15 were associated with rapid production of protective antibodies (HBV vaccine good response) [19, 20] . Functional studies have revealed that in humans, the mechanism of nonresponse for the HLA-DRB1*0301 and HLA-DRB1*0701 alleles was due to the lack of specific T cells that recognized a particular major histocompatibility complex (MHC)/HBV surface antigen peptide complex, rather than due to the failure of the MHC molecule to bind the HBV surface antigen peptide, as shown in a mouse model [21] .

As conventional HLA typing is complex and time-consuming, the use of single nucleotide polymorphisms (SNPs) as vaccinomic markers that can be defined using simple genotyping methods (e.g., TaqMan technology) would be more suitable for clinical implementation. Indeed, genome-wide association studies (GWAS) have detected SNPs in the HLA genes that are in line with the older studies mentioned above and have great potential to be used as vaccinomic markers. In GWAS on an Indonesian cohort, two SNPs in HLA-DQB1 (rs32734227, rs32734289) were identified as markers of HBV vaccine nonresponse [22] , while a genome wide association study (GWAS) on a Han Chinese population revealed rs477515 at HLA-DRB1 to be an independent marker of nonresponse to the HBV vaccine [23] . In a population of adult non-responders to infants given the HBV vaccine, a GWAS by Wu et al. detected a SNP at the HLA-DPB1 locus (rs7770370) as a nonresponse marker to the HBV vaccine booster dose [24] . Recently, a GWAS on Japanese individuals identified HLA alleles of poor response to HBV vaccine as DRB1*0405, DQB1*0401, and DPB1*0501 [25] . As all of these GWASs were performed in Asian populations, the relevance of the genetic markers identified needs to be validated in Caucasians and Africans.

Apart from HLA loci, other genes have been shown to influence HBV vaccine responses. For example, null alleles of the complement component C4A gene (C4AQ0; due to deletion or non-expression) have been associated with poor response to the HBV vaccine [26] . Furthermore, Toll-like receptors (TLRs) and cytokines and their receptors have been identified as potential vaccinomics markers. The interleukin (IL)-1βhaplotype rs1143633 (A) + rs1143627 (G) was associated with HBV vaccine non-response, as well as SNPs in genes encoding IL-13 (rs1295686) and IL-4 (rs2243248). Conversely, rs1805015 for the IL-4 receptor (IL4RA) and rs3804100 for TLR2 were associated with a good response to the HBV vaccine [27] . Several other loci have been connected to HBV vaccine nonresponse in different GWASs, although the studies will need replication before these can be conclusively considered as vaccinomic markers. Interestingly, a SNP in the butyrophilin like (BTNL)2 gene (rs3763316) was among the top hits in two independent GWASs [22, 23] , which indicates its potential as a useful vaccinomics marker. A recent GWAS on the Japanese population identified another SNP in the same gene (rs4248266) as a predictor of HBV vaccine response [25] . This gene encodes an MHC-II-associated, type I transmembrane protein. This is believed to be involved in immune surveillance, by serving as a negative T cell regulator through decreasing T cell proliferation and cytokine release. More details of the genes associated or believed to be associated with HBV vaccine responses are given in Tables 1-4.

Despite the availability of a highly effective measles vaccine, a resurgence of measles has been seen across the developed world. Most of these cases can be attributed to non-vaccinated individuals (usually due to the fears spread by the antivaccine movements), although some of those infected had been fully vaccinated. Numerous studies have shown that host genetic factors result in inter-individual variations in measles-vaccine-induced antibodies, and hence can have a role in vaccine failure. Heritability of an antibody response to measles vaccination was estimated to be 88.5% in a twin study [49] and 49% using the complete genetic information from a GWAS of 935 individuals [50] . This suggests that the inter-individual variations in antibody responses to measles vaccine arise through the combined small effects of many genes [50] .

Early vaccinomic studies on the MMR vaccine mainly involved the genes that encode various cytokines and their receptors. As cytokines are important mediators of both adaptive and innate immune responses, it is not surprising that many SNPs in these genes were associated with responses to the MMR vaccine. Minor alleles of two polymorphisms in the tumor necrosis factor (TNF) (rs1799964) and IL-6 (rs2069849) genes were associated with low rates of expression of measles-specific antibodies after the MMR vaccine in a Somali population [34] , while polymorphisms in rs2228149 in IL-2 receptor subunit α (IL2RA) and rs1143634 in IL-1B have been associated with low rates of expression mumps-specific antibodies [34] . Even more polymorphisms have been identified as associated with the nonresponses to the rubella component of the MMR vaccine: rs7873167, rs3885423, rs1364613 and rs1364612 in interferon (IFN)B1; rs12722713 and rs12722698 in IL2RA; rs228937 in IL2RB; rs2069824 in IL-6; and rs4149650 in 'TNF receptor superfamily member 1A' (TNFRSF1A) [34] . At the same time, several polymorphisms of the HLA loci were addressed in association with the MMR vaccine response. Consistent associations with humoral responses to the measles component after two doses of MMR vaccine were shown in three independent cohorts for the following HLA alleles: B*3503 (good response), and DQA1*0201 and DRB1*0701 (poor responses) [29, 30] . Distinct alleles of poor response were also detected for rubella and mumps components of the MMR vaccine: A*2705, A*5701, DPA1*0201, DPB1*0301, DPB1*1301 (rubella), and DQB1*0303 (mumps) [30] . Of interest, the association of the HLA-DPB1*0301 allele with the poor response to the rubella vaccine was replicated in a recent GWAS that investigated around five million variants in 1000 healthy individuals vaccinated with the MMR vaccine [33] .

Furthermore, genes encoding proteins that interact with the measles virus (the wild and vaccine strains) and augment their entry into host cells were studied in the context of MMR vaccinomics. The most prominent genetic marker from this group of genes was rs2724384 in the CD46 gene, which was associated with low antibody response to MMR vaccination in both a candidate gene study on 764 individuals [47] and a GWAS on 2872 individuals [46] . While CD46 is preferentially used by the vaccine strain measles virus, 'signaling lymphocyte activation molecule' (SLAM) interacts more readily with the wild-type strain. In the candidate gene study, two SNPs in SLAM were associated with measles antibody levels, as well as the IFN-γ and IL-10 responses [47] . In addition to CD46, one of the top hits in the GWAS here [46] was in the 'interferon-induced protein 44 like' (IFI44L) gene (rs273259). Interestingly, in the adversomic study by Feenstra et al. [7] , the A allele of this polymorphism was associated with increased measles-specific antibody titer [46] and increased risk of febrile seizures after the MMR vaccination. The function of IFFI44L is still not clear, but it has been demonstrated that it has a type I IFN response against the hepatitis C virus [51] and appears to be involved in innate immunity.

Toll-like receptors and their intracellular signaling molecules have important roles in innate immunity. Ovsyannikova et al. [48] examined associations between polymorphisms in TLR family genes and measles-vaccine-specific immune responses in 764 individuals vaccinated with two doses of MMR. Two SNPs in the coding regions of the TLR2 (rs3804100) and TLR4 (rs5030710) genes were associated with increases and decreases in measles-specific antibodies, respectively. Furthermore, the minor allele of a SNP in the KIAA1542 gene (rs702966) was associated with low IFN-γ response [48] .

Vitamin A (retinol) and vitamin D and their receptors are essential regulators of immune function, and they have been linked with susceptibility to various viral infections [52] , and they have also been studied in the context of MMR vaccine efficiency [37] . In this study, several SNPs in retinoic acid receptor α (RARB) were associated with variations in both measles antibodies and cytokine secretion levels, including IL-10, IFN-α, and TNF-α. Furthermore, significant associations were shown between multiple vitamin D receptors (VDR) and retinoid X receptor α (RXRA) SNPs/haplotypes and measles-specific IL-2, IL-6, IL-10, IFN-α, IFN-γ, IFNλ-1, and TNF-α cytokine secretion [37] .

The 'tripartite motif containing' TRIM proteins that are activated by type I IFNs were recently shown to be important cellular factors for innate immunity and antiviral defense [53] . Apparently, they also have roles in the MMR vaccine response, as associations have been shown between TRIM5 (rs7122620) and TRIM25 (rs205499) gene polymorphisms and measles-specific antibody levels [38] . In the same study, several SNPs in TRIM5, 22 and 25 were connected with cytokine responses (IFN-γ, IL-2, IL-6, IL-10, TNF-α).

Recently, a GWAS on 1843 individuals identified four polymorphisms in the tumour suppressor gene WT1 (rs4986811, rs5030172, rs5030157, rs5030166) that were associated with decreased rubella-specific IL-6 secretion from peripheral blood mononuclear cells post-MMR vaccine [39] . However, all four of these polymorphisms were identified in a WT1 transcript variant known to be a target of 'nonsense mediated decay', which indicates that its functional role in this vaccine response might be questionable.

In addition to genomic studies, whole transcriptome [54] and micro (mi)RNA profiling [55] have revealed interesting potential new vaccinomic markers. Next-generation sequencing of intracellular mRNA and miRNAs (mRNA-Seq) in measles-virus-stimulated B cells and CD4+ T cells from high and low antibody responders to measles vaccine identified vaccinomic markers from various gene ontology groups: plasma cell survival (CD93, IL6, CXCL12), chemokine/cytokine activity, and cell-cell communication/ adhesion/ migration. The predicted targets for the identified B cell-specific miRNAs included the Fc-receptor and several other signalling pathways, as well as pathways related to transcriptional regulation, viral infection, lipid biosynthesis/metabolism, cytoskeletal protein binding, extracellular matrix-receptor interactions, and apoptosis.

In conclusion here, numerous HLA and non-HLA genetic factors that individually or jointly contribute to the variability seen for the humoral responses to the MMR vaccine have been identified. In the future, new approaches and methods will enable the investigation of immune response mechanisms to the measles vaccine on a new multidimensional level, such as vaccinomics, systems biology, GWAS, epitope prediction, and sophisticated bioinformatics/statistical algorithms, [56] . More details of the genes associated or believed to be associated with MMR vaccine responses are given in Tables 1-4.

Vaccination against HPV is intended for prevention of HPV-associated cancers. Vaccination is especially efficient against virus infection types 16 and 18, which are responsible for 70% of cervical cancers worldwide, and types 6 and 11, which are responsible for 90% of cases of genital warts. The Centers for Disease Control and Prevention recommends that all boys and girls get two doses of the HPV vaccine at 11 to 12 years old. To prevent HPV infection, there are currently three vaccines used for young girls from the age of 9 years: the bivalent vaccine Cervarix (directed at HPV types 16, 18) , the quadrivalent vaccine Gardasil (directed at HPV type 6, 11, 16, 18) and Gardasil 9 (directed at HPV types 6, 11, 16, 18, 31, 33, 45, 52, 58) .

Since the HPV vaccines are relatively novel compared to the HBV and MMR vaccines, no vaccinomics studies have investigated the genetic markers of HPV vaccine efficiency to date. However, HLA and non-HLA genes have dominant roles in the initiation of the antiviral response through innate and humoral immunity networks, and they can also contribute to increased risk of the diseases. Leo et al. associated three haplotypes with increased risk for cervical neoplasia after HPV infection: HLA-DRB1*15/HLA-DQB1*0602/HLA-DQA1*0102, HLA-B*0702/HLA-C*0702 and HLA-DRB1*0401/ HLA-DQA1*0301 [57] . Mainali et al., investigated 191 men infected with HPV, and they identified novel susceptibility genes involved in HPV-16 pathogenesis for genital persistence. The strongest association was for intergenic variant rs1293153 (between the AL133335.1 and DOK5 genes) and rs405103 (in gene AC013565.1) on chromosomes 20 and 15, respectively [58] . Additionally, the genetic variants of transforming growth factor TGFβ1 (T869C, C509T, and G915C) that have important roles in tumour progression and suppression and in immune suppression have been related to HPV-16-positive oropharyngeal cancer, as compared to their wild-type genotypes [59] . This was further supported by Levovitz et al., who reported that TGFβR1 is significantly overexpressed in oropharyngeal and cervical cancer [60] . Altogether, these studies have shown the importance of the TGFβ1 signalling pathway in HPV tumour progression, and again, this highlights the involvement of the genetic background in HPV vaccine immune responses. These loci might thus represent good candidate genetic markers in vaccinomics and might be a starting point for evidence-based vaccination.

Recently, a critical review with novel insights and expert opinion on personalized HPV vaccination by Chambuso et al. identified two important strategies to personalization of HPV vaccination [61] : (a) identification of clinically important genetic markers for long-term HPV vaccine immune responses; and (b) profiling of host genetic risk factors that can contribute to modifications to long-term immune responses. Moreover, they also proposed some important research topics related to vaccine effectiveness and safety that should be addressed in the near future: the association between ethnic origin and impaired HPV vaccine immune responses; the correlation of post-vaccination serum antibodies with vaccine adverse effects; the variable persistence of immunity in different populations; the recognition of acute and chronic adverse events of HPV vaccination according to variations in host immune response genes and the number of vaccine doses administered [61] .

Vaccines are a pharmaceutical product that is subjected to the strictest of safety protocols, as they are administered to healthy individuals, and in most cases to paediatric populations. Although modern vaccines have excellent safety profiles, some individuals do very rarely experience serious side effects after vaccination. These side effects are so rare that they can only be detected in the post-marketing phase of vaccine testing, when it has been administered to a very large number of individuals. According to the vaccine pharmacovigilance working groups of the Council for International Organisations of Medical Sciences/World Health Organisation [62] , vaccine pharmacovigilance is defined as "the science and activities relating to the detection, assessment, understanding and communication of the adverse effects following immunization and other vaccine or immunization-related issues, and to the prevention of unwanted effects of the vaccine immunization". Investigation of the underlying causes of such events is essential not only to improve vaccine safety, but also to maintain public confidence in vaccination.

Generally, the safety of the HBV vaccine has been evaluated as very high [63, 64] . However, over the last two decades following the launch of the massive HBV vaccination campaigns worldwide, there have been reports of the onset of several autoimmune diseases after the HBV administration. These have included: multiple sclerosis, optic neuritis, vasculitis, rheumatoid arthritis, systemic lupus erythematosus, and others. Multiple sclerosis and other demyelinating diseases have been investigated the most, and the majority of these studies have indicated no increased risk of multiple sclerosis after HBV vaccination [65] . However, the doubts about the HBV vaccine risks have not been resolved yet, also because there is good theoretical rationale for HBV-vaccine-induced autoimmunity through various mechanisms [66] . Furthermore, numerous studies on animal and human models have demonstrated activation of transient autoimmunity processes after HBV vaccination in terms of the levels of antibodies [67, 68] and T-regulatory cells [69] .

While a number of genetic markers that can predict HBV vaccine responses are known to date (i.e., for vaccine efficiency), there are very few related to prediction of adverse reactions to the HBV vaccine, as few such adversomics studies have been performed. Indeed, in general, HBV vaccine adversomic studies have been rare and of a more speculative nature. Belloni et al. concluded that there was no statistically significant difference in the levels of auto-antibodies between responders and non-responders for children vaccinated with the HBV vaccine [31] . However, the non-responder children showed a trend towards higher levels of anti-smooth-muscle antibodies (30% vs. 2% for responders). The anti-smooth-muscle antibody-positive non-responders all carried the supratype HLA-C4AQ0, DRB1*0301, DQB1*02, which has been associated with poor HBV response in other studies [17, 18, 20] and is a well-known predisposing factor for autoimmune disorders. Belloni et al. speculated that HBV vaccine non-responders who carry the HLA-C4AQ0, DRB1*0301, DQB1*02 haplotype might thus be more at risk of autoimmune disease [31] . On the basis of the analysis of a series of case reports of various autoimmune diseases, Miller at al. hypothesized that the presence of certain HLA-DRB1 alleles (*01:01, *03:01, *04:01, *13:01, *15:01) that have been identified by others as HBV vaccine response modulators [17] [18] [19] [20] can result in activation of CD8+T cells by HLA-A2-presented HBV surface antigens, which can result in the production of high levels of IFN-studied levels (0.05 to 0.3 mL/min) did not significantly affect the separation, but definitely shorte retention time. The low ionic strength of the buffer had the benefits of improving enantioseparat fastening elution, and preventing precipitation of buffer salt when the aqueous and organic ph meet in the UHPLC mixing chamber. Doubling the rate of the change in %B in the linear grad caused a noticeably shorter retention time due to the increased eluent strength of the mobile ph according to the linear solvent strength theory, indicating that hydrophobic-hydroph interactions are governing analyte (stationary phase) mobile phase interactions. However, high decreased the solubility of CMPA and worsened enantioseparation, likely due to competition o organic modifier with the chiral binding site at the β-CD cavity as reported by Gratz et al [25] . Fin the mobile phase was set at pH = 3.5 and %B change at 0.02 as fixed conditions for all upcoming r of the optimization design.

The linear solvent strength theory governing reversed phase (RP) separations is known to logarithmic relationship where the retention factor K to the base 2 is related to surface tension o aqueous mobile phase modifier (ln K = A + BD+ C ɣ + D(Ke − 1)V2/3ɣ + E + ln RT/PoV, where A, and D are constants). Our regression model for retention (Equations (1) and (2)) has the s logarithmic relationship to the base 2, which infers that hydrophobic interactions predominate in three-point interaction of enantiomers, leading to chiral separation even in the presence of the CM in the mobile phase. Moreover, faster elution was possible with large %B change (0.04%/m supporting the fact that reversed phase mode is followed in the presence of CMPA. On the o hand, selectivity (ɑ) is dependent on buffer and CMPA concentrations in first and second or (Equations (3) and (4)). Hence, careful selection of levels of these two factors is important for qu assurance of the separation, where distancing from the robust optimum values would lead dramatic drop in enantioresolution due to the presence of the quadratic terms.

HP-β-CD ( Figure 5A ) displays free rotating substituted hydroxy groups that promote str and stereospecific interaction via hydrogen bonding with the 3-hydroxy groups of TER, in addi to the inclusion of its aromatic ring into the CD cavity, leading to a hydrophobic interaction contrast, enhanced enantioseparation was only possible at moderate CMPA concentrations (4-7 m of HP-β-CD ( Figure 6A) , where higher concentrations showed decreased column effectiveness to increased column pressure. Despite the fact that the regression model of α with HP-β-CD did show any interaction terms between buffer concentration and CMPA concentration (Equation separation was only achieved at either very low or very high concentrations of buffer due to quadratic order of this factor (Equations (3) and (4)). Moreover, a wider range of CMPA concentra (3-8 mM) , concomitantly with the highest buffer concentration (45-50 mM), leads to accept separation. This may conclude the involvement of both stationary and mobile phases in w interaction and saturation of the stationary phase with HP-β-CD takes place at low bu concentration. On the other hand, at high buffer concentration and ionic strength (45-50 mM) ionized TER predominates in the CMPA rich mobile phase; thus, chiral recognition takes place both surface stationary phase and mobile phase. This also explains the improved resolution w lower retention time at high buffer concentrations when TER elutes with the HP-β-CD in the mo phase.

and TNF, and can promote autoimmune processes [32] . On the similar premise, Mormile hypothesized that infants who are non-responsive to HBV will be at risk for autoimmune disease later in life, and implicated SNPs in the genes for IL-18 and IFN-γ as possible markers of this latent autoimmunity [15] . The concept that certain HLA and HLA-associated alleles and haplotypes might predispose for nonresponse and autoimmunity after HBV vaccination is in part confirmed by a recent whole-genome transcriptomic study that showed that HBV vaccine non-responders had higher pre-vaccination expression levels of pro-inflammatory genes (especially those in the IFN pathways) compared to HBV vaccine responders [70] . More details of the descriptions of the genes associated or believed to be associated with HBV vaccine safety are given in Table 1 .

Fever or elevated body temperature is one of the most frequent adverse events of vaccines. In particular, febrile seizures have been associated with the live-virus vaccines, such as MMR. While numerous studies have investigated the genetic basis of MMR vaccine efficiency, adversomic MMR studies remain scarce. One of the most encompassing was by Feenstra et al., who carried out a series of GWAS that compared the genetic make-up of children who experienced febrile seizures after the MMR vaccine to children with vaccine-unrelated febrile seizures, and to controls without a history of febrile seizures of any kind [7] . Two SNPs in two genes associated with MMR-related febrile seizures were identified: IFI44L (rs273259: P = 5.9 × 10 −12 vs. controls; P = 1.2 × 10 −9 vs. MMR-unrelated febrile seizures) and CD46 (rs1318653: P = 9.6 × 10 −11 vs. controls; P = 1.6 × 10 −9 vs. MMR-unrelated febrile seizures) [7] . Interestingly, as mentioned earlier, both of these genes were associated with efficiency of seroconversion after the MMR vaccine [46, 47] . In the same study by Feenstra et al., three loci were associated with febrile seizures in general, as these loci differed between the controls and the MMR-vaccine-related seizures, as well as the controls and the MMR-vaccine-unrelated seizures, but not between the groups with MMR-vaccine-related and MMR-vaccine-unrelated seizures: SCN1A (rs6432860), SCN2A (rs3769955), and ANO3 (rs114444506) [7] . While SCN1A and SCN2A encode two subunits of a voltage-gated sodium channel and have already been associated with (febrile) seizures, ANO3 encodes one of the anoctamin proteins that form part of at least two Ca 2+ -activated chloride channels. Feenstra et al. also performed electrophysiological recordings on brain slices from wild-type and ANO3-null rats to investigate the mechanisms of ANO3 involvement in the genesis of febrile seizures. They reported that hippocampal neurons of ANO3-null rats showed increased excitability compared to neurons of wild-type rats at both normal and febrile body temperatures [7] .

In extremely rare cases, MMR vaccination can lead to severe or chronic infections in children, which sometimes includes measles inclusion-body encephalitis, with the associated mortality rates as high as 10% to 20% [71] . This typically occurs in immunocompromised individuals, and is mostly due to inborn errors of immunity. T cell deficiencies, such as severe combined immune deficiency, have been associated with severe outcomes after MMR vaccine administration. However, such cases are particularly rare in clinical practice, as most patients with severe combined immune deficiency are diagnosed very early in life, and before administration of their first MMR vaccine dose [71] . Interestingly, in DiGeorge syndrome, where patients have mild-to-moderate T cell lymphopenia but intact T cell function, no severe outcomes after MMR vaccination have been reported [71] .

On the other hand, inborn errors of type I IFN immunity might be more important in this clinical setting considering severe infection after the MMR vaccine, as these conditions have milder presentation and are usually diagnosed later in life. Interestingly, an ontology-based literature mining study identified IFN-γ as one of the top ranking genes in both vaccine and fever networks [72] . There have been a handful of case reports that have described mutations in genes in the IFN networks as a cause for severe adverse reactions to the MMR vaccine. In these case reports, mutations in the following genes have been associated with severe measles infection (and in some cases death) after MMR vaccine administration in otherwise healthy children without overt signs of immune deficiency: STAT1 [40] , STAT2 [41] [42] [43] , INFAR1 [35] , INFAR2 [36] , IRF7 [44] and IRF9 [45] . More detailed descriptions of the genes associated or believed to be associated with MMR vaccine safety are given in Tables 2-4 . Of note, the study that used ontology-based literature mining identified several other genes that have not yet been investigated in the scope of MMR vaccine adversomics as high ranking in both vaccine and fever networks (i.e., IL1B, TNF, IL-6, and CD8A), and in only fever networks (HSPA1A, NFKB1, IL-8, IL-2, MEFV, MAPK1, POMC, CD4, and IL-10), and in only vaccine networks (CSF2, IL7R, ERVWE1, APC, MC4R, IL1R1, and TLR2) [72] . In our opinion, these genes represent good candidates for future adversomic studies of the MMR vaccine.

The HPV vaccines are considered safe and efficient. Since the initial approval of Gardasil by the US Food and Drug Administration in 2006, Gardasil, Gardasil 9 and Cervarix have all been through extensive safety testing in many clinical trials, which have included more than 29,000, 15,000 and 30,000 females and males, respectively [73] . As with other vaccines, the commonly reported side effects are mild. However, there have been a few well-described clinical cases of exacerbation of autoimmunerheumatic conditions after HPV vaccination, with particular reference to manifestation of systemic lupus erythematosus, and systemic-lupus-erythematosus-like disease [74, 75] . The cases that have been reported have included females aged 13 to 32 years with a personal or family history of various autoimmunerheumatic conditions. This again highlights the importance of the genetic predisposition of the individual, with an environmental factor as a second hit to provoke autoimmune diseases. At the same time, it cannot be ignored that women with systemic lupus erythematosus show higher prevalence for HPV infection compared to healthy women, as the disease itself weakens the natural and humoral responses to infections [74, 76] . Risk-benefit considerations need to be taken into account by physicians for each individual case, based on the depth of the anamnesis, the family history, and the genetic background prior to vaccination.

The recent literature also includes reviews of 'HPV vaccination syndrome' as a model for pathogenesis for fibromyalgia [77] [78] [79] . Fibromyalgia is defined as chronic widespread pain, which is commonly accompanied by joint stiffness, fatigue and sleep disturbance, along with cognitive, autonomic and sensory dysfunction. GWAS studies have identified two nonsense mutations here [80, 81] : W32X in C110rf40 and Q100X in ZNFJJ. These have been associated with high levels of the cytokines that are involved in inflammation exacerbation.

It is now widely accepted that biological sex differences influence the type and strength of the immune response to a pathogen infection or in vaccination. Sex differences (e.g., sex, age, reproductive cycle, hormones, and genetic factors) affect all parts of the immune system-innate, humoral, and cellular response [82, 83] . In general, compared to men and relevant for vaccination, women have a higher antibody response, an increased number of B cells, an increased number and activity of Treg and Tc cells and an increased number of macrophages with a higher release of pro-inflammatory cytokines. Despite the qualitative and quantitative senescence of the immune system, these differences persist from childhood to old age. For many vaccines, including HBV, HPV studies in women show increased antibody titres and more adverse reactions than men [84] [85] [86] . Another reason for a more intense immune response in women is the X chromosome, which contains a high density of immune-related genes. Although the second copy of the X chromosome in women is inactivated through the mechanisms of epigenetics, parts of it can be reactivated or skewed X chromosome deactivation occurs [87, 88] . Since both topics are highly relevant for the efficacy and side effects of the vaccine, for further details please refer to Flanagan et. al. [82] and Klein et. al. [89] and the references contained therein.

Despite all this knowledge, not many clinical vaccination studies take immunological differences into account. Data are usually combined, analyses by sex and gender are not considered or are not applicable due to the inequality of the samples. A rational design was proposed by Klein S.L. and Pekosz A. for influenza vaccination, taking into account biological sex differences [90] . In practice, the ultimate goal would be to reduce inflammation and side effects in women while maintaining efficient seroconversion; reducing vaccine doses in women would also mean greater availability of the vaccine within the population. At present, such a "male and female" vaccination approach seems to be a long way off. Many scientists and clinicians agree that such a rational design could have added value in terms of efficacy and adverse events; we also agree that it would be an important step forward in the field of personalized medicine.

The lack of studies in the field of adversomics is at least in part due to the several challenges in the study design. The main problem that the researchers face is the question, which side effects of the vaccine are relevant to its safety and thus worth studying.

Mild side effects are not life threatening and can be an indicator of the efficient immune response to the vaccine, leading to the conclusion that they are not a priority in adversomic studies. On the other hand, even mild side effects can produce anxiety in patients and/or their parents and thus lead to the discontinuation of the vaccination, causing low vaccination rates. From this point of view, the determination of genetic markers of mild side effects of childhood vaccines might be rational. The ability to predict even mild side effects prior to vaccination might alleviate parental fears and lead to the increased trust in vaccination programmes and health system in general. Additional problem in studying mild side effects is that they tend to be underreported by health providers, leading to problems in cohort stratification. Therefore, in order to obtain relevant results the full cooperation of physicians in reporting even mild side effects of vaccinations is of the key importance.

On the other hand, serious side effects are usually promptly reported and the rationale for the identification of genetic markers predicting such side effects is clear. However, because serious side effects of vaccines are very rare in the population, the major challenge in such studies is collecting the sufficient number of cases to reach high enough statistical power. This might be overcame by using multicentric international approach, but unfortunately, different countries have heterogeneous systems and databases of reporting vaccine side effects, which leads to problems in the identification of the relevant cases. What is needed is the international database for vaccine side effects registration, which would enable relevant adversomic studies.

In conclusion, to enable the development of the field of vaccine adversomics the international database for reporting vaccine side effects accessible to researchers globally is needed, as well as full cooperation of health providers in reporting all vaccine side effects.

At least eight types of vaccines are under development against SARS-CoV-2: virus vaccines (inactivated, weakened); viral vector vaccines (replicating, non-replicating); protein-based vaccines (protein subunit, virus-like particles); and nucleic acid vaccines (DNA, RNA). As of 11 May 2020, eight SARS-CoV-2 candidate vaccines were being tested on humans, and among these, the 'genetic immunization' with DNA or RNA vaccines shows potentially the most promising results. This approach is new for human use, and although there are several (non-COVID-19) DNA vaccines available for veterinary use, none of these have been approved for humans. However, important design knowledge has been obtained from several previous threats of such virus This allowed DNA (Moderna) and RNA (mRNA; Inovio) SARS-CoV-2 vaccines to already be tested in clinical trials in April 2020. DNA-based and RNA-based vaccines should overcome some important safety worries and unwanted side effects as these do not contain any virus; instead, they include only the genetic material of the virus immunogenic protein (e.g., the 'spike protein' of SARS-CoV-2), which is inserted into human cells. After successful uptake of DNA, the human cell produces viral proteins that prompt an immune response, and thus protect against infection. Yu et al. have recently published promising results for their SARS CoV-2 DNA vaccine testing on 35 rhesus macaques [91] . The most significant reduction in viral replication was observed in the upper and lower respiratory tract, with full-length S immunogen after three doses of the vaccine (zero, one, and five weeks). The vaccinated animals developed humoral and cellular responses that included neutralizing antibody titres with comparable magnitude to the antibody titres in convalescent macaques and convalescent humans. A recent study on 149 COVID-19 convalescent individuals has also reported encouraging results; namely, even though the plasma titres of the neutralizing antibodies against the receptor binding domain of the SARS-CoV-2 spike protein were relatively low in most patients, they were present in all of the individuals tested, which suggests that a vaccine designed to elicit these antibodies should be broadly effective [92] .

In our opinion, personalized approaches to SARS-CoV-2 vaccine development will be even more important than for other vaccines, as profound differences in the patient responses to SARS-CoV-2 infection have been reported since the early stages of the COVID-19 pandemic. While in most infected individuals relatively mild respiratory symptoms have been noted, a certain subpopulation of patients undergoes a severe cytokine storm that can lead to multi-organ failure and high levels of mortality. Several environmental and demographic factors have been attributed to these interindividual differences in COVID-19 severity, but there are some indications that genetic factors have an important role here. Considering SARS-CoV-2 vaccine development, it should be noted that patients who have severe reactions to the infection might be more at risk for adverse reactions to the vaccine (i.e., cytokine storm). Conversely, individuals with asymptomatic presentation might be potential vaccine non-responders. Thus, the genetic markers identified as predictors of COVID-19 severity should be taken into account during the vaccine development and its later administration.

In May 2020, an important international initiative called the 'COVID-19 Host Genetics Initiative' was started [93] , which aims to determine host genetic factors that can explain interindividual differences in responses to SARS-CoV-2 infection, with particular focus on HLA genes. At time of writing (end of September 2020), 216 studies had joined the initiative, which is both retrospective and prospective in nature, and mostly centred on Europe and USA, with participation still expanding. Retrospective collections are typically biobanks that already include significant genetic data and active connections to health systems, while prospective collections have recently started to directly obtain consent from incoming COVID-19 patients. Retrospective collections enable rapid development of genetic studies on susceptibility and severity, while prospective approaches bring important additional opportunities not only for deeper DNA studies, but also for potentially informative viral and antibody profiling and epitope mapping experiments that can be implemented in vaccine development [93] .

The angiotensin-converting enzyme-2 gene, ACE2, has been hypothesized to have an important role in COVID-19 severity, as it was identified early on as a facilitator of the SARS-CoV-2 virus entry into cells. The ACE2 product, angiotensin-converting enzyme-2, has been shown to be the receptor for both SARS-CoV-2 virus and the NL63 human respiratory coronavirus. A recent study analyzed all of the 1700 variants in the ACE2 gene from the China Metabolic Analytics Project and 1000 Genomes Project databases. No direct evidence was found for coronavirus S-protein binding-resistant ACE2 mutants in different populations [94] . However, East Asian populations have much higher allelic frequencies of genetic variants associated with higher ACE2 expression in tissues, which indicates differential susceptibility to SARS-CoV-2 in different populations [94] .

In contrast, an agnostic approach has revealed several genes of interest that might have roles in SARS-CoV-2 infection responses. A meta-analysis of two GWASs that involved patients with severe COVID-19 (as 835 and 775 individuals in the Italian and Spanish cohorts, respectively) and mostly non-infected controls from the general population (as 1255 and 950 individuals in the Italian and Spanish cohorts, respectively), revealed two critical loci that were responsible for severe COVID-19 presentation: 3p21.31 and 9q34.2. The 9q34.2 locus is associated with the ABO blood groups, which thus confirms the previous anecdotal observations of differential SARS-CoV-2 responses in individuals with different blood groups. In this meta-analysis, blood group A was associated with more severe disease, while blood group 0 was protective. However, the latest multi-institutional study on 7648 COVID-19-positive samples did not support the association of blood type with risk of intubation or death in COVID-19 patients, although it confirmed that blood type O was less common for SARS-CoV-2 infection [95] . Considering the first locus, 3p21.31, six genes have been defined for this area: SLC6A20, LZTFL1, CCR9, FYCO1, CXCR6, and XCR1 [96] , some of which appear directly or indirectly connected to SARS-CoV-2 responses (Figure 1 ). SLC6A20 encodes sodium-proline transporter 1, which can functionally interact with ACE2, the SARS-CoV-2 cell-surface receptor. CCR9, XCR1 and CXCR6 encode chemokine receptors, the last of which regulates the specific location of lung-resident memory CD8 T cells throughout sustained immune responses to airway pathogens [96] . The functions of these other two genes at the 3p21.31 locus include protein trafficking (LZTFL1) and autophagy (FYCO1). Previously mentioned sex differences in immune response and hormonal status should not be neglected also in the infection with SARS-CoV2. In many states, reporting sex-disaggregated data, COVID-19 with severe disease symptoms and higher death rates are seen among men [97, 98] . Recently, Takahashi T et al. provided us with findings, that poor T cell response is associated with worse disease outcome in male patients, but not in female patients. Additionally, higher levels of innate immune cytokines (Il-8 and IL-18) at the disease baseline and during the course of the disease (chemokine CCL5) were associated with worse disease progression in males. Results of this study have shown equality in humoral but diversities in cellular immune response as well as the levels of the effector molecules.

The authors suggested that vaccines and therapies increasing T cell immune response to SARS-CoV-2 might be appropriate for male patients, and those dampening the innate immune activation in the early stage of the disease for females [99] .

Vaccination has been one of the most successful public health interventions in human history, as it continues to save millions of lives every year. However, the effectiveness and safety of some vaccines is limited by the heterogeneity of the responses among different individuals and across different populations. Heterogeneity of vaccine responses is a multifactorial trait, which results from a combination of environmental and genetic factors. Several environmental factors that contribute to this variation have been identified, such as age, gender, ethnicity, body-mass index and health, and smoking status, as well as dose, route of administration and quality of storage of the vaccine. Twin studies have also identified high levels of heritability of vaccine responses. Thus, identification of the genetic factors that modulate responses to vaccine administration is of the utmost importance for the development of safe and effective vaccines, to increase the public trust in vaccines, and consequently to increase the background vaccination rates. 

Precision Medicine Initiative

Application of Pharmacogenomics to Vaccines

From Genomics to Vaccines: Malaria as a Model System

Adversomics, and the Immune Response Network Theory: Individualized Vaccinology in the 21st Century

Personalized Vaccinology: A Review

Adversomics: A New Paradigm for Vaccine Safety and Design

Common Variants Associated With General and MMR Vaccine-Related Febrile Seizures

Vaccines: The Real Issues in Vaccine Safety

Adverse Events and Vaccination-the Lack of Power and Predictability of Infrequent Events in Pre-Licensure Study

Pandemic Influenza a H1N1 Vaccines and Narcolepsy: Vaccine Safety Surveillance in Action

Hepatitis B Vaccination: The Key Towards Elimination and Eradication of Hepatitis B

Recent Advances in Vaccination of Non-Responders to Standard Dose Hepatitis B Virus Vaccine

Hepatitis B Vaccine Ineffective in Most Patients with RA

Hepatitis B Vaccine in Celiac Disease: Yesterday, Today and Tomorrow

Hepatitis B Vaccine Non Response: A Predictor of Latent Autoimmunity

Genetic Regulation of Immune Responses to Vaccines in Early Life

Genetic Prediction of Nonresponse to Hepatitis B Vaccine

The Major Histocompatibility Complex: The Value of Extended Haplotypes in the Analysis of Associated Immune Diseases and Disorders

Differential Genetic Determination of Immune Responsiveness to Hepatitis B Surface Antigen and to Hepatitis a Virus: A Vaccination Study in Twins

Response to Hepatitis B Vaccine: Multiple HLA Genes Are Involved

Hepatitis B Surface Antigen Presentation and HLA-DRB1*-Lessons from Twins and Peptide Binding Studies

New Genetic Associations Detected in a Host Response Study to Hepatitis B Vaccine

A Genome-Wide Association Study Identifies Polymorphisms in the HLA-DR Region Associated with Non-Response to Hepatitis B Vaccination in Chinese Han Populations

SNP rs7770370 inHLA-DPB1 Loci as a Major Genetic Determinant of Response to Booster Hepatitis B Vaccination: Results of a Genome-Wide Association Study

Key HLA-DRB1-DQB1 Haplotypes and Role of the BTNL2 Gene for Response to a Hepatitis B Vaccine

C4 A Deficiency and Nonresponse to Hepatitis B Vaccination

Toll-Like Receptors and Cytokines/Cytokine Receptors Polymorphisms Associate With Non-Response to Hepatitis B Vaccine

Consistency of HLA Associations Between Two Independent Measles Vaccine Cohorts: A Replication Study

Vaccinomics: Current Findings, Challenges and Novel Approaches for Vaccine Development

No Evidence of Autoimmunity in 6-Year-Old Children Immunized at Birth with Recombinant Hepatitis B Vaccine

HypothesisConcurrent HLA-Related Response Factors Mediate Recombinant Hepatitis B Vaccine Major Adverse Events

Human Genetic Variants and Age Are the Strongest Predictors of Humoral Immune Responses to Common Pathogens and Vaccines

Associations Between Cytokine/Cytokine Receptor Single Nucleotide Polymorphisms and Humoral Immunity to Measles, Mumps and Rubella in a Somali Population

Inherited IFNAR1 Deficiency in Otherwise Healthy Patients With Adverse Reaction to Measles and Yellow Fever Live Vaccines

Human IFNAR2 deficiency: Lessons for antiviral immunity

Effects of Vitamin A and D Receptor Gene Polymorphisms/Haplotypes on Immune Responses to Measles Vaccine

Associations Between Polymorphisms in the Antiviral TRIM Genes and Measles Vaccine Immunity

Polymorphisms in the Wilms Tumor Gene Are Associated With Interindividual Variations in Rubella Virus-Specific Cellular Immunity After Measles-Mumps-Rubella II Vaccination

A Novel Presentation of Homozygous Loss-of-Function STAT-1 Mutation in an Infant with Hyperinflammation-A Case Report and Review of the Literature

STAT2 Deficiency and Susceptibility to Viral Illness in Humans. Proc. Natl. Acad. Sci

A Novel Kindred with Inherited STAT2 Deficiency and Severe Viral Illness

Signal Transducer and Activator of Transcription 2 Deficiency Is a Novel Disorder of Mitochondrial Fission

Life-Threatening Influenza and Impaired Interferon Amplification in Human IRF7 Deficiency

Life-Threatening Influenza Pneumonitis in a Child With Inherited IRF9 Deficiency

Genome-Wide Associations of CD46 and IFI44L Genetic Variants with Neutralizing Antibody Response to Measles Vaccine

The Association of CD46, SLAM and CD209 Cellular Receptor Gene SNPs with Variations in Measles Vaccine-Induced Immune Responses: A Replication Study and Examination of Novel Polymorphisms

The Role of Polymorphisms in Toll-Like Receptors and Their Associated Intracellular Signaling Genes in Measles Vaccine Immunity

Twin Studies of Immunogenicity-Determining the Genetic Contribution to Vaccine Failure

Heritability of Vaccine-Induced Measles Neutralizing Antibody Titers

A Diverse Range of Gene Products Are Effectors of the Type I Interferon Antiviral Response

Vitamin Effects on the Immune System: Vitamins A and D Take Centre Stage

TRIM Family Proteins: Retroviral Restriction and Antiviral Defence

Whole Transcriptome Profiling Identifies CD93 and Other Plasma Cell Survival Factor Genes Associated with Measles-Specific Antibody Response after Vaccination

Differential miRNA expression in B cells is associated with inter-individual differences in humoral immune response to measles vaccination

Variability in Humoral Immunity to Measles Vaccine: New Developments

Defining the Genetic Susceptibility to Cervical Neoplasia-A Genome-Wide Association Study

Variants in Immune-Related Genes and Genital HPV 16 Persistence in Men

Association of TGF-Beta1 Genetic Variants with HPV16-Positive Oropharyngeal Cancer

TGFbeta Receptor 1: An Immune Susceptibility Gene in HPV-Associated Cancer

Personalized Human Papillomavirus Vaccination for Persistence of Immunity for Cervical Cancer Prevention: A

Council for International Organisations of Medical Sciences, World Health Organisation

Safety of Vaccines Used for Routine Immunization of U.S. Children: A Systematic Review

Immunization Safety Review: Hepatitis B Vaccine and Demyelinating Neurological Disordres. In Immunization Safety Review: Hepatitis B Vaccine and Demyelinating Neurological Disordres

Vaccines and Multiple Sclerosis: A Systematic Review

Could Autoimmunity Be Induced by Vaccination?

Anti-Phospholipid Antibodies Following Vaccination with Recombinant Hepatitis B Vaccine

Environmental Exposure and Vaccination: Is There a Link? Toxicology

Regulatory T Cell Frequencies and Phenotypes Following Anti-Viral Vaccination

Pre-Vaccination Inflammation and B-Cell Signalling Predict Age-Related Hyporesponse to Hepatitis B Vaccination

Life-Threatening Infections Due to Live-Attenuated Vaccines: Early Manifestations of Inborn Errors of Immunity

Identification of Fever and Vaccine-Associated Gene Interaction Networks Using Ontology-Based Literature Mining

Human Papilloma Virus-HPV Vaccination Is Safe and Effective

Are Women with Lupus at Higher Risk of HPV Infection? Lupus

Systemic Lupus Erythematosus Following HPV Immunization or Infection? Lupus

Corrigendum to 'Prevalence of Cervical HPV Infection in Women with Systemic Lupus Erythematosus: A Systematic Review and Meta-Analysis

Autonomic Dysfunction and HPV Immunization: An Overview

HPV Vaccination Syndrome: A Clinical Mirage, or a New Tragic Fibromyalgia Model

What Role for Autoimmunity, Neuroinflammation, and Small Fiber Neuropathy in Fibromyalgia, Chronic Fatigue Syndrome, and Adverse Events after Human Papillomavirus Vaccination?

Genome-Wide Expression Profiling in the Peripheral Blood of Patients with Fibromyalgia

Carcinogenic Risk of Human Papillomavirus (HPV) Genotypes and Potential Effects of HPV Vaccines in Korea

Sex and Gender Differences in the Outcomes of Vaccination over the Life Course

Sex Differences in Immune Responses

Hepatitis B Vaccination Associated with Higher Female Than Male Mortality in Guinea-Bissau: An Observational Study

Effect of Gender on Reporting of MMR Adverse Events in Saudi Arabia

The Prevalence of Hepatitis B Virus Infection Markers and Socio-Demographic Risk Factors in HIV-Infected Patients in Southern Brazil

The X Cromosome and Immune Associated Genes

The X Chromosome and Sex-Specific Effects in Infectious Disease Susceptibility

The Xs and Y of Immune Responses to Viral Vaccines

Sex-Based Biology and the Rational Design of Influenza Vaccination Strategies

DNA Vaccine Protection Against SARS-Cov-2 in Rhesus Macaques

Convergent antibody responses to SARS-CoV-2 in convalescent individuals

A Global Initiative to Elucidate the Role of Host Genetic Factors in Susceptibility and Severity of the SARS-Cov-2 Virus Pandemic

Comparative Genetic Analysis of the Novel Coronavirus (2019-nCov/SARS-Cov-2) Receptor ACE2 in Different Populations

Blood type and outcomes in patients with COVID-19

Genomewide Association Study of Severe Covid-19 with Respiratory Failure

The COVID-19 Sex-Disaggregated Data Tracker

Considering How Biological Sex Impacts Immune Responses and COVID-19 Outcomes

Sex Differences in Immune Responses That Underlie COVID-19 Disease Outcomes

The authors acknowledge Chris Berrie for scientific editing of the manuscript.

The authors declare that they have no conflicts of interest. The sponsors had no role in the design, execution, interpretation, or writing of the study.