Advances in vaccine adjuvant development and future perspectives

Abstract

Use of highly purified antigens to improve vaccine safety has led to reduced immunogenicity and efficacy, resulting in the need for adjuvants to increase and/or modulate the immunogenicity of the vaccine. Despite the need for potent and safe vaccine adjuvants, currently, there are still very few adjuvants in licensed human vaccines. Advances in immunology and molecular biology, especially in the last decade, have allowed researchers to understand better how the adjuvants work and enhance immune responses. While aluminum salts are still the most widely used adjuvants, research has shifted toward the rational design of adjuvant systems containing immunostimulatory molecules. Application of systems biology, which is based on high-throughput technologies using mathematical and computational modeling, has provided a deeper understanding of the biological events elicited by vaccination as well as the influence of other factors such as sex, age, microbiota, genetics and metabolism on the immune response. By this means, it became possible to tailor potential vaccine adjuvants more precisely for a successful vaccine with enhanced efficacy, safety and protection. In this review, after describing the mechanism of action of the adjuvants, current adjuvants in licensed vaccines, as well as those under clinical development will be mentioned in detail. Finally, new approaches in vaccine adjuvant development using systems biology and artificial intelligence will be reviewed, and future directions in vaccine research in regard to efficacy, safety and quality aspects will be discussed.

1. Introduction

During the last decades, highly purified recombinant proteins or protein subunits, synthetic peptides have been developed as vaccine antigens for safety reasons. However, this resulted in poor immunogenicity, producing rather mild and short immune responses. Therefore, adjuvants are required to boost the potency and longevity of the specific immune response to these antigens. The ideal adjuvant is expected to improve the vaccine immunogenicity without compromising tolerability or safety. Adjuvants have been categorized in several ways according to their origin (e.g. natural, synthetic), physical and chemical characteristics (e.g. size, charge, morphology, etc.) and mode of action (Arca et al. 2009; Reed et al. 2016; Garçon and Di Pasquale 2017; Facciolà et al. 2022; Singleton et al. 2023; Zhao et al. 2023).

The common adjuvants under investigation or included in licensed vaccines have been classified by the World Health Organization (WHO) (WHO Expert Committee on Biological Standardization 2014) into several categories. These include mineral salts or gels (e.g. aluminum hydroxide, aluminum ­phosphate gels); emulsions (e.g. oil-in-water and water-in-oil); amphiphilic molecules and surfactant-based formulations; particulate adjuvants (e.g. liposomes, virosomes, immune-stimulating complexes named as ISCOMS); pathogen-associated molecular patterns (natural and synthetic) (e.g. monophosphoryl lipid A, synthetic acylated monosaccharide RC-529, CpG motifs); endogenous human immunostimulators (e.g. cytokines); inert vehicles (e.g. gold nanoparticles); inulin-based adjuvants, and combination adjuvants or adjuvant systems (e.g. AS01, AS02, CAF01). In general, adjuvants can be broadly divided into two groups as immunostimulators and delivery systems. Delivery systems such as mineral salts, emulsions, liposomes, virosomes and immune stimulating complexes (ISCOM, ISCOMATRIX) present vaccine antigens to the immune system and increase the specific immune response to the antigen. The immunostimulators (e.g. TLR ligands, cytokines, saponins and bacterial exotoxins) directly act on the immune cells to increase responses to antigens. Studies have shown that delivery systems such as liposomes can also generate immunostimulation (Perrie et al. 2016; Feather et al. 2022). Furthermore, immunostimulators are combined with delivery systems to enhance the efficacy and/or safety of the antigen (Brito and O’Hagan 2014; Pirahmadi et al. 2021; Pulendran et al. 2021).

When we look at the history of adjuvants, the first evidence of an adjuvant was proposed by Ramon in the 1920s to indicate substances such as breadcrumbs, agar, tapioca flour, starch oil or saponin that he was using in addition to diphtheria or tetanus toxin to increase the antibody titers (Ramon 1925). This was further confirmed by Glenny reporting that aluminum salts, when mixed with antigens could improve the immune responses (Glenny et al. 1926, 1931). From 1930 through the end of 1990, aluminum was the only adjuvant that was allowed in human vaccines. The oil-in-water emulsion, MF59 was the first new adjuvant approved after aluminum for human use in 1997 (Singh and O’Hagan 1999). It was licensed initially for influenza vaccine in Europe. In the following 20 years, adjuvants systems were developed by combining approved adjuvants, such as aluminum salts, oil-in-water emulsions or liposomes with immunostimulatory molecules that have been demonstrated to have an impact on innate and adaptive immune responses. The search for new adjuvants was initially focused on immunostimulatory molecules derived from microorganisms, specifically the lipopolysaccharide (LPS) derivatives, as well as the saponin derivative Quillaja saponaria-21 (QS21). Among the LPS-derived immunomodulators, 3-0-descyl-4′-monophosphoryl lipid A (MPL) was found to be the most suitable one. The adjuvant systems based on MPL absorbed on aluminum hydroxide or aluminum phosphate were tested in vaccines against different pathogens. The adjuvant systems, including AS04, AS03, AS01 allowed for tailored immune responses adapted to the pathogens and to the target populations. In fact, for a successful adjuvant system with high efficacy and safety, it is important to find the best combination in which each part synergizes with the other to drive a more suitable immune response. In the following sections, examples of combined adjuvant systems, which are currently used or under development will be described in detail.

2. Mechanism of action of adjuvants

Over the years, various mechanisms of actions have been proposed for adjuvants in inducing the immune responses. In early years of the adjuvant studies, little was known about the mechanisms underlying the adjuvant activity, and the adjuvants were even described as “the immunologist”s dirty little secrets’ by Janeway (1989). He proposed that the immune system involved recognition of not only specific antigenic determinants, but also of certain characteristics or patterns common on infectious agents and absent in the host. The further revolutionary works of Akira, Beutler, Hoffmann, and Medzhitov have brought new insights into the immune system (Medzhitov 2001; Hoffmann 2003; Beutler 2009; Kawai and Akira 2010). Beutler and Hoffmann were honored with the 2011 Nobel Prize in Physiology or Medicine for their discoveries on Toll and Toll-like receptor (TLR) activation of innate immunity in fruit fly and mammals. The prize was shared with Steinman in recognition to his efforts in bridging the innate and adaptive systems through his discovery of dendritic cells (Steinman and Cohn 1973). These discoveries provided important milestones in understanding immunology through the recognition of the cellular elements involved in immunity, elucidation of the mechanisms of immune diversity as well as the coordinated immunologic network. This network ensures that when there is an invasion by the pathogen, it is recognized by innate immune system via germline-encoded pathogen-associated molecular patterns (PAMPs) on the pathogen surface. These PAMPs serve as ligands for a broad class of proteins on the immune cells’ surfaces, referred to as pattern recognition receptors (PRRs). Hence, PAMPs are critical in initiating innate immune responses and inducing and directing subsequent adaptive immunity. Several classes of PRRs recognize distinct microbial components and directly activate immune cells. PRRs can also recognize molecules (i.e. proteins, metabolites) released by damaged cells, called as damage-associated molecular patterns (DAMPs) (Schaefer 2014). PRR-induced signals control the adaptive immunity at multiple checkpoints and manage the initiation, magnitude, duration and type of the response, as well as the production of long-term memory (Palm and Medzhitov 2009; Bai et al. 2020). The identification of these pathways has provided new insights into the mechanisms of vaccine adjuvants and since then, studies on adjuvant development have been accelerated and several new adjuvants based on PRRs have been developed that are aimed to instruct the adaptive immune system to respond to a particular infection in the most effective way.

Most PRRs are classified into main families consisting of Toll-like receptors (TLRs) and C-type lectin receptors (CLRs), nucleotide oligomerization domain (NOD)-like receptors (NLRs), retinoic acid-inducible gene-I (RIG-I)-like receptors (RLRs), and AIM2-like receptors (ALRs) (Girardin et al. 2003; Schoenen et al. 2010; Brubaker et al. 2015; Probst et al. 2017; Li and Wu 2021). PRRs show difference in terms of their ability to trigger the adaptive immune response. The largest group of PRRs explored so far are TLRs. They are classified into two subfamilies based on their localization, cell surface TLRs and intracellular (endosomal) TLRs (Hemmi et al. 2000; Kawai et al. 2024). TLRs located on the surface of cell (e.g. TLR1, TLR2, TLR4, TLR5, TLR6, TLR 11) primarily identify components such as protein, lipids and lipoproteins found in microbial membranes. Microbial nucleic acids are recognized by the other category of TLRs (e.g. TLR3, TLR7, TLR8, TLR9), which are located intracellularly (e.g. endoplasmic reticulum, endosomes, lysosomes, endolysomes) (Kawai and Akira 2010).

The molecules that target TLRs have been investigated as vaccine adjuvants (Duthie et al. 2011; Reed et al. 2016; Pulendran et al. 2021; Kaur et al. 2022). These TLR agonists have been combined also with other known adjuvants such as aluminum and saponin to obtain adjuvant systems with synergistic or modulating effects, thus making possible to tailor an adjuvant with specific immune effects. MPL, which is a TLR4 agonist, has the longest history of clinical development among the TLRs. Currently, some of these adjuvant systems have been licensed with the antigen product (Table 1) or are under clinical trials (Table 2).

Table 1.

Adjuvants used in licensed human vaccines.

Adjuvant	Composition	Vaccine
Aluminum	Amorphous aluminum hydroxy phosphate sulfate (AAHS), aluminum hydroxide, aluminum phosphate, potassium aluminum sulfate (alum)	Several inactive vaccines including diphtheria-tetanus-containing vaccines, Hepatitis A, Hepatitis B, Haemophilus influenzae type b, pneumococcal vaccines, COVID-19 vaccines
MF59 MF59.C1 SQBA	Oil-in-water emulsion containing squalene, polysorbate (Tween 80), sorbitan trioleate, sodium citrate, citric acid	Seasonal influenza vaccines(Fluad and Fluad Quadrivalent) COVID-19 (Bimervax)
AS01E TLR4 agonist	Monophosphoryl lipid A (MPL) and QS-21, a natural compound extracted from the Chilean soapbark tree, combined in a liposomal (dioleoyl phosphatidylcholine (DOPC) and cholesterol) formulation	Respiratory syncytial virus (RSV) vaccine (recombinant adjuvanted) (Arexvy) RTS,S/AS01 malaria vaccine (Mosquirix™)
AS01B contains 2-fold more MPL and QS-21 than AS01E TLR4 agonist		Zoster vaccine (Shingrix)
AS04 TLR4 agonist	Monophosphoryl lipid A (MPL) + aluminum salt	Human papillomavirus (Types 16, 18) (Recombinant, adjuvanted, adsorbed) (Cervarix)
		Hepatitis B (rDNA) vaccine (adjuvanted, adsorbed) (Fendrix)
Matrix-M™	40 nm particles based on saponin extracted from the soap bark tree (Quillaja saponaria Molina) Matrix-A™ and Matrix-C™ containing different Q. saponaria saponin fractions (85% Matrix-A + 15% Matrix-C)	COVID-19 vaccine (Novavax) RS21/Matrix-M vaccine for malaria
CpG 1018 TLR9 agonist	B-Class 22-mer phosphothioate oligonucleotide (PS-ODN) comprising microbial DNA-like unmethylated CpG motifs	HepB (Heplisav-B)
		COVID-19 (Valneva VLA2001) Not available on the market anymore
CpG 7909 TLR9 agonist	B-Class CpG ODN based on synthetic DNA molecule, 24 nucleotides in length with a nuclease resistant phosphorothioate backbone	Anthrax vaccine (Cyfendus)
Algel-IMDG TLR7/8 agonist	Aluminum hydroxide-adsorbed imidazoquinolin	COVID-19 vaccine (BBV152 COVAXIN)
RC-529 TLR4 agonist	Synthetic glycolipid (fully synthetic monosaccharide mimetic of monophosphoryl lipid A) adsorbed to aluminum	Hepatitis B vaccine (Supervax) Licensed in Argentina
IRIVs	Spherical, unilamellar vesicles with a mean diameter of 150 nm containing the hemagglutinin (HA) and neuraminidase (NA) envelope glycoproteins of influenza virus	Hepatitis A vaccine (Epaxal) Influenza vaccine (Inflexal V and Invivac) Not available on the market anymore
AS03	Oil-in-water emulsion containing α-tocopherol, squalene, and polysorbate (Tween 80)	Pandemic influenza vaccine Not available on the market anymore

CpG ODN: cytosine-phosphoguanine oligodeoxynucleotides; IRIVs: immunopotentiating reconstituted influenza virosomes; TLR: Toll-like receptors.

Table 2.

New adjuvants in vaccines for infectious diseases under clinical trials for (registered in ClinicalTrials.gov).

Adjuvant	ClinicalTrials.gov ID Trial title Sponsor	Study aim	Phase	Status
Advax-CpG55.2 TLR 9 agonist	NCT06355232 Covid-19 and influenza oral vaccine study Vaxine Pty Ltd	To assess the effectiveness of protein-based COVID-19 or influenza vaccines when given individually or together via oral/ sublingual mucosal route instead of intramuscular delivery by comparing with a seasonal influenza vaccine which will also be administered with Advax-CpG adjuvant via the oral route	Phase I	Recruiting
ALFQ TLR4 agonist	NCT05423418 Safety, tolerability, and immunogenicity of ALFQ in a HIV vaccine containing A244 and B.65321 in healthy adults (RV575) U.S. Army Medical Research and Development Command	To evaluate the safety and tolerability (including reactogenicity) of candidate vaccine A244/B.63521 with Army Liposome Formulation (ALF) mixed with the saponin QS-21 (Quillaja saponaria-21) (ALFQ) adjuvant	Phase I	Recruiting
ALFQ TLR4 agonist	NCT04658667 HIV vaccine in HIV-uninfected adults (RV546) U.S. Army Medical Research and Development Command	To define the safety and immunogenicity of IHV01 and A244/AHFG with and without ALFQ at a full dose and at a fractional dose (one-fifth of a full dose) in a late boost setting for participants who had previously received a late boost of AIDSVAX®B/E with or without ALVAC in RV306	Phase I	Recruiting
ALFQ TLR4 agonist	NCT05500417 Safety and immunogenicity of CJCV2 with and without ALFQ National Institute of Allergy and Infectious Diseases (NIAID)	To evaluate the safety and immunogenicity of an intramuscular Campylobacter Jejuni conjugate vaccine (CJCV2) with and without army liposome formulation containing QS-21 (ALFQ) in a first-in-human trial	Phase I	Completed
Aluminum Hydroxide/LHD153R (AS37) TLR7 agonist	NCT02639351 Safety and immunogenicity of an aluminum hydroxide/LHD153R adjuvanted Meningococcal C-CRM197 conjugate vaccine GlaxoSmithKline	To evaluate the safety and immunogenicity of an aluminum hydroxide/LHD153R adjuvanted Meningococcal C-CRM197 conjugate vaccine compared to aluminum hydroxide adjuvanted Meningococcal C-CRM197 conjugate vaccine in healthy adults (18–45 years of age)	Phase I	Completed
AS37 TLR7 agonist	NCT05561673 A study on the safety and immune response of AS37 together with Hepatitis B antigen in adults aged 18–45 years GlaxoSmithKline	To assess the safety and immunogenicity of AS37 in combination with the Hepatitis B surface antigen (HBAg)	Phase I/IIa	Completed
CAF01	NCT00922363 Trial on the safety of a new liposomal adjuvant system, CAF01, when given with the tuberculosis subunit vaccine Ag85B-ESAT-6 as two injections with two months interval to healthy adult volunteers Statens Serum Institut	To evaluate the safety profile of CAF01, administering 50 µg Ag85B-ESAT-6 alone and 50 µg Ag85B-ESAT-6 with three escalating CAF01 dose levels, in healthy volunteers	Phase I	Completed
CAF01	NCT02787109 Safety of chlamydia vaccine CTH522 in healthy women aged 18 to 45 years Statens Serum Institut	To evaluate the safety of SSI’s adjuvanted chlamydia vaccine CTH522 in healthy women aged 18 to 45 years in a first in human trial	Phase I	Completed
CAF01 CAF09b TLR3 agonist	NCT03926728 Safety and immunogenicity of a chlamydia vaccine CTH522 Statens Serum Institut	To evaluate the safety and immunogenicity of a chlamydia vaccine, CTH522 non-adjuvanted or adjuvanted with Cationic Adjuvant Formulation in healthy adults	Phase I	Completed
CAF10b TLR3 agonist	NCT06050356 First-in-human trial of the novel tuberculosis vaccine candidate, H107e/CAF 10b (nTB-01) Statens Serum Institut	To evaluate the safety, reactogenicity, and immunogenicity of the tuberculosis subunit vaccine H107e/CAF®10b in adults	Phase I	Recruiting
CpG 1018 TLR9 agonist 3M-052-AF TLR7/8 agonist	NCT06334393 Phase 1 trial to assess the safety and immunogenicity of an inactivated, adjuvanted whole Zika virus vaccine candidate (VLA1601) in healthy adults Valneva Austria GmbH	To assess the safety and immunogenicity of an inactivated whole Zika virus vaccine candidate	Phase I	Active, not recruiting
CpG 1018 TLR9 agonist GLA-LSQ TLR4 agonist 3M-052-AF TLR7/8 agonist Aluminum hydroxide	NCT04177355 Evaluating the safety and immunogenicity of HIV-1 BG505 SOSIP.664 gp140 with TLR agonist and/or alum adjuvants in healthy, HIV-uninfected adults National Institute of Allergy and Infectious Diseases (NIAID)	To evaluate the safety and immunogenicity of HIV-1 BG505 SOSIP.664 gp140 with TLR agonist and/or alum adjuvants and VRC HIV Env Trimer 4571 and 3 M-052-AF with alum in healthy, HIV-uninfected adults	Phase I	Completed
3M-052-AF/ Alum  TLR7/8 agonist	NCT06332339 Clinical trial to evaluate the safety, tolerability, and immunogenicity of 16055 NFL Delta Gly4 Env protein trimer and trimer 4571 combined with 3 M-052-AF + alum adjuvant and Ad4-Env145NFL viral particles as heterologous prime-boost regimens in adult participants without HIV National Institute of Allergy and Infectious Diseases (NIAID)	To evaluate the safety, tolerability, and immunogenicity of trimer vaccine with 3M-052-AF + alum adjuvant and Ad4-Env145NFL viral particles as heterologous prime-boost regimens in adult participants without HIV	Phase I	Active, not recruiting
dmLT (LT(R192G/L211A)	NCT05961059 InvaplexAR-Detox and DmLT adjuvant in the Netherlands and Zambia (SUNSHINE) Leiden University Medical Center	To test the safety and efficacy of a new Shigella vaccine (InvaplexAR-DETOX) in combination with a new adjuvant (dmLT) in healthy participants	Phase I	Recruiting
dmLT	NCT01739231 Live attenuated ETEC vaccine ACE527 with and without dmLT adjuvant in adults PATH	To evaluate the safety, immunogenicity and efficacy of a live attenuated ETEC vaccine, ACE527, with and without a mucosal adjuvant, dmLT in healthy adult volunteers	Phase I and II	Completed
Entolimod TLR5 agonist	NCT03063736 Entolimod, an Adjuvant for Vaccine Augmentation Baylor College of Medicine	To evaluate the safety and surrogate efficacy of low dose entolimod in normal, healthy, non-patient subjects by comparing anti-tetanus (TT) antibody levels between patients getting entolimod vs no entolimod combined with tetanus-diphtheria (Td) vaccine	Pilot study	Completed
GLA-AF TLR4 agonist	NCT01385189 Safety and immunogenicity of a human hookworm candidate vaccine with different doses of a novel adjuvant Baylor College of Medicine	To evaluate the safety, reactogenicity, and immunogenicity of Na-GST-1 adsorbed to Alhydrogel® with or without two different dose concentrations of a novel adjuvant, GLA-AF (1 µg or 5 μg) among healthy adult volunteers	Phase I	Completed
GLA-SE MPL-SE TLR4 agonist	NCT01751048 LEISH-F3 + GLA-SE and the LEISH-F3 + MPL-SE vaccine National Institute of Allergy and Infectious Diseases (NIAID)	To evaluate the safety, tolerability, and immunogenicity of the vaccine candidates LEISH-F3 + GLA-SE, LEISH-F3 + MPL-SE, and LEISH-F3 + SE in healthy adult subjects	Phase I	Completed
GLA-SE TLR4 agonist	NCT02465216 Phase 2a ID93 + GLA-SE Vaccine Trial in TB Patients After Treatment Completion Access to Advanced Health Institute (AAHI)	To evaluate the safety and immunogenicity of ID93 + GLA-SE vaccine when administered to adult pulmonary Tuberculosis (TB) patients, following successful completion of TB treatment with confirmed bacteriologic cure, in preparation for a future Phase 2b prevention of TB recurrence trial in the same population	Phase II	Completed
GLA-SE TLR4 agonist	NCT05658614 Anti-Schistosomiasis Sm14-vaccine in Senegal Oswaldo Cruz Foundation	To evaluate the immunogenicity and safety of a new vaccine schedule using the vaccine candidate Sm14 against schistosomiasis in adults with a history of S. Mansoni and / or S. Haematobium infection	Phase II	Unknown status
GLA-SE TLR4 agonist	NCT04927585 Evaluating the Safety and Immunogenicity of Polyvalent DNA/gp120 HIV Vaccine in Healthy, HIV-uninfected adults Worcester HIV Vaccine	To evaluate the safety, tolerability, and immunogenicity of polyvalent env (A,B,C,A/E)/gag (C) DNA and gp120 (A,B,C,A/E) protein vaccines (PDPHV201401) co-administered together with or without adjuvant in repeated doses in healthy, HIV-uninfected adults	Phase I	Completed
IL-12 DNA	NCT00111605 Study of an HIV preventive vaccine given with or without an adjuvant in HIV uninfected adults National Institute of Allergy and Infectious Diseases (NIAID)	To evaluate the safety and tolerability of an experimental HIV vaccine, which will be given with or without IL-12 DNA adjuvant (at three escalating doses of 100, 500, and 1,500 mcg respectively)	Phase I	Completed
IL-12 DNA	NCT03181789 Safety and Immunogenicity of pDNA Vaccines Expressing HIV M Group p24^Gag Conserved Elements and/or p55^Gag, Administered With IL-12 pDNA (HVTN119) National Institute of Allergy and Infectious Diseases (NIAID)	To evaluate the safety, tolerability, and immunogenicity of two HIV-1 pDNA vaccines: p24CE1/2 pDNA and p55^gag pDNA administered with IL-12 pDNA adjuvant, given by intramuscular (IM) injection with electroporation (EP), in healthy, HIV-uninfected adults	Phase I	Completed
SWE TLR4 agonist	NCT06546709 DMID 23-0015; Lassa Fever CVD 1000 Wilbur Chen, MD, MS	To evaluate the safety, reactogenicity, and immunogenicity of two doses of rabies-vectored monovalent first-in-human Lassa fever vaccine	Phase I	Recruiting

Abbreviations: 3 M-052-AF: 3 M-052 aqueous formulation; ALF: army liposome formulation; ALFQ: ALF containing Quillaja saponaria-21 (QS21) saponin; AS37: Adjuvant system 37; CAF01: cationic adjuvant formulation no 1; CJCV2: Campylobacter Jejuni conjugate vaccine; CpG: cytosine-phosphate-guanine; CRM197: Cross-reactive material 197; CTH522: Chlamydia trachomatis recombinant subunit protein; dmLT: double mutant heat labile toxin; ETEC: Enterotoxigenic Escherichia coli; GLA-AF: Glucopyranosylphospho-lipid A aqueous formulation; GLA-LSQ: Glucopyranosyl lipid A liposomal formulation with Quillaja saponaria 21; GLA-SE: glucopyranosyl lipid A in squalene-in-water emulsion; HIV: Human Immunodeficiency Virus; IL-12: Interleukin-12; LEISH-F3: recombinant Leishmania protein subunit antigen; MPL: Monophosphoryl lipid A; MPL-SE: Monophosphoryl lipid A in a stable emulsion; pDNA: Plasmid DNA; SWE: Squalene-in-water emulsion; TLR: Toll-like receptor.

On the other hand, about the mechanism of delivery systems for enhancement of delivery/presentation of the antigen to the immune system, in the initial years, it was suggested that immune activation induced by delivery systems was due to antigen depot and sustained release of antigens, but further studies have shown that an adjuvant can show multiple mechanisms including induction of cytokines, recruitment of immune cells, enhancement of antigen uptake and presentation, and helping antigen transport to draining lymph nodes (Li et al. 2008; Awate et al. 2013; Gatt et al. 2023; Zeng et al. 2023; Zhao et al. 2023; Lan et al. 2024). Henceforward in this review, delivery systems will not be discussed separately, but instead, the existing adjuvant-delivery systems that are either already approved with the product or currently under clinical trials will be mentioned in detail.

Numerous research studies are being conducted on adjuvant systems based on emulsions, and liposomal and polymeric particulate systems against different antigens (Şenel 2012). Recently, our group has developed liposomal and chitosan-based polymeric adjuvant systems in combination with outer membrane proteins (porins) of Salmonella Typhi and evaluated their adjuvant activity in animal model against ovalbumin and Toxoplasma gondii recombinant proteins (Yüksel et al. 2020; Parmaksız and Şenel 2021; Parmaksız et al. 2023). Both polymeric and liposomal adjuvant systems have provided enhanced immune responses following immunization, whilst chitosan-porins-based adjuvant system was found to stimulate higher IgG levels when compared to that of cationic liposomes, indicating long-lasting immunogenicity and protection. Studies on different derivatives of chitosan in micro- and nanoparticulate form have also demonstrated the adjuvant effect of chitosan against various antigens (Sinani et al. 2019, 2024; Parmaksız and Şenel 2023; Sessevmez et al. 2023). For the sake of brevity, the research studies on the development of adjuvants that have not yet reached to clinical studies will not be described in depth in this review.

3. Adjuvants in licensed vaccines

While the application of advanced biotechnology provides opportunity for producing at high quantities well-characterized, highly purified antigens with suitable safety profile, these antigens are poorly immunogenic and thus require addition of adjuvants. However, currently, a very limited number of adjuvants are licensed as component of a particular vaccine (Table 1), and others are being tested in clinical trials (Table 2). To date, there is no single universal adjuvant authorized for use alone, except for aluminum. The rising demand for effective vaccines against emerging infectious diseases and growing investment in innovative vaccine adjuvant research have pushed up the growth of the vaccine adjuvant market despite limiting factors such as high research and development (R&D) cost and safety concerns.

3.1. Aluminum

Aluminum salts have been used safely for almost a century in hundreds of millions of people around the world. The most common aluminum adjuvants used in licensed human vaccines are aluminum hydroxide and aluminum phosphate, and amorphous aluminum hydroxy phosphate sulfate. The term “alum” is often used to refer to aluminum-containing adjuvants but, chemically, alum corresponds to aluminum potassium sulfate. It is important to define whether the aluminum-based adjuvant is aluminum hydroxide or aluminum phosphate, as these adjuvants have very different physical and chemical properties. Aluminum adjuvants are readily available, have relatively low manufacturing costs, and are recognized by regulatory agencies as adjuvants in vaccines. They are commercially available as adsorbent gels under commercial names such as Alhydrogel^®, Rehydragel^TM, and Adju-Phos^® (HogenEsch et al. 2018).

Although there have been concerns about the safety of utilization of aluminum salts, their involvement in vaccines outweighs any theoretical concerns about the potential side effects. A study conducted by FDA has shown that the risk to infants of aluminum in vaccines was not significant (Keith et al. 2002). Furthermore, the Global Advisory Committee on Vaccine Safety (GACVS) has also reviewed the safety of aluminum adjuvants and concluded that current risk assessment studies support the safety of aluminum in vaccines. It was reported that the studies related to the side effects of aluminum concerning autism spectrum disorder were seriously flawed (Global Advisory Committee on Vaccine Safety 2012). The maximum amount of aluminum adjuvant allowed in human vaccines is stated in the pharmacopoeias. In the European Pharmacopeia, 1.25 mg of aluminum (Al) per single human dose is allowed where an aluminum adsorbent has been used in the vaccine, unless otherwise stated (Pharmacopoeia 2025).

Although widely used in currently available vaccines, the mechanisms behind the immune stimulating activity of aluminum-based adjuvants are still not fully elucidated. Aluminum adjuvants selectively stimulate a Th2 immune response, but they do not induce cytotoxic T cell responses and cell-mediated immunity (HogenEsch 2012). The immune stimulating properties of aluminum adjuvants are likely to be mediated by several separate events such as depot effect, phagocytosis, NLRP3 mediated inflammasome activation, host DNA release (Li et al. 2008; McKee et al. 2013; Shardlow et al. 2018; Danielsson and Eriksson 2021; Zhang et al. 2023).

The adsorption of antigens to aluminum adjuvants initially enhances the immune response by facilitating phagocytosis and slowing the diffusion of antigens from the injection site, which allows time for inflammatory cells to accumulate. Hence, the adsorptive strength of the aluminum salt used is important because high-affinity interactions interfere with the immune response. It has been reported that adsorption can also affect the physical and chemical stability of antigens (HogenEsch et al. 2018). In addition, the surface charge of the aluminum affects the electrostatic adsorption with antigen. Aluminum salts show pH-dependent changes on the surface charge, e.g. at neutral pH, aluminum hydroxide has positive charge, whilst aluminum phosphate has negative charge. Furthermore, other components such as ligand exchange (strongest interaction) and hydrophobic interactions, hydrogen bonding and Van der Waals forces and excipients used in the formulation can affect the interaction between aluminum adjuvant and the antigen and influence the potency of vaccines (Jully et al. 2016; Kooijman et al. 2022; Lan et al. 2024). In fact, determination of adsorption of antigen(s) to aluminum hydroxide gels and/or aluminum phosphate gels is one of the parameters tested for quality evaluation of alum-adjuvanted vaccines (WHO Expert Committee on Biological Standardization 2014).

Aluminum adjuvants activate dendritic cells via direct and indirect mechanisms. Phagocytosis of aluminum adjuvants followed by disruption of the phagolysosome activates NLRP3-inflammasomes resulting in the release of active IL-1β and IL-18.

Aluminum-based adjuvants will continue to be a key component of both currently approved and next-generation vaccines. In recent years, aluminum adjuvants have been used as a combination platform with immunomodulatory molecules such as TLRs to enhance immune responses. Due to the large adsorptive surface of aluminum adjuvants, it is possible to adsorb multiple antigens as well as immunomodulatory molecules, resulting in improved efficacy and protection. Such combinations have been licensed in numerous vaccines (Table 1) and are under investigation for newly developed immunomodulating molecules (Table 2).

3.2. MF59

MF59 is the first non-aluminum adjuvant approved for use in human vaccines. Since its approval in 1997 (Fluad^®, Novartis), it is now licensed worldwide in more than 30 countries. It is an oil-in-water (o/w) emulsion, containing fully metabolizable oil squalene (4.3%) (O’Hagan et al. 2012). The oil droplets, which have a mean size of about 160 nm, are stabilized by two nonionic surfactants, Tween 80 (0.5%) and Span 85 (0.5%) (O’Hagan et al. 2013). It has been shown in long-term follow-up studies that there were no significant safety issues, but only common local reactogenicity. The mechanism of action of MF59 has been extensively evaluated and it has been concluded that after injection to the muscle, tissue-resident monocytes, macrophages and dendritic cells are activated and respond by inducing a local immunocompetent environment at the injection site. A mixture of cytokines, chemokines and other factors are produced by these cells, which result in a significant influx of phagocytic cells that take up the antigen and differentiate into APCs (dendritic cells). These cells are responsible for the efficient transport of antigen to the lymph nodes, where the immune response is triggered through the activation of T and B cells and antibody production (O’Hagan et al. 2012). Additionally, MF59 may enhance and accelerate the differentiation of cells toward dendritic cells and alter their phenotype. It is important to note that it is necessary to prepare the emulsion to obtain adjuvant effect as none of the individual components exert an adjuvant effect alone (O’Hagan et al. 2013). MF59.C1 and SQBA are the other names used with the same composition as MF59. A recombinant COVID-19 vaccine (Bimervax^®, Hipra Human Health, S.L.U) containing SQBA has been approved by the EMA (European Medicine Agency Bimervax, 2023).

3.3. AS01

Series of Adjuvants Systems (AS) have been developed over the years by GlaxoSmithKline (GSK) Vaccines, containing two potent immunostimulants, MPL and QS-21 (Quillaja saponaria Molina fraction 21), which potentiate specific cellular immunity besides promoting strong and persistent antibody responses (Garçon et al. 2007; Garçon and Di Pasquale 2017). AS01 refers to this family of adjuvants at two doses, AS01_B (50 μg of MPL and 50 μg of QS-21 per injected dose) and AS01_E (25 μg of MPL and 25 μg of QS-21 per injected dose), both in liposomal formulation (Didierlaurent et al. 2017; Roman et al. 2024). The liposomes are composed of dioleoyl phosphatidylcholine (DOPC) and cholesterol in a phosphate-buffered saline.

MPL directly activates APCs expressing TLR4, stimulating the production of cytokines. QS-21, which is a water-soluble triterpene glycoside with amphiphilic character promotes antigen-specific antibody responses but also stimulates cytotoxic CD8+ T cells. With AS01, the cholesterol in the liposome formulation quenches the intrinsic hemolytic activity of QS-21, hence reducing its undesired effect (Fortpied et al. 2020). In order to attenuate this limitation of free QS-21, as done in liposome based AS01, it was incorporated into particles with cholesterol in the Army Liposome Formulation Q (ALFQ) (Alving et al. 2020) or with a combination of cholesterol and phospholipid, as in immune-stimulating complexes (ISCOMs) or Matrix-M nanoparticles (Lövgren Bengtsson et al. 2011). These adjuvants are further described in the following sections.

Currently, there are several vaccines containing AS01, which reached to the market with acceptable efficacy and safety profiles. RTS,S/AS01E malaria vaccine (Mosquirix) for use in children has received EMA-Article 58 positive opinion and WHO prequalification. The gE/AS01B recombinant herpes zoster vaccine (RZV, Shingrix), and most recently a respiratory syncytial virus (RSV) vaccine (RSVPreF3-AS01E vaccine, AREXVY) received approval from the EMA and FDA. AS01 has been shown to enhance both humoral and cell-mediated immune responses with these vaccines.

The same combination as AS01 (containing MPL and QS-21) was also prepared as an oil-in-water emulsion-based adjuvant, AS02, and was evaluated against several pathogens such as tuberculosis and HIV (Leroux-Roels et al., 2013; Lichterfeld et al. 2012). However, in a Phase II, randomized, double-blind study evaluating the immunogenicity of RTS,S vaccines containing AS01 or AS02 in healthy, malaria-naïve adults, the immune response was found to be significantly higher with RTS,S/AS01 than with RTS,S/AS02 (Leroux-Roels et al. 2014). This experience emphasized the importance of the impact of formulation design on immune response.

3.4. AS03

AS03 is an oil-in-water emulsion adjuvant system composed of α-tocopherol, squalene and polysorbate 80, sharing several ­characteristics with MF59 in regard to mechanism of action (Garçon et al. 2012). The distinction of AS03 is the inclusion of α-tocopherol, which is suggested to have immunomodulatory action itself (Morel et al. 2011). The first licensed AS03-adjuvanted A/H1N1pdm09 influenza vaccines (Pandemrix and Arepanrix H1N1) targeted highly pathogenic avian H5N1 influenza strains. AS03 enhances antibody and T-cell responses to hemagglutinin and other viral proteins in the split antigen. However, post approval data indicated an association between Pandemrix and narcolepsy in pediatric populations as well as adults (Partinen et al. 2012; van der Most et al. 2014). A mimicry-based mechanism was also suggested to explain the association between narcolepsy and A(H1N1)pdm09 influenza infection. In 2019, a comprehensive assessment of the safety of AS03-adjuvanted influenza vaccines was reported based on non-clinical, clinical and post-licensure data in various populations (Cohet et al. 2019). It was concluded that the further use of AS03-adjuvanted inactivated split virion candidate vaccines against seasonal and pandemic influenza infections would be possible, provided that AS03-adjuvanted influenza vaccines do not contain the putative mimicry sequence. During COVID-19 pandemics, in 2022, SARS-CoV-2 spike protein (B.1.351 strain) produced by recombinant DNA technology adjuvanted with AS03 was approved by EMA (VidPrevtyn Beta) (European Medicines Agency VidPrevtyn Beta, 2022). Nevertheless, at present, AS03-containing influenza and COVID-19 vaccines are not available on the market, and their authorizations have been withdrawn.

3.5. AS04

AS04 combines TLR4 agonist MPL adsorbed onto aluminum hydroxide or aluminum phosphate, depending on the vaccine antigen (Didierlaurent et al. 2009). Combination of MPL with aluminum has been shown to improve adaptive immune responses, which is explained by induction of local NF-κB activity and cytokine production (Garçon et al. 2017). The added value of MPL in AS04-based formulation was demonstrated by higher vaccine-elicited antibody responses, as well as the induction of higher levels of memory B cells, when compared to aluminum alone formulations (Garçon and Davies 2010). Currently, AS04 adjuvant is used in two approved vaccines, Hepatitis B (rDNA) vaccine (Fendrix) (European Medicines Agency Fendrix, 2005) and human papillomavirus vaccine (Types 16, 18) (Cervarix) (European Medicines Agency Cervarix, 2007).

3.6. Matrix-M

The Matrix-M adjuvant is based on the immune-stimulating complexes (ISCOM/ISCOMATRIX) technology. The immunostimulating complex (ISCOM) was first described by Morein et al. (1984) as a new structure for antigenic presentation of membrane proteins from enveloped viruses with potent immunomodulatory capability. ISCOMs are stable particles formed by Quillaja saponaria saponins, cholesterol, and phospholipids, and their characteristic structure is based on the strong affinity between saponins and cholesterol. In addition to “quenching” of saponin’s hemolytic activity, the nanoparticulate form allows targeting to phagocytic cells, resulting in activation of appropriate cells. Antigens can be physically incorporated into the matrix of saponins and lipids. This matrix technology was further developed under the trademark ISCOMATRIX, essentially the same structure as ISCOM, but without the incorporated antigen (Drane et al. 2007). ISCOMATRIX was reported to have a much broader application as it is not limited to hydrophobic membrane proteins (Pearse and Drane 2005). ISCOMs and ISCOMATRIXs have been approved for veterinary use.

The Matrix-M adjuvant consists of cage-like spherical nanoparticulate structures 40 nm in size, containing cholesterol, phospholipids and two separate saponin fractions, Matrix-A (85%) and Matrix-C (15%). The Matrix-A and Matrix-C contain different Quillaja saponaria saponin fractions with complementary properties in regard to reduced reactogenicity and enhanced adjuvant activity (Lövgren Bengtsson et al. 2011). Matrix-M has been shown to be stable in aqueous solution at 2–8 °C for long time. Matrix-M adjuvanted vaccine leads to a rapid innate immune response at the injection site as well as the draining lymph nodes by attracting and activating APCs. Activated immune cells results in enhanced antigen presentation to CD4+ and CD8+ T cells. The APCs can then process and present the antigen to CD4+ and CD8+ T cells. Both T-cell types are capable of providing cellular effector functions and CD4+ T cells help to develop a mature adaptive immune response. This acute response leads to the generation of high-affinity and long-lasting antibody and memory responses (Ahlberg et al. 2012; Reimer et al. 2012; Stertman et al. 2023). In a biodistribution study, radiolabeled saponins or cholesterol were incorporated into Matrix-A particles, and labeled Matrix-M adjuvant was given to mice by intramuscular injection with SARS-CoV-2 Spike protein (Carnrot et al. 2023). A rapid distribution of Matrix-M^™ adjuvant from the injection site to the draining lymph nodes was demonstrated, which excluded a depot effect as central to the mechanism of action for this adjuvant. Diverging clearance patterns for saponins and cholesterol were observed, which suggested partial disassembly of the Matrix-particles. This was claimed to imply the downstream effect of Matrix-M adjuvant on adaptive immune responses.

In 2022, Matrix-M adjuvanted protein-based Novavax COVID-19 vaccine received emergency use authorization (EUA) by the FDA. Currently, the Matrix-M adjuvanted Novavax vaccine is waiting for full approval by the FDA, yet it was announced very recently (April 2025) that more information is required before moving forward. In Europe, it was granted a conditional marketing authorization by EMA, under Nuvaxoid name, in December 2021 and a marketing authorization in July 2023 (European Medicine Agency Nuvaxovid, 2021). Another Matrix-M-adjuvanted vaccine is malaria vaccine, R21/Matrix-M for the prevention of malaria in children, which was recommended by the WHO in October 2023, and further prequalified in December of 2023. High efficacy (75%) when administered just before the high transmission season and good efficacy (66%) when given in an age-based schedule was observed (World Health Organization, 2023).

3.7. CpG oligonucleotides

The natural ligand for TLR9, bacterial CpG DNA, can be mimicked through single-stranded unmethylated cytosine-phosphoguanine oligodeoxynucleotides (CpG ODNs) that are stable, inexpensive and easy to produce. With such features, they became attractive to stimulate innate immunity as well as for immunoprotection (Wilson et al. 2006). By mimicking oligonucleotide sequences common in microbial genetic material, they can activate immune cells that express TLR9. While CpG motifs are naturally methylated in mammalian DNA, unmethylated CpGs are recognized as an infection or neoplasm, per se they can stimulate immune responses (Czechtizky et al. 2022). CpG ODNs used as TLR9 agonists induce both humoral and cellular immune responses, characterized by Th1-biased responses (Roman et al. 1997; Vollmer and Krieg 2009). CpG ODNs differ in their structure and immune-inducing properties. A-class (D-type), B-class (K-type), C-class, and P-class CpG ODNs are the main classes of CpG ODNs described (Samulowitz et al. 2010; Bode et al. 2011; Scheiermann and Klinman 2014). They have been evaluated as vaccine adjuvants either alone or in combination with other adjuvants/delivery systems against several infections (Shirota and Klinman 2017; Hoxie et al. 2024).

Currently, there are two CpG ODNs that have been approved in human vaccines. CpG 1018, which is a synthetic single-stranded DNA molecule, has been approved for recombinant Hepatitis B vaccine (Heplisav B) (European Medicine Agency Heplisav B, 2021). The CpG 7909 (ODN 2006) is another CpG ODN that is approved by FDA with anthrax vaccine (Cyfendus) in July 2023 (Food and Drug Administration CYFENDUS, 2023). During pandemics, an inactivated COVID-19 vaccine adjuvanted with CpG 1018 and aluminum was approved by the EMA, but this product is no longer authorized (European Medicines Agency Valneva, 2022).

3.8. Virosomes

The virosomes were first introduced by Almeida et al. (1975). They have removed the surface hemagglutinin and neuraminidase projections of influenza virus from the viral envelope, purified, and relocated on the surface of unilamellar liposomes. The resulting structures were found to resemble the original virus but lack the ability to replicate inside host cells since they do not have viral genetic material. They are able to facilitate antigen uptake and activate the immune system. The same approach was also used to reconstruct antigens from different viruses and bacteria to enhance their immunogenicity (Pellegrino et al. 2015; Kalra and Sharma 2021). Virosomes reconstituted from influenza viruses (immunopotentiating reconstituted influenza virosomes, IRIVs) are the most frequently utilized virosomes for vaccination. The antigen may be chemically attached to a hydrophobic anchor, adsorbed to the IRIV surface, encapsulated within the vesicle, or directly integrated into the membrane, depending on the characteristics of the antigen and the immunological response (Glück et al. 2004; Metcalfe and Glück 2006). In late 1990s, few vaccines adjuvanted with IRIVs, such as Hepatitis A vaccine (Epaxal), influenza vaccines (Inflexal V) and nasal influenza vaccine (Nasalflu) were approved for human use. However, these products are no longer available on the market.

With a similar structure to virosomes, virus-like particles (VLPs) are highly organized spheres that self-assemble from virus-derived structural antigens, whilst virosomes are not produced by host cells but assembled in vitro (Ludwig and Wagner 2007; Mohsen and Bachmann 2022). VLPs can be derived from a variety of non-enveloped and enveloped viruses and are generally manufactured in recombinant expression systems (Moser et al. 2011). VLPs are applied in several licensed vaccines such as Hepatitis B (Engerix-B, Recombivax HB) and human papillomavirus (Gardasil, Cervarix). These stable and versatile subviral particles possess excellent adjuvant properties and are capable of inducing innate and adaptive immune responses. Another VLP-based vaccine is a COVID-19 vaccine (Medicago Covifenz), which is based on a plant-based technology, in which the viral genetic code was delivered to the leaf cell of the plant using bacteria (Health Canada, 2022). The plant’s natural cell process is used to produce a noninfectious VLP that mimics the spike virus that causes COVID-19. In general, the VLP-based vaccines need additional adjuvants to obtain enhanced immune responses (Mohsen and Bachmann 2022).

4. Adjuvants under clinical trials

Ongoing clinical studies on novel adjuvants, although limited in number, can be generalized in two groups: 1-candidate immunostimulatory molecules to manipulate immune responses, which are developed through considering the underlying molecular, cellular, and systemic aspects of natural host defenses and antigen-specific immunity, and 2-advanced delivery systems for these novel adjuvants alone or in combination with a known adjuvant such as aluminum in order to maximize the efficacy while minimizing toxicity. The currently ongoing clinical trials on investigation of new adjuvants are summarized in Table 2. These studies were selected from the online database of clinical research studies (ClinicalTrials.gov), particularly focusing on the studies, which are still ongoing or completed but the results not reported yet.

In formulation development, the type of desired immune response, delivery route and the nature of the antigen should be taken into account. Certainly, while providing these, it is also important to maintain the stability of the antigen. Most of the delivery platforms used for the novel immunostimulatory molecules include aluminum, oil-in-water emulsions and liposomes, containing QS21, which also have adjuvant activity. Yet, it must be kept in mind that each vaccine is a unique case, and different factors may have impact on the suitability of the adjuvant formulation in regard to safety and efficacy.

TLR agonists are amongst the new immunostimulatory molecules used as adjuvants to enhance vaccine effectiveness in clinical development. In recent years, emulsion or liposome-based formulations of the synthetic TLR4 agonists such as glucopyranosyl lipid adjuvant (GLA) and second-generation lipid adjuvant (SLA) have been developed (Reed et al. 2018). GLA and SLA adjuvants incorporated into different formulations such as aqueous nanosuspension (GLA-AF), oil-in-water emulsion (GLA-SE), liposome (GLA-LS), or adsorbed to aluminum oxyhydroxide (GLA-Alum) have been evaluated in clinical studies against numerous diseases such as tuberculosis, schistosomiasis, HIV and leishmaniasis (Treanor et al. 2013; Coler et al. 2015; Day et al. 2021; O’Hagan et al. 2021). When the in vitro and preclinical study results of the formulation effects were compared, it was concluded that the most potent molecules in in vitro systems may not be the most potent in vivo adjuvant formulations (Misquith et al. 2014). In a first-in-human comparative study of a malaria vaccine, differences in immune responses were obtained depending on the formulation, with GLA-SE formulation being the most immunogenic (Mordmüller et al. 2019). Yet, it is important to consider the interaction of the adjuvant formulation with the vaccine antigen (e.g. binding to surface, electrostatic association, etc.), and evaluations should be done according to the antigen.

Currently, different formulations of liposomes-based adjuvants are being evaluated in clinical trials. The cationic adjuvant formulation (CAF) family of adjuvants have been shown to induce antibody, Th1, and Th17 responses (Pedersen et al. 2018). The principal component of the CAF platform is the quaternary ammonium surfactant N,N-dimethyl-N,Ndioctadecylammonium (DDA) formulated into liposomes or emulsions. The adjuvants CAF01, CAF09 and CAF10 are now being evaluated in human clinical trials (Table 2). The CAF01 adjuvant contains the immune stimulating synthetic glycolipic trehalose-dibehenate (TDB) incorporated into cationic DDA liposomes (Agger et al. 2008; van Dissel et al. 2014). Phase I trials evaluating the safety and immunogenicity of CAF01 adjuvanted chlamydia vaccine (NCT02787109) and subunit tuberculosis vaccine (NCT00922363) have been recently completed. Previously, CAF01 adjuvanted vaccine against HIV was investigated in human (Karlsson et al. 2013; Román et al. 2013). CAF09 consists of DDA liposomes stabilized with monomycoloyl glycerol (MMG)-1 combined with Poly (I:C), a TLR3 ligand (Andersen et al. 2009). Another cationic liposomal adjuvant, CAF09b, which consists of dimethyldioctadecyl ammonium bromide liposomes stabilized with monomycoloyl glycerol (MMG)-1 and combined with Poly (I:C), a TLR3 ligand, has been tested in humans for vaccines for prophylaxy or treatment in various cancers. Phase 1 studies comparing two adjuvants, CAF01 and CAF09b in a new chlamydia vaccine were recently completed (NCT03926728). A novel prophylactic tuberculosis vaccine adjuvanted with CAF10b, which is a combined adjuvant system containing DDA/MMG/CpG, is now recruiting volunteers for a first-in-human Phase I studies (NCT06050356).

Another liposomes-based adjuvant, which is in clinical trials, is Army Liposome Formulation (ALF) family. ALF contains saturated phospholipids, cholesterol and MPLA. Although there are similarities between the structure of AS01 and ALF, their lipid composition differs in terms of phospholipid type and cholesterol content. Different types of ALFs, such as ALF adsorbed to aluminum hydroxide (ALFA), ALF with QS21 (ALFQ), and ALFQ adsorbed to aluminum hydroxide (ALFQA) have been developed and tested for e.g. malaria, HIV-1 vaccination (Alving et al. 2020) (Table 2).

The double mutant heat-labile toxin, dmLT is another promising adjuvant, which has been shown to induce mucosal immunity in preclinical and early phase clinical studies following parenteral or mucosal immunization (Seo et al. 2020; Bauer et al. 2023). dmLT is an 84-kDa polymeric protein with an AB₅ structure composed of an enzymatic A-subunit and a pentameric B-subunit. Different than TLR agonists, dmLT activates innate immunity, including MHC-II and cytokine secretion through combined activities of its A- and B- subunits. The dmLT adjuvanted oral live-attenuated or inactivated whole cell enterotoxigenic Escherichia coli (ETEC) vaccines were evaluated in Phase 1 and 2 studies in adults and infants, which were found to be safe and immunogenic (Harro et al. 2019; Kantele et al. 2023). Further clinical studies with dmLT are continued to test the safety and efficacy of a new Shigella vaccine (InvaplexAR-DETOX) (Table 2).

Over the last decade, chemically synthesized low molecular weight TLR7/8 compounds have gained attention as potential adjuvants. Imidazoquinolines, which are synthetic small molecule TLR7/8 agonists that boost and direct adaptive immunity have been evaluated in vivo to increase the effectiveness of numerous vaccine antigens (Patinote et al. 2020). The imidazoquinolines include imiquimod (R837), resiquimod (R848) and telratolimod (3 M-052). During COVID-19 pandemics, a SARS-CoV-2 vaccine adjuvanted with the TLR7/8a agonist, aluminum hydroxide-adsorbed imidazoquinolin (Algel-IMDG) received an emergency use authorization in India. These agonists bind human TLR7 and TLR8 and generate Th1-polarized responses (Patinote et al. 2020). This family of molecules requires tailored formulations to obtain enhanced immune responses. During last decade, these molecules were evaluated in clinical studies against various pathogens such as Hepatitis B and influenza, mainly being applied topically prior to or immediately after vaccine administration. However, such application is no longer applied. In order to reduce systemic dissemination and improve their efficacy at the site of immunization and in draining lymphoid organs, structural properties as well as specific formulations of these molecules could be tailored. Furthermore, higher immune responses were obtained when these small molecules, which are also referred to as small-molecule immune potentiators (SMIPs), are combined with other known adjuvants. In recent years, 3 M-052, which is structurally similar to resiquimod, but bearing a C18 lipid moiety showed higher activity at the site of immunization due to its higher hydrophobicity. It has been evaluated extensively in preclinical and clinical studies (Smirnov et al. 2011). High immune responses were obtained with the 3 M-052 adjuvant with or without GLA formulated in poly(lactic-co-glycolic) (PLGA) based nanoparticles in an HIV vaccine (Kasturi et al. 2020). Currently, Phase 1 studies for vaccines against Zika virus and HIV, adjuvanted with 3 M-052 adjuvant combined with other adjuvants such as CpG1018, GLA and aluminum hydroxide are going on (Table 2).

The AS37 adjuvant system is another small molecule synthetic TLR7a agonist, which is based on benzonaphthyridine (BZN) chemical scaffold, adsorbed to alum, thereby limiting systemic exposure (Wu et al. 2014; Siena et al. 2023). High-throughput screening (HTS) of a library of chemical analogs of BZNs was performed to identify lead candidates. Screening was initially performed in vitro, followed by in vivo screening for favorable adjuvanticity in mouse models (D’Oro and O’Hagan 2024). With the new BZNs, the low solubility of the first-generation SMIPs was overcome, however local retention was increased. To provide reduced localization of modified BZN, adsorption onto alum hydroxide was achieved. AS37 has advanced into clinical evaluation. Phase I clinical data with a meningococcal C conjugate vaccine showed that the AS37-adjuvanted formulation had an acceptable safety profile and was potent in humans and induced expected immune pathways (Siena et al. 2023). Recently, Phase I/IIa was conducted to assess the safety and immunogenicity of HBsAg vaccine adjuvanted with AS37 adjuvant system in healthy, HBs naïve, adults aged 18–45 years, and to compare AS37 adjuvant system with other approved adjuvant systems and an aluminum-based adjuvant (NCT05561673).

Another adjuvant that reached clinical study is Advax, which is a polysaccharide-based adjuvant derived from crystalline particles of delta inulin, a natural plant sugar comprised of fructose and glucose units (Petrovsky and Cooper 2015). It has been shown to enhance humoral and cellular immunity against various viral antigens in animal and human studies (Honda-Okubo et al. 2021; Sajkov et al. 2024). Advax-CpG adjuvant combines delta inulin with CpG55.2, a synthetic human toll-like receptor 9 (TLR9) agonist. CpG55.2 is reported to be the first human vaccine adjuvant to be designed by artificial intelligence (Honda-Okubo et al. 2023). Currently, Phase I study to evaluate the efficacy of COVID-19 and influenza vaccine given via different routes is recruiting volunteers.

5. New approaches in vaccine adjuvant development

Vaccine efficacy and protection are affected by several factors including the type of antigen, adjuvant, delivery system, and host factors such as age, sex, genetics, microbiota, and immune history (Figure 1) (Pollard and Bijker 2021). Although most of these factors are not within our control, in terms of adjuvants, there are many possibilities to tailor novel adjuvants, taking into account these factors, which provide enhanced efficacy and protection.

Figure 1.

Factors affecting the efficacy of vaccines (created by the authors).

In recent years, systems biology has been applied for the development of adjuvants similar to that of vaccine antigens (Rappuoli et al. 2018). Sophisticated computational and mathematical analysis and modeling of complex biological systems represent integral parts of systems biology. Through this interdisciplinary approach, it is possible to reveal the molecular, genetic and environmental interactions as well as the age and sex interactions within the biological system, in order to understand and predict the responses of the immune system to common pathogens and vaccines (Figure 2). However, due to the high complexity of the biological systems, it is not easy to decode systems biology which covers high-throughput, large-scale, and multi-view big data of numerous formats. In this context, artificial intelligence (AI) comes into play with different tools such as machine learning and deep learning to integrate and analyze vast amounts of heterogeneous data (Dasgupta and De 2023). Besides antigen selection and epitope prediction, AI in vaccine development has been applied also for adjuvant identification and optimization strategies (Gonzalez-Dias et al. 2020; Zohdi 2022; Bravi 2024; Olawade et al. 2024; Sahragard et al. 2025; Tang et al. 2025). AI has the potential to accelerate and facilitate the adjuvant research and development, providing a shortcut through merging several steps into a one. In a previously reported study, in-depth profiling of vaccine-induced immune responses, combined with machine-learning has been shown to be promising in identifying adjuvant-specific immune response characteristics that can guide rational adjuvant selection (Chaudhury et al. 2020). In this study, human immune responses induced by malaria vaccine adjuvanted with two adjuvants AS01B and AS02A, having the same components but formulated in different forms (AS01:liposome and AS02:emulsion) were profiled and adjuvant-mediated immune signatures were established by integrating these data. The computational analysis demonstrated significant differences in immune responses between two adjuvants with 71% accuracy, highlighting the impact of adjuvant formulation on vaccine-induced antigen-specific responses. Furthermore, AI-driven strategies provide benefits for clinical trials by optimizing study design and predicting trial outcomes (Olawade et al. 2024). The researchers have also mapped how artificial intelligence has helped optimize manufacturing processes by simulating various production scenarios. It is obvious that machine-learning algorithms and deep-learning systems would be critical in the development of vaccines/vaccine adjuvants especially in case of pandemics and epidemics.

Figure 2.

Systems biology in vaccine adjuvant development through the integration of immunology and genetics knowledge and computational models (created by the authors).

It has been demonstrated that genetic variations in humans affect the susceptibility to infection and quality of response to immunizations through serum immunoglobulin levels, seroconversion rates, and level of antigen-specific immune responses (Tsang 2015). In a study published by Milieu Intérieur Consortium at Institut Pasteur, Paris, which conducted a project on a population-based study to dissect the interplay between genetics and environment and their impact on the immune system, the serological response to various common human pathogens or vaccine (cytomegalovirus, Epstein–Barr virus, herpes simplex virus 1 and 2, varicella zoster virus, Helicobacter pylori, Toxoplasma gondii, influenza A virus, measles, mumps, rubella, and hepatitis B virus) in an age- and sex-stratified cohort of 1000 healthy individuals were characterized (Scepanovic et al. 2018). Following extensive serological testing and genome-wide association analyses, it was demonstrated that age, sex, and specific human genetic variants contribute to inter-individual variability in humoral immunity. Systems analysis has also been applied on vaccine adjuvants that are already included in licensed human vaccines or in clinical trials (Harandi 2018). Nakaya et al. (2016) has applied systems biology to investigate the innate and adaptive responses to MF59-adjuvanted versus nonadjuvanted trivalent seasonal influenza vaccines in early childhood (14- to 24-month-old). It was demonstrated that MF59 enhances the magnitude, kinetics, and consistency of the innate and adaptive response to vaccination with the trivalent seasonal influenza vaccine during early childhood.

During the past decade, trained immunity, which is long-term functional reprogramming of innate cells and their progenitors, has been confirmed using various stimuli in animal and human models through transcriptomic, epigenetic, and metabolic profiling and/or functional analysis of cytokine production (Lee et al. 2022b). Trained immunity is based on the epigenetic and metabolic reprogramming of cells (Fanucchi et al. 2021).

Epigenetics includes genetic modifications in gene expression without altering the DNA sequence (Megdiche and Salerno-Gonçalves 2025). These epigenetic changes play an important role in regulating gene activity, influencing immune pathways, and modulating the strength and duration of immune responses. The capacity of adjuvants to induce durable epigenetic reprogramming of the innate immune system has been demonstrated to stimulate improved resistance against pathogens. Hence, in addition to stimulating robust T and B cell responses, epigenetic adjuvants can provide broad protection against diverse pathogens by training the innate immune system. Comprehensive characterization of the transcriptomics and epigenomics of the early and late immune responses during vaccination has been performed with various adjuvant systems such as 3 M-052, AS37, AS03 (Wimmers et al. 2021; Lee et al. 2022a, 2022b).

The recently described subset of memory T cells, named as tissue resident memory T cells (TRM cells) that occupy epithelial, mucosal and other tissues (skin, brain etc.) without recirculating, have been suggested to play a key role in vaccine-induced protective immunity. This discovery has altered our understanding of adaptive immunity including site-specific responses mediated by tissue-adapted memory T cells throughout the body (Szabo et al. 2019). Fundamentally, they have been shown to provide enhanced immunity against re-infection and to accelerate pathogen clearance (Rotrosen and Kupper 2023). Studies are still ongoing to elucidate the systemic and local signals needed to establish and maintain protective mucosal TRM subsets without inducing pathogenic populations. Systems biology can be applied to predict antigen epitopes as well as the adjuvants that are able to induce and tailor tissue-resident memory responses at mucosal sites. Some promising adjuvants administered by mucosal route have been already identified in preclinical models. The route of vaccine administration, the choice of antigenic epitopes, and the impact of microenvironment appeared to be crucial parameters in the development of vaccine-induced mucosal tissue-resident memory responses in animal models (Longet and Paul 2023). Furthermore, it has been suggested that sex and age also would have impact on tissue-resident memory response profile and functionality (Booth et al. 2021; Bachnak et al. 2022).

Recent studies have highlighted the connection between alteration in epigenetic background and metabolic rewiring in different physiological and pathological conditions (Ferreira et al. 2024). The immunomodulatory properties of an adjuvant based on the modulation of metabolic regulators can improve the memory and effector function of immune cells to a given vaccine (Lee et al. 2021; Rotrosen and Kupper 2023; Mone et al. 2025). Metabolic control of dendritic cell functions are being investigated and recently, dendritic cell activation by TLR agonists has been demonstrated to trigger a metabolic switch from catabolic metabolism to anabolic metabolism to accommodate increasing immune responses (Wculek et al. 2019).

In addition to the above-mentioned factors, which have the potential to influence vaccine efficacy, increasing evidence from preclinical and clinical studies suggests that the composition and function of the gut microbiota have also significant impact on modulating immune responses to vaccination (Lynn et al. 2022). Although not completely explained, the proposed mechanisms behind the influence of the intestinal microbiota on vaccine immunogenicity include cross-reactive epitopes between microbes and vaccine antigens, modulation of B cell responses through microbial metabolites, and the provision of natural adjuvants through certain microbes (Zimmermann 2023). In a study where systems biological analysis of immunity to the trivalent inactivated influenza vaccine (TIV) in humans was applied, a correlation between early expression of TLR5 and the magnitude of the antibody response was found and TLR5-mediated sensing of the microbiota was demonstrated to impact antibody responses to seasonal influenza vaccination vaccine (Oh et al. 2014). Integrating the cross-talk between the microbiota and immune system through a systems biology approach offers the opportunity to evaluate the responsiveness to vaccines as well as design of vaccine adjuvants. Manipulation of microbiota via different means such a diet alteration may enable the control of the vaccine responses (Collins and Belkaid 2018). Furthermore, vaccines designed according to the inherent adjuvant properties of the microbiota would have a significant impact on successful immunization (Ciabattini et al. 2019). Recently, a systematic review on the role of intestinal microbiota in vaccinations against COVID-19, Salmonella typhi, Vibrio cholerae, and rotavirus vaccinations in humans was conducted (Loddo et al. 2025). It was concluded that the intestinal microbiota surely plays a role in enhanced immunogenicity, especially in younger people. In a recent paper, applications of genetic and genomic approaches in vaccine development and in studying interactions with microbiota was evaluated in malaria vaccine research (Su et al. 2023). Microbiota has been shown to influence disease outcomes and vaccine efficacy. Yet, more studies are needed to elucidate the complex interactions between microbiota and the host immune system to optimize vaccine design.

6. Conclusion and future perspectives

Since the approval of aluminum as an adjuvant in 1930, advances in vaccine adjuvant development have gone through different stages, and currently a new dimension in vaccine adjuvant research is taking shape as a result of advances in the analysis of immune responses, applying systems biology and population-based studies. Currently, we know that the immunogenicity of a vaccine is not a complete predictor of its efficacy, but other factors such as age, sex, infection history, genetics, metabolism, gut microbiota can also affect vaccination outcomes. Systems biology approach for the development of new antigens as well as adjuvants provides valuable knowledge about safety and efficacy, yet such system analysis approaches still have to be further tested in relevant animal models and validated in clinical studies. Such approaches have the potential to save resources and accelerate vaccine research and development. However, it appears to be that there are still some challenges that need to be overcome.

Currently, new adjuvants systems licensed with vaccines or under clinical investigation are based on natural or synthetic immunostimulating molecules, mainly the TLRs agonists. In general, these molecules are combined with known adjuvants, mostly with aluminum, to provide enhanced immune responses. The vaccine platforms such as emulsions, liposomes, lipid nanoparticles which also show immunogenic activity themselves, are being utilized to deliver the antigen and the adjuvant. Studies have shown that type of the formulation as well as the formulation components and their concentration significantly affect the outcome of the vaccines. Systems biology is being applied not only to new molecules under investigation but also to those that are already licensed to further elucidate the impact of extrinsic and intrinsic factors on their activity. These molecules designed as adjuvants are evaluated both for enhancing the T-cell response and B-cell responses and for training the immune system, as metabolic and epigenetic adjuvants, which will prolong the protection of the vaccine.

Although such advanced knowledge and technologies are being applied in vaccine adjuvant research and development to obtain adjuvants providing desired efficacy and protection, currently, there are still very limited number of adjuvants that have reached to the market, licensed with the vaccine. The main reason for this is the safety of these vaccines in humans. Furthermore, the quality of the adjuvant is another important aspect in regard to regulatory requirements. The consistency and capacity of manufacturing as well as the supply chain of the components of the adjuvant also affects the availability of the adjuvant. During COVID-19 pandemics, these issues have been experienced even with the approved adjuvants.

The knowledge gained from the research to date will certainly benefit future vaccine adjuvant development and formulation. Nonetheless, despite the many accomplishments in vaccine research, much remains to be done to develop pathogen-specific antigens and adjuvants with high efficacy and safety, as well as long-term protection. So, we will continue working!

Funding Statement

This research received no external funding.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.