Recent progress in mRNA cancer vaccines

ABSTRACT

The research and development of messenger RNA (mRNA) cancer vaccines have gradually overcome numerous challenges through the application of personalized cancer antigens, structural optimization of mRNA, and the development of alternative RNA-based vectors and efficient targeted delivery vectors. Clinical trials are currently underway for various cancer vaccines that encode tumor-associated antigens (TAAs), tumor-specific antigens (TSAs), or immunomodulators. In this paper, we summarize the optimization of mRNA and the emergence of RNA-based expression vectors in cancer vaccines. We begin by reviewing the advancement and utilization of state-of-the-art targeted lipid nanoparticles (LNPs), followed by presenting the primary classifications and clinical applications of mRNA cancer vaccines. Collectively, mRNA vaccines are emerging as a central focus in cancer immunotherapy, offering the potential to address multiple challenges in cancer treatment, either as standalone therapies or in combination with current cancer treatments.

KEYWORDS:

• mRNA vaccine
• lipid nanoparticles
• cancer vaccines
• tumor antigens
• personalized therapy

Introduction

The field of oncology has recently witnessed a significant advancement with the introduction of mRNA-based tumor vaccines, signaling a new epoch in cancer immunotherapy.¹ These vaccines utilize mRNA’s versatility to elicit potent immune responses against cancer by encoding tumor antigens. This approach prompts the immune system to recognize and eliminate cancer cells displaying these antigens.²

One of the key advantages of mRNA vaccines is their exceptional safety profile. Unlike DNA-based therapies, mRNA vaccines do not integrate into the host genome, thereby negating the risk of insertional mutagenesis. Furthermore, mRNA’s ability to concurrently encode multiple antigens boosts vaccine efficacy and paves the way for the development of personalized vaccines. The transient nature of mRNA expression ensures that antigen stimulation is temporary, effectively reducing the likelihood of chronic inflammations or autoimmune responses.³ Additionally, the rapid design and synthesis of mRNA vaccines enable swift responses to evolving oncological challenges, including customizing vaccines to the unique tumor characteristics of individual patients, marking a significant stride toward personalized cancer therapy.⁴

Although mRNA cancer vaccines are showing considerable promise, their broader application is hindered by several challenges. Challenges among these are the inherent instability of mRNA, the need for efficient and targeted delivery systems, and the possibility of immune tolerance, as well as evasion by tumor cells. These challenges present significant obstacles that are the focus of current research efforts.⁵ Consequently, the optimization of mRNA sequences for maximal protein expression, ensuring mRNA stability, and enhancing delivery methods to achieve optimal cellular uptake and antigen presentation are pivotal areas of ongoing research.

The successful development of mRNA vaccines for Corona Virus Disease 2019 (COVID-19) has overcome technical bottlenecks via a comprehensive optimization of the mRNA construct. This achievement was made possible through the redesign and enhancement of each aspect of the mRNA, thereby improving its stability and expression efficiency.^6–9 Furthermore, the advancement of diverse mRNA delivery systems has facilitated the efficient and targeted delivery of mRNA molecules.¹⁰ mRNA cancer vaccine benefits from all of these advancements as well, and as clinical trials progress, mRNA cancer vaccines are emerging as a promising tool in the fight against cancer.

This review delves into the development and application of mRNA vaccines in the field of oncology. We provide a comprehensive summary of the development history of mRNA cancer vaccines, highlighting the structural optimization of antigen expression vectors to enhance mRNA stability and translation efficiency. Additionally, we discuss the application and future prospects of other RNA-based antigen expression vectors in cancer vaccine development. An overview of the LNP delivery system used for mRNA vaccines is presented, with an emphasis on targeting strategies. We also elucidate various types of mRNA cancer vaccines and their current state of advancement. It is anticipated that mRNA vaccines will surmount various challenges in cancer treatment and emerge as a major modality in cancer immunotherapy.

The concept of mRNA cancer vaccine as an innovative therapeutic approach

The popularity of mRNA vaccines has significantly surged with the application of two COVID-19 vaccines, marking a pivotal moment in their development and public recognition.^6,7,9 However, the investigation and development of mRNA vaccines long predates the COVID-19 pandemic. The journey began following the successful synthesis of mRNA in vitro and its expression in vivo (Table 1).¹¹ In the early 1990s, efforts to develop mRNA vaccines for tumor prevention commenced. A landmark study by Martinon F. et al. demonstrated the induction of anti-influenza cytotoxic T lymphocytes (CTLs) in mice using liposome-encapsulated mRNA encoding the influenza virus nucleoprotein (NP), showcasing the first instance of using mRNA to stimulate in vivo antigen-specific immune responses.¹² Boczkowski et al. furthered this research by transfecting mRNA encoding tumor antigens into dendritic cells, leading to the production of CTLs both in vitro and in vivo. This study was pivotal in demonstrating mRNA’s potential to induce anti-tumor immune responses.¹³ However, early mRNA molecules faced limitations due to instability and low expression efficiency. Unmodified mRNA was initially noted for its high immunogenicity, activating innate immune responses through toll-like receptors (TLR3, TLR7, TLR8, and TLR9).^14–17 The subsequent incorporation of modified nucleosides (such as m5C, m6A, m5U, s2U, and pseudouridine) into mRNA formulations markedly reduced their immunogenicity, with the complete replacement of uridine with pseudouridine improving mRNA stability and translation efficiency.^18,19 The scientists who made these breakthrough findings, Drs. Katalin Karikó and Drew Weissman, were awarded the 2023 Nobel Prize in Physiology or Medicine for their discoveries concerning nucleoside base modifications that enabled the development of effective mRNA vaccines against COVID-19.

Table 1. Recent milestone events in the development of mRNA vaccines.

Year	Milestone Events
1990	Discovery of in vitro transcription and In vivo expression of mRNA. (Wolff JA et al.)
2000	Exploration of mRNA as a novel vaccine approach. (Hoerr I et al.)
2005–2008	mRNA modifications to enhance stability and efficacy. (Karikó K et al.)
2010	Development of lipid nanoparticles (LNPs) for mRNA delivery. (Pardi N et al.)
2010	Significant Progress of mRNA technology in infectious disease vaccines. (Pardi N et al.)
2020–2022	FDA approval of two mRNA vaccines for COVID-19. (Polack FP, Baden LR et al.)
late 2020s	Research and clinical trials on personalized mRNA cancer vaccines. (Miao L et al.)

The investigation into liposomal delivery of siRNA provided valuable insights for the advancement of LNP systems.²⁰ LNPs have since been validated as effective carriers for in vivo mRNA delivery in various preventive and therapeutic vaccination strategies.²¹ The COVID-19 pandemic has accentuated the immense potential of mRNA technology,^6,7,9 offering valuable insights for the development of mRNA-based cancer vaccines. Furthermore, the advent of personalized neoantigen vaccines has yielded substantial clinical benefits for patients suffering from a variety of malignancies.²² The progression of mRNA-4157 in combination with pembrolizumab to Phase III clinical trials for melanoma patients represents a significant milestone in the field of personalized mRNA cancer vaccines. Current research on mRNA cancer vaccines primarily focuses on eliminating the immunogenicity and instability of mRNA, developing efficient and targeted mRNA delivery systems, and advancing the development of personalized mRNA cancer vaccines. The subsequent sections of this paper will provide a more comprehensive exploration of these critical aspects.

RNA-based vectors for antigen expression

Design strategies for in vitro transcribed mRNA

The structure of in vitro transcribed (IVT) mRNA, mirroring eukaryotic mRNA, includes a 5’ cap, a protein-coding open reading frame (ORF), 5’ and 3’ untranslated regions (UTRs), and a 3’ poly(A) tail. The structural complexity of each mRNA segment significantly impacts translation efficiency, properties, and immunogenicity,^23,24 making optimization crucial for the efficacy of mRNA technologies.

For IVT mRNA, the 5’-cap structure (m7GpppN) is vital for stability and translation efficiency.²⁵ Two main methods for cap formation are Vaccinia capping enzyme (VCE)-mediated capping and co-transcriptional capping. VCE capping offers specificity but at higher cost and complexity, while co-transcriptional capping is simpler but less effective.²⁶ Recent advancements include the use of VCE to introduce GTP analogs at the 5’ end, creating modified mRNAs.²⁷ Moreover, Integrating T7 RNA polymerase with VCE simplifies cap-modified mRNA synthesis, reducing the need for expensive analogs.²⁸ Continuous cap analog optimization has significantly improved mRNA stability and transcription, enhancing its research and application potential (Table 2).^29–31 Furthermore, a recent study shows a shift from cap1 to cap2 in mRNA caps with age, decreasing RIG-I receptor activation and increasing mRNA stability.³² Incorporating cap2 may reduce immunogenicity and extend mRNA lifespan.

Table 2. Different generation of mRNA cap analogs.

Cap Analogs	Description	Rate of capping	Advantages	Disadvantages
First-Generation Cap Analogs: m7G Cap	The most basic form of cap analogs, characterized by the 7-methylguanosine (m7G) structure.	40%	Fundamental protection and translation enhancement for mRNA.	Limited stability and susceptibility to degradation by RNase.
Second-Generation Cap Analogs: ARCA	An improved version of the m7G Cap with anti-reverse orientation features, ensuring correct attachment to the 5’ end of mRNA.	80%	Better stability and translation efficiency compared to the first generation.	Potential for further improvement.
Third-Generation Cap Analogs: CleanCap®	The latest generation of cap analogs, employing advanced synthesis techniques for a more efficient and cleaner cap structure.	70–99.5%	Enhanced translation efficiency and stability, leading to a reduction in by-product formation.	Increases in costs and complexities associated with production processes

The 5’ untranslated region (UTR) of mRNA, crucial for gene expression regulation, can be optimized for enhanced translation efficiency.³³ For instance, adjustment like altering the AUG codon position or incorporating unstructured sequences can improve translation efficiency.^34,35 A shorter 5’ UTR, around 17 nucleotides, also aids in reducing variability and translation issues.³⁶

Similar to the 5‘UTR, the 3’ UTR also harbors several regulatory elements and plays a pivotal role in regulating mRNA stability, subcellular localization, and translation efficiency.³⁷ Key strategies include avoiding miRNA binding sites, eliminating adenine/uridine-rich elements, and minimizing its length.^38–40 Despite challenges in designing optimal UTRs, natural UTRs, such as the 5’ UTR of human α-globin mRNA used in the BNT162b2 COVID-19 vaccine, are often preferred for their proven gene expression efficiency.⁴¹ However, UTRs impact different cell types variably, requiring tailored optimization for specific target cells. The rapid advancements in high-throughput screening (HTS) technology and genetic algorithms have facilitated the identification of potential optimization sequences from a diverse range of 5‘UTR and 3‘UTR libraries derived from cellular or viral genomes.^42,43

Optimization of the coding sequence (CDS) in mRNA is crucial for effective gene expression. Common strategies involve replacing rare codons with more prevalent synonymous ones in host cells.⁴⁴ In 2005, Karikó et al. addressed the challenges of high mRNA immunogenicity and low protein expression via incorporating the modified nucleosides, as mentioned earlier.^18,19 Additionally, factors like GC content and mRNA secondary structure significantly influence mRNA translation.^45,46 However, these optimization strategies, while effective, have limitations in fully harnessing the potential of mRNA sequences for optimal performance. Within the field of precision medicine, the development of mRNA design algorithms is a promising approach to overcome these challenges. These algorithms focus on identifying mRNA sequences with stable secondary structures, enhanced translation efficiency, and optimized codons, expected to significantly contribute to advancing mRNA-based technologies.^47,48 We have listed a few here with noted features and advantages/disadvantages (Table 3).

Table 3. mRNA design algorithms.

mRNA design algorithms	Features	Application	Advantages	Disadvantages
Gene Optimizer	Multi-parameter optimization algorithm	Gene expression optimization	Considers multiple sequence parameters related to gene expression	Requires advanced bioinformatics knowledge
LinearDesign	Optimizes codon usage and RNA stability	mRNA vaccines, gene therapy	Fast and efficient, operable on personal computers	Relatively new, limited case studies
Vector Builder’s Codon Optimization Tool	Offers optimal Codon Adaptation Index (CAI)	Gene synthesis, gene therapy	A comprehensive species list included	Less consideration for other mRNA optimization factors
NovoPro’s Exp Optimizer	Online codon optimization tool,	Recombinant gene expression	Considers key parameters in transcription and translation	Results may vary due to random sampling
RNA structure	Analyzes and predicts RNA secondary structures	mRNA secondary structure optimization	Enhances mRNA translation efficiency and stability,	Predictions may not always be accurate
Eukaryotic Pathway Prediction (EPP)	Predicts post-transcriptional modification pathways	Post-transcriptional regulation of mRNA	Facilitates fine-tuning of gene expression	Requires specialized background knowledge

The Poly(A) tail, characterized by a long repetitive polyadenylate sequence, plays a crucial role in defending mRNA against enzymatic deacetylation and degradation, thereby augmenting its stability and translational efficacy.³¹ The incorporation of short sequences or thiophosphate between polyadenylate sequences has been shown to further enhance stability. For example, the Poly(A) tail sequence of BNT162b2 comprises a structure of 30A + 10GCAUAUGACU +70A.^49,50 Additionally, a recent study demonstrated that a Poly(A) tail containing 20% cytidine can protect mRNA from deacetylase activity and significantly improve the translation efficiency and half-life of mRNA.^51,52

Optimization strategies for mRNA structure can be synergistically employed to enhance protein production. The advancement of mRNA vaccines underscores the need for a universally standardized mRNA design and screening protocol. Comprehensive screening and precise customization of target-specific sequences hold the promise of diverse enhancements in future mRNA designs. This approach enables personalized adaptations to specific cell types and unique microenvironment, thereby maximizing the protein synthesis efficiency of each mRNA transcript.

Other RNA-based antigen expression vectors

The current instability of mRNA presents significant limitations in its application as a therapeutic agent, prompting recent research to focus on the exploration of alternative RNA vectors. Besides linear mRNA, several other types of RNA vectors have advanced quickly (Figure 1). One such promising vector is the self-amplifying RNA (saRNA), which is derived from a virus and carries the RNA polymerase sequence. This saRNA exhibits self-amplification and translation capabilities, enabling it to elicit a detectable protective immune response with only minimal quantities.⁵³ This characteristic suggests that saRNA could be effectively applied in cancer vaccines.⁵⁴ Several preclinical studies exploring this potential are already underway. For instance, a comparative study demonstrated that saRNA targeting the human papillomavirus (HPV) exhibited stronger anti-tumor effects compared to both unmodified and modified non-replicating mRNAs.⁵⁵ However, it is currently unclear whether RNA polymerase can trigger an immune response in vivo. Furthermore, the larger size of saRNA may pose new challenges for existing delivery systems. One potential strategy to address this involves spatially segregating the sequence encoding RNA polymerase from the sequence of the target protein.

Figure 1. Different types of RNA expression vectors used in mRNA cancer vaccines.

Circular RNA (circRNA), with its covalent circular structure, offers enhanced stability over mRNA due to its resistance to RNase. This characteristic provides distinct advantages for RNA vaccines.^56,57 The circRNA vaccine platform developed by Qu et al. has shown protective efficacy against multiple COVID-19 variants, contributing valuable insights to the development of circRNA-based cancer vaccines.⁵⁸ Studies have demonstrated that circRNA molecules encoding antigens or cytokines are more effective in tumor suppression than mRNA vaccines.^59,60 Additionally, Amaya et al. highlighted circRNA’s role as an adjuvant in enhancing anti-tumor effects.⁶¹ However, the precise mechanisms behind these effects are yet to be fully understood. In a recent study, Huang et al. showed that circFAM53B, a tumor-specific circRNA, initiates anti-tumor immunity by expressing a specific peptide, underlining circRNA’s potential in immunotherapy.⁶²

Despite these developments, challenges remain in circRNA synthesis, including suboptimal cyclization efficiency and high costs of reagents like enzymes. Nonetheless, circRNA is emerging as a promising class of RNA-based therapeutics and vaccines with significant future potential.

A recent advancement involves hybridizing short double-stranded RNA (dsRNA) with single-stranded mRNA to form a dendritic, comb-like structure, serving both as an antigen expression vector and immune adjuvant. This method allows precise control over immune stimulation strength in mRNA vaccines by manipulating dsRNA’s length, sequence, and quantity.⁶³ This comb-structured RNA design, applicable to various mRNA vaccines under development, is anticipated to significantly enhance their efficacy and safety.

Delivery platforms

A crucial component of mRNA therapy is the delivery system. Common delivery systems include protamine, lipoplex (LPX),⁶⁴ lipopolyplex (LPP),⁶⁵ virus-like particles (VLPs),⁶⁶ dendritic cell,⁶⁷ and LNPs.⁶⁸ Due to its most advanced status in current application, this review primarily focuses on the application of LNP technology.

LNPs typically consist of cationic or ionizable lipids, helper lipids, cholesterol, and PEGylated lipids.⁶⁹ Cationic or ionizable lipids, positively charged in acidic conditions, bind negatively charged mRNA through electrostatic interactions. In endosomes, these lipids protonate, facilitating mRNA escape for cytoplasmic antigen expression.⁷⁰ Cholesterol improves LNP stability, aiding endocytosis and endosomal escape.⁷¹ while Phospholipids promote mRNA encapsulation and stability, while PEGylation prevents in vivo protein-LNP interactions.⁷²

Originally developed for delivering siRNA,⁷³ LNP technology has been adapted for mRNA encapsulation and delivery.⁷⁴ While offering efficient mRNA packaging and biocompatibility, most LNPs accumulate in the liver post-administration, potentially leading to hepatotoxicity.⁷⁵ The current focus in LNP-mRNA cancer vaccines is on targeting strategies for selective mRNA delivery to specific organs or cells.

Recent advancements in LNP targeting include passive or active methods to tumors.⁷⁶ Effective anti-tumor immunity hinges on delivering antigen-encoding mRNA to antigen-presenting cells (APCs), with LNPs designed to target immune cells expected to bolster anti-tumor responses and minimize systemic side effects.⁸ Subcutaneous injections near lymph nodes, as a passive targeting approach, have proven effective in various studies.^77–79 However, subcutaneously administered LNPs can also accumulate in organs like the liver and spleen, potentially impacting vaccine efficacy and safety.⁸⁰

Modifying lipid structure of LNP is another targeting strategy, as LNPs naturally interact with plasma proteins in vivo, forming a “protein corona” that alters their interactions with cells and organs.⁸¹ Preclinical studies show that these modifications can significantly affect mRNA delivery and LNP targeting.^82–84 Lipid 113-O12B, for example, specifically targets lymph nodes and is uptaken by a significant proportion of dendritic cells and macrophages, inducing robust immunity and, when combined with anti-PD1 antibody, achieving a 40% complete tumor remission rate.⁸⁵ However, in vivo screening has limitations, including the ethical concerns of animal use and the inability of in vitro screening to accurately predict in vivo delivery efficiency.⁸⁶ Moreover, the specific mechanism of these lipids remains unclear, and the limited understanding of the interaction between LNPs and biological components hampers the prediction of the in vivo targeting behavior of novel LNPs.

Regulating the internal and/or external charges of LNPs is crucial for targeting.^87,88 SORT nanoparticles, developed for tissue-specific mRNA delivery and CRISPR/Cas gene editing, interact with serum proteins that bind to receptors on target organ cells, allowing selective organ targeting.⁸⁹ However, reliance on endogenous patient proteins for targeting may lead to off-target effects and inter-individual variability.^90,91

Passive targeting of LNPs is popular for its simplicity and broad applicability, but it lacks specificity. Active targeting, while more specific and efficient, involves higher costs and potential immunogenicity risks. Targeted LNPs, therefore, represent a promising avenue in mRNA delivery, with future research needed to optimize both passive and active strategies to enhance LNP performance and safety in mRNA and drug delivery.

Classifying mRNA cancer vaccines

mRNA-based cancer vaccines that encode viral antigens

The development of mRNA-based cancer vaccines targeting viral antigens is an emerging field. These vaccines are particularly relevant for malignancies associated with viruses such as Human Papillomavirus (HPV), Hepatitis B (HBV), Epstein-Barr (EBV), and HIV, which are implicated in cervical, liver, and nasopharyngeal cancers, among others.^92–94 Traditional vaccines against these viruses have predominantly been prophylactic.^95–97 However, mRNA-based vaccines offer a therapeutic approach by stimulating the body’s immune system to recognize and attack cells expressing these viral antigens.¹ For example, a recent study showed that a messenger RNA-HPV therapeutic vaccine (mHTV) exhibited strong immunogenicity and anti-tumor effects in mice and non-human primates, positioning mHTV as a promising therapeutic and prophylactic vaccine for HPV.⁹⁸

The flexibility of mRNA vaccine platforms is particularly valuable in personalized medicine. Customizing vaccines to match individual patient’s viral antigen profiles can address the challenge posed by the evolving nature of viral oncogenes, offering more effective treatment. HIV, known for its rapid mutation and diversity, presents a significant challenge in vaccine development.⁹⁹ An mRNA-based HIV vaccine induced both humoral and cellular immune responses in mice and primates, with Rhesus monkeys showing a 79% reduction in SHIV exposure after receiving priming and booster vaccines.¹⁰⁰ A recent phase I clinical trial of the mRNA vaccine eOD-GT8 60mer against HIV showed good safety and induced broad neutralizing antibodies (bnAbs) in 97% of participants (35 out of 36), demonstrating the potential of mRNA vaccines in HIV prevention.¹⁰¹ Additionally, mRNA vaccines offer a novel approach for chronic virus infections such as HBV.¹⁰² A mRNA-based vaccine for treating Hepatitis B virus-associated hepatocellular carcinoma (HCC) (NCT05738447) is currently undergoing clinical trial.

EBV is linked to various cancers, including nasopharyngeal carcinoma, Hodgkin’s lymphoma, and gastric cancer.¹⁰³ The complex nature of EBV and its interactions with the host immune system present significant challenges for vaccine development. Notably, EBV’s latency within host cells renders it difficult for the immune system to detect and eradicate. To date, no vaccines have been approved for EBV infection, and the development of mRNA vaccines for EBV remains in the preclinical stage, though they have shown promising potential.^104,105

Despite the potential of mRNA vaccines to prevent viral infections linked to cancer, vaccines targeting cancer-related viruses predominantly remain in preclinical stages. This is due to challenges such as viral mutation, immune evasion, and immune tolerance stemming from long-term infection.^106–108 Future research needs to focus on overcoming these hurdles to develop more effective mRNA vaccines against these viruses.

mRNA vaccines encoding tumor antigens

Tumor-associated antigens (TAAs)

TAAs are a class of antigens that are expressed at significantly higher levels in tumors compared to normal tissues. This overexpression designates them as prime targets for cancer vaccines. These vaccines, customizable based on individual TAAs, offer enhanced efficacy and reduced side effects compared to traditional therapies like chemotherapy or radiotherapy.¹⁰⁹ This personalized approach marks a significant advancement in oncology, presenting a more targeted and patient-specific treatment strategy. Some of the well-known TAAs are listed in the Table 4.

Table 4. A list of well-studied TAAs for different cancer types.

Tumor-Associated Antigen (TAA)	Associated Cancer Types	References
CEA (Carcinoembryonic Antigen)	Colorectal and other adenocarcinomas	Gold and Freedman^Citation110
HER2/neu	Breast cancer, gastric cancer, colorectal cancer, and bladder cancer.	Ménard et al.^Citation111, Park et al.^Citation112
MAGE-A3	Melanoma, non-small cell lung cancer, breast cancer	van der Bruggen et al.^Citation113
MUC1 (Mucin 1)	Breast, ovarian, and other epithelial cancers	Gendler et al.^Citation114
PSA (Prostate-Specific Antigen)	Prostate cancer	Wang et al.^Citation115
NY-ESO-1	Melanoma, breast cancer, esophageal cancer, etc	Chen et al.^Citation116
WT1 (Wilms’ Tumor 1)	Leukemia, solid tumors	Oji et al.^Citation117
AFP (Alpha-fetoprotein)	Hepatocellular carcinoma	Crandall and Lau^Citation118
GPC3 (Glypican 3)	Hepatocellular carcinoma	Nakatsura et al.^Citation119
MART-1/Melan-A	Melanoma	Ribas et al.^Citation120

Despite their potential, mRNA vaccines targeting TAAs have limitations. TAAs, being non-mutated self-antigens, can lead to poor T cell responses and immune tolerance in clinical immunotherapy.¹²¹ Additionally, TAA expression in normal cells might cause collateral damage. These vaccines may also be less effective in cancers where TAAs have low expression levels or mutations.¹²² Strategies to overcome these limitations include co-delivering cancer antigens with immune-stimulatory molecules, enhancing immune activation in the tumor microenvironment, and combining vaccines with conventional treatments like chemotherapy or radiotherapy.^123–126 BNT111, targeting four common TAAs (NY-ESO-1, MAGE-A3, Tyrosinase, TPTE), is exemplary. In over 90% of cutaneous melanoma patients, BNT111 induced novel and enhanced preexisting immune responses against these antigens (Phase I clinical trial, NCT02410733). A phase II trial (BNT111–01, NCT04526899) is underway, combining BNT111 with Cemiplimab (an anti-PD1 antibody made by Regeneron Inc.) for advanced unresectable stage III or IV melanoma patients unresponsive to anti-PD-1 therapies.

In mRNA vaccines encoding TAAs, priority should be given to TAAs exclusively expressed in tumor cells, minimizing expression in normal cells. Comprehensive studies on TAAs’ expression patterns across tumors and normal tissues are crucial for identifying safer and more effective targets.

Tumor-specific antigens (TSAs)

Tumor-specific antigens (TSAs) are proteins or peptide fragments uniquely expressed or significantly overexpressed in cancer cells compared to normal cells. These antigens result from various oncogenic processes, including mutations, abnormal gene expression, and post-translational modifications.¹ Neoantigens are presented on the surface of cancer cells via Major Histocompatibility Complex (MHC) molecules. These neoantigens are absent in normal tissues, making them ideal targets for immunotherapeutic strategies. Their uniqueness to individual tumors circumvents the issues of tolerance and autoimmunity often associated with targeting shared tumor antigens. In 2014, Tran et al. successfully treated a bile duct cancer patient with ERBB2IP-specific CD4+ cells, achieving complete tumor regression,¹²⁷ demonstrating TSAs’ potential in inducing anti-tumor responses. They also found gene mutations in 9 out of 10 gastrointestinal tumor patients,¹²⁸ which indicates the widespread presence of tumor heterogeneity, with TSAs varying among patients even within the same tumor type. The development of personalized mRNA vaccines involves the identification and selection of neoantigens through advanced sequencing and bioinformatics approaches. Next-generation sequencing (NGS), high throughput screening and machine learning algorithms have facilitated TSA prediction by identifying mutations and aberrant transcription/translation events in individual tumor genome exons.^129–132

The clinical potential of personalized mRNA neoantigen vaccines is underpinned by their ability to induce robust and specific immune responses. Early-phase clinical trials have shown promising results, with evidence of vaccine-induced T cell responses and potential clinical benefits (Table 5). Ugur Sahin et al., identified mutations most likely to provoke immune responses in 13 melanoma patients and developed mRNA encoding up to 10 novel antigens. The clinical trial showed that among 13 patients, 8 had no recurrence within a year after vaccination, and of the remaining 5, 2 experienced tumor shrinkage, with 1 achieving complete remission combined with PD-1 antibody treatment.¹³³ mRNA-4157 and BNT122 are two typical personalized mRNA vaccines underway for clinical trials. mRNA-4157, encoding a repertoire of 34 antigens targeting unique mutations in individual patients’ tumor DNA, is undergoing clinical trials for melanoma, non-small cell lung cancer, and other solid tumors. The clinical trial testing mRNA-4157 combined with pembrolizumab (anti-PD1 made by Merck Inc.) in head and neck squamous cell carcinoma patients showed a 50% overall response rate and a 90% disease control rate. Two phase IIb clinical trials combining mRNA-4157 and pembrolizumab for high-risk melanoma patients reduced the risk of recurrence or death by 44% and 49%, respectively, suggesting more pronounced benefits with longer treatment durations.¹³⁴ The combination therapy of mRNA-4157 and pembrolizumab has advanced to Phase III clinical trials (NCT05933577), marking it as the first mRNA cancer vaccine to reach this stage.

Table 5. Selected representative mRNA vaccine clinical trials with published results.

NCT Number	Trial phase	Target antigen	Conditions	Patients	Interventions	Clinical response
NCT04503278	Phase I/Phase II	Claudin 6 (CLDN6)	Solid Tumor	22	Biological: CLDN6 CAR-T/CLDN6 CAR-T(A) Biological: CLDN6 unmodified RNA-LPX/CLDN6 modified RNA-LPX	The unconfirmed ORR in 21 evaluable patients was 33% (7 of 21), including one complete response. The disease control rate was 67% (14 of 21), with stable disease in seven patients. Patients with germ cell tumors treated at the higher DL exhibited the highest response rate (ORR 57% (4 of 7)).
NCT03897881	Phase II	Individualized neoantigen	Melanoma	157	Drug: mRNA-4157 Biological: Pembrolizumab	mRNA-4157 (V940) in combination with KEYTRUDA reduced the risk of recurrence or death by 49% (HR = 0.510 [95% CI, 0.288–0.906]; one-sided nominal p = .0095) and the risk of distant metastasis or death by 62%
NCT03480152	Phase I/Phase II	Individualized neoantigen	Melanoma, Colon cancer, Gastrointestinal cancer, Genitourinary cancer, Hepatocellular cancer	5	Biological: National cancer institute (NCI)-4650, a messenger ribonucleic acid (mRNA)-based, Personalized cancer vaccine	The vaccine was safe and elicited mutation-specific T cell responses against predicted neoepitopes not detected before vaccination, and potential future combination of such vaccines with checkpoint inhibitors or adoptive T cell therapy should be evaluated for possible clinical benefit in patients with common epithelial cancers.
NCT03164772	Phase I/Phase II	5T4, MAGEC1, MAGEC2, MUC1, NY-ESO-1, survivin	Metastatic non-small cell lung cancer	61	Drug: Durvalumab Drug: Tremelimumab Biological: BI 1,361,849 Device: PharmaJet Tropis device	Compared to baseline, 80% of patients exhibited an elevation in antigen-specific antibody levels, 40% demonstrated an increase in functional T cells, and 52% showed evident multi-antigen specific responses. In combination therapy with pembrolizumab, one case achieved partial response (PR), and 46.2% of patients experienced stable disease (SD).
NCT02366728	Phase II		Glioblastoma, Astrocytoma, Grade IV, Giant cell glioblastoma, Glioblastoma multiforme	64	Biological: Unpulsed DCs Biological: Td Biological: Human CMV pp65-LAMP mRNA-pulsed autologous DCs Biological: 111In-labeled DCs Drug: Temozolomide Drug: Saline Drug: Basiliximab	The 3-year OS for Td-treated patients was 34% (CI95 19–63%) compared to 6% given unpulsed DCs (CI95 1–42%), The CMV pp65-LAMP RNA-pulsed dendritic cell vaccination was associated with positive immunologic and clinical response in patients with glioblastoma (GBM).
NCT01817738	Phase I/Phase II	PSA, PSMA, PSCA,	Prostate cancer	197	Biological: CV9104 Biological: Placebo	No significant differences in progressionfree survival
		STEAP1, PAP, and MUC1
NCT00923312	Phase I/Phase II	MAGE-C1, MAGE-C2, NY-ESO-1, survivin,	Non small cell lung cancer	46	Biological: CV9201	No objective responses; progression-free survival and overall survival not improved
NCT04161755	Phase I	a maximum of 20 neoantigens per patient	Pancreatic Cancer	29	Drug: Atezolizumab Biological: RO7198457 Drug: mFOLFIRINOX	At 18-month median follow-up, patients with vaccine-expanded T cells (responders) had a longer median recurrence-free survival (not reached) compared with patients without vaccine-expanded T cells (non-responders; 13.4 months, p = .003).

Personalized mRNA vaccines offer promising treatments for cancers resistant to current immunotherapy, such like pancreatic ductal adenocarcinoma (PDAC), which shows resistance to immune checkpoint inhibitors (ICIs).¹³⁵ The abundance of neoantigens in pancreatic cancer makes neoantigen-based mRNA vaccines a viable therapeutic strategy. BNT122 employs mRNA to express 20 novel antigens in PDAC patients. When combined with chemotherapy and immune checkpoint therapy, BNT122 exhibits potential in decelerating the recurrence in PDAC patients, with a follow-up global randomized trial (IMCODE 003, BNT122) forthcoming.¹³⁶

TSA mRNA vaccines represent a highly personalized approach, tailoring each vaccine to an individual’s tumor antigen. Combined with immune checkpoint inhibitors, these vaccines offer new avenues for treating diseases long challenging to medical science.

mRNA-encoded immunomodulators

The immunosuppressive tumor microenvironment (TME) is characterized by limited T cell infiltration or an abundance of immune inhibitory cells, contributing to resistance against current therapies. Cytokines, crucial for regulating intercellular communication among immune cells, can efficiently transduce immune signals and orchestrate robust immune responses against target antigens. The delivery of cytokines, co-stimulating ligands, or other immune regulatory factors via mRNA can effectively remodel the TME and enhance tumor sensitivity to various immunotherapies.

Innovative approaches include the delivery of mRNA encoding the tumor suppressor p53 to liver cancer cells using nanoparticles, combined with anti-PD-1 monoclonal antibodies. This strategy has shown effectiveness in modulating the TME and achieving anti-tumor effects.¹³⁷ Additionally, previous studies have exploited mRNA encoding caTLR4 and co-stimulatory ligands CD70 and CD40L to stimulate immune responses in APCs. The followed enhancement in T cell functionality has been demonstrated to combat tumors effectively and suppress early resectable breast cancer lesions in mice.¹³⁸ mRNA-2752, an mRNA vaccine encoding three immune regulatory factors, is currently under Phase I and Phase II clinical trials to assess its safety and tolerability, both as a standalone treatment and in combination with fixed-dose durvalumab (anti-PDL1 antibody made by AstraZeneca Inc.) in patients with advanced malignancies.¹²⁹

Conclusions and future perspectives

The successful deployment of COVID-19 vaccines has significantly accelerated the advancement of mRNA technology in oncology. Utilizing innovative RNA-based vectors and targeted strategies, new mRNA vaccine platforms are being developed to target an extensive range of tumor antigens. Notably, the progress in personalized mRNA vaccine marks a considerable breakthrough in cancer treatment. These vaccines show enhanced potential in clinical trials, particularly when combined with existing therapies, including ICI and traditional treatments.

Despite comprehensive development, mRNA cancer vaccines still confront multiple challenges. Further research is essential to refine the development and application of these vaccines. Continued improvements in vaccine design, delivery technologies, and rapid identification of tumor neoantigens are crucial, positioning mRNA vaccines as a promising tool in personalized cancer therapy.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [Grant No. 32370923, 81972692].