mRNA-based therapeutic strategies for cancer treatment

Abstract

In the rapidly evolving landscape of medical research, the emergence of RNA-based therapeutics is paradigm shifting. It is mainly driven by the molecular adaptability and capacity to provide precision in targeting. The coronavirus disease 2019 pandemic crisis underscored the effectiveness of the mRNA therapeutic development platform and brought it to the forefront of RNA-based interventions. These RNA-based therapeutic approaches can reshape gene expression, manipulate cellular functions, and correct the aberrant molecular processes underlying various diseases. The new technologies hold the potential to engineer and deliver tailored therapeutic agents to tackle genetic disorders, cancers, and infectious diseases in a highly personalized and precisely tuned manner. The review discusses the most recent advancements in the field of mRNA therapeutics for cancer treatment, with a focus on the features of the most utilized RNA-based therapeutic interventions, current pre-clinical and clinical developments, and the remaining challenges in delivery strategies, effectiveness, and safety considerations.

Graphical abstract

Vu and colleagues offer an insightful overview of RNA-based therapy, a recently emerged as an exciting therapeutic approach. The review describes the rational design of synthetic mRNA and strategies to deliver mRNAs, and offers an extensive discussion of current developments, as well as safety and regulatory considerations for clinical applications of these technologies.

Introduction

RNA-based therapies encompass a diverse array of strategies aiming at manipulating the expression of pathogenic genes and proteins for treatment of diseases (summarized in Table 1 and Figure 1). Small interfering RNA (siRNA) and microRNA capitalize on the regulatory mechanisms of RNA interference to silence gene expression.^1,2 The small RNA molecules are loaded onto the RNA-induced silencing complex to guide the complex toward complementary mRNAs, leading to initiation of mRNA degradation or translation inhibition.^1,3 Currently, only siRNA based pharmaceuticals have been successfully implemented for clinical applications. To date, four siRNA based drugs have been approved by the U.S. Food and Drug Administration (FDA) between 2018 and 2020.^4,5 These are patisiran for polyneuropathy, givosiran for acute hepatic porphyria, lumasiran for primary hyperoxaluria type 1, and inclisaran for the treatment of high low-density lipoprotein cholesterol.

Table 1.

Types of RNA-based therapeutics

	Description	Clinical applications	Main caveats	Drugs
siRNA	siRNA: targets and bind highly specific complementary RNAs through the RISC to induce its degradation and prevent protein production	genetic disorders, hypercholesterolemia, cancers (4 FDA-approved drugs up to date)	effective and safe delivery to target cells and tissues, avoiding off-target effects, managing immunogenicity	patisiran, givosiran, lumasiran, and inclisaran (4)
miRNA	miRNA: targets and bind various complementary RNAs through RISC to induce its degradation and prevent protein production	cancers, infections	effective and safe delivery to target cells and tissues, avoiding off-target effects, managingimmunogenicity	N/A
CRISPR	sgRNA guides Cas9 or Cas13 to specific DNA or RNA sequences for genome of transcription editing	genetic disorders, cancers, adoptive cell therapies, infectious diseases	potential off-target effects, risks of immune responses to the Cas proteins, usage of viral vectors for delivery	N/A
saRNA	saRNA: binds promoter regions and allow overexpression of endogenous therapeutic genes such as tumor suppressor	genetic disorders, cancers	effective and safe delivery to target cells and tissues, avoiding off-targets, managing immunogenicity	N/A
Ribozymes	RNA molecules with catalytic properties, cleaves specific RNA molecules, altering gene expression or inhibiting pathogenic processes	genetic disorders, viral infections, cancers	optimizing ribozyme activity and specificity, minimizing off-target effect	RZ-OOI (FTD) (9)
Oligo nucleotides	ASOs bind to specific RNA molecules, promoting degradation or altering splicing processes	rare genetic disorders, neurodegenerative diseases (e.g., ALS), cancers	effective and safe delivery to target cells and tissues, avoiding off-target effects, managing immunogenicity	Fomivirsen, pegaptanib, mipomersen, nusinersen, inotersen, defibrotide, eteplirsen, golodirsen, viltolarsen, casimersen (10)
Aptamers	binds to disease-related, small molecules, proteins, cells and tissues, inhibiting their activity or help delivering drugs	Cancers, anticoagulant therapies, targeted drug delivery	effective and safe delivery to target tissues, optimizing binding affinity, specificity of aptamers, avoiding off-targets	pegaptanib, avacincaptad pegol (14,15)
mRNA	deliver synthetic mRNA molecules to cells, produce therapeutic proteins or antigens	vaccines, cancer immunotherapy, regenerative medicine	effective and safe delivery to target cells and tissues, avoiding off-targets, managing immunogenicity	SARS-COV-2 vaccines

miRNA, microRNA; RISC, RNA-induced silencing complex.

Figure 1.

Types of RNA-based therap

Illustrations of various types of RNA therapy, including siRNA, microRNA (miRNA), CRISPR-Cas9/Cas13, saRNA, ribozymes, ASOs, aptamer, and mRNA.

Besides RNA interference, the CRISPR-Cas9-based method also utilizes RNA to guide the Cas9 enzyme to targeted genome regions for gene editing. The technology with the ability to edit the genome at the single nucleotide level holds the potential for treating SNP-related genetic diseases. Further improvements, including the development of fast-acting CRISPR and the use of different Cas enzymes such as Cas13, which directly edit RNAs,⁶ will broaden its uses and applications. To date, 39 CRISPR-based clinical trials for the treatment of different cancer types have been initiated according to clinicaltrials.gov.

Another form of RNA therapeutics are ribozymes and small activating RNAs (saRNAs). Ribozymes are catalytic RNAs that can be used to target and cleave RNAs. A first drug based on ribozymes, RZ-001, FDA Fast Track Designation for glioblastoma has gained FDA fast-track designation in 2023 to treat glioblastoma.⁷

mRNA-based therapeutics enable the induction of exogenous therapeutic proteins. One notable advantage of this method is its versatility; in principle, it allows for the expression of any desired protein in target cells.⁸

A wide range of other ribonucleotide acids plays a role in cancer development and are currently under development as potential treatments. saRNAs are short, double-stranded RNA molecules that target promoters to induce the expression of endogenous target genes.⁹ Overexpression of PIWI-interacting RNAs have been shown to significantly stall tumor growth in gliobastoma.¹⁰ Circular RNA circCDYL2 could be leveraged to answer drug resistance issues in HER2⁺ breast cancer.¹¹ Transfer RNA-derived fragment expression levels affects patients prognosis and T cell activation and could be used as therapeutic targets.^12,13

In addition, DNA-based short oligonucleotides can be used to target endogenous RNAs degradation through complementary base pairing. Antisense oligonucleotides (ASOs) are the treatment with the highest number (10) of FDA-approved drugs to date (fomivirsen, pegaptanib, mipomersen, nusinersen, inotersen, defibrotide, eteplirsen, golodirsen, viltolarsen, and casimersen).¹⁴ ASOs utilize two primary mechanisms of action: directly targeting mRNAs (gap-mer) or modulating splicing processes.^15,16 These therapeutic approaches show potential in tackling rare genetic diseases, as well as neurodegenerative disorders such as amyotrophic lateral sclerosis.¹⁷ Another use case of oligonucleotides is aptamers, short single-stranded DNA or RNA molecules with a complex three-dimensional structure that binds directly to a wide range of molecular target (small molecules, cell surface receptors, etc.). To date, two aptamer drugs have been approved by FDA, including pegaptanib to treat age-related macular degeneration and avacincaptad pegol, which is used in geographic atrophy secondary to age-related macular degeneration.^18,19

While diverse in their mechanisms of actions, all of these strategies share similar caveats and challenges. Delivery strategies need to keep small exogenous RNA molecules or oligonucleotides protected from RNAse and the immune system, while ensuring specific intracellular delivery to the targeted cells. In particular, RNA molecules, compared with DNA, are very fragile and sensitive to the RNAse action present everywhere in the body. Tolerability issues also have to be addressed to avoid deleterious immune reactions to these exogenous molecules, which has already stalled progress of various clinical trials.²⁰ The safety features along with a wide range of applications makes mRNA-based therapeutics a favourable approach.

mRNA was first discovered by pioneer scientists François Jacob, Jacques Monod, and Matthew Meselson.^21,22 mRNA was identified in the T4 bacteriophage, where it was characterized as the intermediary that carries genetic information from the nucleus to the ribosomes, where proteins are synthesized. A significant milestone came in 1990 when in vitro transcription (IVT) mRNA was successfully injected into mice, eliciting the expression of the encoded protein in the mouse muscle cells.²³ Clinical trials of mRNA applications began a few years after the initial mice injection in 1990. It emerged as a promising approach for prophylactic immunization delivery of synthetic mRNA-encoding antigens to elicit immune responses. In 1995, the first mRNA cancer vaccine was developed, encoding the carcinoembryonic antigen to elicit an anti-tumor immune response in mice.²⁴ Subsequently, in 2009, the first mRNA-based immunotherapy was tested in melanoma patients, along with the first chimeric antigen receptor (CAR) adoptive immunotherapy designed with mRNA encoding the chimeric receptor.^25,26

Today, mRNA-based approaches are revolutionizing various fields of medicine, ranging from infectious diseases to cancer treatment, regenerative medicine, and disease diagnosis. The groundbreaking potential of mRNA-based medicines emerged during the coronavirus disease 2019 (COVID-19) pandemic, where mRNA vaccines demonstrated unprecedented efficacy in eliciting immune responses and long-term antibody production to neutralize the virus. The rapid development of the mRNA vaccines by Moderna and Pfizer-BioNTech not only sparked interest in mRNA for various prophylactic and therapeutic applications, but also validated the platform’s viability in production, quality control, and distribution.²⁷

With the focus on mRNA therapeutics in cancer, this review aims to provide an overview of rational design of synthetic mRNA and strategies to deliver mRNAs. The review covers current applications of mRNA therapy in cancer treatment, while bringing in a discussion of additional considerations of safety and regulatory for broad applications of the technologies. The remarkable potential held by mRNA-based drugs can transform the landscape of medicine, ushering in a new era of targeted and effective therapies.

Synthetic mRNA as a therapeutic reagent

The ideal mRNA therapeutic would embody several essential features. First, it should encode for a functional, non-pathogenic protein, carrying all the necessary elements for translation and post-translational modifications, while possessing a sufficiently long half-life to carry out its therapeutic functions effectively. Delivery of the mRNA therapeutics should ensure the administration of enough molecules to guarantee the desired therapeutic effects while minimizing unwanted immune responses. Moreover, a delivery strategy must be highly targeted, exclusively affecting the intended cells, and carefully designed to avoid any off-target effects. Therefore, the two most important aspects are the effective design of an mRNA transcript and the packaging and delivery strategy.

Production and design of synthetic mRNAs

A synthetic mRNA is typically transcribed and amplified in vitro in a PCR by T7 or SP6 RNA-polymerase from a plasmid carrying a specific DNA sequence. The sequence is designed with an open reading frame (ORF) that encodes the desired protein, framed by start and stop codons along with 5′ and 3′ UTRs. A poly T-tail of 120 thymidines (T) is added to the amplicons by a reverse primer with T₁₂₀ extension. The resulting RNA is purified, and any remaining 5′ triphosphates in the PCR are removed to prevent potential immune responses and recircularization.²⁸

Because the therapeutic effects of mRNAs depend on its in vivo translation efficiency, optimization of regulatory elements of mRNA to improve proteins synthesis is key.²⁹ The requirement for a higher level of protein expression cannot be resolved by administering a larger load. This is due to the heightened innate immunogenicity of mRNAs, which hinders protein production.^8,30 Therefore, researchers have focused on an optimal design of a mRNA cargo (Figure 2) with a modified 5′ cap, poly(A) tail, ORFs, 5′ and 3′ UTRs, and coding sequences.^29,31 These elements influence protein expression by enhancing mRNA stability and translation efficacy to achieve the protein expression levels required for mRNA therapy.^8,29

Figure 2.

Modification of mRNA structure for therapeutic use

mRNA is composed of an ORF carrying the code sequences, flanked by two UTRs. The 3′ end is adenylated forming the poly A tail and the 5′ end is capped by a modified guanine. In vitro transcribed mRNA can be engineered to express a wanted sequence, carry some post-transcriptional modifications or viral replicase to amplify mRNA translation in targeted cells.

The 5′-m7G cap protects the mRNA from 5’→3′ exonuclease mediated degradation, thus prolonging the half-life of the mRNA.³² The 5′ cap is also recognized by the eukaryotic initiation factor 4F (eIF4F) during the cap-dependent translation initiation step, which is the rate-limiting step of protein synthesis.^32,33 To enhance the stability of mRNA and translation efficiency, 5′ cap analogs were created through modifications to the 5′ cap structure. This allows for a higher affinity for eIF4E (a subunit of eIF4F) while decreasing its affinity to decapping enzymes.³⁴ The 5′ cap analogs also increase IVT efficiency by improving the capping efficiency by capping enzymes.^32,35 Among them, 3′-O-Me-m7G(5′)ppp(5′)G is an anti-reverse cap analogue (ARCA) that only allows for capping in the correct orientation and thus is highly effective in preventing the rapid degradation and poor translation associated with reverse capping.³⁶ ARCAs used in liposomes have been showed to lead to a 100-fold enhancement in translation activities.³⁶

The 3′-poly(A) tail size plays a central role in mRNA quality control and affects translation efficiency.³⁷ Contrary to popular belief of the optimal requirement of 100–250 adenosines, only mRNAs containing fewer than 20 adenosines showed decreased translational efficacy of exogenous and endogenous mRNA.^37,38 Therefore, a typical design is to add approximately 120 adenosines to create a stable poly(A) tail for all IVT mRNAs.

mRNAs contain 5′ and 3′ UTRs that involve in multiple processes of mRNA degradation and translation.³⁹ The 5′ UTR contains elements that initiate translation either by cap-dependent or cap-independent mechanisms.⁴⁰ On the other end, the 3′ UTR contains sequences highly enriched in adenosine and uridine called AU-rich elements, which can accelerate the rate of mRNA degradation.⁴¹

Depending on the protein, codon optimization has been shown to enhance eukaryotic translational efficacy and translation rates.^42,43 Different organisms show preferential biases toward certain codons that encode for the same amino acid.⁴⁴ However, other proteins may require a slow translation rate to properly fold and maintain its folded state.⁴⁵ To maximize both the translation rate and efficacy, the mRNA sequence needs to be examined and an appropriate codon optimization strategy needs to be implemented.⁴⁵ Another recent advance in the field includes the use of self-amplifying mRNA vaccines. Developed from RNA viruses, these vaccines deliver to the host cells mRNA encoding for the wanted immunogen, as well as viral replication machineries, which enable the amplification of the number of mRNAs available for translation and increase the production of immunogens.⁴⁶ Self-adjuvant mRNA-based cancer vaccines designed with a major histocompatibility complex (MHC) I-targeting domain allow maximization of the adaptative immune response against cancer cells.⁴⁷

Strategies for in vivo delivery of mRNA therapeutics

To obtain in vivo activity, mRNA molecules must be encapsulated for delivery.^48,49 An ideal method would have minimal immunogenicity and toxicity, a high loading capacity, efficient cytoplasmic delivery by escaping endosomal degradation, and defined tropism and tissue targeting, while being inexpensive and non-labor intensive to produce, as well as having a long storage life. Current delivery methods of mRNAs can be classified into two main groups: viral and non-viral vectors. Although viral vectors offer advantages in delivery efficacy, their innate drawbacks limit their potential clinical applications.⁵⁰ The use of viral vectors carry unfavorable immunological side effects, such as strong cytotoxic T lymphocyte responses, vector dose-dependent direct cytotoxicity, lack of specific tissue targeting, genome insertional mutagenesis,⁵¹ difficulty in controlling the gene expression level,⁵¹ vector size limitation,⁵¹ high immunogenicity, restricted tissue tropism, high production costa, and biosafety constraints.^50,51,52 In recent years, the development of novel and improvements in non-viral delivery methods have overcome many of the shortcomings of viral vector-based approaches.

Lipids and lipids-based nanoparticles (LNPs) are the predominant non-viral method for mRNA delivery. Lipids are organic, water-insoluble compounds.⁵³ Lipids have a hydrophilic head group and two hydrophilic tails that can form different structures via self-assembly based on the relative size ratio between the head group and the tails.⁵³ A typical LNP formulation is are structurally composed of four components: cationic or ionizable lipids, a helper lipid, poly(ethylene glycol) (PEG) lipids and zwitterionic phospholipids (PLs).^52,53,54 These different classes of lipids are discussed in details below and shown in Figure 3.

Figure 3.

LNP structure and formulation

There are three types of lipids constituting LNPs: ionizable cationic lipids, zwitterionic PLs, and helper lipids. Various types of lipids can provide distinct features affecting encapsulation of mRNA, tissue targeting, cell delivery, and immunogenicity.

Cationic and ionizable lipids

Cationic lipids and ionizable lipids are the main components to formulate LNPs for mRNA delivery. Cationic lipids are able to retain their positive charge or remain cationic in a pH-independent manner. Ionizable lipids, in contrast, have a high enough pKa value that these lipids are positively charged at a low pH during LNP formation and also a low enough pKa to be neutral at physiological pH.^52,53,55 During LNP formation in an aqueous environment, cationic and ionizable lipids form LNP-mRNA complexes spontaneously via hydrophobic interactions and electrostatic interactions between the positively charged lipids and negatively charged mRNA.^52,53 The positive charge of the cationic and ionizable lipids allow interactions with the anionic lipid bilayers of the endosome, forming a cone-shaped structure (hexagonal H_II phase structure), which disrupts the endosome bilayer allowing for the delivery of mRNA cargos into the cytoplasm.^56,57,58 Early studies widely used cationic lipids for the in vitro transfection of mRNA due to their high delivery efficacy, cytotoxicity, and high immunogenicity.^52,53 However, the positive charge of cationic lipids causes LNPs to be opsonized with serum proteins, resulting in uptake by macrophages in the reticuloendothelial system and subsequent clearance from circulation.^59,60,61 Furthermore, cationic lipids cause Toll-like receptor 4-mediated pro-inflammatory responses and produce reactive oxygen species, prompting the development of ionizable cationic lipids.^61,62,63 Despite these caveats, cationic lipids still are used to develop mRNA vaccinations, as they are able to stimulate anti-viral immune responses in antigen presenting cells and act as an adjuvant.⁶⁴ Optimization over two decades from the discovery of the first ionizable lipid 1,2-dioleoyl-3-dimethylaminopropane resulted in an 8,000-fold improvement in the therapeutic index.⁶⁵

Helper lipids

The term helper lipids has been used to define a broad category of lipids that include PLs and glycerolipids, as well as sterols such as cholesterol. Cholesterol is commonly incorporated into LNP formulations, as it is essential for encapsulating nucleic acids and improving vesicle stability.^52,65 Compared with cholesterol-free liposomes, incorporating cholesterol and varying its ratio in LNP formulations can alter the binding to serum proteins after injection, affecting clearance, as well as modulating stability, permeability, fluidity, and resistance to aggregation.^66,67,68,69 Additionally, the incorporation of phytosterols, or cholesterol analogues, has been shown to significantly affect LNP morphology and gene transfection.⁷⁰

PEG lipids

PEGylated lipid groups (PEG lipids) are used in LNP formulation primarily to stabilize LNP particles and to prevent excessive binding and opsonization by serum protein, as well as non-specific uptake to ultimately increase the half-live of LNP-mRNAs in circulation.^52,53 By changing the PEG content, the size of the LNP particle can be adjusted, which has shown to affect mRNA translation efficiency.^71,72 PEG lipids at the surface of the liposome prevent aggregation and fusion of LNP particles, increasing storage stability.⁷³ However, PEG lipids on the surface of LNPs may result in anti-PEG IgM production, which hampers mRNA therapeutic efficacy.^74,75,76

Zwitterionic PLs

The last group of lipids in LNP formulation are the zwitterionic PLs. The most commonly used types of PLs are saturated phosphatidylcholine (PC) and unsaturated phosphatidylethanolamine (PE) containing lipids.^52,65 Following their long usage in liposomes for drug delivery, 1,2-distearoyl-sn-glycero-3-phosphocholine is being used in LNP formulations as well, used in the formulation of patisiran (Onpattro) and severe acute respiratory synbdrome coronavirus 2 (SARS-CoV-2) mRNA vaccines.^77,78 While PC-containing PLs showed increased uptake levels by endocytosis compared with PE-containing PLs, replacing with PE-containing PLs increased mRNA delivery by 3- to 5-fold.^65,78,79 The increase in mRNA delivery is due to the headgroups of PE can form a cone-shaped structure with membrane bilayers similar to cationic ionizable lipids, allowing improved intracellular delivery of mRNA.^58,65,78

Targeting specific organs has been a major challenge in LNP uses; intravenously injected LNPs display liver tropism due to apolipoproteins (ApoE) preferentially binding to lipoprotein-rich LNPs in the blood serum.^53,78 ApoE regulates plasma lipoprotein and cholesterol levels, mediating the transport and clearance of ApoE-bound LNPs by interacting with surface low-density lipoprotein receptors in the liver.^65,80,81,82 To overcome this challenge, a strategy called selective organ targeting (SORT) was developed in recent years.⁷⁸ The integration of SORT lipids into LNP formulations forms SORT LNPs that allow gene delivery to the lungs, spleen, or liver.^78,83 LNP formulations with anionic SORT molecules such as 20%–30% DMP or 10%–40% 18PA incorporation resulted in spleen-specific delivery while the addition of increasing percentage of cationic 1,2-dioleoyl-3-trimethylammonium-propane showed lung tropism.^78,83

In addition, it is critical to achieve tumor-specific gene delivery. Anchored secondary scFV-enabling targeting was developed, which coats LNPs with cell-specific antibodies to enhance on-target efficacy.^48,84 A study successfully inhibited tumor growth of OV8 peritoneal xenografts by specifically targeting epidermal growth factor receptor (EGFR), which is overexpressed in OV8 tumor cells.⁸⁵ Intraperitoneal injection of LNPs coated with EGFR antibodies preferentially targeted OV8 tumor cells.⁸⁵ This approach at least allows for specific targeting of tumors that display a unique surface expression profile that are targetable by antibodies.⁴⁸ While targeting the tumor tissue directly remains an unsolved problem, researchers are using LNPs for cancer vaccines.^86,87 A recent study developed a novel mRNA cancer vaccine against murine melanoma model by using a novel lymph node-specific ionizable lipid (113-O12B) as a delivery method.⁸⁶ Indeed, 113-O12B resulted in higher mRNA expression in the lymph node and lower mRNA expression in the liver when administered intramuscularly.⁸⁶ In addition, 113-O12B showed enhanced expression in antigen-presenting cells (APCs), increased antibody and CD8⁺ T cell responses, and greater infiltration of APCs to the tumor site, which lead to an overall improvement in therapeutic efficacy.⁸⁶

Overall, the major obstacles in implementing mRNA therapeutics are to limit unwanted immune responses, improve targeting accuracy to avoid off-target events, enhance protein yield and half-life, and improve RNA purity. Addressing these critical fields holds the key to advancements in increasing efficacy and safety of mRNA-based therapeutic interventions.

mRNA-based therapeutics in clinical developments for cancer treatment

mRNA-based approaches are proving to be revolutionary in tackling a wide array of health challenges. In cancer, they have been used in several major applications: (1) encoding tumor antigens or immune stimulatory molecules to boost the endogenous immune response; (2) as biomarkers for diagnosis or responses to treatment; (3) cancer vaccines; and most predominantly in (4) CAR-T cell therapy.

Encoding tumor antigens or immune stimulatory molecules

The delivery of target mRNAs can elicit robust immune responses toward cancer cells. In fact, several clinical trials aiming to harness the body’s immune response to combat different cancer types are already underway. LNPs encapsulated with mRNAs encoding cytokines (IL-12, IL-27, and granulocyte macrophage colony stimulating factor) were developed to direct immune-stimulatory therapy in tumor microenvironments. Intratumoral injection of LNPs carrying IL-12 mRNAs proved most effective in inhibiting melanoma tumor growth, and a combination of IL-12 and IL-27 mRNA showed a synergistic effect without causing systemic toxicity. This approach demonstrates a promising new strategy for cancer treatment, notably by promoting immune cell infiltration into tumors.⁸⁸ Antibodies can be used in a synergistic approach to augment the efficacy of antigen-loaded dendritic cells. mRNA encoding for two immune modulatory antibodies were utilized to target T lymphocytes membrane receptors CTL4 and GITR, respectively. GITR is commonly expressed on regulatory T cells and both receptors are known for their involvement in inducing tolerance to self-antigens and exerting immune suppression. The administration of antibodies against these receptors led to enhanced anti-tumoral activity, resulting in improved survival in melanoma-bearing mice.⁸⁹

RNA molecules as biomarkers

mRNA molecules are also gaining prominence as molecular biomarkers for the diagnosis and monitoring of diseases. A cancer diagnosis is usually based on solid biopsy specimens, which can be highly invasive and require significant turnaround time for a complete analysis. The future of cancer diagnosis is shifting toward liquid biopsy, where circulating cancer cells or circulating tumoral DNA can be detected in a blood sample and allow the detection of genomic markers with minimal invasiveness.⁹⁰ Blood is now the commonly used surrogate tissue for biopsy and analysis. RNA, including mRNAs, can also be effectively extracted from whole blood specimens. Importantly, it has been demonstrated that, although the mRNA content in blood is mainly derived from erythroid cells, the transcriptomic profile exhibits a remarkable 80% similarity when compared with other major tissues.⁹¹ In colorectal cancer (CRC), a broad spectrum of mRNAs, including BIRC5, MLH1 or cytokeratins, is detectable at high levels in patients’ peripheral blood. These molecules serve as specific diagnostic markers for the disease.⁹² Elevated levels of cell free BIRC5 mRNA in CRC patients’ serum is also a significant predictor for invasiveness and aggressiveness and is associated with poor clinical outcomes.⁹³ Stratification of CRC patients based on survival prognosis following surgical intervention was also allowed by a panel of five mRNA biomarkers.⁹⁴ In challenging to diagnose cancers, the effective identification of mRNA biomarkers provides hope for early disease detection. In the example of ovarian cancer, elevated levels of KRAS, c-FOS, PUMA, and EGFR mRNAs have been detected in patients in comparison with healthy controls.⁹⁵

Cancer vaccines

Vaccine strategies for cancer treatment includes cell vaccines, protein/peptide vaccines, and nucleic acid vaccines (DNA, RNA).⁹⁶ Understanding the involvement of the immune system in cancer progression and deciphering how cancer cells evade immune responses have paved the way for the emergence of immunotherapies. The results obtained with check-point inhibitors or adoptive T cells encourage research efforts in cancer vaccines to stimulate host immune system to better recognize neoantigens and target cancer cells. This breakthrough has significantly altered the landscape for treatments of cancer. mRNA vaccines employ a synthetic form of mRNA to instruct target cells to produce a harmless protein able to trigger an immune response in the recipient. This protein is recognized by the immune system as foreign, prompting the production of antibodies and the activation of immune cells to combat potential future infections (Figure 4).

Figure 4.

Mechanism of action of mRNA vaccine for cancer treatment

LNPs carrying the therapeutic mRNA is delivered to the cell through endocytosis and translated into a protein using the host cell translation machinery. This protein can be secreted and enter the cell again and produce an antigen that will be presented on the class II MHC, activating CD4⁺ T cells and B cells, triggering an antibody-dependent immunity response. The protein can also go through proteasome degradation and the subsequent production of peptides can allow the loading of class I MHC and presentation of the antigen to cytotoxic LT CD8⁺.

While DNA vaccines trials showed promising results, the use of RNA and mRNA alleviates carcinogenicity issues due to DNA genome incorporation. RNA-based vaccines are functional in the cytoplasm, which also allows faster clearance and less intense side effects. Non-prophylactic mRNA vaccines in cancer are similar in principle to their virus infection counterparts, with the viral antigen replaced by a cancer neoantigen. In the same way that viruses do, cancer cells mutate to adapt to the immune response and progressively become better at escaping immune surveillance as the disease progresses. An mRNA formulation allows cells to translate, secrete, or express a surface immunogen capable of eliciting an adaptative and specific immune response against the cancerous cells. Delivery strategies can vary and some current clinical trials use naked mRNAs, self-amplifying RNAs, or dendritic cells loaded with mRNAs ex vivo.^97,98

The main difficulty associated with vaccines for cancer is determining the specific antigen. One possibility is to use mRNA encoding for a unique tumor-associated antigen (TAA) or a tumor-specific antigen. Cancer cells produce neoantigens, typically mutated proteins mostly absent in healthy cells, rendering them highly specific targets. mRNA vaccines can be used to train the immune system to recognize one or several of those TAAs to selectively kill cancer cells. Several TAAs cocktail are currently in the clinical pipeline and gave interesting results in the treatment of metastatic melanoma and non-small cell lung cancer.⁹⁷

A potential issue with this technique is the differential expression of TAAs between different patients whose cancer cells showed different genetic signatures. Moreover, some TAAs are also expressed by healthy cells, which exposes patient to deleterious adverse effects of auto-immune reactions. A way to make this strategy more personalized is to use neoantigens specific to each patient. These neoantigens can be identified through the sequencing of cancerous tissues and infiltrating T lymphocytes (TILs) from individual patients. In fact, a personalized mRNA vaccine combining more than 20 antigens was developed for metastatic gastrointestinal patients.⁹⁹ These antigens were identified from mutations found in common oncogene (TP53, KRAS, etc.), and HLA class I candidate neoantigens that were predicted in silico to bind to each patient’s MHC alleles. Neoantigens efficiently recognized by the patient’s own TILs cells were also included. This highly personalized vaccine platform triggered T cell reactivity against some of the targets predicted during vaccine design. While clinical improvement was minimal, this study provides an interesting starting point to develop safe and immunogenic personalized vaccines to treat cancer patients.⁹⁹ Another approach is the development of mRNA vaccines encoding for immunostimulants.¹⁰⁰ In combination with the personalized neoantigens platform, this could improve the immune response and cytotoxicity of successfully elicited T cells.

RNA editing in CAR T cell therapy

Different RNA therapeutics and a combination of RNA therapeutics and genetic engineering are currently utilized to improve the safety and efficacy of autologous and allogeneic CAR T cell therapy in pre-clinical or clinical developments. These approaches are aimed at redirecting T cell specificity, overcoming HLA barriers, addressing host-mediated rejection, increasing persistence of CAR T cells, and boosting the potency or preventing CAR T cell fratricide.

To direct T cell specificity

The TCR repertoire can be abrogated by disrupting either TCR α or β chains and represents a major advance for generating allogeneic universal CAR T cells. Eyquem et al.¹⁰¹ showed that the integration of a CD19-specific CAR into the TRAC locus of human peripheral blood T cells resulted in uniform CAR expression and enhanced T cell potency, possibly due to decreased tonic CAR signaling, effective internalization, and re-expression of the CAR after antigen encounter. In addition, triple knockout of β2 microglobulin (B2m), class II major histocompatibility complex transactivator, and TRAC was shown to improve the persistence of CD19 CAR T cells and maintained antitumor efficacy.¹⁰²

To overcome HLA barriers

Disruption of either the TCR α or β chains, or the parts of the CD3 complex can be used to prevent cell surface assembly of a functional multimeric TCR. For the removal of TCRα or β from allogeneic CAR T cells, a number of studies have reported the use of CRISPR-Cas9 for B cell malignancies,^103,104 as well as T cell cancers¹⁰⁵ and base editing for T cell acute lymphoblastic leukemia (T-ALL).^106,107,108

To address host-mediated rejection

The inclusion of alemtuzumab targeting CD52 provides an advantage to CD52 knocked-out CAR T cells for a period of approximately 4 weeks.^104,109,110 Alternative approaches directly remove HLA molecules on infused cells, targeting the conserved B2m domain of HLA class I molecules that interact with CD8 T cells, are also in early clinical phase testing.¹¹¹

To prevent T cell dysfunction

After antigen recognition, negative regulators of T cell activation are transiently upregulated, preventing T cells from becoming overly stimulated. However, chronic antigen exposure can lead to sustained overexpression of these negative regulators, leading to T cell exhaustion and/or dysfunction. Significant effort has focused on preventing T cell dysfunction by inhibiting immunosuppressive signals in T cells. As the programmed cell death 1 (PD1)/programmed death ligand 1 (PDL1) axis is considered key in driving T cell exhaustion, antibody-mediated blockade of the PD1/PDL1 axis has been combined with engineered T cells in preclinical studies and in clinical trials, with some signs of efficacy.^112,113 An alternative is to use genome editing to ablate the gene encoding PD1. This strategy has the advantage of specifically inhibiting the PD1/PDL1 axis only in engineered tumor-specific T cells, perhaps mitigating toxicities related to autoreactivity. CD19-CAR T cells in immunocompetent mice suggests that PD1 knockout CAR T cells can differentiate, form memory, and persist in the presence of constant antigen exposure provided by continuous B cell renewal, with no evidence of malignant transformation. Early clinical trials with PD1-disrupted T cells have shown feasibility and safety in the absence of unconstrained proliferation or persistence, but also in the absence of objective antitumor responses.^114,115,116

Other pathways that have been targeted with genome editing to enhance CAR T cell function include gene disruption of transforming growth factor (TGF)-β receptor II in CAR T cells.¹¹⁷ Gene disruption of Fas decreased activation-induced cell death in engineered T cells and increased the antitumor effect of CAR T cells¹¹⁸ Triple gene disruption of NR4A transcription factors in CAR T cells promotes tumor regression and prolongs the survival of tumor-bearing mice.¹¹⁹ With the recent development of next-generation sequencing, high-throughput genetic perturbation technologies, and genetic screens, new candidate genes have been identified for roles driving T cell exhaustion. Examples include CD39, identified as a major driver of T cell exhaustion in both primary and metastatic colorectal tumor models testing human epidermal growth factor receptor 2 (HER-2)-specific TCR-edited T cells for eliminating HER-2-expressing patient-derived organoids in vitro and in vitro.¹²⁰

CRISPR-Cas9-mediated ID3 and SOX4 knockouts prevented or delayed T cell dysfunction and improved CAR T cell efficacy against solid tumors.¹²¹ Genome-editing disruption of DNA methyltransferase 3 alpha has reversed exhausted epigenetic states in vitro and in animal models.^122,123 Knocking out the gene encoding the EGR2 transcriptional regulator improves memory differentiation.¹²⁴ Finally, a recent study used an in vitro model of T cell exhaustion mediated by tonic signaling to identify the mediator complex subunit 12 (MED12) and cyclin C genes, which encode proteins in the kinase module of the multiprotein mediator complex, as regulators of CAR T cell effector functions. Indeed, MED12-deficient T cells showed enhanced potency in mediating antitumor effects.¹²⁵ Although challenging in primary T cells and in vitro models, these types of CRISPR screenings can enable the unbiased discovery and functional characterization of gene targets and pathways with key roles in T cell function.¹²⁶

To boost the potency of adoptively transferred cells

Disrupting TET2 in a patient altered the differentiation state and proliferative capacity of CAR-expressing T cells, resulting in considerable and durable therapeutic effects without transformation over more than 4 years. However, further studies have indicated that biallelic TET2 ablation also enables antigen-independent CAR T cell clonal expansion, raising potential safety concerns associated with disruptions of genes linked to broad epigenetic changes.¹²⁷

To prevent CAR T cell fratricide

Fratricide has been observed in CAR T cell product development for T cell and non-T cell malignancies targeting CD3, TCRβ, CD7, CD38, CD70, and NKG2D ligands. To avoid fratricide, one solution is to use genome editing to eliminate the targeted antigen from the surface of CAR T cells. Functional CAR T cells targeting CD3 or CD7 have been successfully generated by knocking out CD3 or CD7, using CRISPR-Cas9¹²⁸ or base editors.^129,130 Off-the-shelf anti-CD7 CAR T cells, which are resistant to fratricide through genome-editing¹³¹ or base-editing¹⁰⁸ approaches, have induced deep and durable responses in a small number of selected patients with relapsed or refractory T-ALL.

Various clinical studies explored the efficacy of mRNA CAR-T against different antigens like BCMA, CD19, or CD123 for various hematological malignancies. Some trials were also conducted to treat solid tumors like c-met or NKG2D-ligand. Mesothelin CARs led to promising anti-tumor results in pancreatic ductal adenocarcinoma.¹³²

Due to the transient expression of mRNA, its use alleviates integration-related safety issues and on-target off-tumor effects, which have been an important source of toxicity associated with CAR-T constitutively expressing the CAR receptors. This characteristic also obliges multiple rounds of treatment and production of a large quantity. Improvements to deal with that matter include the usage of nanoparticles like LNP instead of electroporation, which has been shown to decrease cell toxicity and increase mRNA delivery.¹³³ A clinical platform for engineering T cells with transient CAR expression using mRNA encoding a CAR with CD3-ζ and 4-1BB co-stimulatory domains was established to overcome off-target toxicity in normal tissues. Two case reports from ongoing trials show that administering mRNA CAR T cells targeting mesothelin (CAR-Tmeso cells) is safe and feasible, with no apparent off-tumor on-target toxicity. The CAR-Tmeso cells persisted briefly in the blood after intravenous administration, migrated to tumor sites, and demonstrated antitumor activity. They also triggered the development of novel anti-self-antibodies, indicating an antitumor immune response. These results suggest the potential of using mRNA engineered T cells to evaluate potential off-tumor on-target toxicities and support the development of mRNA CAR-based strategies for carcinoma and other solid tumors.¹³⁴

Safety and regulatory issues

Moderna’s chief scientific officer Melissa J Moore, in her 2021 TED talk, popularized the principles and applications of mRNA therapeutics.¹³⁵ When describing the vast range of clinical application harnessing mRNA biology, she spoke about an impending tsunami of mRNA medicines. Faced with the rapid and vast development in this field, addressing safety, regulatory issues, and public reception becomes imperative. The first question raised is whether or not mRNA medicine should fall under the same regulatory rules as gene therapy products. Although mRNA therapy does not integrate into targeted cells genome and is only expressed transiently, it still falls under certain aspects of gene therapy definition.

The FDA describes gene therapy simply as a medical intervention based on the modification of the genetic material of living cells.¹³⁶ In contrast, the European directives provide a more precise definition: (1) it contains an active substance which consists of a recombinant nucleic acid used in or administered to human beings with the aim of regulating, repairing, replacing, adding or deleting a genetic sequence; and (2) in its therapeutic, prophylactic, or diagnostic effects relate directly to the recombinant nucleic acid sequence it contains or to the product of the genetic expression of this sequence.¹³⁷ However, some organizations like the World Health Organization consider that mRNA vaccines fall under the definition of a vaccine more than gene therapy, which creates ambiguity in the regulatory guidelines that should be applied especially for mRNA vaccines.

To date, most RNA-based therapeutics that have been approved are siRNAs and demanded significant preclinical analysis to predict biodistribution and pharmacokinetics. These studies are particularly challenging for mRNA-based drugs, as they possess different pharmacological activities. It is important to understand how the product is distributed to the body, what by-products it can generate during metabolism, and how it is eliminated by the body. Also, most RNA therapeutics are delivered via LNPs, which adds another layer of pharmacological analysis.¹³⁸ Quality control before the encapsulation is also a major consideration. For mRNA, the IVT product has to be free of DNA and RNA debris remaining from the DNA digestion steps. Such contaminations could impact the efficiency of the product and trigger an immune response against the foreign nucleic acids, possibly escalating to a cytokine storm.¹³⁹

mRNA vaccines do not require the same quality control and safety reviews.¹⁴⁰ Although mRNA vaccines in clinical trial showed to be safe and trigger very few adverse events, much remains unknown in the field of mRNA pharmacology. At the same time, considering mRNA vaccines as regular vaccines can be misleading. An mRNA vaccine does not contain an antigen, but the material to induce the expression of an antigen, thus acting as a pro-vaccine. This can potentially require specific regulatory guidelines to ensure the safety of such pro-drugs. These issues can present obstacles to gather a positive public reception to this new type of drugs. A number of people still consider mRNA vaccines as gene therapy, which is often regarded as long-lasting genetic engineering on human beings. The emergency situation during the SARS-COV-2 pandemic forced the safety review for new drug development to be accelerated and some important step could have been overlooked. The public opinion of mRNA therapies could also have been affected by that fact. Well thought out and complete guidelines for RNA-based therapeutics are needed to reassure the public and allow mRNA vaccination to be accepted as a standard solution for disease control.

Perspectives

mRNA therapeutics have emerged as a ground-breaking approach to treating various diseases and conditions. With the success of mRNA-based COVID-19 vaccines, this field has gathered significant attention and investment.¹⁴¹ From the current stage, there are several key areas where improvements can be made to further enhance the efficacy, safety, and applicability of mRNA therapeutics.¹⁴²

mRNA therapeutics faces multiple challenges, one of which is the efficient delivery of mRNA molecules to target cells or tissues. To overcome this challenge, LNPs are utilized heavily compared with other possible delivery methods and have seen great advances in recent years.¹⁴³ There are possible avenues to investigate in terms of increasing the delivery efficiency, decreasing off-target effects, and minimizing immune responses triggered by LNPs. Even with the improvements in LNP formulations, researchers are exploring novel delivery systems, such as polymer-based nanoparticles, hybrid LNPs, and cell-penetrating peptides to address these limitations and improve the targeted delivery of mRNA therapeutics.^144,145 Exosome delivery methods offer a human-derived and safer option that have been developed during the SARS-COV-2 pandemic and could be leveraged for more tolerable and specific cancer treatments.^146,147 Exosomes delivery of therapeutic mRNA in HER2⁺ and glioma cancer cells drastically impaired cell and tumor growth,^148,149 enhanced cell uptake, and prolonged patient survival.¹⁵⁰

Another potential avenue of improvement is increasing the stability of the mRNA molecule itself.^151,152 mRNA is inherently unstable and can be rapidly degraded by RNAses, which limits its therapeutic potential.^32,36 Researchers are working on optimizing the chemical structure of the mRNA to enhance its stability and prolong its half-life within the body.^34,36 This could improve the stability of mRNA therapeutics, allowing for less frequent dosing and potentially decreasing the overall treatment burden.¹⁴²

Although mRNA therapeutics have shown a good safety profile overall, there have always been concerns regarding the safety of mRNA therapeutics in the public’s eye.¹⁵³ There is ongoing research to further decrease potential immunogenicity and off-target effects through modifying the components of LNPs.⁴⁵ Optimizing the mRNA sequence, modifying the nucleotide backbone, and refining the formulation of delivery systems and changing the formulation of the LNPs can help to mitigate immune responses and enhance the safety of mRNA-based therapies.^{29,36,42,45,74}

Fine-tuning LNP formulations to selectively target specific cell types or tissues is an area of great interest. By incorporating cell-specific targeting ligands, engineering mRNA sequences to have cell type-specific expression profiles, or incorporating specific lipids with organ-specific tropism, researchers aim to enhance the precision and efficacy of mRNA therapeutics.^48,53,86 This approach could minimize potential side effects and increase the therapeutic index of mRNA-based treatments.

mRNA therapeutics have demonstrated effectiveness in vaccine development and infectious diseases. Current and future efforts are aiming at potential applications in other disease area, such as cancer, genetic and autoimmune disorders, cardiovascular diseases, and neurodegenerative conditions with many clinical trials on-going.¹⁵⁴ Combining mRNA therapeutics with other treatment methods, such as small molecules, antibodies, or gene editing technologies, could unlock synergistic effects and improve therapeutic outcomes.⁹⁸ By leveraging the unique properties of mRNA, researchers can explore innovative combination approaches to address complex diseases with multifactorial mechanisms.