Table 1 presents a summary of registered clinical trials based on clinicaltrials.gov for mRNA cancer vaccines encoding TSAs. One of these vaccines, mRNA-4157, is a personalized mRNA vaccine that has the capacity to encode up to 34 antigens and is currently being assessed for its effectiveness in treating melanoma (NCT03897881). Results from the phase IIb KEYNOTE-942 trial indicate that mRNA-4157/V940 combined with pembrolizumab could be a potential adjuvant therapy for melanoma, as patients receiving this combination had a significant reduction in disease recurrence risk compared to those who only received PD-1 inhibitor, showing promising results (52). In a phase 1 clinical trial, autogene cevumeran (BNT 122) containing up to 20 neoantigens was tested for its ability to stimulate immunity against neoantigens in patients with resected pancreas ductal adenocarcinoma (PDAC). The results revealed that patients who responded to the vaccine exhibited a longer recurrence-free survival than non-responders at an early median follow-up of 15 months, suggesting that vaccine-induced neoantigen-specific immunity may be associated with improved outcomes in PDAC (53). However, evidence for the effectiveness of mRNA-4650 in treating gastrointestinal cancer appears to be discouraging. A study by Cafri et al. evaluated the immunogenicity and clinical efficacy of mRNA-4650 and only observed an increase in the frequency of cancer-specific T cells, but no clinical benefit (22).

Currently, several institutions such as BioNTech and Moderna have joined forces in the pursuit of developing personalized mRNA cancer vaccines, with numerous clinical trials currently underway. Despite having a theoretically proven efficacy, these vaccines face obstacles in areas such as design, production, and cost. Thus, additional optimization efforts are necessary to overcome these challenges. Furthermore, further preclinical and clinical trials are indispensable in validating the effectiveness of personalized mRNA cancer vaccines. These trials can identify opportunities for improvement and expedite the translation of this innovative approach from bench to bedside, ultimately benefiting cancer patients.

Fifteen clinical trials are currently in progress to assess mRNA cancer vaccines that encode TAAs, as demonstrated in Table 1 . The BNT111 vaccine developed by BioNTech is an example of an mRNA cancer vaccine that encodes four melanoma-associated antigens (MAAs) including New York esophageal squamous cell carcinoma 1 (NY-ESO-1), melanoma-associated antigen A3 (MAGE-A3), tyrosinase, and transmembrane phosphatase with tensin homology (TPTE) (44). These RNAs are encapsulated in liposomes and administered intravenously to patients. In a phase I trial, this vaccine alone and in combination with ICIs induced durable objective responses and had a favorable safety profile among patients with advanced melanoma (44). Currently, a phase II trial is underway evaluating the vaccine candidate in combination with the anti-PD-1 antibody cemiplimab for patients with unresectable stage III or stage IV melanoma who are refractory to or have relapsed after anti-PD-1 therapy. These findings suggest that the BNT111 vaccine holds great promise as a treatment option for melanoma and may provide new hope for patients with advanced forms of the disease.

mRNA-5671 is a tetra-valent vaccine that has been formulated with lipid nanoparticle (LNP) technology and is based on messenger RNA (36). This innovative vaccine targets four of the most frequent KRAS mutations - G12D, G13D, G12C, and G12V (36). Pre-clinical investigations have indicated a substantial improvement in CD8 T cell responses to KRAS antigens post-immunization with mRNA encoding KRAS mutations (36). In a phase I trial, patients suffering from advanced or metastatic NSCLC, colorectal cancer, or pancreatic adenocarcinoma, and having KRAS mutations, are being enrolled to determine the efficacy of mRNA-5671 with or without pembrolizumab (NCT03948763).

A phase 1/2 clinical trial (NCT04382898) is currently underway to evaluate the effectiveness of the cancer vaccine BNT112. This vaccine encodes five different tumour-associated antigens and is being administered alone or in combination with cemiplimab to patients with metastatic castration-resistant prostate cancer. Another trial is a randomised phase 2 study (NCT04534205) evaluating the anti-human papillomavirus (HPV)-16-derived oncoprotein-encoding mRNA BNT113 vaccine in HPV16-positive, PD-L1-positive head and neck squamous cell carcinoma. BNT113 is also being tested in a two-arm phase 1/2 vaccine dose-escalation study (NCT03418480) for patients with previously treated or advanced HPV16-positive head and neck squamous cell carcinoma. Another phase 1 study (NCT04163094) is being conducted to evaluate the BNT115 that encodes ovarian-specific tumour-associated antigens. This vaccine is being administered both before and in combination with adjuvant and neoadjuvant chemotherapies to patients with ovarian cancer. Finally, a fifth clinical trial (NCT05142189) is evaluating the FixVac vaccine BNT116 in combination with cemiplimab or docetaxel in a phase 1 study for patients with advanced or metastatic non-small cell lung cancer.

The profound potential of mRNA cancer vaccines stems from their ability to encode a wide range of proteins, including immunostimulants that can modify the tumor immune microenvironment (TME) and enhance the efficacy of immune checkpoint inhibitors (51). This promising avenue of mRNA vaccine research has spurred clinical trials of cytokine-encoding mRNA products by BioNTech and Moderna, with highly encouraging results (13, 54). For instance, intratumorally administered mRNA-2416 produced by Moderna, which encodes OX40L, demonstrated safety and tolerability in a phase I trial and elicited broad proinflammatory activity with desirable changes in the TME (54). These findings provide critical support for its further investigation in combination therapy with anti-PD-L1 inhibitor durvalumab in solid tumors. Similarly, co-administration of mRNA-2752 encoding OX40L/IL23/IL36g with durvalumab in a dose escalation study (NCT03739931) exhibited antitumor effects, validating the potential of mRNA cancer vaccines as a therapeutic modality. Furthermore, BioNTech’s BNT131 (SAR441000), which encodes IL-12sc, IL-15sushi, IFN-α for intratumoral injection, is being tested as monotherapy and in combination with cemiplimab for patients with advanced solid tumors to alter the TME. Several other mRNA products have also shown promise, such as ECI-006, a combination of TriMix and melanoma-specific TAAs administered intranodularly and being tested in a phase 1 study of resected melanoma (NCT03394937); and MEDI1191, an immunomodulatory fusion protein containing IL-12α and IL-12β subunits developed for intratumoral injection (55). While only seven product candidates are currently undergoing clinical trials, the results thus far demonstrate the immense potential of mRNA cancer vaccines in improving cancer immunotherapy outcomes ( Table 1 ). With these promising results, further research is needed to determine the optimal administration route and maximize vaccine efficacy.

The complexity of treatment decisions is enhanced by tumour heterogeneity, which can be divided into spatial and temporal heterogeneity (56). Temporal heterogeneity refers to the dynamic evolution of the genomes through the tumour progression course, whereas spatial heterogeneity refers to the phenomenon that a tumour is composed of subclones of different genetic backgrounds (57–59). This heterogeneity leads to variable responses to therapies among individuals with cancers of the same tumour subtype and is believed to be one of the major causes of progressive disease and failure of therapy (60, 61). Typically, only a small fraction of a specimen is sampled for mRNA cancer vaccine design, which is unlikely to provide full information on tumour gene profile, contributing to increased uncertainties for the clinical efficacy of mRNA cancer vaccine (11). Thus, tumour heterogeneity largely limits the efficacy of mRNA cancer vaccine and how to overcome tumor heterogeneity is a difficult challenge for its clinical application.

To overcome spatial heterogeneity in tumours, one strategy is to use tumour tissue multipoint sampling to identify differences between tumour regions and inform the design of personalised mRNA cancer vaccines. Another approach involves using mRNA cancer vaccines that target multiple antigens expressed across various tumour regions, thereby compensating for spatial heterogeneity. Meanwhile, monitoring disease progression and adjusting treatment plans accordingly may help to address temporal heterogeneity. However, these strategies not only increase the complexity of vaccine design and administration but also raise costs and increase treatment time for patients. Another promising method is the use of artificial intelligence algorithms such as MHC-binding prediction, quantification of mutated transcript expression, and clonality of the mutation to predict neoantigens based on tumour genomic data, then used to prioritize these mutations as vaccine candidates based on their likelihood to elicit a T cell response, which could improve the efficiency and accuracy of vaccine design and overcome the heterogeneity of tumour (62–64). However, to date, there is still limited information on artificial intelligence for mRNA cancer vaccine design. This is certainly an important and interesting area worthy of future investigations.

Tumour initiation, progression and maintenance depend highly on interactions between the tumour and the associated microenvironment (65). Tumour microenvironment refers to the surrounding microenvironment of tumour cells, including surrounding cells, signalling molecules, and extracellular matrix (66). The importance of the tumor microenvironment in cancer has been recognized since the late 1800s and then accumulating evidences suggested immunosuppressive TME not only promote immune evasion and tumor growth, but also lead to decrease the efficacy of immunotherapy (67–69). Immunosuppressive TME could decrease the efficacy of immunotherapy by inhibiting the function and activation of immune cells, such as T cells and natural killer cells, which are critical for attacking and eliminating cancer cells (14). Although mRNA cancer vaccine may have the capability to elicit a cellular immune response, the inhibitory tumor microenvironment can impede T cell infiltration into tumors and result in T cell exhaustion (50). As with many other immunotherapies, overcoming immunosuppressive TME is one of the most challenging and unsolved problems for mRNA cancer vaccine.

Combining mRNA cancer vaccines with other anti-cancer treatments, which is being constantly tried in cancer patients, can be an effective strategy for overcoming the immunosuppressive microenvironment. One commonly used combination therapy is to combine mRNA vaccines with immune checkpoint inhibitors, which can “release the brakes” on the immune system, allowing it to attack cancer cells more effectively (55, 70). Adoptive T-cell therapy, another type of immunotherapy, can also be used alongside mRNA vaccines. This innovative treatment involves extracting T-cells from a patient’s blood or tumor, modifying them in a laboratory to target specific cancer antigens, and then reintroducing them into the patient’s body. Combining these modified T-cells with mRNA vaccines can significantly enhance the immune response against cancer cells, thus improving the efficacy of this therapy (71). The combination of mRNA vaccines and radiotherapy has also shown promising therapeutic effects in preclinical and clinical models (72, 73). Overall, combining mRNA cancer vaccines with other anti-cancer treatments can be a powerful approach to overcoming the immunosuppressive microenvironment in tumors. Continued research and development of these combination therapies will be critical to improving outcomes for cancer patients.

Poor performance in any step of the mRNA delivery process would compromise the therapeutic efficacy, and administration routes should be a first-order consideration for the clinical usage of mRNA cancer vaccine (74, 75). The administration route of mRNA vaccines strongly influences the translation efficiency of the target protein and the distribution of mRNA cancer vaccine in vivo ( 16, 76). Currently, there is no consensus on the optimal route of administration for mRNA vaccines, although the SARS-CoV-2 mRNA vaccines that have received approval utilize intramuscular injection (50). Various routes, including intramuscular, subcutaneous, intranodal, intradermal, and intranasal vaccination, are actively being explored in the search for an optimal administration route for mRNA cancer vaccines (55). Each administration route for mRNA cancer vaccines has unique advantages and limitations. As depicted in Figure 5 , the strengths and weaknesses of diverse mRNA cancer vaccines administration routes are presented.

Intramuscular and intravenous injections are the two most frequently used administration routes in clinical trials for mRNA cancer vaccines ( Table 1 ). Intramuscular administration is a widely used and feasible vaccination route that involves the direct injection of a vaccine into muscle tissue (14, 77). This method has demonstrated efficacy in inducing an immune response for mRNA cancer vaccines, being both easily executed and well-tolerated with the added benefit of offering flexible dosing options, while also causing minimal side effects at the injection site (78). Consequently, its use has become prevalent (14, 77, 78). Additionally, intravenous injection is the most frequently used direct administration route in current clinical trials for active therapeutic mRNA cancer vaccines (79). It has been shown to be safe, well-tolerated, and capable of inducing an immune response against cancer cells (80). Furthermore, it enables the vaccine to reach multiple lymphoid organs and allows for repeated dosing to maintain immunity over time (55, 80).

Intranasal administration of mRNA cancer vaccines can effectively deliver the vaccine to antigen-presenting cells in the peripheral lymph nodes (81, 82). However, intranodal injections provide a more direct route to reach lymphatic antigen-presenting cells (83). While both delivery methods have their advantages, they also face significant limitations. Intranasal administration may offer non-invasive delivery but is restricted by nasal cavity volume constraints (81, 82). Conversely, intranodal injections require specialized equipment and highly skilled personnel, and the injection volumes are small, which may not be sufficient for larger tumors or inducing a robust immune response (84). Accordingly, further research is necessary to optimize these methods for mRNA cancer vaccines.

Intratumoral injections of mRNA cancer vaccines, which can rapidly activate immune cells and minimize off-target, have been investigated in clinical and preclinical trials (85, 86). Although this approach aims to induce local inflammation with mRNA encoding immunostimulatory, its effectiveness is limited by the size and location of the tumor, which may restrict the amount of vaccine that can be delivered (87). Repeated injections may be required to maintain the immune response over time, and combining intratumoral injections with systemic delivery methods such as immune checkpoint inhibitors may offer a more comprehensive approach to treating cancer (55). Despite these limitations, intratumoral injections remain a promising area of research for directly interact with immune cells to activate an immune response mRNA cancer vaccine. Intradermal and subcutaneous injections are two common administration routes for mRNA cancer vaccines (74). Both methods allow regional antigen-presenting cells to easily process the mRNA, which is essential for eliciting an immune response against cancer cells (74). However, these administration routes often induce severe local injection-site reactions than intramuscular administration, which can negatively impact patient compliance and overall treatment efficacy (14, 74).

In view of the immense potential offered by various routes of administration for successful cancer therapy, it is of paramount importance to thoroughly assess the advantages and disadvantages of each vaccination route. Recently, investigators have utilized a non-invasive method involving a dual radionuclide near-infrared probe to track the spatiotemporal trafficking of the vaccine following intramuscular injection, offering essential guidance in precisely evaluating the dose, injection site, and biological distribution of the vaccine (88). In order to determine the most effective vaccination route and optimize vaccine efficacy, there remains a pressing need for novel approaches that can accurately monitor and analyze the spatiotemporal kinetics of vaccines (88). By carefully weighing the pros and cons of each administration route, we can identify the most promising strategies for delivering immunotherapies that can target cancer cells effectively while minimizing off-target effects. Furthermore, by leveraging advanced tools to track and analyze vaccine movement within the body, we can acquire critical insights into the factors that govern their biological activity and efficacy. Therefore, these innovative methods hold tremendous promise for advancing our understanding of how to optimize mRNA cancer vaccination routes and maximize the impact of cancer immunotherapy.

The introduction of mRNA cancer vaccines represents an exciting new frontier in cancer treatment, but also underscores the need for novel efficacy evaluation standards. Unlike traditional cancer treatments, such as chemotherapy, targeted therapy, and radiation therapy, which targeting the tumor cells directly and emphasizing tumor shrinkage as a successful response to treatment evaluated by radiological imaging, mRNA vaccines stimulate the immune system to produce an anti-tumor response indirectly that standard radiographic imaging techniques do not account for this unique mechanism of action and may not reflect the true clinical benefit of mRNA vaccines (89). mRNA cancer vaccines may lead to inflammatory reactions resulting in tumour swelling, which results in increased complexity of evaluating the efficacy of mRNA cancer vaccines (90). Moreover, cancer vaccines, particularly when used as monotherapy, may exhibit greater efficacy in cases of low disease burden and may not produce the striking radiographic responses typically observed with cytotoxic therapy (91). As a result, the development of cancer vaccines may encounter difficulties in demonstrating effectiveness when evaluated in late-stage disease using traditional assessment methods such as standard radiographic response evaluation criteria (92). Due to their highly individualized and customized nature, personalized mRNA cancer vaccines require different treatment plans and dosages tailored to each patient’s unique condition and cancer type (89). To meaningfully identify which mRNA cancer vaccines should advance beyond early phase trials and into larger phase III clinical trials, novel biomarkers need to establish that can accurately monitor the treatment response to these vaccines.

Quantitative measurement of immune cell responses is essential for establishing surrogate biomarkers of efficacy, especially for tumor antigen-specific T cell responses, which are critical in selecting optimal doses for cancer vaccine trials and have the potential to lead to tumor rejection. Multiple studies are evaluating the effectiveness of mRNA therapies by measuring immune response indicators, such as cytokines, chemokines, or immune cells in the blood. A variety of immune monitoring techniques, including flow cytometry analysis of cell populations’ phenotypes, functionalities, and activation status, are employed (93). Additionally, enzyme-linked immunospot (ELISPOT) assays measure cytokine release after antigen-specific immune responses, while peripheral cytokine profiling using enzyme-linked immunosorbent assay evaluates innate immune responses (93–95). Another method, tetramer analysis with MHC multimers loaded with antigen peptides, is used to measure antigen-specific CD8 T cells, and T cell receptor analysis through sequencing and polymerase chain reaction helps to elucidate the immune repertoire, including genetic arrangement and specificity. These assays are applied to immune cells in peripheral blood mononuclear cells acquired from patients to detect and describe T cell responses. Although ELISPOT is one of the most commonly used techniques to identify CD8 T cell responses to a given antigen, its quantitative output cannot determine the absolute number of antigen-specific T cells (94). It is noteworthy that currently available immunological assays, such as ELISPOT, flow cytometry-based multimer staining, and intracellular cytokine staining, have been found to exhibit technical inconsistencies across different laboratories (96). For example, one study reported inter-laboratory variations of up to 50% in ELISPOT (96). Standardized and harmonized procedures, from specimen banking to assay validation and result reporting, are thus necessary for successful clinical development.

Over the past few decades, single-cell RNA sequencing (scRNA-seq) technology has garnered significant attention as a novel tool in assessing the immunotherapy response of cancer patients (97). The scRNA-seq technology refers to a collection of techniques that enable the untargeted quantification of transcripts present in individual cells, facilitating the identification of cell types and states associated with immunotherapy response (97, 98). The scRNA-seq technology allows for the comprehensive analysis of cellular transcriptomes, facilitating the identification of distinct immune cell subsets and their gene expression patterns correlated with immunotherapy response (99). Furthermore, this technique provides insights into the dynamic changes within the tumor microenvironment during immunotherapy, aiding in the understanding of mechanisms underlying treatment response (99). Recently, multiple studies have reported utilized scRNA-seq to explore immune cell subset infiltration and regulation changes pre- and post-mRNA vaccine administration to understand therapeutic response and analyze underlying mechanisms to determine mRNA vaccine efficacy. With further advancements in technology and deeper applications, scRNA-seq technology holds the promise of providing valuable insights into mRNA cancer vaccine efficacy detection.

Recently, the circulating tumour DNA (ctDNA) test has the advantages of sensitivity, flexibility, repeatability, and safe gain much attention to help the evaluation of treatment responses, and the application of ctDNA detection in monitoring mRNA cancer vaccine therapeutic response is a worthwhile attempt (100). Accumulating evidence suggests that changes in ctDNA levels during treatment compared to baseline can serve as a valuable biomarker for stratifying patients as molecular responders or nonresponders and distinguishing those with a favorable prognosis, especially those displaying stable disease according to RECIST v1.1 criteria (101). Palmer et al. investigated the utility of ctDNA as a potential biomarker for monitoring therapeutic responses to mRNA cancer vaccines (70). Their study demonstrated that a decrease in ctDNA levels among patients undergoing treatment was positively correlated with prolonged overall survival, thereby underscoring ctDNA’s potential as a promising biomarker for predicting treatment response (70). The future exploitation of efficacy evaluation system integration of immunologic and radiologic endpoints, the establishment of biomarkers, and standardization of evaluation protocols is needed to develop in the future, which will be necessary to realize the full potential of this promising technology in the fight against cancer.