mRNA-Based Neoantigen Vaccines in Pancreatic Ductal Adenocarcinoma (PDAC)—A Promising Avenue in Cancer Immunotherapy

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most aggressive and lethal malignancies, characterized by an exceptionally poor prognosis and a mortality rate nearly equivalent to its incidence. Despite decades of research and clinical advancements, the five-year survival rate for patients diagnosed with PDAC remains below 5%, similar to statistics from thirty years ago [1]. According to Cancer Statistics 2021, the American Cancer Society reported approximately 60,430 new cases and 48,220 deaths for PDAC in the United States, ranking third after lung and bronchus cancer and colorectal cancer. In the 28 countries of the European Union, it was estimated that approximately 111,500 people (55,000 in males and 56,500 in females) will die from PDAC by 2025 [2,3]. Global Cancer Statistics 2018 showed that the incidence and mortality of PDAC were 458,918 and 432,242 in the world, respectively, and deaths account for about 94.2% of new cases [4,5].

Multiple determinants of PDAC risk have been delineated, encompassing demographic factors like male sex or black race, environmental and lifestyle exposures (e.g., any history of tobacco use, obesity), comorbid conditions (e.g., diabetes mellitus, chronic pancreatitis, Helicobacter pylori infection), and inherited susceptibility (e.g., pathogenic variants in hereditary pancreatic cancer genes and selected cancer predisposition syndromes). PDAC may also develop from a precancerous lesion of pancreatic intraepithelial neoplasia, a phenomenon often arising from acinar-to-ductal metaplasia, which can occur as a result of cellular injury or inflammation in the pancreas [1,6,7,8,9]. Distinct molecular features characterize early-onset PDAC: relative enrichment of RAS wild-type tumors versus the broader PDAC population, a higher prevalence of biallelic CDKN2 inactivation via homozygous copy loss with overexpression of FOXC2—concordant with upregulation of VIM, CDH11, and CDH2 that drive epithelial–mesenchymal transition—compared with cases of average onset (>55 years), and a lower frequency of SMAD4 alterations than in late-onset disease (≥70 years) [2,3,4]. From a prevention and public-health standpoint, roughly one-third of the overall PDAC burden is attributable to the two-thirds of recognized modifiable lifestyle risks factors [10,11]. Many such exposures plausibly promote tumorigenesis through oxidative stress and chronic inflammation, and environmental pollutants have been associated with higher risks across multiple cancers, including PDAC [12,13]. Growing population-level exposure to these factors over recent decades may further amplify risk among genetically predisposed individuals. Nevertheless, PDAC etiology remains incompletely defined, underscoring the need for additional studies and evidence syntheses to refine risk stratification and optimize screening and primary prevention strategies [8,10,14,15].

One of the primary factors contributing to the high mortality of PDAC is its anatomical location. Situated deep within the retroperitoneal space and surrounded by vital vascular structures and adjacent organs, pancreatic tumors are often challenging to resect, particularly in cases of local tissue infiltration. Furthermore, due to the central position of the pancreas and the nonspecific nature of early symptoms, diagnosis frequently occurs at an advanced stage of disease progression [10,14,16,17]. As a result, at the time of diagnosis, tumors commonly exhibit local invasion or distant metastases, rendering them inoperable [18,19]. Another key factor underlying the dismal prognosis of PDAC is its pronounced resistance to systemic therapies. This resistance is driven by multiple biological barriers that limit the efficacy of anticancer treatments (Figure 1). These barriers arise from three hallmark features of PDAC: a highly immunosuppressive tumor microenvironment (TME), pronounced genetic heterogeneity, and a propensity for early metastatic spread [16,20,21,22].

The current standard of PDAC therapy involves surgical resection combined with systemic chemotherapy. However, only approximately 20% of patients are eligible for surgery at the time of diagnosis due to advanced disease stage [7]. Even among those undergoing curative-intent resection, disease recurrence occurs in nearly 80% of cases within two years postoperatively [8]. For patients with unresectable, locally advanced, or metastatic PDAC, systemic chemotherapy regimens such as gemcitabine plus nab-paclitaxel or FOLFIRINOX (a combination of fluorouracil, oxaliplatin, irinotecan, and leucovorin) are commonly employed. While these regimens have modestly improved survival, their use is often limited by substantial toxicity, deterioration in quality of life, and the eventual emergence of drug resistance [9]. Alternative treatment approaches—including radiotherapy, targeted therapies, and immunotherapy—have also been investigated in this patient population; however, their clinical efficacy remains limited [2,3]. Given these challenges, PDAC continues to represent one of the most formidable obstacles in modern oncology, necessitating the development of more effective and durable therapeutic strategies.

In light of the limited efficacy of currently available treatment modalities, there is a growing interest in innovative therapeutic strategies aimed at overcoming the inherent resistance mechanisms of PDAC. Among these, cancer vaccines have emerged as a promising approach, offering several advantages including multifaceted mechanisms of action, a favorable toxicity profile, broad therapeutic potential, and the ability to induce long-lasting immunological memory. A wide range of vaccine platforms have been evaluated in both preclinical and clinical settings, including peptide-based vaccines, dendritic cell-based vaccines, and viral vector platforms. Although encouraging results have been obtained in preclinical models, translating these findings into clinical efficacy remains a significant challenge. Recently, mRNA-based vaccines have garnered particular attention as a novel and versatile modality in cancer immunotherapy, including in the treatment of PDAC.

Thus, the aim of this review is to summarize the current state of knowledge and available evidence on the application of mRNA vaccines in the PDAC therapy.

PDAC is distinguished by explicit molecular heterogeneity and an immunosuppressive TME. Genomic analyses have defined distinct PDAC subtypes (e.g., classical versus basal-like) with divergent biology, reflecting the extensive inter- and intratumoral molecular variability that complicates efficient treatment [4,21]. A determinative feature of PDAC is a dense desmoplastic stroma, rich in cancer-associated fibroblasts and extracellular matrix, which creates a physical barrier to infiltration of antitumoral immune cells [22,23]. Moreover, this fibrotic and hypovascular microenvironment promotes tumor hypoxia and limits drug delivery, while also actively contributing to immune evasion. Tumor and stromal cells engage in aberrant signaling that shifts the immunoregulatory balance toward a pro-tumor state [21]. PDAC is often termed a "cold" tumor, as they exhibit low immunogenicity and unfavorable miroenvironment [4,23,24].

Instead, immunosuppressive populations predominate—including tumor-associated macrophages (TAMs) polarized to an M2 phenotype, myeloid-derived suppressor cells (MDSCs), and regulatory T cells (Tregs)—all of which secrete factors that inhibit effector T-cell function [22,25]. Elevated levels of anti-inflammatory cytokines such as transforming growth factor-β (TGF-β) and interleukin-10 (IL-10), as well as pro-angiogenic vascular endothelial growth factor (VEGF) are prevalent among PDAC, further reinforcing immune tolerance [23]. PDAC cells and infiltrating immune cells may express and upregulate multiple immune checkpoint molecules (e.g., PD-L1 and PD-1, CTLA-4, TIM-3, LAG-3, and VISTA) that inhibit antitumoral T-cell activation in the TME. These intrinsic and extrinsic mechanisms of immunosuppression underlie PDAC's notorious resistance to immunotherapy [26,27]. Notably, single-agent checkpoint inhibitors (e.g., anti-PD-1) have shown minimal efficacy in PDAC, largely due to the paucity of pre-existing T-cell infiltration and the multitude of active, parallel immunosuppressive pathways. Overall, the hostile immune landscape—combining pronounced tumor heterogeneity, a barrier-rich stroma, and active immune suppression—presents a major hurdle to conventional therapies. This context provides a strong rationale for novel immunotherapeutic approaches. By understanding and eventually modulating the PDAC microenvironment, mRNA vaccine strategies aim to overcome these barriers [23].

Effective cancer vaccines depend on identifying appropriate antigenic targets expressed by tumor cells. These targets belong to two broad categories: tumor-associated antigens (TAAs) and tumor-specific antigens (TSAs). TAAs are proteins that are expressed in healthy tissue, however also aberrantly present in cancer [4]. In PDAC they include, e.g., mucin-1 (MUC1), carcinoembryonic antigen (CEA), and glypican-1 (GPC1) [12,13,28,29]. For instance, GPC1 is frequently overexpressed in PDAC and has been correlated with poor prognosis and contributing to pro-tumor state [30]. While the presence of TAAs on tumor cells can be recognized by the immune system, their expression in healthy tissue raises concerns of on-target/off-tumor effects and induction of self-tolerance that may limit vaccine potency [4,31]. By contrast, TSAs, also called neoantigenes, are molecules arising from somatic mutations exclusive to cancer. Actually, there is no expression of them on healthy cells, making these targets highly attractive. However, TSAs are rather patient-specific and result from the individual's unique mutational signatures, meaning the landscape of suitable neoepitopes can vastly differ from case to case [32].

The process of neoantigen identification begins with the next-generation sequencing of tumor tissue. Computational analysis is then employed to predict which mutations are likely to be recognized by the patient's immune system. Key steps include searching for mutations that alter protein coding, assessing peptide binding affinity to HLA class I and II (using tools such as NetMHCpan and MHCflurry), and evaluating stability of the peptide–MHC (major histocompatibility complex) molecules. High transcript expression of the mutant gene can be an important criterion, as abundant expression may compensate for moderate HLA binding affinity and increase the chance of immunogenic T-cell responses. Candidate neoantigens are typically validated through in vitro assays—for example, by assessing the ability of mutant peptides to be recognized by the patient's T cells or by mass spectrometry-based immunopeptidomics to confirm their natural presentation on tumor HLA molecules [33,34,35]. However, experimental validation at large scale remains a challenge, mainly due to the costs and limited number of patient samples, which restricts the amount of neoantigens that can be tested.

PDAC typically exhibits a relatively low tumor mutational burden, and thus there remains a limited number of potentially targeted neoantigens, especially compared to malignancies like melanoma or lung cancer. Indeed, around 1% of PDAC patients harbor high tumor mutational burden, mismatch-repair deficiency and high microsatellite instability, and are more prone to response to checkpoint inhibition, whereas the majority of PDAC patients remain unresponsive [36,37]. Tumor heterogeneity in PDAC adds another layer of complexity to antigen selection. Different subclones within the same tumor may carry distinct mutations, meaning an immunogenic neoepitope from one region of the cancer might not be ubiquitous among all tumor cells [4]. Consequently, vaccines must target a broad panel of neoantigens to cover the clonal diversity of a patient's tumor. Indeed, the trials we have discussed in Section 3. encoded as many as 20 different neoantigens per patient in an effort to maximize coverage. However, despite these challenges, our ability to predict and validate true neoantigens improves and the hope is that mRNA-based vaccines encoding patient-specific neoepitopes will elicit robust T cells that recognize and destroy PDAC cells while sparing normal tissue.

mRNA vaccine technology offers a flexible and powerful platform for presenting antigens, but its success relies on sophisticated molecular design and accurate delivery methods. Broadly, mRNA cancer vaccines come in two forms: conventional (non-replicating) mRNA and self-amplifying mRNA (saRNA). Conventional mRNAs encode only the antigen of interest and typically include regulatory elements such as a 5′ cap, 5′ and 3′ untranslated regions (UTRs), and a poly(A) tail, all optimized for stability and efficient translation. Self-amplifying mRNAs, by contrast, are larger constructs that encode not only the target antigen but also viral replicase proteins (commonly derived from alphaviruses) that enable the RNA to autonomously replicate in the cytoplasm [38,39,40]. By producing multiple copies of themselves in situ, saRNA vaccines can greatly amplify antigen expression. Preclinical studies have shown that saRNA platforms can achieve equivalent immune responses at lower doses compared to conventional mRNA vaccines. Of course, this feature has implications for manufacturing and cost [41,42]. On the other hand, the larger size of replicating mRNAs may pose challenges for delivery and can overly activate innate immune responses, which must be carefully managed [39].

Regardless of the mRNA type, a number of structural optimizations are applied to enhance vaccine performance. Codon optimization of the open reading frame (ORF) is routinely performed to match human tRNA abundance and avoid rare codons, thereby improving translation efficiency. Similarly, the UTRs are engineered to stabilize the mRNA and facilitate ribosome recruitment. For example, the poly(A) tail, human β-globin 5′ and 3′ UTRs are often used for their known contribution to mRNA stability and translational efficiency [43]. Chemically modified nucleosides represent another novel innovation: substituting uridine with pseudouridine or 1-methylpseudouridine in the mRNA can significantly reduce activation of pattern recognition receptors (like TLR7/8, RIG-I) that could trigger innate responses and inhibit protein translation [44]. The described modifications enable mRNA vaccines to avoid the paradox of self-sabotaging their own expression. At the same time, a certain degree of innate immune stimulation is desirable to act as a built-in adjuvant. An elegant approach to achieve this balance has been the development of TriMix: a cocktail of mRNAs encoding three immune-stimulatory proteins (CD70, CD40L, and constitutively active TLR4). When co-delivered with an antigen mRNA, TriMix provides maturation signals to dendritic cells and co-stimulation for T cells, thereby enhancing vaccine-induced immunity [40,45]. This strategy highlights the versatility of mRNA platforms—multiple mRNAs can be combined to simultaneously encode antigens and adjuvant signals.

Naked mRNA is susceptible to rapid degradation by RNases and struggles to cross cell membranes due to its large size and polyanionic charge [46,47]. The advent of nanoparticle carriers has largely solved this issue. Lipid nanoparticles (LNPs) have become the leading system for mRNA delivery. They typically consist of an ionizable lipid (crucial for cellular intake and facilitates endosomal escape), non-cationic phospholipids, cholesterol to provide stability, and a polyethylene glycol-lipid to prolong circulation and storage [48,49,50]. LNPs form particles capable of protecting mRNA and efficiently fusing with cell membranes to deliver the mRNA into the cytosol. These carriers facilitate efficient transport of mRNA into immune cells, particularly dendritic cells, which subsequently present the translated antigens on their surface via MHCs—an essential step for initiating adaptive immune responses [48,51]. Moreover, the incorporation of adjuvants that stimulate immune activation has been shown to further enhance the immunogenicity and therapeutic efficacy of mRNA vaccines, as demonstrated in multiple preclinical and clinical studies [52,53]. Alternative delivery systems under exploration include polyplexes (polymers that bind and condense mRNA), peptide-based nanoparticles, and nanoemulsions [54,55]. In addition, chemical modifications of mRNA—such as the incorporation of modified nucleotides—can improve its stability and bioavailability, leading to more sustained and efficient antigen expression in vivo. These strategies have been shown to enhance cytotoxic T cell responses and antibody production, ultimately promoting the effective elimination of tumor cells [56]. Another promising strategy involves the use of tumor-derived exosomes (TEXs)—membrane-bound vesicles released by tumor cells that contain tumor-associated antigens and adhesion molecules [57,58]. Protein-based vaccines, particularly those utilizing heat shock protein-peptide complexes (e.g., HSPPC-96), carry tumor-derived peptides and activate the immune system through antigen presentation by antigen presenting cells (APCs) [59]. Peptide-based cancer vaccines are designed to target specific tumor epitopes—such as KRAS, telomerase, or survivin—yet their efficacy is often limited by MHC restriction and the typically monoclonal nature of the T cell response [60]. DNA vaccines, which encode selected tumor antigens (e.g., MUC1, survivin, ENO1, VEGFR-2), can induce cytotoxic responses through endogenous antigen expression in host cells. However, their clinical application is constrained by safety concerns, including the risk of anti-DNA antibody induction and potential genomic integration [4]. Approaches to cancer vaccines are summarized in Figure 2.

Finally, the route of administration can influence vaccine performance and there is a wide range of available approaches: from intradermal and intramuscular—the most common routes, to more localized, such as intratumoral injections [61]. In summary, due to improvement of nucleic acid engineering and development of advanced delivery systems, mRNA vaccines have overcome many initial limitations (like unacceptable instability or low in vivo potency), however further research is needed to fully establish their clinical translation in the treatment of PDAC.

Numerous vaccine modalities have been explored in PDAC over the last decades, including those based on peptides, viral vectors, dendritic cells, or exosomes. While some of them have shown immunogenicity in preclinical or early clinical studies, their overall therapeutic impact in PDAC has been limited, highlighting the need for more advanced and improved approaches [4]. Peptide-based vaccines typically involve one or several short peptides derived from TAAs or patient-specific neoantigens, administered with an immunologic adjuvant (such as GM-CSF or Toll-like receptor agonists) [62,63]. These vaccines are relatively easy to manufacture and well-tolerated, but they often elicit only modest T-cell responses. One issue is that short peptides usually trigger predominantly CD8⁺ T-cell responses restricted to a particular HLA allele, which limits their full potential [64]. Additionally, because TAAs are self-antigens, activity of peptide vaccines can be hindered by immune tolerance [65]. In PDAC, peptide vaccines against common antigens like telomerase, or MUC1 have demonstrated safety and immune activation, but no significant improvement in survival as monotherapies has been observed [66,67].

DNA vaccines (plasmid DNA encoding a tumor antigen) have also been tested in PDAC treatment. One of the examples is a plasmid vaccine encoding enolase-1 (ENO1), an antigen overexpressed in PDAC, which has successfully increased survival in a mice model [68]. DNA vaccines are inexpensive and thermostable, and they can drive endogenous antigen production like mRNA does. However, DNA must enter the cell nucleus to be further transcribed, which might be inefficient in vivo; methods like electroporation are often required to improve DNA uptake by cells [69,70]. More importantly, there are safety concerns regarding DNA vaccines, including the risk of genomic integration or development of autoimmune reactions [69].

Viral vector vaccines, using platforms such as adenovirus or poxvirus engineered to express antigens, have elicited antitumoral immunogenicity in some cancers [71]. One notable example is GVAX (an irradiated, GM-CSF-secreting allogeneic pancreatic tumor cell vaccine) used in combination with a Listeria monocytogenes vector expressing mesothelin (CRS-207). This modality has improved survival of metastatic PDAC patients and shown minimal toxicity [72]. Viral vectors can induce strong T-cell responses, however pre-existing anti-vector immunity (against common viruses like adenovirus) may limit the widespread application of these vaccines [73].

Dendritic cell (DC) vaccines involve autologous dendritic cells loaded with tumor antigens. They can also be personalized and have shown some success in cancer treatment. However, the designing process is labor-intensive and patient-specific, and the immunosuppressive microenvironment of PDAC can impair the function of DCs [74,75,76]. Still, a few trials (e.g., using DCs loaded with MUC1 peptides) have reported vaccine-induced T-cell responses in PDAC, though without remarkable clinical outcomes [77].

Compared to above-mentioned platforms, mRNA vaccines offer compelling advantages. They do not integrate with the host genome and therefore carry no risk of genomic modifications, highlighting their distinguished safety profile [61,78]. They can be produced rapidly, which is ideal for large-scale personalized neoantigen vaccine distribution [79]. Moreover, a single mRNA vaccine can encode numerous epitopes (even entire sets of neoantigens or multiple TAAs), providing a multi-targeted attack in one product. mRNA vaccines inherently engage both the CD8⁺ and CD4⁺ T-cells [53,80,81]. Finally, unlike protein or peptide vaccines which may be limited by HLA specificity, mRNA vaccines may elicit broader immune responses [82]. Taken together, these features explain the rising enthusiasm for mRNA vaccines over more traditional cancer vaccine approaches [78].

mRNA vaccines trigger both humoral and cytotoxic immune responses against PDAC cells. The core principle of mRNA-based cancer vaccines lies in their capacity to direct the immune system toward TAAs. By introducing mRNA that encodes these antigens, the vaccines are designed to generate a focused immune reaction against cancerous cells while minimizing damage to normal tissues [4]. Once administered, the mRNA is internalized by APCs—especially DCs—where it is translated into the corresponding antigenic proteins. After endocytosis, LNPs release the mRNA into the cytoplasm of APCs, allowing translation into TAAs and activating antigen processing and presentation pathways that promote an anticancer immune response [11,83].

After mRNA escapes the endosome, ribosomes begin translating it into proteins, which are subsequently processed by the ubiquitin–proteasome system to generate peptides suitable for antigen presentation. These peptide fragments are then transported via secretory vesicles to the endoplasmic reticulum, where they are loaded onto MHC class I and II molecules and displayed on the surface of APCs. This process leads to T-cell activation. Cytotoxic CD8+ T cells (Tc) recognize antigens presented by MHC class I, while MHC class II molecules activate helper CD4+ T cells (Th). Through T-cell receptors (TCRs), T cells identify these antigens and perform multiple interactions between immune checkpoint molecules on APCs and lymphocytes. The immune response is further strengthened by pro-inflammatory cytokines such as IL-1, IL-2, and IL-12, which support T-cell proliferation and effector differentiation. Activated CD8+ T cells release granzymes, perforin, TNF-α, and interferon-γ, promoting inflammation and inducing apoptosis in neoplastic cells [84,85].

The antigens can also stimulate B cells, which, with the help of CD4+ T cells, differentiate into plasma cells that secrete antigen-specific antibodies. This humoral immune response works synergistically with cytotoxic CD8+ T-cell activity to enhance the destruction of cancer cells. Additionally, the engagement of co-stimulatory immune checkpoints further drives the activation of antigen-specific T cells, while B lymphocytes are simultaneously induced to produce anticancer antibodies [86,87]. The mechanism of mRNA vaccines in PDAC is similar to the general mode of action observed for mRNA-based vaccines. Figure 3 illustrates a schematic overview of these activities.

Given the immunosuppressive features of PDAC, it seems inevitable that combination strategies, rather than mRNA vaccines alone will be required to unleash the full potential of mRNA vaccines [4]. Therapeutic cancer vaccines aim to generate a pool of tumor-specific T cells, however their efficacy in monotherapy may be blunted by checkpoints and suppressive cells in the TME. Combining mRNA vaccines with immune checkpoint inhibitors (ICIs) is a particularly compelling approach. ICIs such as anti-PD-1, anti-PD-L1, or anti-CTLA-4 can prevent or reverse T-cell inhibition, thereby amplifying the function of vaccine-induced T cells [88]. Vaccines prime critical antigen-specific T-cells, while ICIs sustain the effector function of those cells, by modulating a "cold" tumor—a clear synergistic interaction [4]. Checkpoint blockade then keeps those T cells active upon encountering tumor antigen. Consistent with this rationale, ongoing trials in PDAC are now evaluating mRNA neoantigen vaccines in combination with PD-1/PD-L1 inhibitors. A first-in-human study of a personalized mRNA neoantigen vaccine (autogene cevumeran, by BioNTech) combined with atezolizumab (anti–PD-L1 antibody) and chemotherapy in patients with resectable PDAC has reported promising results. 50% of vaccinated patients have developed neoantigen-specific T-cell responses, and those responders experienced a longer median recurrence-free survival compared to non-responders. While the number of participants was small (n = 16), this finding has confirmed that a mRNA neoantigen vaccine can result in improved outcomes when appropriately supported by other therapies [4,89].

Combining immunotherapy with chemotherapy may be advantageous, as certain chemotherapeutics (like gemcitabine or cyclophosphamide) can regulate MDSCs or Tregs, thereby decreasing immunosuppressive tendencies [90]. Standard PDAC chemotherapy regimens such as FOLFIRINOX can induce immunogenic cell death, releasing tumor antigens and damage-associated molecular patterns (DAMPs), which enhances and promotes the expansion of cancer-targeting T cells [91]. Indeed, the autogene cevumeran trial mentioned above included chemotherapy, and it seems possible that cytotoxic therapy had an impact on altering the TME in ways that synergized with the autogene cevumeran vaccine.

Another combination strategy is stromal remodeling. Agents targeting the desmoplastic stroma or cytokines might improve lymphocyte infiltration. For instance, inhibitors of the TGF-β pathway, which is a key driver of fibrosis and immune evasion in PDAC, could be combined with a vaccine to enhance T-cell penetration [92]. Enzymatic degradation of stromal barriers (e.g., PEGylated hyaluronidase against hyaluronic acid in the matrix) is another approach that has shown promising results in preclinical studies and could logically be extended to vaccine modalities [93]. Leveraging the mRNA technology, some approaches employ multiple therapeutic modalities into one vaccine. For example, an mRNA mixture encoding immune modulators can impact the TME. A recent example is mRNA-2752, an intratumorally administered mRNA vaccine encoding the T-cell stimulator OX40L and pro-inflammatory cytokines IL-23 and IL-36γ. This therapy has shown the activation of dendritic cells and CD8+ T cells locally and has been tested alone or combined with durvalumab in patients with solid tumors [94,95].

Therefore, mRNA vaccines, especially when combined with synergistic immunotherapies, have a potential to break through PDAC's prominent resistance mechanisms. Ongoing clinical research is focused not only on potential drug combinations but also on identifying predictive biomarkers that could find patients who are most likely to benefit from a mRNA vaccine combined with another medication [4]. In conclusion, while mRNA-based neoantigen vaccines offer a highly personalized approach against PDAC, their greatest potential likely lies in integration into combination therapy. With several clinical trials assessing mRNA vaccine combinations in PDAC, the coming years will shed light on their avenue to overcome tumor-induced immunosuppression and improve patient outcomes.