Engineered exosomes silence oncogenic KRAS^G12D and awaken immunity in pancreatic cancer

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal and debilitating malignancy and is currently the second leading cause of cancer-related mortality in the United States, with a median survival of approximately 6 months and a 5-year survival rate of only 5% to 9%. Over the past three decades, significant advances in understanding mechanisms of PDAC initiation and progression have emerged, particularly from genetics, epigenetics, metabolomics, and immunology. These insights have driven extensive efforts to translate this knowledge into effective therapies. However, PDAC remains one of the most formidable and humbling challenges in modern medicine. In this sobering landscape, the most difficult areas to overcome remain in prevention and therapy. In this regard, reviewing and highlighting the work of Kalluri and colleagues is a promising endeavor, as it represents a significant advance in oncology, specifically targeting the previously undruggable KRAS^G12D oncogene and yielding responses far better than expected. This article, therefore, seeks to underscore the precision of the study design, the rigor of the experimentation, and its novel contributions to the mechanism of disease, novel therapeutics, and biomarker discovery. Readers are invited to view this study as a landmark in the field, establishing new benchmarks for future scientific studies and their medical applications.

Rewriting KRAS^G12D targeting through RNA-centric therapeutics

Among human cancers, PDAC is characterized by the near-universal presence of oncogenic KRAS mutations, which occur in over 90% of patients, with KRAS^G12D accounting for approximately 50% of cases (1,2). Notably, until very recently, KRAS was considered “undruggable” because the protein produced by this oncogene lacks structural features that could be exploited for drug discovery. However, recently, the first breakthroughs in this field occurred with covalent inhibitors targeting the less common KRAS^G12C oncogene, which is more common in lung cancer patients (3). In contrast, for KRAS^G12D mutations, several selective inhibitors and degraders are advancing through early-phase clinical trials with encouraging preliminary activity (4), although no targeted therapies are currently approved and optimal clinical strategies remain to be defined.

Kalluri’s team overcomes this challenge by targeting oncogenic KRAS messenger RNA (mRNA) rather than the mutant protein (5). Specifically, these investigators introduce small extracellular vesicles (sEVs) as an innovative platform for in vivo small interfering RNA (siRNA) delivery, using exosomes derived from mesenchymal stromal cells. When reaching PDAC cells, these exosomes selectively antagonize the KRAS^G12D transcript without affecting its wild-type counterpart. To our knowledge, this approach represents not only a conceptual advance by leveraging endogenous intercellular communication machinery for the delivery of precision therapeutics but also a significant translational achievement. Besides its elegance and clinical relevance, the work is distinguished by its rigor in overcoming the major challenge of producing exosomes at a clinical scale with reproducible quality. The authors establish Good Manufacturing Practice (GMP)-compliant production, coupled with stringent quality control, using the high-resolution and specificity of cryo-electron microscopy. Moreover, biodistribution studies in non-human primates set a new benchmark for laboratories pursuing exosome-based therapies, not only for PDAC but also for other solid tumors harboring the same mutation. Following loading of purified exosomes with KRAS^G12D siRNA and systemic delivery to mice and Rhesus macaques, these vesicles preferentially accumulated in the pancreas and liver with negligible toxicity. Mechanistically, the authors discover a remarkable natural tropism mediated in part by the endogenous CD47 signal on the exosome surface, which enhances uptake through macropinocytosis and provides superior performance compared with synthetic nanoparticle platforms. While these findings highlight favorable biodistribution, further optimization of delivery efficiency and tumor-specific uptake will be important to ensure consistent performance within the heterogeneous and densely fibrotic PDAC tumor microenvironment (TME) (6).

Meeting gold-standard criteria in phase I clinical evaluation

Patients with PDAC enrolled in the phase I trial of siRNA-loaded exosomes were highly representative of the typical population with advanced, metastatic disease and extensive prior treatment, including resistance to multiple standard therapies (5). Despite this, the therapy demonstrated a remarkable safety profile, characterized by the absence of infusion-related reactions, no dose-limiting toxicities, and no treatment delays due to adverse events. Notably, the maximum tolerated dose was not reached, even under an escalating-dose protocol ranging from 0.15 to 4.8 mg per infusion, with cumulative exposure reaching up to 28.8 mg of the therapeutic siRNA over a 6-week treatment period. This favorable safety profile contrasts with that of many existing therapies for PDAC, which often exhibit modest efficacy but substantial toxicity, including cutaneous reactions, severe gastrointestinal symptoms, and metabolic disturbances that limit dosing, shorten treatment duration, and compromise patient quality of life. Accordingly, the tolerability of iExoKras^G12D may open the door to the design of rational combinations aimed at maintaining physiological function in patients with PDAC. Beyond safety, the trial also demonstrated high target specificity and mechanistic engagement. Liquid biopsy analyses revealed reductions in circulating KRAS^G12D levels, accompanied by decreased phosphorylated extracellular signal-regulated kinase (ERK) and fewer cytokeratin 19-positive circulating tumor cells. Collectively, the success of this phase I study reflects the achievement of key gold-standard endpoints, including mechanistic validation, identification of pharmacodynamic biomarkers, and a solid rationale for dose escalation and future combinatorial approaches. As a phase I study, these findings are appropriately focused on safety and mechanistic engagement, and while clinical efficacy remains to be established, they provide a strong foundation for subsequent studies in larger, controlled trials.

Revealing the unexpected immune checkpoint function of oncogenic KRAS

A particularly important contribution of this study lies in the deep mechanistic insight it provides into how oncogenic KRAS reprograms the immune microenvironment (Figure 1). Through a series of well-designed preclinical experiments, the authors demonstrate that KRAS^G12D signals to the nucleus to epigenetically repress FAS expression in PDAC cells (5). The data convincingly show that KRAS^G12D induces the coordinated recruitment of two complementary epigenetic silencing mechanisms, namely DNA methylation and H3K27 trimethylation mediated by DNMT1 and EZH2, respectively, at the FAS promoter, resulting in gene repression. Loss of FAS expression on tumor cells disrupts FAS-FASL signaling and impairs CD8⁺ T cell-mediated cytotoxicity, thereby promoting immune evasion. These novel findings provide a mechanistic explanation for how oncogenic KRAS cooperates with this death receptor pathway to confer immunotherapy resistance in PDAC.

Figure 1 Molecular mechanism illustrating KRAS-driven epigenetic repression of FAS and immune evasion in PDAC cells before treatment (left), and restoration of FAS-mediated immune recognition following exosome-mediated delivery of KRAS-targeting siRNA (right). Image was created using Biorender.com. mRNA, messenger RNA; PDAC, pancreatic ductal adenocarcinoma; siRNA, small interfering RNA.

Importantly, this epigenetic program is reversible. Delivery of Kras^G12D-targeting siRNA via iExoKras^G12D antagonizes KRAS-driven repression, restores FAS expression on the cancer cell surface, and re-sensitizes tumor cells to FASL-mediated CD8⁺ T cell killing. The functional relevance of this mechanism is elegantly demonstrated in genetically engineered KPC (Kras/Trp53 mutant) mouse models, in which iExoKras^G12D treatment significantly extends survival in immunocompetent animals. This survival benefit is lost in mice deficient in CD8⁺ T cells, whereas enhanced therapeutic efficacy is found in CD4⁺ T cell-deficient animals, consistent with reduced regulatory T cell-mediated immunosuppression. Together, these results establish CD8⁺ T cells as critical effectors of immune surveillance following silencing of oncogenic KRAS and uncover an unexpected role for KRAS as an immune checkpoint regulator that can be therapeutically targeted through exosome-mediated siRNA delivery.

Transforming immunologically cold tumors creates a window of opportunity for rational immunotherapy combinations

Among its many challenges, PDAC is also well recognized for its resistance to immune checkpoint blockade, including programmed cell death protein 1 (PD-1) and programmed death-ligand 1 (PD-L1) inhibitors, therapies that have transformed the treatment landscape for several other malignancies (7). In this context, the work by Kalluri and colleagues is particularly promising for the field, since it demonstrates that treatment with iExoKras^G12D increases intratumoral infiltration of CD8⁺ T cells, CD4⁺ T cells, and Foxp3⁺ regulatory T cells in patient biopsies, hallmarks of robust immunosurveillance activation (5). At present, this immune remodeling can be viewed as a “priming event”, in which iExoKras^G12D initiates the reprogramming of the TME from an immunologically cold state to one that is permissive of anti-tumor immune responses. This transition creates a window for rational combination strategies with immunotherapies designed to further amplify immune activation. Consistent with this concept, the authors provide proof of principle by showing that iExoKras^G12D synergizes with cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) blockade, but not with PD-1 inhibition, in the preclinical setting. Specifically, combined treatment with iExoKras^G12D and anti-CTLA-4 antibodies induces significant tumor regression and improves survival in orthotopic PDAC models, whereas PD-1 blockade failed to display similar benefits.

The mechanistic basis for this differential response likely reflects fundamental differences between immune checkpoint therapies. CTLA-4 blockade modulates both effector T cells and regulatory T cells (8), thereby amplifying the immune consequences of oncogenic KRAS suppression and enabling more robust antitumor T cell responses (Figure 2). In contrast, PD-1 blockade may be insufficient in the absence of prior immune priming (9). Further strengthening these findings, the authors report that combining iExoKRAS^G12D with CTLA-4 blockade promotes the formation of tertiary lymphoid structures within tumors that resemble ectopic lymph nodes (5). Such structures have been associated with improved responses to immunotherapy and favorable prognosis across multiple cancer types, as they function as localized foci for antigen presentation, T cell priming, and the establishment of durable immunologic memory (10). Therefore, it is tempting to speculate that, when combined, the described effects may support more durable tumor immune surveillance and help prevent, at least in part, disease relapse, an unmet need of paramount importance for extending both the duration and quality of life of patients affected by pancreatic cancer.

Figure 2 Combination of iExoKras^G12D therapy with anti-CTLA-4 blockade transforms tumor immunity and induces tertiary lymphoid structures. Image was created using Biorender.com. CTLA-4, cytotoxic T-lymphocyte-associated protein 4; siRNA, small interfering RNA; Treg, regulatory T cell.

Notwithstanding these promising findings, it is important to consider that PDAC is highly prone to TME-driven adaptive resistance, which may ultimately limit the durability of KRAS-targeted interventions (2). The dense desmoplastic stroma and complex cellular composition not only constrain drug delivery but also shape tumor cell signaling. Attenuation of oncogenic KRAS may trigger compensatory feedback loops, including reactivation of receptor tyrosine kinases and downstream MAPK and PI3K-AKT pathways, sustaining tumor cell survival despite effective target engagement (11,12). Stromal-tumor interactions and immunosuppressive populations may further reinforce these responses, while intratumoral heterogeneity and clonal evolution enable resistant subpopulations to emerge over time (13). Together, these considerations underscore the need for rational combination strategies that address both tumor-intrinsic and microenvironmental resistance mechanisms. Integrating iExoKras^G12D with inhibitors of compensatory pathways or approaches that remodel the TME may enhance response durability. In parallel, immunotherapeutic combinations such as CTLA-4 blockade may help sustain antitumor immunity (14). At the same time, the broader consequences of sustained immune modulation remain to be defined, including potential off-target effects and unintended immune activation with prolonged treatment.

Leveraging the exosome platform to broaden therapeutic reach

Notably, the implications of this work extend beyond KRAS^G12D and pancreatic cancer. Several other cancers, including colorectal cancer, lung adenocarcinoma, and hepatobiliary tumors, among others, harbor the KRAS^G12D variant and display immunosuppressive features similar to those observed in PDAC (15). More broadly, the exosome-based delivery platform described here has the potential to serve as a versatile, adaptable therapeutic system (16). These biological sEVs can be engineered to package diverse siRNAs, enabling targeting of naturally occurring or synthetic RNAs across a wide range of malignant and nonmalignant diseases. Importantly, this study also establishes a conceptual link between oncogenic drivers, epigenetic regulation, and immune evasion in tumors that have silenced FAS and become resistant to effective T cell-mediated surveillance, even in settings where abundant, active T cells populate the TME (5). Thus, this approach may be broadly applicable for restoring FAS function and enhancing immune recognition, including in combination with emerging cellular immunotherapies such as chimeric antigen receptor (CAR)-T and T cell receptor (TCR)-engineered T cell therapies.

Beyond its biological insights, this work presents a bona fide platform for the development of exosome-based therapeutics that are both safe and effective. The extensive characterization of this therapeutic tool highlights a rigorous approach to manufacturing, quality control, and toxicology, as well as a carefully controlled translational path from preclinical studies to human trials. An important consideration for clinical translation is the potential for batch-to-batch variability in sEV production (17). Because these sEVs are derived from human bone marrow-derived mesenchymal stromal cells, donor-dependent and process-related factors can influence vesicle composition and functional properties, with implications for consistency and potency across production lots (18). Addressing this will require standardized donor or cell bank selection, rigorous characterization protocols that define critical quality attributes and potency assays, and potentially the use of more uniform or engineered mesenchymal stromal cell sources or culture systems to support scalable, reproducible manufacturing (19). Looking ahead, future studies could integrate circulating tumor DNA monitoring with paired tissue biopsies in early-phase clinical trials to ensure robust on-target activity and inform optimal dose selection. In this regard, the study serves as a blueprint for team-based translational research, beginning with mechanistic insights derived from patient samples, progressing to the generation and optimization of relevant preclinical models, and culminating in a rational clinical trial design. Overall, this approach effectively closes the translational loop, linking fundamental laboratory discovery to the delivery of safe and effective therapies for patients.

Opening a road ahead as precision oncology engages tumor immunity

Phase II clinical trials shift the primary focus from safety to evaluating whether a therapeutic intervention meaningfully engages its target and demonstrates biological activity in patients with the disease of interest (20). In this regard, a logical next step following the current study would be to initiate a phase II trial evaluating iExoKras^G12D in combination with CTLA-4 blockade in patients with pancreatic cancer. Such a study could be particularly informative if conducted in the second-line metastatic or neoadjuvant setting, where patients may retain a greater capacity to mount effective immune responses. An optimal trial design would incorporate adaptive features, including circulating tumor DNA monitoring, detailed immune profiling, and assessment of tertiary lymphoid structure formation as pharmacodynamic readouts. Careful attention to patient selection will also be critical, with stratification based on baseline immune status and KRAS^G12D allele burden to identify tumors most likely to benefit from this therapeutic strategy. Demonstration of meaningful biological activity together with an acceptable safety profile would provide a strong rationale for advancing to phase III studies and, ultimately, toward regulatory approval. Despite all the promise of this approach, several key challenges remain to be addressed. Further work will be required to determine whether exosome platforms can be engineered to achieve enhanced tumor specificity through surface modification with defined targeting ligands. In addition, the durability of the immune memory elicited by this therapy, as well as the extent to which therapy-induced tertiary lymphoid structures correlate with long-term disease control, remains to be established. It will also be critical to determine whether resistance mechanisms emerge over time and, if so, to identify strategies to anticipate or counteract such adaptations. Nevertheless, the work by Kalluri and colleagues represents a critical advance not only for pancreatic cancer but for the broader field of precision oncology. The study elegantly demonstrates that exosomes can join the expanding arsenal of precision delivery platforms for human therapy, that oncogenic KRAS variants long considered “undruggable” can be safely targeted in the clinical setting, and that such interventions can actively reprogram tumor immunity. Therefore, their study outlines a path toward new therapeutic opportunities for historically incurable cancers by directly targeting the molecular drivers of tumor initiation and progression while engaging the immune system as the final effector of durable antitumor responses.

Redefining paradigms at the interface of precision oncology and immunity

The work serves as a powerful reminder of the principle articulated by Thomas Kuhn in The Structure of Scientific Revolutions, namely that scientific progress often requires challenging established paradigms and embracing the uncertainty and complexity of new ones (21). Indeed, not long ago, the idea that inhibitory RNA could be delivered to otherwise incurable tumors using biologically derived vesicles from mesenchymal stromal cells would have seemed aspirational. Even more remote was the notion that silencing oncogenic drivers could simultaneously reprogram the immune microenvironment. In light of this, it is particularly notable that Kalluri and colleagues pursued and ultimately realized a concept that directly challenged the long-standing view of targeted therapy and immunotherapy as separate domains within precision oncology (22). These investigators demonstrate that these approaches are not competing strategies but rather complementary paradigms that can be integrated to achieve therapeutic outcomes that neither could accomplish alone. As presented, their work brings renewed hope to patients with pancreatic cancer and their families, many of whom have endured a prolonged and difficult odyssey for meaningful advances in treatment. While the journey from bench to bedside is long, complex, and fraught with obstacles, this work exemplifies why sustained investment in mechanistic insight and translational innovation remains essential. Ultimately, it reinforces the possibility that even the most lethal malignancies may, in time, be transformed into manageable diseases, and that durable cures, once considered unattainable, may increasingly come within reach.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Annals of Pancreatic Cancer. The article has undergone external peer review.

Peer Review File: Available at https://apc.amegroups.com/article/view/10.21037/apc-26-0008/prf

Funding: This work was supported by the National Institutes of Health(No. R01DK52913 to R.A.U. and G.L., and No. R01CA247898 to G.L.), the Advancing a Healthier Wisconsin Endowment (to R.A.U.), the Linda T. and John A. Mellowes Endowed Innovation and Discovery Fund (to R.A.U.), and the Markus Family Funds for Discovery and Innovation Family Funds to the Mellowes Center for Genomic Sciences and Precision Medicine.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://apc.amegroups.com/article/view/10.21037/apc-26-0008/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bannoura SF, Khan HY, Azmi AS. KRAS G12D targeted therapies for pancreatic cancer: Has the fortress been conquered? Front Oncol 2022;12:1013902. [Crossref] [PubMed]
2. Linehan A, O’Reilly M, McDermott R, et al. Targeting KRAS mutations in pancreatic cancer: opportunities for future strategies. Front Med (Lausanne) 2024;11:1369136. [Crossref] [PubMed]
3. Choucair K, Imtiaz H, Uddin MH, et al. Targeting KRAS mutations: orchestrating cancer evolution and therapeutic challenges. Signal Transduct Target Ther 2025;10:385. [Crossref] [PubMed]
4. Shi K, Li A. Targeting KRAS G12D: Advances in Inhibitor Design. Thorac Cancer 2025;16:e70203. [Crossref] [PubMed]
5. Kalluri VS, Smaglo BG, Mahadevan KK, et al. Engineered exosomes with Kras(G12D) specific siRNA in pancreatic cancer: a phase I study with immunological correlates. Nat Commun 2025;16:8696. [Crossref] [PubMed]
6. Zhou Y, Ma Y, Sheng J, et al. Breaking Down Barriers in Drug Delivery by Stromal Remodeling Approaches in Pancreatic Cancer. Mol Pharm 2024;21:3764-76. [Crossref] [PubMed]
7. Ju Y, Xu D, Liao MM, et al. Barriers and opportunities in pancreatic cancer immunotherapy. NPJ Precis Oncol 2024;8:199. [Crossref] [PubMed]
8. Peggs KS, Quezada SA, Chambers CA, et al. Blockade of CTLA-4 on both effector and regulatory T cell compartments contributes to the antitumor activity of anti-CTLA-4 antibodies. J Exp Med 2009;206:1717-25. [Crossref] [PubMed]
9. Tumeh PC, Harview CL, Yearley JH, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature 2014;515:568-71. [Crossref] [PubMed]
10. Baxevanis CN, Sofopoulos M, Tsitsilonis OE, et al. Exploring the Pivotal Functions of Tertiary Lymphoid Structures in Cancer Prognosis and Immunotherapy Outcomes. Cancers (Basel) 2025;17:3754. [Crossref] [PubMed]
11. Ryan MB, Coker O, Sorokin A, et al. KRAS(G12C)-independent feedback activation of wild-type RAS constrains KRAS(G12C) inhibitor efficacy. Cell Rep 2022;39:110993. [Crossref] [PubMed]
12. Lin Q, Morales-Giron AA, Sander C, et al. Overcoming Adaptive Resistance to KRASG12D Blockade in Pancreatic Cancer through Vertical Pathway Inhibition. Clin Cancer Res 2026;32:2279-92. [Crossref] [PubMed]
13. Petrenko O, Kirillov V, D’Amico S, et al. Intratumor heterogeneity in KRAS signaling shapes treatment resistance. iScience 2025;28:111662. [Crossref] [PubMed]
14. Liu Y, Han J, Hsu WH, et al. Combined KRAS Inhibition and Immune Therapy Generates Durable Complete Responses in an Autochthonous PDAC Model. Cancer Discov 2025;15:162-78. [Crossref] [PubMed]
15. Zeissig MN, Ashwood LM, Kondrashova O, et al. Next batter up! Targeting cancers with KRAS-G12D mutations. Trends Cancer 2023;9:955-67.
16. Serrano DR, Juste F, Anaya BJ, et al. Exosome-Based Drug Delivery: A Next-Generation Platform for Cancer, Infection, Neurological and Immunological Diseases, Gene Therapy and Regenerative Medicine. Pharmaceutics 2025;17:1336. [Crossref] [PubMed]
17. Adlerz K, Patel D, Rowley J, et al. Strategies for scalable manufacturing and translation of MSC-derived extracellular vesicles. Stem Cell Res 2020;48:101978. [Crossref] [PubMed]
18. Witwer KW, Van Balkom BWM, Bruno S, et al. Defining mesenchymal stromal cell (MSC)-derived small extracellular vesicles for therapeutic applications. J Extracell Vesicles 2019;8:1609206. [Crossref] [PubMed]
19. Maumus M, Rozier P, Boulestreau J, et al. Mesenchymal Stem Cell-Derived Extracellular Vesicles: Opportunities and Challenges for Clinical Translation. Front Bioeng Biotechnol 2020;8:997. [Crossref] [PubMed]
20. Torres-Saavedra PA, Winter KA. An Overview of Phase 2 Clinical Trial Designs. Int J Radiat Oncol Biol Phys 2022;112:22-9. [Crossref] [PubMed]
21. Kuhn TS. The structure of scientific revolutions. Chicago: University of Chicago Press; 1970.
22. Vanneman M, Dranoff G. Combining immunotherapy and targeted therapies in cancer treatment. Nat Rev Cancer 2012;12:237-51. [Crossref] [PubMed]