Exosomes in personalized medicine for pancreatic cancer

Image

The story begins with an unexpected finding: when exosomes, nanometer-sized lipid bilayer vesicles shed by all cells, were administered into mice, they did not simply disappear into circulation or accumulate in the usual reticuloendothelial organs such as liver, spleen, lymph nodes. Instead, they distributed across multiple tissues — including, strikingly, the pancreas1. This singular observation sparked a growing curiosity about whether exosomes, being the body’s intercellular couriers, could be repurposed into therapeutic delivery vehicles. If nature had already engineered them for stability, immune compatibility, and efficient cargo transport, perhaps they could succeed where synthetic nanoparticles can fall short.

Compared with liposomes or polymer-based nanoparticles, exosomes offer an array of advantages rooted in evolution rather than engineering2. These vesicles carry surface proteins such as integrins, tetraspanins, and CD47 that allow them to circulate longer, evade macrophage clearance, engage in membrane fusion, and home to specific tissues. Unlike synthetic platforms, they integrate seamlessly into endogenous uptake pathways — including macropinocytosis (a process used by cells to uptake nutrients)— and likely exhibit lower immunogenicity. Exosomes indeed possess an intrinsic ability for shuttling RNA between cells, with stability and biodistribution features that make them compelling candidates for clinical translation.

Against this conceptual backdrop, and with the catalytic support of the MD Anderson Cancer Center’s Moon Shot initiative, a bold idea took shape: to develop engineered exosomes as a new type of biologic drug for pancreatic cancer. In 2017, Kamerkar and colleagues demonstrated that exosomes derived from fibroblast-like mesenchymal cells could be engineered to deliver siRNA targeting the KRASG12D mutation — one of the most pervasive oncogenic drivers in pancreatic ductal adenocarcinoma. This work revealed two key mechanistic advantages1. First, the exosomes naturally displayed CD47, the “don’t eat me” signal that shielded them from rapid monocyte and macrophage clearance. Second, KRAS-mutant pancreatic cancer cells were shown to rely on macropinocytosis, a form of bulk nutrient uptake that incidentally enhanced their ability to internalize therapeutic exosomes. Together, these features enabled the engineered “iExosomes” to achieve substantial survival benefits in genetically engineered mouse models, outperforming synthetic delivery systems and marking the first convincing demonstration that exosomes could be mobilized to target a historically undruggable oncogene.

Image

But elegant biology is only the starting point for a clinical therapy. The next challenge was to build a manufacturing and regulatory framework capable of supporting a first-in-human trial. In 2018, Mendt and colleagues described the development of a full GMP production pipeline using bone marrow–derived mesenchymal stem cells as an exosome factory3. This effort encompassed bioreactor scale-up, standardized purification methods, release criteria, stability testing, and rigorous biodistribution and toxicology studies in mice. By the time this work was complete, the team had transformed a conceptual technology into a clinical-grade biologic with a well-defined regulatory process.

The culmination of this decade-long effort arrived in 2025, with the publication of the first-in-human trial of engineered exosomes carrying KRASG12D-specific siRNA4. The work reported on toxicology and biodistribution studies of the GMP-grade iExosomes in mice and monkeys. This was followed by human testing. In this Phase I study (iEXPLORE, NCT03608631), patients with advanced, heavily pretreated pancreatic cancer received intravenous infusions of bone marrow mesenchymal stromal cells (MSC)-derived iExosomes. The findings were both reassuring and encouraging: no dose-limiting toxicities were observed, the maximum tolerated dose was not reached, and early biological signals of activity emerged. Tumor biopsies showed reductions in KRASG12D alleles (in liquid biopsies), suppression of downstream pERK signaling, and increased infiltration of CD8⁺ T cells — an immunologic awakening not typically seen in late-stage PDAC. These outcomes, built on other mechanistic work, established iExosomes as the first engineered exosome therapeutic to target an oncogenic driver in human cancer.

Image

This progress was made possible by the collaboration of many involved in the published study, which formed the multidisciplinary backbone of the project — spanning exosome biology, translational oncology, manufacturing, biostatistics, and clinical trial design. It is rare for a concept to travel this far from initial observation to first-in-human testing, and rarer still for it to do so with such a tightly integrated team. 

As the landscape of KRAS therapeutics evolves, small-molecule inhibitors now stand beside iExosomes as complementary approaches. KRASG12C inhibitors are already clinically available, and KRASG12D inhibitors are rapidly advancing. Yet tumors inevitably develop resistance through compensatory signaling, alternative KRAS isoforms, and microenvironmental escape pathways. Engineered exosomes, with their modular RNA cargo capacity and low toxicity, are poised to play an important role in these emerging resistance networks — delivering siRNA or other nucleic acids to silence escape nodes, eliminate residual KRAS transcripts, or reshape the immune microenvironment.

Indeed, the 2025 Nature Communications paper revealed that iExosomes do more than knock down KRAS: they also remodel the tumor microenvironment, changing the immune composition and enhancing susceptibility to checkpoint blockade4. In preclinical and clinical analyses, iExosomes treatment was associated with polarization of inflammatory pathways and increased anti-tumor cytotoxic CD8⁺ T-cell presence, which led to a tumor milieu more receptive to anti-CTLA-4 therapy. Because iExosomes exhibited no systemic toxicity in Phase I, these findings laid the groundwork for a Phase II trial now recruiting patients to evaluate iExosomes combined with anti-CTLA-4 in pancreatic cancer — a mechanistically grounded, opportunistic therapeutic alliance.

Image

In summary, the journey of engineered exosomes spans more than a decade of scientific insight, technical innovation, and clinical perseverance. What began with an unexpected biodistribution study grew into a new therapeutic platform rooted in the biology of intercellular communication. The publication4 of the first-in-human trial in late 2025 marks a turning point: engineered exosomes have moved from conceptual possibility to clinical feasibility, opening the door to a new category of biologics for precision oncology. The work is far from finished, but the path is now clear. 

COI disclosure (as noted in associated publication4): AM is named on a patent that has been licensed to Exact Sciences. CH reports research funding to the institution from Sanofi, Avenge, Iovance, KSQ, Theolytics, BTG, Novartis, 280Bio, Astrazeneca, EMD Serono, Takeda, Obsidian, Genentech, BMS, Summit Therapeutics, Artidis, and Immunogenesis, and personal fees from Regeneron and stock options from Briacell outside the submitted work. LMSS reports research support from Theolytics, advisory role/consulting fees from BioNTech, and travel support from 10x Genomics, all outside the scope of this work. MD Anderson Cancer Center, VSK, ES, MM, RK hold a patent in the area of exosomes biology. Patents related to EVs and Exosomes have been licensed to PranaX, Inc. VSK and RK are stock equity holders in PranaX, Inc. VSK and RK are founders, stock owners, and scientific consultants to PranaX, Inc. The remaining authors declare no competing interests.

Author note: ChatGPT was used for figure generation and assistance in writing and editing

References

1. Kamerkar, S., et al. Exosomes facilitate therapeutic targeting of oncogenic KRAS in pancreatic cancer. Nature 546, 498-503 (2017).
2. Kalluri, R. & LeBleu, V.S. The biology, function, and biomedical applications of exosomes. Science 367(2020).
3. Mendt, M., et al. Generation and testing of clinical-grade exosomes for pancreatic cancer. JCI Insight 3, e99263 (2018).
4. Kalluri, V.S., et al. Engineered exosomes with Kras(G12D) specific siRNA in pancreatic cancer: a phase I study with immunological correlates. Nat Commun 16, 8696 (2025).