Often diagnosed when surgery is no longer an option, pancreatic cancer is one of the most difficult cancers to treat and has one of the lowest rates of survival among major malignancies.
Like many solid tumors, the most common type of pancreatic cancer, pancreatic ductal adenocarcinoma (PDAC), is shielded by the desmoplastic matrix—a dense barrier of connective tissue, structural proteins, and specialized cells called cancer-associated fibroblasts (CAFs)—that also suppresses the immune response.
Importantly, this microenvironment has limited the use of chimeric antigen receptor (CAR) T cell therapy, a form of cancer immunotherapy that has achieved notable success in treating blood cancers, such as lymphoma, leukemia, and multiple myeloma.
Now, researchers led by Ellen Puré of the School of Veterinary Medicine have used lipid nanoparticles (LNPs)—tiny fat-based particles that can serve as efficient drug delivery systems—to generate CAR T cells directed at CAFs in vivo to “melt away” this barrier in a preclinical model. These findings, published in Cancer Immunology Research , pave the way for a potentially safer, more accessible, and cost-effective method for using CAR T cell therapy to treat solid tumors.
“The expansion of CAR Ts has been enormous,” says Puré, a professor and director of the Penn Vet Cancer Center. “But they've really only been successful in liquid tumors.”
Conventional CAR T cell therapy is also, she notes, complex to administer, requiring a patient’s T cells to be removed, engineered, and then reinfused. The process also often requires lymphodepletion, which temporarily lowers the number of immune cells that might compete with or inhibit the infused CAR T cells. By comparison, LNP–based approaches may be simpler and less expensive to deliver and have shown encouraging results in other conditions.
LNPs act as tiny delivery vehicles, carrying the CAR instructions directly to the patient’s T cells so that CAR production happens in the patient’s body, explains Khuloud Bajbouj, a senior research investigator in Puré’s lab.
The team used LNPs to equip T cells to target fibroblast activation protein, or FAP, because it is present at high levels on a subset of cancer‑associated fibroblasts —cells known to be critical for PDAC progression. “It is like you are equipping T cells with a laser-focused approach,” says Bajbouj.
They chose to focus on the FAP as their previous studies targeting these cells using conventional CAR T delayed tumor growth, says Puré.
In this study, they found that a single dose of targeted LNPs (tLNPs) in a preclinical model of PDAC was as good, if not better, at inhibiting tumor growth than the conventional approach.
“It's usually less than 10% of all the T cells that go into the tumor that are armed,” says Puré. “But when we do this with the tLNPs, we’re finding 40 to 60% of the T cells are expressing the CAR.”
These armored T cells are there for a short time compared with what is seen with the conventional approach, she adds. “It’s like the entire army just comes in, all at once, instead of in waves.”
And the effect of this surprised even the researchers. “We expected it to get rid of the FAP-positive cells,” says Puré. “But we didn't expect the desmoplastic matrix to just melt away.”
This result paves the way for the use of this approach with other therapies to improve their efficacy in solid tumors, says Bajbouj. “It can be used alongside conventional therapies such as chemotherapy or immunotherapy such as immune checkpoint inhibitors, antibody drug conjugates, and the like.”
“If you’re not in at the table, you can’t negotiate, right?” says Puré, adding that these tLNPs could also be used to test new therapies or even CAR T therapies that have not yet been shown to work. “Once the door is open, a lot of therapies could go in and do well.”
Their data also suggest that this approach might be effective for treating metastatic cancer, which, she adds, is what kills most cancer patients.
“FAP-positive cells can facilitate tumor cell spread to other places, helping them survive in transit and enhancing establishment of metastatic lesions,” says Puré. Tumor cells also send signals to activate FAP in those other places, which she likens to preparing garden beds before planting seeds.
“We think these cells are very important in the metastatic phase,” she says. “If we target that, we know that we will have even a bigger impact.”
Ultimately, they believe that targeting FAP-positive cells using LNPs has the potential to treat other conditions, such as fibrosis, autoimmunity, arthritis, or even wound scarring.
“You are not going to put patients through conventional CAR T cell therapy for all of these indications,” says Puré. “But the tLNP mRNA approach may be more palatable from a risk-benefit perspective.”
Ellen Puré is a professor in the Department of Biomedical Sciences at Penn Vet and the director of the Penn Vet Cancer Center.
Khuloud Bajbouj is a senior research investigator in the Puré lab.
Other authors are Li Huang, Leslie Todd, and Zebin Xiao of Penn Vet; Steven M. Albelda, Faris Halilovic, Carl June, Hamideh Parhiz, Tyler E. Papp, Jayalakshmi Ramani, and Drew Weissman of the Perelman School of Medicine ; and Haig Aghajanian, Yanjie Bao, Adrian Bot, and Matthew Butcher of Capstan Therapeutics.
This work was supported by sponsored research agreements from Capstan Therapeutics to E.P, S.M.A, H.P., D.W., C.H.R. E.P was also supported by a grant from Alliance for Cancer Gene Therapy (ACGT), Project Number: ACGT 53927.
Cancer Immunology Research
Experimental study
Animals
Targeted Lipid Nanoparticle Delivery of FAP-CAR mRNA Enables Potent In Vivo T-cell Engineering against Pancreatic Tumors
17-Mar-2026
K. Bajbouj reports grants from Capstan Therapeutics during the conduct of the study. L. Todd reports grants from Capstan Therapeutics during the conduct of the study, as well as a patent for Disrupting tumor tissues by targeting fibroblast activation protein (FAP) patent number: 12116418 issued. L. Huang reports grants from Capstan Therapeutics during the conduct of the study. H. Aghajanian reports other support from Capstan Therapeutics during the conduct of the study, as well as other support from Capstan Therapeutics outside the submitted work. C.H. June reports grants and personal fees from Capstan Therapeutics outside the submitted work. D. Weissman reports a patent for Use of in vivo FAP CAR pending. H. Parhiz reports grants from Capstan Therapeutics during the conduct of the study; grants and other support from Capstan Therapeutics outside the submitted work; and a patent for anti–CD5-targeted FAP-CAR T mRNA-LNP (US-20230203538-A1) licensed. S.M. Albelda reports grants from Capstan Therapeutics during the conduct of the study; grants and other support from Capstan Therapeutics outside the submitted work; and a patent for fibroblast activation protein (FAP)–targeted CAR-expressing cells to deplete FAP-expressing cells in protumorigenic niches, primary tumors, and metastatic disease and disrupt tumor desmoplastic matrix in solid tumors issued and licensed to Capstan Therapeutics. E. Puré reports grants from Capstan Therapeutics during the conduct of the study; grants from Capstan Therapeutics outside the submitted work; and a patent for 12,116,418 B2 issued. No disclosures were reported by the other authors.