2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

SESSION

702. CAR-T Cell Therapies: Basic and Translational: Poster II

Program: Poster Abstract Session
Sunday, December 13, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Assays, Molecular Testing, Omics Technologies, Pathology, Bioinformatics, Cell Expansion, Computational Biology, Emerging Technologies, Gene Editing, Profiling, Treatment Considerations, Combination Therapy, Drug Development, Biological Therapies, Chimeric Antigen Receptor, (CAR)-T Cell Therapies, Immunotherapy, Adverse Events, Artificial Intelligence (AI), Deep Learning, Generative AI, Machine Learning, Biological Processes, Apoptosis, Pathogenesis, Genomics, Immune Mechanism, Metabolism, Microbiome, Molecular Biology, Multi-Systemic Interactions, Clinical Practice (Health Services and Quality), Quality Improvement, Diseases, Immunology, Autoimmune Disorders, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, Plasma Cell Disorders, Myeloid Malignancies, Acute Myeloid Malignancies, ALL, Non-Hodgkin Lymphoma, B Cell Lymphoma, Aggressive Lymphoma, Research, Fundamental Science, AML, Translational Research, Clinical Research, Clinical Trials, Real-World Evidence, Registries, Survivorship, Study Population, Human, Adult, Animal Model
PAPER

4102 Transient, non-integrating CAR-T cells generated in vivo with targeted lipid nanoparticles control acute myeloid leukemia in humanized PDX models.

Thomas Andrew Fox, MD PhD, Marta Freschi2, Giada Zambonini, M.Sc.3, Edoardo Campodonico, MD4, Francesco Romano, MD5, Francesco Medici, MD6, Andrea Cosentino, MD7, Joe Yu8, Gabriele Casirati, M.D., Ph.D.9, Pietro Genovese, PhD10
1 Dana-Farber Cancer Institute, Boston, Massachusetts, United States, Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States, University College London Hospitals NHS Foundation Trust, London, United Kingdom, University College London, London, United Kingdom
2 Dana-Farber Cancer Institute, Boston, Massachusetts, United States, Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States
3 Dana-Farber Cancer Institute, Boston, Massachusetts, United States, Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States
4 Dana-Farber Cancer Institute, Boston, Massachusetts, United States, Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States
5 Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States
6 Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States
7 Dana-Farber Cancer Institute, Boston, Massachusetts, United States, Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States
8 Tiva Biosciences, Woburn, Massachusetts, United States
9 Dana-Farber Cancer Institute, Boston, Massachusetts, United States, Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States
10 Dana-Farber Cancer Institute, Boston, Massachusetts, United States, Dana-Farber/Boston Children's Cancer and Blood Disorders Center, Boston, Massachusetts, United States, Harvard Medical School, Boston, Massachusetts, United States

Background

CAR-T therapy has transformed treatment of selected hematologic malignancies, but broader application is constrained by complex ex vivo manufacturing and limited control of CAR expression after administration. Targeted lipid nanoparticles (tLNPs) offer a non-integrating strategy to deliver CAR mRNA directly to T cells in vivo, enabling rapid, transient, re-dosable CAR expression. We developed anti-FLT3 CAR mRNA tLNPs and tested whether this platform generates functional CAR-T cells in vivo and controls AML in humanized patient-derived xenograft (PDX) models.

Methods

CAR mRNA was synthesized by in vitro transcription and formulated into anti-CD3- or anti-CD8-targeted tLNPs by microfluidic mixing; in vivo targeting was evaluated in PBMC-humanized mice. To assess antitumor and on-target hematopoietic activity, NSGBW mice were engrafted with CD34-positive HSPCs, reconstituted for 12 weeks, challenged with AML PDXs, and injected with syngeneic human T cells before tLNP or CAR-T treatment.

Results

CD3- and CD8-targeted tLNPs carrying GFP mRNA efficiently transduced stimulated and unstimulated human T cells in vitro: 750 ng tLNP added to 5 × 105 T cells yielded 95% ± 3% GFP-positive cells at 24 hours (n=3), undetectable by day 6, confirming transient expression. In PBMC-humanized mice, CD3-targeted GFP tLNPs generated GFP expression in 47.0% of circulating CD4-positive and 33.8% of CD8-positive T cells at 24 hours, versus 12.9% and 17.2% with untargeted LNPs. CD8-targeted GFP tLNPs showed high specificity, inducing GFP expression in 65% vs 13% of CD8- vs CD4-positive splenic T cells. Engagement of the CD3 molecule by CD3-tLNPs induced in vivo T-cell activation (CD69), whereas CD8-tLNPs did not.

We next tested whether tLNP-mediated delivery of anti-FLT3 CAR mRNA could generate functional CAR-T cells. FLT3 CAR tLNPs mediated CAR expression in 21% ± 4% of T cells at 24 hours and induced antigen-specific killing of FLT3-expressing K562 cells in vitro. In HSPC-humanized AML PDX mice, cohorts received PBS, T cells alone, ex vivo-transduced FLT3 CAR-T cells, or T cells followed by two CD3-targeted tLNP doses one week apart. Two mice died within 24 hours of the second CD3-targeted dose, likely from CD3-associated T-cell over-activation. tLNPs generated circulating CAR-positive T cells at 24 hours (G4S-positive, 76% ± 12% vs 29% with ex vivo CAR-T) and expanded the human T-cell compartment (20% ± 10% vs 9% ± 3% of hCD45+ cells). Already at an early endpoint (10 days), both CD3-targeted tLNP and ex vivo CAR-T treatment reduced splenic AML burden (0.4% ± 0.5% and 4.7% ± 8.9% of hCD45-positive cells) versus PBS (25.0% ± 8.1%) or T cells alone (8.0% ± 3.2%). In bone marrow, instead, CD3-targeted tLNP treatment reduced PDX burden to 4.0% ± 2.4% whereas ex vivo CAR-T treatment showed less clearance (30.5% ± 17.9%), indicating faster marrow clearance with tLNPs.

To mitigate CD3-associated toxicity, we evaluated CD8-targeted tLNPs. Since faster marrow clearance could reflect local CAR-T generation, we added P2A-NanoLuciferase to the CAR mRNA and performed bioluminescence imaging at 24 hours, showing robust CAR expression at hematopoietic/disease sites (104-fold above controls). Unlike CD3-targeted tLNPs, two CD8-targeted doses were well tolerated and expanded splenic CD8-positive T cells to 2.1 × 106 cells (~100-fold above tumor-only controls; 63% of CD3-positive cells), producing near-complete splenic PDX clearance (0.13% vs 2.75% of hCD45-positive cells; ~98% reduction) and partial marrow clearance. Hematopoietic effects varied by subset: primitive CD34+ CD38- HSCs, LMPPs, and MPPs were reduced in both tLNP and ex vivo CAR-T cohorts, whereas rapidly turning-over CD34+ CD38+ progenitors, including GMPs, were better preserved with tLNP than with ex vivo CAR-T, suggesting that on-target off-tumor effects of transient tLNP CAR-T may be better tolerated by fast-turnover progenitors.

Conclusions

Targeted LNPs enable efficient, transient, re-inducible generation of functional anti-FLT3 CAR-T cells in vivo, with robust antileukemic activity in humanized AML PDX models, establishing transient in vivo CAR-T generation as a re-dosable platform for AML. Studies exploring additional CAR targets, dose optimization, and powered cohorts are ongoing.
Thomas Andrew Fox, MD PhD: No relevant conflicts of interest to declare.