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

Paper Details

Abstracts, authors, affiliations, and disclosures.

Sunday, December 13, 2026
04:30 p.m. - 06:00 p.m.
MCCNO - La Nouvelle C
Treatment Considerations, Combination Therapy, Drug Development, Biological Therapies, Antibody Therapy, Bispecific Antibody Therapy, Chimeric Antigen Receptor, (CAR)-T Cell Therapies, Immunotherapy, Biological Processes, Immune Mechanism, Diseases, Lymphoid Malignancies, Plasma Cell Disorders, Research, Translational Research, Study Population, Human, Animal Model
PAPER

685 Self-driving tanks: Engineering bone marrow-resident, T-cell engager-secreting CAR T-cells for multiple myeloma

Matthew Ho, MBBS, PhD, Owen Koucky2, Robert Bartoszek3, Angel Ramirez-Fernandez4, Patrizia Porazzi, PhD5, Antonio Imparato, MD6, Luca Paruzzo, MD7, Miriam Velasco-Sidro8, Shannon E Mcgettigan, PhD9, Fei Miao10, Wei Liu, PhD11, Janna Minehart12, Gregory Chen, MD PhD13, Inyoung Jung, PhD14, Malihatosadat Gholamzadeh, PhD15, Nico Gonzalez16, Erik Williams17, Julia Han Noll, PhD18, Ankita Jain19, Devin Dersh, PhD20, Aastha Pal21, Shan Liu, PhD22, Ranjani Ramasubramanian, BS23, Alberto Carturan, MD24, Neil Sheppard, PhD25, John Scholler26, David Allman, PhD27, Marco Ruella, MD28, Zainul Hasanali, MD, PhD29, Caroline Diorio, MD30, Adam Waxman, MD, MS31, Shivani Kapur, MD32, Sandra Susanibar-Adaniya, MD, MS33, Edward A. Stadtmauer, MD34, Dan T. Vogl, MD, MSCE35, Adam D. Cohen, MD36, Don Siegel, MD PhD37, Alfred L. Garfall, MD38, Joseph Fraietta, PhD39
1 University of Pennsylvania, Philadelphia, United States
2 University of Pennsylvania, Philadelphia, United States
3 University of Pennsylvania, Philadelphia, United States
4 University of Pennsylvania, Philadelphia, United States
5 University of Pennsylvania, Philadelphia, United States
6 University of Pennsylvania, Philadelphia, United States
7 University of Pennsylvania, Philadelphia, United States
8 CNIO-HMRIB Cancer Immunotherapy Clinical Research Unit, Spanish National Cancer Research Centre (CNIO), Madrid, Spain
9 University of Pennsylvania, Philadelphia, United States
10 University of Pennsylvania, Philadelphia, United States
11 University of Pennsylvania, Philadelphia, United States
12 University of Pennsylvania, Philadelphia, United States
13 University of Pennsylvania, Philadelphia, United States
14 University of Pennsylvania, Philadelphia, United States
15 University of Pennsylvania, Philadelphia, United States
16 University of Pennsylvania, Philadelphia, United States
17 University of Pennsylvania, Philadelphia, United States
18 University of Pennsylvania, Philadelphia, United States
19 University of Pennsylvania, Philadelphia, United States
20 University of Pennsylvania, Philadelphia, United States
21 University of Pennsylvania, Philadelphia, United States
22 University of Pennsylvania, Philadelphia, United States
23 University of Pennsylvania, Philadelphia, United States
24 University of Pennsylvania, Philadelphia, United States
25 University of Pennsylvania, Philadelphia, United States
26 University of Pennsylvania, Philadelphia, United States
27 University of Pennsylvania, Philadelphia, United States
28 University of Pennsylvania, Philadelphia, United States
29 Division of Hematology and Oncology, Department of Medicine, Indiana University School of Medicine, Indianapolis, United States
30 Children's Hospital of Philadelphia, Philadelphia, United States
31 University of Pennsylvania, Philadelphia, United States
32 University of Pennsylvania, Philadelphia, United States
33 University of Pennsylvania, Philadelphia, United States
34 University of Pennsylvania, Philadelphia, United States
35 University of Pennsylvania, Philadelphia, United States
36 University of Pennsylvania, Philadelphia, United States
37 University of Pennsylvania, Philadelphia, United States
38 University of Pennsylvania, Philadelphia, United States
39 University of Pennsylvania, Philadelphia, United States

Background: BCMA CAR T-cell therapy is not yet curative in most patients with multiple myeloma (MM). Correlative analyses from a clinical trial at Penn demonstrated consistently higher CAR transgene levels in peripheral blood (PB) than in bone marrow (BM) at multiple timepoints (PMID: 36413381), suggesting that insufficient BM residency may limit durable immune surveillance at the primary site of disease. However, systemically increasing CAR T-cell potency may increase toxicity because CAR T-cells traffic to normal tissues and can cause severe on-target, off-tumor effects (PMID: 41540109). To address these limitations, we translated complementary mechanisms of leukocyte homing and retention into therapeutic design. Gain-of-function CXCR4WHIM mutations, which cause WHIM syndrome, increase responsiveness to the CXCR4 ligand CXCL12 and promote BM homing of leukocytes, whereas fingolimod, an S1PR1functional antagonist used to treat multiple sclerosis, provides clinical proof that S1PR1 inhibition can restrict lymphocyte egress.

Methods: Accordingly, we developed self-driving TANKs (T-cells Armored for eNgager-augmented Killing), which integrate the novel BCMA binder VHH20A, CXCR4WHIM-mediated BM homing, S1PR1KO-mediated BM retention, and localized secretion of a GPRC5DxCD3 T-cell engager (TCE) to concentrate cytotoxic activity within the BM while enabling dual-antigen targeting and bystander T-cell recruitment.

Results: Using an isogenic CAR backbone and standardized manufacturing protocol, we benchmarked VHH20A against the BCMA binders used in ide-cel, cilta-cel, and anito-cel. Across BCMAWT and 4 clinically relevant BCMAescape variants, VHH20A demonstrated 2- to 5-fold greater in vitro cytotoxicity than all three clinical binders, with minimal cytotoxicity against BCMA-negative cells. In chronic antigen-restimulation assays, VHH20A also achieved 3- to 44-fold greater peak expansion.

To optimize BM homing, we evaluated 4 CXCR4WHIM variants (R334X, S338X, S339Ffs*6, L329Qfs*13). CXCR4L329Q exhibited the greatest CXCL12 sensitivity (EC50, 95.5 vs 467.7pM for CXCR4WT) and produced the highest CXCL12-directed migration (~4.5-fold greater than CXCR4WT) in in vitro transwell migration assays. In competitive in vivo homing studies, differentially labeled CXCR4WT and CXCR4WHIM CAR T-cell populations were co-infused at equal ratios into the same recipient mice. CXCR4L329Q showed the greatest BM enrichment, comprising 68.7% of recovered BM CAR T-cells, compared with 23.4%, 5.5%, 2.3%, and <0.1% for S339F, R334X, S338X, and WT, respectively.

To further enhance BM residency, we performed CRISPR-mediated S1PR1KO, consistently achieving >90% predicted knockout efficiency by ICE across 5 independent donors. Beyond reducing S1P-mediated chemotaxis by ~3-fold, S1PR1KO increased surface CXCR4 expression and further enhanced CXCL12-mediated chemotaxis by approximately 2-fold. In vivo, combining CXCR4L329Q with S1PR1KO produced the greatest BM enrichment relative to either modification alone or unmodified CAR T cells, supporting complementary contributions of BM homing and retention. Importantly, the dual modifications did not compromise anti-MM cytotoxicity. At a dose of 400,000 cells, dual-modified CAR T-cells produced ~3-fold lower median intramedullary tumor burden than dose-matched conventional VHH20A CAR T-cells on days 20 and 27 after infusion.

Finally, to enhance anti-MM cytotoxicity and enable dual-antigen targeting, we engineered self-driving TANKs to locally secrete a short-acting, Fc-free GPRC5DxCD3 TCE to spatially restrict T-cell redirection to the BM. TCE secretion significantly enhanced anti-MM cytotoxicity compared with conventional BCMA CAR T-cells (cytotoxicity AUC, 227 vs 136; P<0.001). In bystander T-cell killing assays, GPRC5DxCD3 TCE-secreting self-driving TANKKs produced ~4- to 6-fold greater bystander-mediated cytotoxicity than TANKs secreting a nontargeting control TCE.

Conclusion: Self-driving TANKs support spatial engineering as a design principle for improving the therapeutic index of CAR T-cell therapy. Because spatial localization is orthogonal to antigen recognition, this modular platform may be adaptable across targets and payloads, with potential applications beyond MM in other BM-resident malignancies and lymphoma by exploiting CXCL12-rich stromal niches in the BM and lymph nodes to facilitate self-driving TANK localization.

Matthew Ho, MBBS, PhD: No relevant conflicts of interest to declare.