Back

Reprogramming BCMA-Targeted CAR-T Cells through γ-Secretase Modulation Blocks Antigen Shedding and Extends CAR-T Longevity

Sharma, S.; Chauhan, V.; Bhat, Z. A.; Aich, J.; Madaan, D.; Gupta, S.; Rashmi, D.; Alam, R.; Dave, D.; Patel, P.; Chaudhary, N.; Singh, D. K.; Singh, V. P.; Husain, M.; Srivastava, A. K.; Mabalirajan, U.; Pendharkar, D.; Murugaiyan, N.; Ramalingam, S. p.; Kharya, G.; Verma, A. K.; Ahmad, T.

2026-01-23 immunology
10.64898/2026.01.20.700575 bioRxiv
Show abstract

B-cell maturation antigen (BCMA) shedding by {gamma}-secretase generates soluble BCMA (sBCMA), which diminishes membrane antigen density, and limits the durability of BCMA-directed immunotherapies in multiple myeloma (MM). Here, using AI-driven diffusion modeling, we identify peptide inhibitors that selectively block {gamma}-secretase-mediated BCMA cleavage, stabilizing membrane-bound BCMA (mBCMA) without compromising cellular viability. The lead peptide, P5, suppresses sBCMA and restores mBCMA in vitro and in xenograft models. To create sustained, cell-intrinsic inhibition, we engineered Shedding Deterrent Locked-in (SHEDLOCK) CAR-T cells that locally secrete {gamma}-secretase-modulatory peptides. SHEDLOCK-1 CAR-T cells secreting P5 exhibits enhanced cytotoxicity and persistence relative to conventional CAR-T cells. Building on this, SHEDLOCK-2 CAR-T cells were generated, which secrete a modified, naturally derived peptide (nxP) that directly engages the {gamma}-secretase catalytic site, concurrently preventing BCMA shedding, and surprisingly enhancing CAR-T cell longevity by preserving telomere integrity and metabolic fitness. In MM xenograft and patient-derived models, SHEDLOCK-2 CAR-T cells demonstrate durable antitumor activity with a favorable safety profile. Together, these findings establish SHEDLOCK as a next-generation CAR-T platform and provide a strong preclinical foundation for Phase I clinical evaluation. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=128 SRC="FIGDIR/small/700575v1_ufig1.gif" ALT="Figure 1"> View larger version (69K): org.highwire.dtl.DTLVardef@3cd38aorg.highwire.dtl.DTLVardef@164167borg.highwire.dtl.DTLVardef@1ffed21org.highwire.dtl.DTLVardef@1294d06_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.