We identify dominant anticancer T cell clonotypes from patients who clear metastatic cancer. While some recognize HLA-restricted neoantigens, others use a single TCR to target multiple different shared tumor-associated antigens across diverse cancers. Remarkably, some clonotypes recognize many tumor types without HLA restriction. These HLA-unrestricted TCRs and their ligands overcome a central barrier in T cell immunotherapy, opening new paths toward broadly applicable treatments across patients and cancer types.

T cell homeostasis is regulated in part by the extrinsic cell death pathway mediated by fas ligand (FasL) binding to the Fas receptor. Tumor-specific T cells are triggered to undergo apoptosis following interaction with FasL in the tumor microenvironment. In preclinical studies, blockade of FasL improved T cell persistence and tumor infiltration as well as antitumor efficacy. Importantly, FasL is not required for tumor killing by T cells. FasL inhibitors are now being developed through clinical testing. Potent FasL inhibition may represent a novel immunotherapy for cancer and could augment the efficacy of immune checkpoint inhibitors and adoptive cell therapies.

Bispecific antibodies (bsAbs) have emerged as one of the most promising therapeutic modalities in cancer. We have developed a high-throughput platform to engineer and test the function of bsAbs at scale. Our approach has been successfully applied to discover novel bsAbs that target cancer cells, macrophages, or T cells. In the future, this system could be applied to test libraries of 100,000s of bsAbs for cancer or other life-threatening diseases.

This talk will outline a platform designed to support the development of NK cell-based immunotherapies. It will look at how these therapies can be applied across different diseases and where they may offer advantages. The session will also touch on opportunities to combine NK cell therapies with NK cell engagers, highlighting how these approaches may work together to expand treatment options in the future.

I will present our "tune-and-track" CAR T cell platform, engineered with affinity-tuned receptors for selective targeting of tumors overexpressing ICAM1. These CAR T cells also co-express SSTR2, enabling noninvasive imaging via PET/CT. This therapeutic approach has been evaluated in a Phase I clinical trial for patients with advanced thyroid cancer. In addition, I will discuss complementary strategies aimed at enhancing the safety and efficacy of CAR T for solid tumors.

In solid tumors cell therapy, our preclinical and clinical data to date show that CD28 and 4-1BB costimulatory domains in CAR T cells are insufficient to achieve cytotoxicity against heterogenous antigen expressing cancer cells. We exploited KITv costimulatory domain that functions through IFN-gamma pathway through an IL-2 independent mechanism. Mesothelin-targeted CD28 and KITv costimulated CAR T cells are effective against high, low, and non-antigen expressing cancer cells within the tumor.

• Identifying optimal target combinations: what approaches are most promising for nominating pairs that could meaningfully widen the therapeutic window?
  
• Choosing the right Boolean logic (AND, OR, NOT): when does each gate fit the biology, and what are the trade-offs?
• Antibody engineering for gated activity: which design levers matter most for delivering clean logic in vivo?
• Translational hurdles: how do we preclinically validate multi-target specificity and the putative therapeutic index of multispecifics?

• What is “Dark Genome” and how is it associated with diseases?
• What are the challenges of identifying druggable targets in the dark genome?
• Why is AI the best tool for developing dark-genome-related therapeutics?
• A forward-looking perspective on drug development​

This presentation will introduce peptide-centric CAR T strategies designed to recognize intracellular oncoproteins via naturally presented, unmutated peptides. Using recent findings as context, it will outline how immunopeptidome profiling and computational modelling can guide receptor design and support cross-allotype targeting. The talk will broadly discuss preclinical evidence, potential advantages for low-mutation cancers, and key considerations for advancing these approaches toward clinical translation.

This presentation will explore how artificial intelligence can support the design and optimization of T cell therapies. Current T cell therapy discovery processes rely on preclinical models that poorly recapitulate patient biology and often lead to failure in clinical translation. We present tcellMIL, an attention-based multiple instance learning model that predicts patient response from single-cell data and simulates engineered cell states. Applied to CD19-CAR T products, it improves response prediction and identifies beneficial transcriptional programs, which we validate experimentally. This framework enables clinical data-driven optimization of cellular therapies.

We have developed one-step construction of allogeneic CAR-NK cells that evade host-mediated rejection by combining selective interference of HLA-I expression, a CAR and PD-L1. CAR-NK cells overexpressing PD-L1 exhibit enhanced anti-tumor activity through upregulation of cytotoxic genes and reduced exhaustion and enhanced safety profile due to the decreased production of inflammatory cytokines involved in cytokine release syndrome. Thus, our approach represents a promising strategy in enabling “off-the-shelf” allogeneic cellular immunotherapies.

CUE-401, our lead autoimmune asset, is a novel bifunctional fusion molecule integrating a clinically validated IL-2 mutein with an attenuated TGF-ß domain to act as a master switch for Treg differentiation and immune tolerance. It expands existing Tregs, induces new iTreg subsets, and demonstrates in vivo efficacy with favorable manufacturing readiness as it advances toward clinical evaluation.

This talk will present preclinical data demonstrating that CUE-401, a novel bispecific fusion protein comprising an IL-2/TGF-ß fusion protein, selectively expands, activates, and induces regulatory T cells (Tregs)—including both natural (nTregs) and induced (iTregs)—that are functionally suppressive, phenotypically stable, and effective in ameliorating disease in murine models of autoimmunity.

Therapies based on enhancing the function and numbers of Tregs represent one of the most promising approaches to restore tolerance in many immune-mediated diseases. The aim of our study is to reprogram the effector T cells to become Treg-like phenotype, which has not been fully characterized.  We also explored novel engineering to increase the stability of Treg reprogramming. Among several constructs tested, some show enhanced Treg-like phenotype and functional profile.

This presentation reviews cell therapies in autoimmunity and the mechanistic basis for cell therapies in mediating deep cell depletion. I will discuss various targets, including CD19, BCMA, emerging targets, and dual-targeting strategies. Modalities such as autologous and allogeneic CAR T, CAR NK, in vivo CAR, and T cell engagers will be examined across different disease indications, highlighting future directions that shape the field.