Monoclonal antibodies have become a major class of cancer therapeutics through their capacity to block oncogenic signaling, recruit immune effector mechanisms, or deliver linked payloads. Antibody-drug conjugates (ADC) couple tumor targeting to potent cytotoxic agents, whereas immune-stimulating antibody conjugates (ISAC) deliver innate immune agonists intended to remodel the tumor microenvironment and strengthen adaptive antitumor immunity.
\r\rBoth ADC and ISAC formats generally depend on antigens displayed at the surface of viable tumor cells. Yet, prominent surface targets such as HER2, TROP2, and nectin-4 are not tumor-specific and are expressed at variable levels in normal tissues, limiting selectivity and, in some settings, the therapeutic window. The requirement for surface expression also excludes a large repertoire of intracellular proteins, including oncofetal antigens whose expression is more restricted to malignant tissues and often shared among different cancers. This surface-targeting paradigm reflects the conventional assumption that an antibody must bind a viable cancer cell to promote tumor destruction. Our studies of natural antibody responses in patients with lung cancer suggest a fundamentally different mechanism in which antibodies recognize intracellular tumor antigens after their release into the tumor microenvironment and may direct therapeutic activity to sites of tumor turnover.
\r\rWe identified mAb3.4, a fully human IgG1 reconstructed from the unique immunoglobulin gene sequences of a clonally expanded tumor-infiltrating plasma cell in an untreated non-small cell lung cancer (NSCLC) specimen. mAb3.4 recognizes insulin-like growth factor 2 mRNA-binding protein 3 (IGF2BP3, also known as IMP3), an oncofetal RNA-binding protein expressed in NSCLC and multiple other solid tumors but virtually absent in most healthy adult tissues. This naturally selected human antibody provides a potential route to target tumor-derived material containing an intracellular antigen without dependence on peptide-HLA presentation.
\r\rThe specificity of mAb3.4 was established by immunoprecipitation and mass spectrometry and confirmed using complementary strategies, including an independent anti-IGF2BP3 antibody and loss of reactivity following IGF2BP3 knockdown. We also detected IGF2BP3 by Western blotting in conditioned medium from untreated NSCLC cells, demonstrating its extracellular presence without induced tumor-cell injury. Ongoing studies will determine whether extracellular IGF2BP3 is soluble or associated with extracellular vesicles. Importantly, the expansion of an IGF2BP3-specific plasma cell clone within an untreated NSCLC specimen provides evidence that this intracellular antigen was encountered by immune cells in vivo.
\r\rmAb3.4 creates two complementary conjugate opportunities. First, an ISAC could concentrate a potent innate immune agonist near tumor-infiltrating antigen-presenting cells, promoting local myeloid activation, antigen presentation, and amplification of tumor-specific T cell responses, particularly in combination with immune checkpoint blockade. Second, an ADC-like construct incorporating an extracellularly cleavable linker and a membrane-permeable payload could exploit local antigen retention and bystander killing without requiring conventional target-mediated internalization. These strategies can be evaluated using matched control conjugates and IGF2BP3-positive and negative tumor systems to establish antigen dependence, local pharmacology, and therapeutic window.
\r\rWe are seeking a development collaboration to optimize antibody format, conjugation chemistry, linker and payload selection, and to evaluate biodistribution, pharmacokinetics, safety, and ultimately antitumor activity. Preclinical validation will use a bone marrow-liver-thymus (BLT) humanized mouse model available at our center. BLT mice are immunodeficient mice reconstituted with human hematopoietic stem and progenitor cells together with human liver and thymic tissue, supporting broad multilineage human immune reconstitution and T cell development within a human thymic organoid implanted under the kidney capsule. This robust model contains human T cells, B cells, monocytes, and dendritic cells, enabling assessment of myeloid activation, antigen presentation, and adaptive antitumor immunity in vivo.
\r\rBeyond NSCLC, the broad tumor expression of IGF2BP3 creates the potential for a multi-tumor program and, more generally, for a new class of antibody conjugates directed to extracellularly accessible material derived from intracellular oncofetal antigens.
\r\rIR CU27062
Licensing Contact: Jerry Kokoshka