Optimizing Anti-NMDAR Antibodies for Triple-Negative Breast Cancer Therapy

PROJECT SUMMARY/ABSTRACT Triple-negative breast cancer (TNBC), which lacks hormone receptor and HER2 expression, is associated with a high risk of metastatic recurrence and high mortality, necessitating novel therapeutic strategies. A high percentage of TNBC (>70%) exhibit upregulation of N-methyl-D-aspartate (NMDA) receptor (NMDAR), which are normally exclusively expressed in the central nervous system, and high NMDAR expression is associated with perineural invasion and poor prognosis. Despite the potential of anti-NMDAR antibodies to kill NMDAR- positive cells, cancer patients with spontaneous anti-NMDAR production often experience severe neurological toxicity manifesting as psychosis and seizures (termed anti-NMDAR encephalitis). However, it remains unknown whether it is possible to uncouple the anti-tumor effects of anti-NMDAR encephalitis from their toxicity, thereby establishing a safe and effective therapeutic strategy for TNBC. A key innovation of this proposal is the use of single-particle cryo-EM to screen for antibody-induced structural alterations in NMDAR, which are necessary and sufficient for neurological toxicity. We propose a workflow for the identification, characterization, and validation of anti-NMDAR antibody therapies. First, we will establish an orthotopic mouse model based on inoculation of 4T1 (triple-negative breast cancer) cells, engineered to have doxycycline-inducible expression of NMDAR in the mammary fat pad (Aim 1). The NMDAR-positive models will first be used to immunize mice with NMDAR and isolate NMDAR-reactive antibodies by single-cell RNA sequencing of intratumoral B cells. Recombinant expression of candidate NMDAR-reactive antibodies will be used to generate cryo-EM models of antibody-receptor complexes, thereby screening for antibody-induced structural changes (Aim 2). Finally, candidate anti-NMDAR antibodies with low predicted pathogenicity due to minimal structural effects will be used to treat mice with NMDAR-positive orthotopic tumors to elucidate maximal therapeutic efficacy and minimal toxicity (Aim 3). In preliminary data generated for this proposal, we have performed a proof-of-principle experimental sequence demonstrating that doxycycline treatment in vivo is sufficient to induce NMDAR expression in 4T1-NMDAR tumors and elevated anti-NMDAR antibody titers in plasma. scRNA-seq of intratumoral B cells, was enriched for NMDAR binders and we have identified multiple sequences with high- affinity for NMDAR, two of which were characterized by cryo-EM displaying distinct epitopes. Altogether, this proposal will leverage anti-NMDAR antibodies as an effective and non-toxic therapy for NMDAR-positive TNBC, addressing a key unmet need for the 40,000 patients per year diagnosed with TNBC in the USA.