A Novel Lentivector for Enhanced In Vivo Delivery of T Cell Receptors for Cell Therapy for Stage IV Melanoma

PROJECT SUMMARY Project Title: A Novel Lentivector for Enhanced In Vivo Delivery of T Cell Receptors for Cell Therapy for Stage IV Melanoma Organization: GigaMune Inc. PI: David Scott Johnson, Ph.D. Melanoma patients rely on a host of therapies, for example, anti-programmed cell death protein 1 (anti-PD-1) and anti-cytotoxic T lymphocyte associate protein 4 (anti-CTLA-4) checkpoint inhibitor antibodies. Still, the 5- year survival rate for Stage IV metastatic melanoma is only 22%. Recently, the first T cell therapy (Amtagvi) for solid tumors, which comprises tumor infiltrating lymphocytes (TILs), was approved for metastatic melanoma. Amtagvi proved the utility of autologous polyclonal anti-tumor T cell receptors (TCRs) in melanoma patients. However, the $515,000 per patient cost, 22-day manufacturing time, and limited capacity raise concerns regarding the extent of patient access to this effective therapy. Engineering TCRs ex vivo into autologous patient-derived T cells (i.e., “TCR-Ts”) is also under clinical investigation for late stage melanoma with NY-ESO-1 reactive TCRs demonstrating about 50% overall response rates. However, unlike TIL therapy, these methods are “TCR-monoclonal” and therefore may be complicated by tumor recurrence due to preexisting tumor heterogeneity, human leukocyte antigen (HLA) loss of heterozygosity, or antigen loss. To address this issue, at least one company has initiated clinical studies using multiplexed, or polyclonal, TCRs. In a previous Phase I SBIR project, we identified five novel high avidity TCRs against common melanoma- associated targets. While these TCRs have potential as TCR-T cell therapies for melanoma, many barriers, including treatment costs and severely limited ex vivo manufacturing slots, restrict patient access to conventional ex vivo-engineered cell therapies. A revolutionary alternative approach would be to deliver TCR genes to T cells in vivo, sidestepping ex vivo manufacturing entirely. Lentivector approaches for in vivo chimeric antigen receptor (CAR) gene delivery are in early clinical trials with candidates from other commercial groups. However, all these clinical candidates comprise close variants of VSV-G, which may not be optimal for in vivo delivery. To discover pseudotypes more optimized for in vivo cell therapy, we previously performed a large-scale bioinformatic sequence search for novel viral G-proteins. We identified 166 G-proteins (median 49% identity to VSV-G) and then tested 9 in vitro for efficiency, specificity, and sensitivity of gene delivery to various human immune cell types. We used the results from this screen to select three G-protein candidates for a pilot GFP gene delivery study in PBMC-humanized mice, using anti-human CD3 antibody fragments for T cell tropism. One candidate G-protein performed significantly better than VSV-G. This candidate G-protein was able to deliver an anti-CD19 CAR gene to T cells in peripheral blood mononuclear cell (PBMC)-humanized mice, which subsequently cleared 100% of B cell lymphoma tumor burden in 100% of mice. In this Fast Track SBIR project, our over-arching goal is to assess the translational potential of a combination of our novel in vivo T cell-tropic delivery technology with our novel high-avidity anti-tumor TCRs, both as individual TCRs or as polyclonal pools of TCRs.