First-in-human clinical trial of personalized neoantigen vaccines as early intervention in untreated patients with lymphoplasmacytic lymphoma

Abstract

Abstract Lymphoplasmacytic lymphoma (LPL) is an incurable low-grade B-cell lymphoma of the bone marrow. Despite a cumulative risk of progression, there is no approved therapy for patients in the asymptomatic phase. We conducted a first-in-human clinical trial of a novel therapeutic DNA idiotype neoantigen vaccine in nine patients with asymptomatic LPL. Treatment was well tolerated with no dose limiting toxicities. One patient achieved a minor response, and all remaining patients experienced stable disease, with median time to disease progression of 61+ months. Direct interrogation of the tumor microenvironment by single-cell transcriptome analysis revealed an unexpected dichotomous antitumor response, with significantly reduced numbers of clonal tumor mature B-cells, tracked by their unique BCR, and downregulation of genes involved in signaling pathways critical for B-cell survival post-vaccine, but no change in clonal plasma cell subpopulations. Downregulation of HLA class II molecule expression suggested intrinsic resistance by tumor plasma cell subpopulations and cell-cell interaction analyses predicted paradoxical upregulation of IGF signaling post vaccine by plasma cell, but not mature B-cell subpopulations, suggesting a potential mechanism of acquired resistance. Vaccine therapy induced dynamic changes in bone marrow T-cells, including upregulation of signaling pathways involved in T-cell activation, expansion of T-cell clonotypes, increased T-cell clonal diversity, and functional tumor antigen-specific cytokine production, with little change in co-inhibitory pathways or Treg. Vaccine therapy also globally altered cell-cell communication networks across various bone marrow cell types and was associated with reduction of protumoral signaling by myeloid cells, principally non-classical monocytes. These results suggest that this prototype neoantigen vaccine favorably perturbed the tumor immune microenvironment, resulting in reduction of clonal tumor mature B-cell, but not plasma cell subpopulations. Future strategies to improve clinical efficacy may require combinations of neoantigen vaccines with agents which specifically target LPL plasma cell subpopulations, or enable blockade of IGF-1 signaling or myeloid cell checkpoints.