Mutated neo-antigens as targets for individualized cancer immunotherapy
Date issued
Authors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
License
Abstract
Carcinogenesis is largely driven by somatic gene mutations. Accumulating evidence suggests that a significant subset of mutations result in neo-epitopes recognized by tumor-specific T cells and thus may constitute the Achilles' heel of malignant cells. T cells directed against mutations have been shown to have a key role in spontaneous immune responses against cancer and in the clinical efficacy of potent cancer immunotherapy modalities, such as adoptive transfer of autologous tumor infiltrating lymphocytes and immune checkpoint inhibitors. Whereas these findings strengthen the idea of a prominent role of neo-epitopes in tumor rejection, the systematic therapeutic exploitation of mutations was hampered until recently by the uniqueness of the repertoire of mutations (“the mutanome”) in every patient’s tumor. Our group pioneered to set up a process for an individualized immunotherapy approach to target the full spectrum of a patient’s personal tumor-specific mutations by combination of exome and transcriptome sequencing, bioinformatic target identification and selection followed by RNA based tumor vaccination.
In this thesis, it was shown that tumor mutations, in particular single nucleotide variations (SNVs), insertions, deletions (indels) and gene fusions are frequently immunogenic. Moreover, it was demonstrated for the first time that SNVs are predominantly recognized by CD4+ T cells. Neo-epitopes can be targeted by customized RNA-based monotope or poly-neo-epitope vaccines with substantial therapeutic effects in mouse tumor models. Notably, CD4 neo-epitope vaccination was shown to positively affect the tumor microenvironment of the CT26 colon carcinoma by decreasing the ratio of FoxP3+ regulatory T cells to CD4+ T cells and augmenting CD4+ and CD8+ T-cell infiltration. In addition, anti-tumoral efficacy of neo-epitope specific CD4+ T cells in CT26 colon carcinoma tumors was shown to critically depend on CD8+ T cells and CD40L signaling. CD40-CD40L interaction between activated CD4+ T cells and dendritic cells (DCs) mediates the upregulation of costimulatory molecules on DCs which in turn supports the priming of naive, tumor specific CD8+ T cells. A meta-analysis revealed that immunogenic SNVs and indels have a significantly better binding prediction for MHC class II than their non-immunogenic counterparts. Vaccination against mutations selected for favorable MHC class II binding prediction and abundant mutated mRNA expression resulted in tumor-control without prior immunogenicity testing.
The preclinical data shown here for three different mouse tumor models established the feasibility of individualized cancer vaccines and raised hopes that this concept will be effective in humans as well. This study paved the way and directly influenced the design of clinical studies in melanoma and triple-negative breast cancer patients.