In the latest step toward breaking the barrier to identifying neoantigens displayed on MHCII, Neon has generated a systems biology model that outperforms publicly available predictors.
The development brings immuno-oncology closer to personalized therapies directing CD4+ T cells against tumor neoantigens.
A growing number of studies suggest that like MHCI-responsive cytotoxic CD8+ T cells, CD4+ T cells can directly kill tumors by responding to antigens presented on MHCII. At least one reported more powerful and longer lasting responses from CD4+ cells (see "T Cell Balancing Act").
But most attempts to trigger antitumor T cell responses have focused on MHCI, in part because MHCII has a more complex peptide-binding mechanism. The imbalance has contributed to poorer accuracy for tools to predict which antigens will be displayed on MHCII, particularly for molecules outside of the most commonly studied MHCII subtype, HLA-DR.
In an Immunity paper published late Thursday, Neon Therapeutics Inc. (NASDAQ:NTGN) developed a mass spec-based tool -- mono-allelic purification with tagged allele constructs (MAPTAC) -- that captured peptide-binding motifs for over 40 MHCII alleles, including 15 previously uncharacterized ones, and covered the HLA-DR, HLA-DP and HLA-DQ subtypes.
The motif data were then used to train algorithms packaged in a modeling tool dubbed neonmhc2.
Neon researchers showed that among 12 neoantigens that were predicted to bind a well-studied MHCII allele by neonmhc2, but not by a publicly available MHCII prediction tool, eight triggered CD4+ T cell responses in culture.
Updating neonmhc2 to incorporate cellular mechanisms that modulate antigen presentation -- including peptide processing, MHCII loading, phagocytosis of tumor cells and autophagy -- led to predictive accuracies 7.4 to 61 times better than the public tool.
Michael Rooney, a lab head at Neon, and his colleagues told BioCentury the company has profiled enough HLA-DR variants to cover 95% of the U.S population and is pursuing similar population coverage for HLA-DP and HLA-DQ. The company is further mining clinical trial data to understand which MHCII-displayed peptides will elicit immune responses.
Neon plans to incorporate the tool into its RECON bioinformatics platform for predicting and selecting neoantigens for personalized immunotherapies.
Neon gained $0.42 (14%) to $3.33 Friday.
In April, Gritstone Oncology Inc. (NASDAQ:GRTS) gained 10% after reporting data showing its EDGE platform improved predictive accuracy for antigen presentation on HLA-DR molecules 20-fold over public tools. Results were presented at the American Association for Cancer Research (AACR) meeting.
Targets: MHCI - Major histocompatibility complex class I; MHCII - Major histocompatibility complex class II; HLA-DP - Major histocompatibility complex class II DP; HLA-DQ - Major histocompatibility complex class II DQ; HLA-DR - Major histocompatibility complex class II DR