AI Drug

DeepMatcher™

"Syntekabio pursues innovation in various areas of independent AI drug development."

DeepMatcher™, AI platform derives active substances through “search for available hundreds of millions of compound libraries”. In addition, it enables accurate and efficient discovery of new drug candidates as a platform for discovery of lead substances through optimization, secondary target and off-target analysis, and development of drug resistance biomarkers.

NEO-ARS™

“Contribution to development of immuno-oncology drugs through AI-based discovery of new antigens.”

NEO-ARS is an automatic report system for predicting neoantigens. It is known that hundreds of gene mutations appear in the course of cancer development and proliferation, and the frequency of occurrence and the form of the mutation sequence vary depending on the cancer type and patient. These genetic mutations produce abnormal protein fragments called neoantigens (tumor specific antigens), which do not exist in normal cells but are expressed only in cancer cells.

When neoantigens are expressed in cancer cells, they appear in a form bound to major histocompatibility complex type I (MHC-1, Major Histocompatibility Complex-I) on the surface of cancer cells, and as a result of the interaction between antigen-presenting cells and cytotoxic T cells, T cells selectively recognize cancer cells with neoantigens and induce anticancer immunity.

As such, new antigens activate the unique immune system of cancer patients and attack only cancer cells, making it safe and possible to develop customized treatments that reflect the unique characteristics of each patient's genetic mutation, suggesting a new paradigm for anticancer treatment.

Neoantigen-based immunotherapy can be developed in the form of an anticancer vaccine that directly administers the new antigen or a cell therapy that administers the immune cells of a patient activated outside the body with the neoantigen. It is expected to greatly improve the response rate.

However, high technology is required to accurately and quickly find different neoantigens for each patient.

Syntekabio developed NEO-ARS to predict neoantigens with high accuracy from tumor and blood genome data of cancer patients. Unlike algorithms that predict neoantigens based on two-dimensional amino acid sequence information, NEO-ARS improves accuracy by predicting immunogenicity on three-dimensional protein structures.

NEOscan performance evaluation: Using NEOscan from TCGA/cosmic oncogenome data, a common neoantigen candidate was derived and experimental validation was performed.(Left) MHC-neoantigen peptide in vitro binding assay result. For each of the three HLA-A types, the binding of neoantigen candidates predicted by NEOscan was experimentally evaluated. Confirmed that most neoantigen candidates actually bind to MHC. Orange dotted line: Criteria for determining whether to combine. If it exceeds the reference value, it is interpreted as combining. (Right) The immunogenicity of neoantigen candidates derived from NEOscan was evaluated by ex vivo ELIspot experiments using healthy PBMC samples. The graph shows the results for HLA-A*33:03 type, 11 neoantigen candidates as an example. Higher SFC means more T cells that are activated by antigen and secrete IFNγ (* indicates a statistically significantly higher SFC level compared to DMSO control).

Currently, we are continuously upgrading NEO-ARS performance by discovering new antigens in cancer patients and evaluating the immunogenicity of predicted neoantigens (T cell activation: IFNγ secretion) in connection with domestic hospitals. For development, we are seeking collaboration with related drug developers.

GBL-ARS™

"Syntekabio develops Genomic Bigdata-based Biomarker Discovery Technology"

GBL-ARS is an automatic report system for genetic biomarker labeling. Using drug reactivity and genomic data obtained from early-stage clinical trials, we discover biomarkers from the perspective of genetic mutations to predict drug efficacy. The discovered biomarkers are applied to patient selection to greatly improve the success rate of late clinical trials, and it also enables the expansion of indications based on biomarkers, repositioning of drugs, and recovery of clinically failed drugs.