Machine learning-driven identification of DLL3 as a molecular target and development of a DLL3-binding cyclic peptide for glioblastoma

1

DLL3 was identified as a promising therapeutic target in glioblastoma due to its high expression in tumors and low presence in normal brain tissue.
2

An integrated approach combining machine learning, transcriptomic analysis, and structural modeling was used to identify DLL3 and develop a cyclic peptide ligand.
3

The cyclic peptide candidate, IMP-3, showed favorable binding to DLL3 and selective accumulation in GBM cells compared to normal astrocytes.
4

The study highlights the potential of targeted radionuclide therapy in GBM, leveraging DLL3 as a specific membrane target for therapeutic delivery.
5

Cyclic peptides offer advantages such as improved stability and binding affinity, making them suitable for targeted therapies in challenging environments like the brain.

Frontiers In Oncology

by Dan Xu, Daqing Huang, Zhijie Li, Yan Yan, Zhencun Cui, Zhongfang Zhao, Xiaoju Chen, Maolong Chen, Xiongxiong Liu, Zhaobo Zhou, Qianxi Ni, Taofeng Zhang, Hui Wang, Qi Zeng, Xi’an Xiong, Bin Liu
July 14, 2026

1