ProteinQure Secures $11M to Advance AI-Designed Peptide-Drug Conjugate to Clinical Trials

ProteinQure, a biotech company, is utilizing artificial intelligence (AI) and machine learning to further develop antibody-drug conjugates (ADCs). The company recently secured $11 million in series A funding to initiate a first-in-human trial of its pioneering peptide-drug conjugate (PDC). This funding round, which increased ProteinQure's total funding to $16 million, was led by Tom Williams of Heron Rock Fund with contributions from Golden Ventures and Kensington Capital.

Lucas Siow, the Chief Executive of ProteinQure, stated, “We are progressing what we believe to be the first AI-designed peptide therapeutic into clinical trials. We're prepared to validate clinically how our advanced peptide design platform can target previously inaccessible areas.” The primary candidate from ProteinQure, known as PQ203, is a PDC targeting the sortilin receptor, aimed at treating tumors resistant to the inhibition of topoisomerase. Topoisomerase is a protein targeted by various ADCs, such as Trodelvy (sacituzumab govitecan-hziy) by Gilead Sciences. The Phase I trial is anticipated to start recruiting patients with triple-negative breast cancer across the United States and Canada in the third quarter, targeting an enrollment of 70 to 100 participants.

ProteinQure notes that PDCs offer several benefits over ADCs, particularly in terms of size. Due to their smaller size, peptide-based therapies may penetrate tissues more effectively, delivering therapeutic payloads more efficiently to the tumor microenvironment. Additionally, the company believes that PDCs might present a more tolerable safety profile due to their reduced immunogenicity.

The development of PQ203 was accomplished using ProteinQure’s ProteinStudio platform. This platform integrates custom-trained AI models with molecular simulations and internal laboratory experiments to expedite the development cycle from conceptual design to a clinically relevant PDC. ProteinQure has assembled a library of machine learning models based on sequence and structure. These models have been trained on data from both public and proprietary research studies and are designed to support thousands of synthetic or non-canonical amino acids.

Furthermore, ProteinQure employs 3D modeling and computational biology to examine and optimize peptides and linkers, predicting essential properties of therapeutic candidates such as affinity, solubility, specificity, and functional outcomes. Heron Rock's Williams remarked, “ProteinQure has demonstrated genuine platform capabilities for designing non-canonical peptide therapeutics with outstanding specificity and drug-like qualities, progressing from concept to clinic with impressive capital efficiency.”

This strategic approach positions ProteinQure as a leader in the innovative use of AI and machine learning for the development of peptide-based therapeutics. The company's efforts to bring AI-designed peptide therapeutics into clinical settings underscore its commitment to transforming treatment paradigms and addressing unmet needs in oncology. By advancing its lead candidate, PQ203, into human trials, ProteinQure aims to provide new therapeutic options for patients with difficult-to-treat cancers, reflecting the potential of its cutting-edge platform to revolutionize drug development.