Cancer cells vis
Credit: koto_feja/Getty Images

Year-old biotech company Meliora Therapeutics announced on Thursday it has raised $11 million in a seed financing round that will fund further development of its machine learning platform that will seek to catalog and detail the exact mechanisms of action (MOA) of existing oncology drugs. The financing round was led by HOF Capital and ZhenFund, with participation from Obvious Ventures, Village Global, BrightEdge, Pebblebed, Axial, Olive Capital, and individual investors Michael Polansky, Wes Sterman, among others.

According to a press release announcing the financing, Meliora aims to improve the approval rate of cancer drugs in development—currently only about three percent—by understanding of their true mechanisms of action via application of its machine-learning platforms to more precisely determine each drug’s mechanisms and the resulting impact on cancer biology.

“Modern cancer drug development is often not mechanism driven, leading to clinical development failures down the line,” said company CEO and co-founder David Li. “Meliora is using a novel molecular fingerprint matching method to determine the true mechanism of action of precision cancer therapies. By using machine learning to connect MOA with specific therapeutic molecules, we will increase the overall probability of success. We’re thrilled to have the support of this top-tier investor syndicate, and the funds will be used to expand our team and accelerate development of our proprietary computational platform.”

Scientific co-founder Jason Sheltzer, PhD, is a faculty member in Genetics and Surgical Oncology at the Yale School of Medicine, where his laboratory studies cancer genomics and tumor evolution. His June, 2019 paper “Off-target toxicity is a common mechanism of action of cancer drugs undergoing clinical trials” published in Science Translational Medicine brought this problem to light. Sheltzer’s research studied a set of drugs that were in various stages of clinical testing and found, according to the paper’s abstract “the proteins ostensibly targeted by these drugs are nonessential for cancer cell proliferation. Moreover, the efficacy of each drug that we tested was unaffected by the loss of its putative target, indicating that these compounds kill cells via off-target effects.”

Meliora’s platform is called AnchorOmics, which the company says is able to capture drug activity at the molecular, cellular, and organismal level and then transforms these data into a standardized digital format. It refers to this as a digital fingerprint that captures the true impact of a drug on cellular function as indicated by the transcriptome, methylome, proteome, morphological state, and more.

“We use advanced machine learning techniques to generate each perturbation’s fingerprint, and locate that fingerprint in our atlas, thus mapping an anti-cancer drug’s mechanisms of action. Fundamentally, AnchorOmics allows Meliora to combine insights from a wide range of modalities into a single vector space atlas,” according to the company’s website.

As Sheltzer sees it, this deeper set of information should allow for more effective cancer drug development.

“Meliora’s computational platform allows companies to identify the true mechanism of action in small molecule compounds, and uncover mischaracterized compounds in the process,” he said in a press release. “By expanding the therapeutic window of these molecules with known safety profiles, the Meliora team has built a flywheel of precision medicine oncology assets with a higher potential for clinical success by understanding how the drug really works.”

This site uses Akismet to reduce spam. Learn how your comment data is processed.