Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster Display session

103P - Using AI to break new ground in oncological drug discovery: Rapid identification of novel targets and polypharmacological compounds for effective liposarcoma treatment

Date

15 Mar 2024

Session

Poster Display session

Presenters

Nik Subramanian

Citation

Annals of Oncology (2024) 9 (suppl_2): 1-32. 10.1016/esmoop/esmoop102441

Authors

N. Subramanian1, N. Maignan1, G. Tieo2, D. Papazian1, J. Dawe1, M. Georges1, C. Hudson Queiroz de Souza1, R. Ravinder1, S. Martin1, M. Jeitany2

Author affiliations

  • 1 Kantify, Kantify, 1050 - BRUSSELS/BE
  • 2 School Of Biological Sciences, NTU - Nanyang Technological University, 639798 - Singapore/SG

Resources

This content is available to ESMO members and event participants.

Abstract 103P

Background

Liposarcoma (LPS), a rare and aggressive soft tissue cancer, lacks effective treatment options. Traditional drug discovery methods have been largely ineffective against LPS. Our study represents a paradigm shift in cancer treatment discovery, employing advanced AI algorithms to identify novel druggable targets and predict compounds with polypharmacological potential for LPS therapy.

Methods

Utilizing Kantify's proprietary AI-based drug discovery platform, Zeptomics, we embarked on a twofold approach: first, identifying novel targets shared across LPS subtypes based on published drug screening data, and second, conducting an in-silico screen of approximately 10,000 drugs and natural compounds that haven't previously been considered for LPS. This approach prioritized compounds predicted to modulate simultaneously multiple novel predicted targets, addressing cancer cell compensation mechanisms. We validated these findings through in-vitro screenings of nine leading compounds on dedifferentiated and myxoid LPS cell lines, coupled with CRISPR-Cas9 depletion of the identified targets.

Results

Over the course of just three months, we ran the in-silico screening, followed by in-vitro validation. The validation resulted in an unprecedented hit rate (∼77%) with seven compounds demonstrating significant efficacy in reducing LPS cell viability, with IC50 ranging from 1.5uM to nM scales. This method's efficiency far exceeds traditional drug repurposing rates. Among the effective compounds, two are oncology drugs, three are prescription medications with minimal side effects, one is a natural compound, and one is a veterinary drug. Furthermore, the CRISPR-Cas9 gene depletion substantiated the importance of seven novel targets for LPS survival.

Conclusions

This study not only unveils novel targets for LPS treatment but also exemplifies the power of AI in predicting multi-targeted therapeutic compounds. This groundbreaking approach, transcending traditional drug discovery methods, holds immense potential for broadening cancer treatment strategies and is poised to make a significant impact in oncological drug development and repurposing approaches.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Kantify.

Disclosure

N. Subramanian, S. Martin: Financial Interests, Personal and Institutional, Ownership Interest, financial interest in Kantify (either through employment, board seats, or stock ownership): Kantify. N. Maignan, D. Papazian, J. Dawe, M. Georges, C. Hudson Queiroz de Souza, R. Ravinder: Financial Interests, Personal and Institutional, Full or part-time Employment, financial interest in Kantify (either through employment, board seats, or stock ownership): Kantify. All other authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.