Advancements in Therapeutic Leads for HCC and PDAC: A Computational and Preclinical Approach

The survival rates for hepatocellular carcinoma (HCC) and pancreatic ductal adenocarcinoma (PDAC) are notably low, with only 18% and 9% respectively surviving beyond five years. Despite the FDA's approval of multi-kinase inhibitors like cabozantinib, regorafenib, lenvatinib, and the conditional approval of the immune checkpoint inhibitor nivolumab for HCC, the prognosis remains grim. PDAC faces an even more urgent need for innovative treatments. Our research introduces a cutting-edge computational drug discovery tool that can swiftly identify potential drug candidates with unique mechanisms of action and validate them through preclinical studies.

The twoXAR platform, powered by artificial intelligence, amalgamates a variety of patient-derived datasets to create comprehensive and unbiased models of human diseases. It employs a range of proprietary algorithms and deep learning techniques to identify intricate connections between disease biology and biomedical data. This integration with a library of existing drugs allows for the identification of new, valuable drug candidates.

Using the twoXAR platform, we developed virtual disease models for HCC and PDAC and identified 10 and 11 molecules, respectively, predicted to be effective against these diseases. These novel drug candidates were not previously considered for clinical therapy for HCC or PDAC. Notably, TXR-311 and TXR-312 were validated for HCC, and TXR-405 and TXR-411 for PDAC, using in vitro assays with tumor cell lines. TXR-311, in particular, demonstrated a significant reduction in IC50 values compared to sorafenib and showed high selectivity for HCC tumor cells over healthy hepatocytes. In vivo studies with patient-derived xenograft (PDX) models further confirmed TXR-311's efficacy and tolerability, making it a promising candidate for further development as a potential HCC treatment.

The study was presented by Isaac Hakim and colleagues at the Annual Meeting of the American Association for Cancer Research in 2020, with the abstract published in the Cancer Research journal.