AI-Enhanced ND Biosensor Automates Sensitivity Prediction to OXPHOS Inhibitors

Hepatocellular carcinoma (HCC) is the one of leading causes of cancer-related mortality worldwide. Spalt-like transcription factor 4 (SALL4) is an oncofetal protein in HCC, and high SALL4 expression level is correlated to poor prognosis. However, SALL4 lacks well-structured small molecule binding pockets, making it difficult to design targeted inhibitors. There was a therapeutic strategy based on the increased oxidative phosphorylation (OXPHOS) in SALL4^hiHCC cells. SALL4-induced high expression level of OXPHOS serves as a therapeutically targetable vulnerability in HCC that specific OXPHOS inhibitors can be applied.<br/>Here, we developed a workflow that utilized molecular beacon, a nucleic-acid-based, activatable sensor with high specificity to the target mRNA, delivered by nanodiamond, to establish an artificial intelligence (AI)-based platform for rapid evaluation of patient-specific drug sensitivity. Specifically, when the HCC cells were treated with the nanodiamond-medicated OXPHOS biosensor, high sensitivity and specificity of the senor allowed the identification of OXPHOS expression in cells. Assisted by a trained convolutional neural network, drug sensitivity of cells towards an OXPHOS inhibitor, IACS-010759, can be accurately predicted. The whole assessment could be accomplished within two days, enabling rapid and efficient clinical decision support for HCC treatment.