High-sensitivity photoelectric sensing of apatinib for recurrent hepatocellular carcinoma in TACE.

Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality, with transcatheter arterial chemoembolization (TACE) being a widely used treatment for intermediate-stage patients. However, TACE is often associated with high recurrence rates, necessitating reliable methods for early detection and prediction of tumor recurrence. Current diagnostic tools rely heavily on imaging techniques and biomarkers, which often lack sensitivity and specificity, particularly in the early stages of disease progression. To address these challenges, we developed an innovative integrated platform combining a high-sensitivity photoelectric sensor based on Bi/h-Bi2Te3 heterostructures with machine learning models for the detection and prediction of HCC recurrence following TACE. The photoelectric sensor exhibits exceptional sensitivity and stability, enabling precise detection of apatinib, a key therapeutic agent used in HCC management. The machine learning models demonstrated superior performance in predicting HCC recurrence, with an area under the receiver operating characteristic curve (AUC) of 0.840, accuracy of 78.0%, sensitivity of 72.4%, and specificity of 80.3% on the independent test set. Overall, this work establishes a practical materials-informatics framework that links sensitive therapeutic drug monitoring (LOD = 0.08 µM) with individualized recurrence risk stratification after TACE, providing a promising route toward precision management of HCC.