Abstract
Monitoring biomarkers offers insights for early disease (e.g., cancer, chronic diseases) screening, treatment guidance and response evaluation. To tackle challenges in precise biomarker detection in complex blood samples, we propose a metal-organic framework (MOF)-enhanced anti-fouling immunoprobe (MAIP). This sensing platform integrates several key features, including an anti-fouling surface, fully oriented antibody immobilization, and an enhanced electron transfer layer. This platform demonstrates remarkable sensitivity and specificity, capable of detecting trace concentrations of cancer markers, ranging from pg/mL to ng/mL. Comprehensive analysis of clinical serum samples, combined with machine learning (ML) evaluations, underscores the platform's reliability and effectiveness in cancer biomarker detection. Using a combination of three biomarkers, the algorithm demonstrated satisfactory classification performance with an accuracy of 100% on the test set across 64 serum samples. Moreover, when evaluated using 10-fold cross-validation, the optimized model maintained a robust accuracy of 0.955 ± 0.09, reinforcing its strong stability and good generalization ability across different data partitions. The MAIP-based platform developed here offers swift analysis, high accuracy, and a simplified testing process. This makes it a promising approach for directly profiling cancer markers from blood, thereby facilitating timely and effective patient care.