Abstract
Solid tissue biopsy is fundamental in guiding surgeons during intraoperative and peri-operative management of cancer patients. However, conventional histopathologic methods depend heavily on the expertise of trained pathologists, facing challenges in accuracy and efficiency. Methods: Here, we show that unbiased labeling of proteins within tissue sections using tumor-selective dyes enhances tumor-specific signals, enabling robust and accurate differentiation of tumors from normal tissues in less than 45 min. This diagnostic approach combines a tumor-selective dye labeling strategy and a three-dimensional (3D) histological electrophoresis separation strategy to visualize protein differences between tissues and exclude off-target interference. Results: We successfully diagnose and delineate malignant tissue from frozen and fresh surgical specimens from 34 patients across six types of cancer (mean AUC = 0.93). Furthermore, we apply this method to distinguish different histological characteristics in liver cancer surgical specimens, as well as identify and quantify the degree of inflammation in tumor-surrounding tissues. Conclusion: This rapid, accurate, unbiased, and marker-free approach may enhance intraoperative detection of multiple types of tumor specimens.