Radical surgery remains the primary treatment for gastric cancer, but intraoperative tumor margin assessment relies on surgeons' visual inspection, limiting accuracy. There is thus an urgent clinical need for real-time visualisation of tumour margins.

In recent years, near-infrared (NIR) fluorescence imaging has emerged as a critical tool for precision tumor resection. However, existing probes like indocyanine green (ICG) lack tumor-targeting specificity. Ferritin (FTn), with its unique nanocage structure, excellent biosafety, and well-defined in vivo behavior, presents an attractive platform for targeted molecular probes.

Yet, translational challenges persist, including animal model limitations and clinical validation bottlenecks. To address this, our study employs freshly resected human gastric tissue in an ex vivo perfusion system, simulating the circulatory dynamics of the humanized ferritin-based probe FTn-ICG in vivo. Using a prospective clinical sample cohort, we aim to validate its diagnostic efficacy in delineating gastric cancer margins, ultimately overcoming the critical barrier of precise tumor boundary identification.

Inclusion Criteria:

• Pathologically or cytologically confirmed gastric or gastroesophageal junction adenocarcinoma eligible for radical resection, with histologically verified predominant adenocarcinoma component; Age ≥ 18 years; No gender restriction ; Voluntary participation with written informed consent.

Exclusion Criteria:

• Patients who have received neoadjuvant therapy; Patients deemed ineligible for participation by the investigator's assessment.

The study population was recruited from the Nanfang Hospital of Southern Medical University in Guangzhou, China.