Objective: To investigate the therapeutic effect of a dual-network hydrogel-microneedle (MN) on inhibiting oral squamous cell carcinoma (OSCC) recurrence and promoting tissue repair, thereby providing a foundation for the clinical treatment of OSCC. Methods: A Schiff base hydrogelformed between chitosan (CS) and poly(ethylene glycol) dibenzaldehyde (PEG-CHO), combined witha coordination-crosslinked hydrogel formed between sodium alginate (SA) and Cu²+, were jointly integrated to form the CS/PEG-CHO/SA/Cu²+ (CPSC) dual-network hydrogel base. The hydrogel formulation was optimized based on characterization by scanning electron microscopy, Fourier transform infrared spectroscopy, rheology, and swelling tests. Subsequently, 5-fluorouracil (5-FU) and gold nanorod (GNR) were incorporated into the CPSC hydrogel to fabricate an MN loaded with 5-FU/GNR (FG). The mechanical strength and dissolution behavior of FG MNs were characterized, and their photothermal performance was evaluated under near-infrared (NIR) irradiation using thermal imaging. The mouse fibroblast cell line L929 cells and the human oral mucosal epithelial-derived cell line GMSM-K cells were co-cultured with FG MNs for 2 and 4 days, respectively, and cell viability was assessed by the Cell Counting Kit-8 (CCK-8) assay and dead/live staining. Separately, the human tongue OSCC cell line Cal-27 cells were divided into four groups: a control group (cultured in complete medium), 5-FU MN group (co-cultured with 5-FU-loaded MNs), GNR MN+ NIR group (co-cultured with GNR-loaded MNs followed by NIR irradiation at 0.4 W/cm2 for 5 min), and FG MN+ NIR group (co-cultured with FG MNs followed by identical NIR irradiation). After 24 and 48 h of co-culture, Cal-27 cell viability was assessed using both CCK-8 assay and dead/live staining. All animal experiments were approved by the Institutional Animal Ethics Committee. A subcutaneous Cal-27-derived OSCC postoperative recurrence model was established in BALB/c nude mice. The mice were divided into four groups: a control group, 5-FU MN group, GNR MN+ NIR group, and FG MN+ NIR group. Mice in the control group received no intervention, while the other groups were treated with the corresponding MN patches. Additionally, the GNR MN+ NIR and FG MN+ NIR groups received NIR irradiation (0.4 W/cm2, 5 min). Tumors were excised and their volumes measured on day 7, and tumor recurrence and wound healing were assessed on day 14. Results: A dual-network hydrogel-MN was successfully constructed, demonstrating favorable mechanical and dissolution properties. Upon NIR irradiation for 5 min, the temperature rapidly increased to 48.4°C, reaching the threshold for effective tumor ablation. CCK-8 and dead/live staining results showed that FG MNs exhibited no cytotoxicity toward L929 and GMSM-K cells, indicating good cytocompatibility. However, Cal-27 cell viability in the chemotherapy-photothermal combination group was lower than that in the other three groups, with a statistical difference (P<0.0001). In the nude mouse model, subcutaneous Cal-27 tumors were resected 7 days post-implantation. No statistical difference in tumor volume was observed among groups at this timepoint, confirming successful model establishment. By postoperative day 14, the control group displayed complete tumor recurrence with unhealed wounds, while the 5-FU and GNR MN+ NIR groups showed mild recurrence and partial wound healing. The FG MN+ NIR group demonstrated superior efficacy in both inhibiting tumor recurrence and promoting tissue repair compared with all other groups, with a statistical difference (P<0.0001). Conclusion The dual-network hydrogel-MN system effectively suppresses postoperative recurrence of Cal-27-derived subcutaneous OSCC and promotes tissue repair through chemo-photothermal combined therapy.