Tumor microenvironment-simulated organoids for personalized therapy prediction in head and neck squamous cell carcinoma.

Abstract

Patient-derived organoids (PDOs) have emerged as promising models for predicting personalized drug responses in cancer therapy. However, the absence of essential immune and stromal components limits their ability to recapitulate the tumor microenvironment. Here, we established a total of 30 patient-derived organoids (PDOs) from 79 patients with locally advanced (LA) and recurrent/metastatic (R/M) head and neck squamous cell carcinoma (HNSCC). These PDOs maintained sustained expansion capacity and preserved the histopathological characteristics and genomic heterogeneity of their parental tumors. By integrating autologous immune cells and cancer-associated fibroblasts (CAFs) into PDOs, respectively, microenvironment-simulated PDOs (MS-PDOs) were established using a feasible co-culture condition. Compared with conventional PDOs, MS-PDOs-PBMC exhibited specific cytotoxicity and responses to PD-1/PD-L1 inhibitors, while MS-PDOs-CAFs showed enhanced tolerance to chemotherapy drugs, indicating that microenvironment components modulate therapeutic responses in HNSCC. The drug response profiles of MS-PDOs exhibited diverse sensitivity to PD-1/PD-L1 inhibitors, chemotherapy drugs, and combination regimens. Notably, the therapeutic predictions of MS-PDOs were consistent with clinical treatment outcomes, supporting their translational relevance. Collectively, MS-PDOs serve as a robust platform for modeling the tumor microenvironment and predicting therapeutic responses, supporting precision medicine-guided clinical decision-making and offering personalized treatment strategies for HNSCC patients.