10.1002/9781118797914

Vollständiger Abstract

Worum geht es in dieser Arbeit?

<h4>Background</h4>Head and neck squamous cell carcinoma (HNSCC) is characterized by clinically undetectable early metastasis and a poor prognosis at locally advanced and recurrent metastatic stages. Platinum-based resistance remains a major limitation to treatment efficacy. Currently, no reliable biomarkers are available to identify lymph node metastasis or to predict therapeutic sensitivity. Therefore, this study aimed to develop a predictive model to address these clinical challenges.<h4>Methods</h4>Four cell lines (CAL-27, FaDu, Leuk1, NOEC) were employed to establish zebrafish cell-derived xenograft (zCDX) models and assess the differential metastatic potential. Surgical or biopsy samples from, respectively, 20 and 14 HNSCC patients were used for building and validating the zebrafish patient-derived xenograft (zPDX) models. Cisplatin-sensitive HNSCC cell lines (CAL-27 and FaDu) and their cisplatin-resistant sublines (CAL-27-DDP and FaDu-DDP) were established, and the differential responses to cisplatin were evaluated in both murine and zCDX models. Additionally, to assess the patient-specific therapeutic efficacy of cisplatin, zPDX models were generated from four patient-derived surgical specimens and a supplementary biopsy for preliminary validation.<h4>Results</h4>The zebrafish xenograft model recapitulated the invasive phenotype of the HNSCC cell lines. The model distinguished between metastatic and non-metastatic disease with high accuracy (AUC = 0.98). It faithfully reproduced the cisplatin sensitivity and resistance patterns identified in corresponding in vitro and murine models. Moreover, it showed strong concordance with patient-specific cisplatin responses.<h4>Conclusions</h4>The zPDX model provides a rapid, efficient, clinically applicable platform for predicting metastasis and evaluating platinum-based therapy responses in HNSCC, facilitating the design of scientifically grounded personalized treatment.

Abstract: PubMed · Datensatz