Harter, V. | bioRxiv Science

bioRxiv2026

Predicting Head and Neck Squamous Cell Carcinoma outcomes using long-term Patient-Derived Tumor Organoids

Head and neck squamous cell carcinoma (HNSCC) remains associated with substantial morbidity and a 5-year overall survival rate of approximately 60%, reflecting persistent radio- and chemo-resistance and the lack of effective precision medicine strategies. Patient-Derived Tumor Organoids (PDTO) constitute promising functional models that may predict individual treatment response. In this study, we generated PDTO from surgically resected HNSCC of the oral cavity, oropharynx, larynx, and hypopharynx. A total of 20 long-term PDTO lines were established, maintaining growth over seven passages and successfully cryopreserved, capturing the molecular and clinical diversity of the patient cohort. These PDTO faithfully recapitulated histological features, major tumor marker expression, and the genomic and transcriptomic landscapes of their tumors of origin, with stability over time. Functional assays revealed heterogeneous responses to cisplatin and X-rays. Importantly, in vitro sensitivity of PDTO was associated with clinical outcome of patients at 24 months. Cisplatin response of PDTO predicted prognosis with 66.7% sensitivity and 100% specificity, while X-ray response showed 91.7% sensitivity and 75% specificity. Notably, all patients whose PDTO were classified as resistant to both cisplatin and X-rays experienced relapse and/or death within 24 months. Collectively, the successful long-term expansion and cryopreservation of HNSCC PDTO establish a stable and scalable preclinical resource that captures the molecular and clinical heterogeneity of the disease. This biobank provides a valuable platform for mechanistic studies and for the evaluation of innovative therapeutic strategies. This cohort represents one of the largest clinically annotated HNSCC PDTO collections to date, demonstrating a robust association between PDTO response to cisplatin and X-rays and patient prognosis. These findings support the predictive potential of PDTO-based functional assays and argue for their integration into standardized, rapid, and miniaturized precision oncology workflows for HNSCC.

cancer biology↗

bioRxiv2025

Deep Learning Bridges Histology and Transcriptomics to Predict Molecular Subtypes and Outcomes in Muscle-Invasive Bladder Cancer

Muscle-Invasive Bladder Cancer (MIBC) is a heterogeneous disease with distinct molecular subtypes influencing prognosis and therapeutic response. However, molecular profiling through RNA sequencing remains costly, time-consuming and complicated by intratumoral heterogeneity. We developed a Deep Learning (DL) approach to infer molecular subtypes from routine histopathological slides and to evaluate its prognostic value in patients treated with neoadjuvant chemotherapy (NAC). We developed an DL-model predicting the expression of 848 subtype-associated genes from histological images of transurethral resection of bladder tumor, enabling spatial molecular subtyping at tile level. The model was trained on 297 NAC-treated patients from the VESPER clinical trial and evaluated on three independent cohorts (COBLAnCE, n=224; Saint-Louis, n=30 and TCGA, n=315), covering diverse staining protocols and scanner types. Spatial transcriptomics from six VESPER patients confirmed the spatial consistency of the inferred expression profiles. Our approach achieved a ROC AUC of 0.94 for molecular subtype prediction, with 95% of genes significantly predicted, demonstrating its ability to capture transcriptomic dysregulations from histological morphology. Predicted expression maps revealed spatially coherent patterns and intratumoral molecular heterogeneity. Importantly, tumors predicted with basal/squamous features (pure or mixed), were associated with significantly worse progression-free and overall survival after NAC (log-rank p=0.014 and 0.037, respectively). This DL-based framework enables accurate and spatially resolved inference of gene expression and molecular subtypes in MIBC without sequencing. These findings could improve patient stratification in clinical practice and support the design of more targeted clinical trials. Further validation in larger cohorts is needed before routine clinical implementation.

cancer biology↗