Advancing Education, Research, and Quality of Care for the Head and Neck oncology patient.
Introduction: Invasiveness is a key property of cancer cells, but its clinical relevance and variability across patients remains poorly understood. Head and neck squamous cell carcinoma (HNSCC) cancer cell lines are often used to model invasiveness, but the ability of well-established, two-dimensional cell lines to model in vivo tumor behavior is limited. Here, we utilize a bank of early passage HNSCC patient-derived three-dimensional organoids (PDOs) to model invasiveness across a cohort of patients, uncovering a prognostically relevant gene signature associated with this functional property.
Methods: To model tumor invasiveness in HNSCC, we utilized a bank of 14 PDOs. Matrigel trans-well invasion assays were performed on each PDO line, using serum-containing media as a chemoattractant. Invaded cells were fixed, stained with 0.4% crystal violet, imaged at 20x magnification, and quantified. Organoids were classified as highly invasive or less invasive based on the average number of invaded cells, normalized to the most invasive organoid line. RNA-seq and whole exome sequencing (WES) data from PDOs were used to perform differential gene expression and mutational analyses. Genes of interest were filtered by level of expression as well as biological and prognostic relevance. CRISPR-mediated knockdowns of select genes were performed in two highly invasive PDO lines to further understand their role in invasiveness.
Results: Organoids were classified as highly invasive (n=6) or less invasive (n=8). Differential gene expression analysis identified 106 statistically significant genes that were upregulated in invasive organoids (Log2FC>1.5, padj<0.05). Genes were then filtered by negative overall survival associations (HR>2.25, p-value<0.05) in TCGA HPV negative head and neck squamous cell carcinoma patients, revealing a set of 14 genes of interest: MAFF, SH3TC2, ACOT9, PLAUR, SCG5, ANGPTL4, NNMT, LAMB3, CSF2, SRPX, TRIML2, ANXA3, ITGA5, and CTGF. Of these genes, 5/14 (35.7%) also had negative associations with disease free survival (HR>2.1, p-value<0.05). Notably, LAMB3 and ITGA5 are key genes within the previously established partial epithelial to mesenchymal transition (p-EMT) program, which is believed to play a critical role in cancer cell invasion and metastasis.
Conclusions: Functional and transcriptional characterization of our HNSCC patient-derived organoid bank revealed several genes associated with invasiveness, which hold prognostic significance and may reflect novel therapeutic targets in this patient population. Further work aimed at pharmacologically targeting these markers is needed to fully understand their therapeutic potential.