A functional gene expression analysis in epithelial sinonasal cancer: Biology and clinical relevance behind three histological subtypes

• Published in: Oral oncology 2019-03, Vol.90, p.94-101
• Main Authors: De Cecco, Loris, Serafini, Mara Serena, Facco, Carla, Granata, Roberta, Orlandi, Ester, Fallai, Carlo, Licitra, Lisa, Marchesi, Edoardo, Perrone, Federica, Pilotti, Silvana, Quattrone, Pasquale, Piazza, Cesare, Sessa, Fausto, Turri-Zanoni, Mario, Battaglia, Paolo, Castelnuovo, Paolo, Antognoni, Paolo, Canevari, Silvana, Bossi, Paolo
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Biomarkers, Tumor - genetics; Cancer biology; Carcinoma - classification; Carcinoma - genetics; Carcinoma, Neuroendocrine - classification; Carcinoma, Neuroendocrine - genetics; Carcinoma, Squamous Cell - classification; Carcinoma, Squamous Cell - genetics; Female; Gene expression; Gene Expression Profiling; Humans; Male; Maxillary Sinus Neoplasms - classification; Maxillary Sinus Neoplasms - genetics; Middle Aged; Neuroendocrine signature; Prognosis; Rare Diseases - classification; Rare Diseases - genetics; Retrospective Studies; Sinonasal epithelial cancer; Transcriptome; Tumor microenvironment; Tumor Microenvironment - genetics; Young Adult
• ISSN: 1368-8375; 1879-0593
• DOI: 10.1016/j.oraloncology.2019.02.003

•Gene expression profile discriminates sinonasal cancer histological subtypes.•NKSCCs show up-regulation of epithelial tumor pathways and moderate immune component.•SNECs have a very low immune component.•SNUCs have low epithelial/endocrine differentiation and high immune component. Epithelial sinonasal cancers (SNCs) are rare diseases with overlapping morphological features and a dismal prognosis. We aimed to investigate the expression differences among the histological subtypes for discerning their molecular characteristics. We selected 47 SNCs: (i) 21 nonkeratinizing squamous cell carcinomas (NKSCCs), (ii) 13 sinonasal neuroendocrine cancers (SNECs), and (iii) 13 sinonasal undifferentiated cancers (SNUCs). Gene expression profiling was performed by DASL (cDNA-mediated annealing, selection, extension, and ligation) microarray analysis with internal validation by quantitative RT-PCR (RT-qPCR). Relevant molecular patterns were uncovered by sparse partial-least squares discriminant analysis (sPLS-DA), microenvironment cell type (xCell), CIBERSORT, and gene set enrichment (GSEA) analyses. The first two sPLS-DA components stratified samples by histological subtypes. xCell highlighted increased expression of immune components (CD8+ effector memory cells, in SNUC) and “other cells”: keratinocytes and neurons in NKSCC and SNEC, respectively. Pathway enrichment was observed in NKSCC (six gene sets, proliferation related), SNEC (one gene set, pancreatic β-cells), and SNUC (twenty gene sets, some of them immune-system related). Major neuroendocrine involvement was observed in all the SNEC samples. Our high-throughput analysis revealed a good diagnostic ability to differentiate NKSCC, SNEC, and SNUC, but indicated that the neuroendocrine pathway, typical and pathognomonic of SNEC is also present at lower expression levels in the other two histological subtypes. The different and specific profiles may be exploited for elucidating their biology and could help to identify prognostic and therapeutic opportunities.