All HPV-negative head and neck cancers are not the same: Analysis of the TCGA dataset reveals that anatomical sites have distinct mutation, transcriptome, hypoxia, and tumor microenvironment profiles. (May 2021)

• Record Type: Journal Article
• Authors: Kim, Hugh Andrew Jinwook; Zeng, Peter Y.F.; Shaikh, Mushfiq Hassan; Mundi, Neil; Ghasemi, Farhad; Di Gravio, Eric; Khan, Halema; MacNeil, Danielle; Khan, Mohammed Imran; Patel, Krupal; Mendez, Adrian; Yoo, John; Fung, Kevin; Lang, Pencilla; Palma, David A.; Mymryk, Joe S.; Barrett, John W.; Boutros, Paul C.; Nichols, Anthony C.
• Is Part Of: Oral oncology. Volume 116(2021)
• Publication Date: 2021-05
• Subjects: Oral cancer -- Genomics -- Mutational status -- Larynx cancer -- HNSCC -- HPV-negative -- Oral cavity -- Hypoxia -- mRNA abundance -- Pathway analysis -- Immune landscape; Mouth -- Cancer -- Periodicals; Mouth -- Tumors -- Periodicals; Mouth Diseases -- Periodicals; Mouth Neoplasms -- Periodicals; Bouche -- Cancer -- Périodiques; Bouche -- Tumeurs -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9943105
• DOI: 10.1016/j.oraloncology.2021.105260
• Languages: English
• ISSNs: 1368-8375

Abstract

Highlights: Differences in survival and molecular biology exist between HNSCC anatomical sites. Laryngeal cancer has high point mutation burden. Glycosylation was enriched in laryngeal cancer, which may be targeted by antifungals. Tumor microenvironments of anatomical sites may guide immune and radiation therapies. Methylation patterns of Hox genes relate to de-differentiation of laryngeal cancer. Abstract: Purpose: Head and neck squamous cell carcinoma (HNSCC) affects various anatomical sites, which often dictates whether the cancer is managed with primary surgery or radiation. This study aimed to assess differences in single nucleotide variation (SNV), copy number, mRNA abundance, methylation, and tumor microenvironment (TME) between HPV-negative oral cavity (OC), oropharyngeal (OPC), hypopharyngeal (HPC), and laryngeal (LC) cancers within The Cancer Genome Atlas (TCGA). Methods: We downloaded the clinical information and molecular data for the TCGA HNSCC cohort from the data portal and published literature. The TME was estimated using mRNA abundance data. We conducted our analyses within the Bioconductor statistical framework in the R environment. CNA and mRNA abundance results were correlated and grouped with SNV results for downstream pathway analysis. Results: LC had a higher mutational burden than OC and OPC (p <10 −4 ). LC tumors were enriched in CSMD3, NSD1, DCHS2 and ANK2 SNVs, while OC tumors were enriched in CASP8 SNVs (FDR < 0.1). LCs were enriched for neuronal and glycosylation pathways, while OCs were enriched for extracellular matrix pathways. B cells and endothelial cells were more abundant in LC while monocytes were more abundant in OC (FDR < 0.1). OPC was the most hypoxic, followed by OC then LC (FDR < 0.05). OC had greater methylation of Hox genes than LC. Subsite analysis revealed that oral tongue cancers had fewer CASP8 and FBN2 mutations and higher dendritic cell abundance than other oral cavity cancers. Conclusions: We identified significant genomic, transcriptional, and microenvironmental differences between HPV-negative HNSCC. Further study is warranted to determine if these findings portend differential response to specific treatment modalities.