Advancing Education, Research, and Quality of Care for the Head and Neck oncology patient.
Background: Head and neck squamous cell carcinoma (HNSCC) is an aggressive malignancy commonly treated with surgery and radiotherapy. Given high rates of morbidity and lack of precision therapeutics for HPV-negative disease, there is a need to identify novel immunotherapy targets which may be leveraged for therapy. To accomplish this, we first performed bulk RNA sequencing (RNA-seq) analyses through our institutional tumor registry and identified upregulation of cancer-testis antigens (CTAs) in radiation-recurrent HNSCCs. CTAs represent potential targets for immuno-oncology given their localized expression to immune-privileged tissues with no (testis-restricted) or minimal (testis-selective) expression in other human body sites. We then validate these findings across previously untreated (de novo) and recurrent tumors to define the CTA landscape in HNSCC through analyses of external datasets, including single-cell RNA sequencing (scRNA-seq) and The Cancer Genome Atlas (TCGA).
Methods: To identify molecular mechanisms in HNSCC patients which may be leveraged for therapy, bulk RNA-seq was performed on tumor samples comparing de novo vs recurrent tumors after prior radiotherapy. The Cancer-Testis Antigen Database (CTdatabase) was used to determine the distribution of physiologic expression of each identified CTA. An external scRNA-seq dataset (GSE234933) of de novo and recurrent HNSCCs was then re-analyzed to validate findings from bulk RNA-seq. To define the CTA landscape across HNSCCs, we searched multiple datasets – institutional bulk RNA-seq, scRNA-seq, and TCGA – for normalized CTA expression. We further re-analyzed a scRNA-seq dataset of normal oral mucosa (GSE164241) to exclude CTAs expressed in benign tissues. Finally, to demonstrate that our findings are not limited to comparisons between non-matched HNSCC tumors and normal oral mucosa samples, we employed the CopyKAT R package to separate HNSCC epithelium into malignant and benign populations, analyzing CTA expression within these populations in the same patients.
Results: Initial differential expression analysis on bulk RNA-seq and scRNA-seq datasets identified several testis-restricted and testis-selective CTAs as significantly upregulated in radiation-recurrent tumors, suggesting potential as immuno-oncologic targets in the recurrent setting. Our expanded analysis then demonstrated CTAs across HNSCCs, including robust expression not only in radiation-recurrent disease, but in de novo tumors as well. In bulk RNA-seq data, several testis-restricted CTAs (MAGEA1, MAGEB2, SPANXB1, SPANXD, DKKL1, CSAG2) and multiple testis-selective CTAs from the MAGE family, among others, were prominently expressed. Subsequent analysis of scRNA-seq data corroborated expression of these testis-restricted CTAs and additionally identified unique testis-selective CTAs. TCGA analysis further expanded this list, identifying new testis-restricted CTAs (CXorf48 and TEX14) and additional testis-selective CTAs. Exclusionary analysis using scRNA-seq normal oral mucosa data refined our total list to fourteen unique testis-restricted and twenty-two testis-selective CTAs exclusively expressed in HNSCC. Finally, using CopyKAT, we demonstrate that CTAs are specifically enriched in malignant epithelial populations as compared to benign populations within the same HNSCC patients.
Conclusion: We present a set of fourteen testis-restricted and twenty-two testis-selective CTAs orthogonally validated across multiple tumor datasets. Their tumor-specificity and expression in both de novo and recurrent HNSCCs underscores their potential as promising precision immuno-oncologic targets for future T cell receptor engineering efforts against HPV-negative disease.