Immune microenvironment of the gene signature reflecting the standardised uptake value on 18F-fluorodeoxyglucose positron emission tomography/computed tomography in head and neck squamous cell carcinoma.

Abstract

A comprehensive understanding of the link between 18F-FDG PET/CT and the tumor immune microenvironment (TIM) is lacking. We, therefore, investigated the TIM in regard to the gene signature of 18F-FDG PET/CT in head and neck squamous cell carcinoma (HNSC). The mRNA sequence data of 480 HNSC patients on The Cancer Genome Atlas portal were used to explore genes showing high associations with maximum standardised uptake value (SUVmax) on 18F-FDG PET/CT based on Pearson correlation test. Hierarchical clustering of the selected gene signature was performed and divided patients into high and low SUV clusters. Principal component analysis was performed to derive the summarised expression profile of the gene signature and defined the first principal component scores as the SUV signature scores (SUVSSs). The SUV clusters and SUVSS based on the gene signature were characterised by overall survival, clinical variables, and the immune microenvironment in terms of overall immune score, immune cell type enrichment score, expression of immunomodulator genes as well as somatic copy number alterations (SCNA) possibly contributing to immune cell recognition. The high SUV cluster classified by the gene signature (191 genes) was an independent predictor of overall survival (adjusted hazard ratio 1.40, p = 0.022). The SUVSS values differed across the molecular subtypes of HNSC (p < 0.001), and HPV status (p = 0.024). Tumors in the low SUV cluster exhibited significantly higher overall immune score and lower SCNA scores (p < 0.05 for all) compared with tumors in the high SUV cluster. The low SUV cluster showed an immune cell composition consisting of high levels of T cells, B cells, mast cells, neutrophils, monocytes, and eosinophils, but lower basophils and similar macrophage levels to the high SUV cluster. Differential gene expression analysis demonstrated SUV cluster-distinct expression of several immunomodulators including PD-1, CD40LG, IL2RA, TLR4, BTLA, and TIGIT. HNSC exhibited the distinct TIM according to the gene signature reflecting SUVmax on 18F-FDG PET/CT. Our results support an understanding of the close relationship between FDG uptake and tumor immune response, and suggest that 18F-FDG PET/CT could be clinically usable as a biomarker for assisting immunotherapy.