Precision oncology in head and neck squamous cell carcinoma: from circulating biomarkers of immunotherapy response to advanced spatial profiling in clinically relevant samples

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer worldwide, characterized by metastatic evolution and poor prognosis. Immune Checkpoint Inhibitors (ICIs) have recently been approved as first-line treatment for recurrent or metastatic HNSCC. However, the effectiveness of immunotherapy is heterogeneous and varies among patients and HNSCC types, which highlights the need to identify reliable predictive biomarkers. Due to their remarkable stability in plasma, microRNAs (miRNAs) present significant potential as a novel category of minimally invasive biomarkers. In this study, I evaluated plasma miRNAs at baseline and investigated their role as biomarkers to select and monitor patients with HNSCC who would benefit from immunotherapy treatment. Furthermore, to gain a comprehensive understanding of therapy resistance mechanisms within the intricate tumor microenvironment, I adapted a high-resolution spatial transcriptomics platform, Open-ST, for the analysis of formalin-fixed paraffin-embedded (FFPE) tissue samples. This approach is uniquely suited for retrospective analyses, enabling pre-treatment profiling based on known clinical outcomes as well as comparisons between pre- and post-treatment samples. Circulating miRNAs expression profiles were determined from plasma samples of a discovery cohort of 22 HNSCC patients treated with the anti-PD-1 monoclonal antibody Pembrolizumab, alone or in combination with chemotherapy. Differentially expressed (DE) miRNAs between Responder (R) and Non-Responder (NR) patients to immunotherapy were identified. The DE miRNAs were validated in an independent cohort of 23 HNSCC patients. Validated miRNAs were evaluated to assess their potential use as biomarkers to predict both ICIs response and patients’ prognosis. Moreover, their biological role was investigated through enrichment analysis using miRNAs’ target genes. The new Open-ST protocol was benchmarked on FFPE samples using a mouse model, with fresh-frozen tissue processed in parallel to evaluate performance differences and improvements. MiRNAs profiling of 800 miRNAs identified a signature of 152 miRNAs that characterized HNSCC patients. 18 miRNAs resulted differentially expressed between R and NR patients, in particular, 7 DE miRNAs were upregulated, while 11 ones were downregulated in R vs NR. Among those DE miRNAs, miR-122-5p, miR-146a-5p, miR-16-5p, miR-198, miR-451a and miR-628-3p were confirmed to be upregulated, while miR-548v was confirmed to be downregulated in R compared to NR patients within the validation cohort. These seven circulating miRNAs effectively classified R and NR patients, both as single biomarkers and in combination. Their combination is also associated with patients' Progression Free Survival (PFS) and Overall Survival (OS), underlining their potential prognostic value. Furthermore, Kyoto Encyclopedia of Genes and Genomes (KEGG) Enrichment Analysis shed light on the biological processes in which miR-122-5p, miR-146a-5p, miR-16-5p, miR-198 and miR-451a target genes may be involved. The adaptation of Open-ST on FFPE samples led to the identification of more than 40000 genes and up to 1400 unique molecules in adult mouse cerebellum, in line with what found in FF samples. The analysis revealed a commonality in spatial gene composition and expression, confirming the applicability of the method to FFPE-preserved tissues. The combination of circulating miR-122-5p, miR-146a-5p, miR-16-5p, miR-198, miR-451a, miR-628-3p and miR-548v, is a predictive biomarker of Pembrolizumab response in HNSCC patients and can assist clinicians in selecting appropriate treatment strategies. The results obtained using Open-ST on FFPE samples pave the way for new studies aimed at comprehensively analyzing the tumor microenvironment of HNSCC in the context of immunotherapy resistance, thereby contributing to the advancement of personalized medicine.