Head and Neck Advanced Research for Multi-Omics and Optimized Immunotherapy

1. Background 1.1 Research Background Immune checkpoint inhibitors (ICIs) have emerged as a new class of anticancer therapy following cytotoxic chemotherapy and targeted therapy. They have demonstrated survival benefits in various solid tumors, including head and neck cancers, thereby shifting the paradigm of cancer treatment. Tumors evade immune surveillance through checkpoint molecules such as PD-1 and CTLA-4, and the development of ICIs such as pembrolizumab and nivolumab has shown efficacy in head and neck cancers as well.

Advanced and metastatic head and neck cancers not only pose life-threatening risks but also severely affect speech, swallowing, and appearance, leading to a higher suicide rate compared to the general population. In Korea, the incidence of head and neck cancer increased by about 30% between 2010 and 2020, with tongue cancer showing a particularly steep rise of 7.7% per year in young adults aged 20-30.

In the KEYNOTE-048 trial, pembrolizumab combined with platinum-based chemotherapy significantly prolonged overall survival compared to cetuximab-based regimens, with greater benefit observed in patients with PD-L1 CPS ≥20 or ≥1. Subsequent analyses showed that the survival benefit correlated with PD-L1 CPS only in the pembrolizumab-combination group, and about 20% of patients were identified as long responders. However, the overall response rate to ICIs remains only 15-26%, and PD-L1 CPS alone does not fully explain treatment outcomes.

The TIME (tumor immune microenvironment) study on head and neck cancer classified tumors into fully infiltrated, stroma-restricted, immune-excluded, and immune-desert subtypes, with 48% being fully infiltrated. Tumors with high expression of cytokines such as CXCL13, IFN-γ, and IL-10 exhibited more active immune cell infiltration. However, these results were based on pre-treatment tissues, and there is a lack of studies investigating TIME changes after treatment.

In recurrent head and neck cancer, intratumoral and stromal TIL density significantly decreased, while the proportion of immune-desert tumors increased. Stromal TIL enrichment correlated with better response to ICIs. However, this study was based on AI-driven image analysis and did not capture detailed immune microenvironment composition or post-treatment dynamics.

In Japan, spatial transcriptomics analysis was performed in one head and neck cancer patient treated with nivolumab, revealing that post-treatment loss of MHC class I gene expression was associated with immune resistance.

High CXCL13 expression is linked to the formation of tertiary lymphoid structures (TLS) and strongly associated with immune cell distribution in the TIME. TCGA analyses showed that CXCL13 expression is significantly higher in head and neck cancers compared to normal tissues, and correlates with the abundance of CD8+ T cells, B cells, macrophages, and other immune cells. Thus, it is necessary to investigate whether CXCL13 and TLS change following ICI treatment. In addition, CD44+ cancer stem cells in head and neck cancer are associated with poor prognosis, and their dynamic changes before and after immunotherapy warrant investigation.

Peripheral blood-based approaches to predict immune responses have also been explored. Retrospective studies using inflammatory markers such as NLR, MLR, and PLR developed predictive models that outperformed PD-L1 CPS in accuracy. Another study reported that PD-L1 expression on monocytes after chemoradiotherapy correlated with plasma CXCL11 levels. Gene expression analyses of pre- and post-surgical samples identified significant differences in EIF2, EIF4, and mTOR signaling pathways. Moreover, cytokine levels in the serum of head and neck cancer patients were significantly higher than in healthy controls, and an increase in immature neutrophils was suggested to be associated with tumor infiltration.

In a follow-up analysis of the KEYNOTE-048 trial, patients treated with pembrolizumab plus chemotherapy exhibited longer progression-free survival with subsequent taxane therapy (PFS2), independent of PD-L1 CPS (HR 0.67, P=0.00788). This suggests that ICI treatment may enhance responsiveness to later taxane-based chemotherapy. Biomarker studies are needed to explain these interactions, and given regional and ethnic variations, research based on Korean patients is particularly important.

1.2 Research Summary Head and neck squamous cell carcinoma (HNSCC) is a heterogeneous malignancy with prognosis varying by anatomical site and tumor characteristics, but advanced disease generally carries poor outcomes. This study aims to investigate dynamic changes in both the tumor immune microenvironment (TIME) and systemic immunity before and after first-line ICI plus chemotherapy in HNSCC patients.

The rationale for this project is as follows:

1. ICIs have improved survival in HNSCC but remain effective in only about 15% of patients. Predictive biomarkers to select responders remain unclear.
2. Although many studies have investigated TIME in HNSCC, no study has performed paired analyses of tumor tissue and blood before and after ICI treatment to assess spatial and temporal immune changes.
3. The extent to which TIME changes are reflected in peripheral blood remains unknown, limiting non-invasive monitoring of treatment response.
4. ICIs are known to enhance the efficacy of subsequent taxane chemotherapy, but the underlying mechanism is not understood.

Therefore, this study seeks to identify predictive biomarkers for treatment response and develop personalized therapeutic strategies for patients with advanced or metastatic HNSCC receiving ICI plus chemotherapy.

• Study Objectives and Goals To identify predictive biomarkers of treatment response and develop personalized therapeutic strategies in patients with advanced/metastatic HNSCC treated with ICIs plus chemotherapy. ________________________________________
• Study Duration The study will be conducted for 3 years from the date of IRB approval. Participants will be followed for 3 years from enrollment. Analyses will begin in the second year, focusing on dynamic changes in the TIME and systemic immunity before and after treatment. In the third year, biomarkers of ICI response and resistance will be identified, and correlations between TIME and systemic immune changes will be evaluated. ________________________________________
• Study Participants 4.1 Inclusion and Sample Size Justification The study aims to explore immune microenvironment (TME) changes observed in tumor samples before and after ICI-based treatment. The primary outcome is to assess changes in immune cell marker expression and their association with treatment response. This exploratory design seeks to generate clinically meaningful immunologic patterns as preliminary data for future studies (e.g., biomarker-driven predictive model development, prospective validation).

Eligible patients will be those diagnosed with advanced/metastatic HNSCC receiving first-line ICI plus chemotherapy. Approximately 30 paired pre- and post-treatment samples will be collected.

The target sample size of 30 patients is based on feasibility considerations (availability of paired samples, recruitment within 1.5-2 years) and aligns with prior studies of similar design, which typically included 20-40 patients. This is deemed an appropriate exploratory cohort size for the study objectives.

At Samsung Medical Center, about 600 HNSCC patients are treated annually, with over 50 newly diagnosed advanced/metastatic cases receiving palliative chemotherapy each year, ensuring sufficient patient accrual for this study.

5 Study Design and Methods 5.1 Dynamic Analysis of Immune Changes Before and After Treatment

• Tumor Immune Microenvironment (TIME) Analysis Paired tumor biopsies or FFPE samples will be collected before and after treatment, sectioned, and analyzed. Immune biomarkers to be evaluated include tumor cells, T cells, B cells, neutrophils, macrophages, immune gene signatures (e.g., interferons such as IFN-γ, and interleukin families), T-cell subsets (CD4+, CD7+, CD8+, CD16, CD34+, CD38+, CD56+), cancer stem cells (CD44), CXCL13, and PD-1/PD-L1 expression. Changes in these biomarkers will be compared pre- and post-treatment. Interactions between tumor and immune cells will be examined, as well as alterations in specific immune cell clusters and spatial patterns (e.g., tertiary lymphoid structures, immune hubs). Additionally, extracellular vesicle changes will be evaluated to explore mechanisms underlying enhanced responsiveness to subsequent taxane therapy after ICI treatment.
• Systemic Immunity Analysis Peripheral blood mononuclear cells (PBMCs) will be isolated from blood samples and analyzed using flow cytometry. Existing validated panels in the PI's laboratory will be adapted for this study. Quantitative changes in immune cell populations and cytokines (e.g., IFN-γ) before and after treatment will be assessed.

5.2 Development of a Personalized Treatment Strategy for ICIs in Head and Neck Cancer

• Identification of Predictive Biomarkers Clinical response data to ICI therapy will be integrated with tissue and blood-based analyses to identify candidate predictive biomarkers. Significant immune cell changes will be defined as ≥1.5-fold differences. Predictive models will be developed using multivariate and regression analyses, with the goal of achieving an Area Under the Curve (AUC) ≥0.80.
• Correlation Between TIME and Systemic Immunity Associations between tumor microenvironmental changes and peripheral blood changes will be examined. A Pearson correlation coefficient ≥0.5 will be considered meaningful. Further, the relationship between these immunologic changes and patient prognosis will be evaluated.
• Mechanisms of ICI Resistance Comparative analyses between responders and non-responders will be conducted to elucidate mechanisms of primary resistance. When feasible, paired tumor and blood samples at the time of disease progression will also be analyzed to investigate mechanisms of acquired resistance.