Abstract 1401: A systems immunology analysis to detect prognostic biomarkers in patients with squamous cell carcinomas of the head and neck

Abstract

Abstract In 2015, an estimated 59,340 people will develop head and neck cancer and an estimated 12,290 deaths will occur. Recent evidence has demonstrated that the immune system plays a key role in the development, establishment, and progression of head and neck squamous cell carcinoma (HNSCC). Here, we conduct comprehensive immune profiling to better understand the immunophenotype of HNSCC patients, assess the applicability of immune-modulatory drugs, and identify prognostic biomarkers of response to cetuximab treatment. Our systems immunology approach is composed of three complimentary, high-dimensional technologies: time-of-flight mass cytometry (CyTOF), luminex, and the HIMChip microarray platform. Mass cytometry quantifies protein expression on a single-cell basis by utilizing transition element, isotope-tagged antibodies. The Luminex immunoassay measures plasma cytokine levels. The “HIMChip” microarray is a custom Agilent SurePrint HD 8×15k format array containing over 7,000 unique probes for over 4,274 human immune-related genes. We employed these technologies to study the global immune status of thirty HNSCC patients receiving treatment with cetuximab, an IgG1 monoclonal antibody (mAb) targeting the epidermal growth factor receptor (EGFR). Patients were consented to participate on the Phase 0 biomarker-focused clinical trial (NCT01114256) and peripheral blood was obtained before and after cetuximab treatment. Peripheral blood mononuclear cells and plasma were isolated from each time point and utilized in downstream analysis. Via manual gating, our CyTOF-based phenotyping measured 24 distinct populations and the expression of 5 activation markers and 5 checkpoint receptors. The results of manual gating were confirmed with an automated, unsupervised algorithmic analysis combining stochastic neighbor embedding and k-means clustering. The resulting 2D representation of our single-cell data preserved global geometries and facilitated exploration of unique subsets within the natural killer (NK) compartment. To assess the prognostic value of these NK cell subsets, we applied a lasso-regularized logistic regression model to stratify patients based on their clinical response to cetuximab therapy. We identified an NK cell subset that is associated with clinical benefit to therapy. Luminex analysis revealed a cetuximab-induced cytokine signature composed of VCAM1, IP-10, and VEGFD that may play a role in increasing NK cell trafficking to tumor (T-statistics of 1.77, 2.17, and 1.55 respectively). The HIMChip microarray results support our proteomic findings. These results provide the first known comprehensive immune profiling of HNSCC patients as they undergo treatment with cetuximab. Validation in a large cohort of patients is needed and future mechanistic studies will investigate the efficacy of our novel NK cell population in disease control. Citation Format: Cariad Chester, Ajay Fernandez, Atsushi Yonezawa, Xing Zhao, Naren Rajasekaran, Holbrook Kohrt. A systems immunology analysis to detect prognostic biomarkers in patients with squamous cell carcinomas of the head and neck. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1401.