Abstract Background: Immune-checkpoint blockade (ICB), with or without chemotherapy, is the standard of care for recurrent/metastatic head and neck squamous cell carcinoma (HNSCC). However, only a minority of patients get clinical benefit from it. The combined expression of PD-L1 in a tumor section is currently the only biomarker approved for response prediction. Yet, its invasive nature and suboptimal predictive performance underscores the need for new, non- invasive surrogates of immune activity in patients undergoing ICB. Extracellular vesicles (EVs) carry key molecules critical for cell-cell communication and have the potential to be used as a non-invasive, real-time biomarker of systemic immune activity, essential for success in ICB- based therapy. Methods: We used a multichannel, microfluidic device that employs an immune- affinity approach, ‘herringbone chip’ (EVHB-Chip), to retrospectively isolate EVs from twelve HNSCC patients before and after the first cycle of ICB. One mL of plasma was used per patient per time point. For each patient, two devices were serially used, with each sample continuously following through each chip in a non-destructive manner—the first device aimed to isolate B cell EVs (antibodies against CD19 and CD20). The second device was optimized to capture T cell EVs (against CD274/PD-L1). Digital-droplet PCR was performed after EVs were isolated to detect the RNA transcripts in the EV cargo. Serial sampling was performed for six out of the twelve patients. Response assessment was conducted via RECIST1.1 criteria or clinical evaluation. Results: Most of our cohort was male 7/12 (57%) and had oral cavity 4/12 (33%) or oropharynx 4/12 (33%) tumors. Additionally, 11/12 (91%) patients received ICB monotherapy. Most of the cohort had a PD-L1 ≤ 1 (7/12; 58%), 7/12 patients experienced disease progression (PD), 4/12 experienced stable disease (SD), and one patient had a partial response (PR). Most of the RNA signal in pre-ICB samples was obtained from CD20 (12/12; 100%), followed by CD45 (10/12; 83%), CD19 (9/12; 75%), and CD274/PD-L1 (5/12; 41%). No difference was found in the RNA signal pre-ICB between responders and non-responders (p=0.1). For the patients with serial sampling available, the signal from CD274/PD-L1 increased after one infusion of ICB in 4/6 patients and remained negative in one. A trend towards a high amount of T cell-derived EVs was observed for the entire cohort (p=0.07; p=0.08) when comparing the amount of RNA via our EV marker (GAPDH, ACTB) between T and B cell devices. Patients with higher than 11.4/mL copies of CD19 (median value for the cohort) in the pre-ICB plasma sample tended to have a higher overall survival (HR 0.26, 95% CI: 0.06-1.13; p-value: 0.05) and progression-free survival (HR 0.28, 95% CI: 0.07-1.16; p-value: 0.06). Conclusion: The EVHB-Chip allowed for the detection and profiling of T and B cell-derived EVs in pre-ICB plasma samples of HNSCC patients undergoing ICB in this pilot cohort. Higher CD19 signal pre-ICB could potentially impact response and survival outcomes in ICB-based therapies. Citation Format: Andrew R. Levy, Daniel A Ruiz Torres, Uma Paithankar, Raheel Ahmad, Daniel C Rabe, Daniel L Faden, Shannon L Stott. Microfluidic isolation of immune cell-derived extracellular vesicles in head and neck cancer patients on immune-checkpoint blockade [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr B048.
Read full abstract