MSACL 2026 Abstract

Clinically Adapted Rapid Evaporative Ionization Mass Spectrometry for Ex Vivo Tumor–Healthy Tissue Discrimination in Head and Neck Cancer

Gabriel Stefan Horkovics-Kováts (1), Nathalie Gumpert (2), István Pap (1), Zahra Nozari (1) Daniel Simon (1), Tamás Karancsi (3), Luisa Symeou (2), Julian Künzel (2), Kathrin Renner (2), Christopher Bohr (2), Zoltán Takáts (1)
(1) Department of Immunomedicine, University of Regensburg, Regensburg, Germany, (2) Department of Otorhinolaryngology, University Hospital Regensburg, Regensburg, Germany, (3) AmbiMass, Budapest, Hungary

Gabriel Stefan Horkovics-Kováts (Presenter) University of Regensburg

Relevant Financial Disclosures (within past 24 months, reported on Apr 22, 2026) No relevant financial relationship(s) to disclose.

INTRODUCTION:
Head and neck squamous cell carcinoma (HNSCC) remains associated with poor prognosis, with five-year survival rates of 40–50%. Achieving complete tumor resection while preserving functional tissue is critical, particularly in anatomically complex regions such as the tongue. Rapid Evaporative Ionization Mass Spectrometry (REIMS) is a well-established technology for real-time molecular characterization of surgical aerosol and has demonstrated value for intraoperative tissue identification in multiple surgical applications. However, its application in HNSCC remains limited, and robust spectral datasets are lacking. Furthermore, successful clinical translation requires mass spectrometry systems that are compatible with the constraints of the operating theatre, including mobility, low noise levels, and rapid readiness. Establishing well-controlled datasets in combination with such clinically adaptable instrumentation is therefore essential to enable reliable tissue classification and support intraoperative implementation.

METHODS:
An ex vivo study was performed using paired tissue samples from 10 HNSCC patients, including tumor and corresponding adjacent healthy tissue. Samples were analyzed using a REIMS ion source coupled to an RDa TOF mass spectrometer (Waters). The system was integrated into a mobile, fully enclosed and movable platform equipped with a multistage roots/diaphragm pump (AmbiMass), enabling rapid venting and pump-down (~5 minutes) and reduced operational noise (~55 dB), thereby improving suitability for operating room environments. Electrosurgical aerosol was generated and collected under optimized conditions established in prior work, including controlled sampling position, suction configuration, and gas flow parameters. Spectral data was acquired using standardized acquisition settings to ensure reproducibility across all samples and to support downstream multivariate statistical analysis.

RESULTS:
A total of 20 tissue specimens (10 tumor, 10 healthy) were successfully analyzed, generating a high-quality and internally consistent spectral dataset under controlled ex vivo conditions. The optimized and enclosed REIMS platform enabled stable signal acquisition across all samples with minimal environmental interference. Preliminary inspection of the spectra indicates reproducible molecular differences between tumor and healthy tissues, particularly within lipid-associated mass regions. Compared to prior in vivo observations, ex vivo measurements demonstrated improved signal-to-noise ratios and reduced spectral variability, facilitating data quality suitable for supervised model building and later real-time in vivo tissue classification.

CONCLUSION:
This study establishes a robust ex vivo REIMS dataset for HNSCC using a clinically adaptable, low-noise mass spectrometry platform. The combination of optimized sampling conditions and improved instrument integration enables reproducible, high-quality spectral acquisition without disruption to routine operating room conditions. The observed molecular differences between tumor and healthy tissue support the feasibility of accurate tissue discrimination. Ongoing database building and multivariate analysis will provide improved classification performance and supports future translation of REIMS toward real-time intraoperative decision support in head and neck surgery.