MSACL 2019 EU Abstract

Brain Tumour Characterization Using Laser Desorption Imaging – Rapid Evaporative Ionisation Mass Spectrometry

Hanifa J.A Koguna (1, 2), Daniel Simon (1, 2), Julia Abda (1), Josephine Bunch (1, 2), Zoltan Takats (1)
(1) Department of Surgery and Cancer, Imperial College London, London, United Kingdom (2) National Centre of Excellence in Mass Spectrometry Imaging, National Physical Laboratory, Teddington, United Kingdom

Warning: Undefined variable $headshot in /var/www/html/view_abstract/view_abstract_in_program.php on line 704

Hanifa Koguna (Presenter) Imperial College London / NPL

Presenter Bio: I am a Ph.D. student at Imperial College London and National Physical Laboratory. I am part of the Cancer Research UK Grand Challenge Rosetta Team. My background is clinical (junior neurosurgeon) and my interests include brain tumours, MSI, REIMS, medical robotics and surgical innovation in general.

Relevant Financial Disclosures (within past 24 months) No relevant financial relationship(s) to disclose. Conflict mitigation NOT REQUIRED.

INTRODUCTION: Brain tumours have a very poor prognosis with 10-year survival currently at 14%. The extent of tumour excision is positively correlated with outcomes, however current neuro-navigation methods offer limited information at the tumour margin. Rapid Evaporative Ionisation Mass Spectrometry (REIMS) coupled with electrosurgery known as Intelligent Knife (iKnife) has been previously shown to provide near real-time tissue characterisation with accuracy, comparable to histopathology in different tumour types. Mass spectrometry imaging (MSI) methods provide the means to detect molecules in a spatially resolved manner and can provide insights into tumour metabolism. Laser Desorption Imaging – Rapid Evaporative Ionisation Mass Spectrometry (LDI-REIMS) is one of such MSI techniques that allow for analysis in ambient conditions.
OBJECTIVES: This study aims to build spatially resolved diagnostic ex vivo models using LDI-REIMS that can potentially be used for in vivo tissue diagnosis with the iKnife, hence potentially improving the extent of tumour resection.
METHODS: Fresh frozen brain tumours including grade IV glioblastoma, meningioma and metastatic tumour (lung adenocarcinoma origin) were obtained from patients (consented for research) following surgical resection. Tumours were sectioned (10 μm), mounted on Superfrost slides (Thermo Fisher Scientific, Waltham, USA) and stored at -80°C prior to analysis. A Xevo G2-S QToF (Waters, Wilmslow UK) mass spectrometer was used in negative mode with an Opolette HE2731 OPO laser (Opotek, Carlsbad, USA) at 2.9µm wavelength for the LDI-REIMS to achieve a 70 μm pixel size. Post analysis, slides were sent for H&E staining and histopathological validation. Data analysis was conducted with multivariate statistical methods including principal components analysis (PCA) and linear discriminant analysis (LDA)
RESULTS: Preliminary results reveal distinct classification of tumours with high accuracy, sensitivity, specificity on cross-validation. In addition, insights into intra-tumour heterogeneity were gained by observing the spatial distribution of different metabolites in parts of the tumours. Next steps include using a spectral identification algorithm based on the ex vivo LDI-REIMS models to characterize a novel sample. Furthermore, co-registration of single ion images with histologically annotated H&E images is also planned.
CONCLUSIONS: Although the sample size is limited, results are promising. Further analysis is warranted as well as expanding the dataset.