Abstract
INTRODUCTION: Biobanked brain tumor specimens share a critical shortcoming: the inability to efficiently screen tissue specimens prior to banking, potentially leading to storage of poor-quality samples. Standard histologic approaches to evaluate cellularity of tissue, such as frozen sectioning, may alter molecular characteristics and damage tissue for downstream analyses. Here, utilizing a novel ex vivo approach, we evaluate the feasibility of interrogating tissue cellularity and cytoarchitecture with label-free confocal reflectance microscopy (CRM) prior to biobanking. METHODS: Biopsies from patients with intracranial neoplasms (n = 3 for each; glioma, meningioma, pituitary adenoma, and schwannoma) were transported to the pathology department and placed in ice-cold saline. Samples were immediately positioned on the stage of our pathology-based Zeiss LSM710 confocal microscope and visualized with CRM. CRM images were evaluated for tumor cellularity, architecture and morphology, and then compared to corresponding permanent sections. RESULTS: CRM could contrast cellularity and stroma in all samples, as compared to corresponding permanent sections. Additionally, CRM contrasted vasculature and regions of necrosis in glioblastomas, tumor architecture in meningiomas and schwannomas, and sheets of neoplastic cells in pituitary ademonas. DISCUSSION: We found confocal reflectance microscopy to be a simple approach to screening biospecimens prior to biobanking. CRM quickly generated distinct ex vivo microscopic images from biopsies without tissue processing or application of exogenous dyes. CRM contrasted histopathological features present in our samples, and produced images that could be digitally stored and recalled with the matched specimen in the biobank. In our previous studies, we reported CRM did not alter the molecular integrity of fresh biopsies from brain tumor animal models. We hypothesize that wider use of this technique could improve the quality of banked tissue.