Detection of human brain tumor infiltration with quantitative stimulated Raman scattering microscopy

Abstract

Differentiating tumor from normal brain is a major barrier to achieving optimal outcome in brain tumor surgery. New imaging techniques for visualizing tumor margins during surgery are needed to improve surgical results. We recently demonstrated the ability of stimulated Raman scattering (SRS) microscopy, a non-destructive, label-free optical method, to reveal glioma infiltration in animal models. Here we show that SRS reveals human brain tumor infiltration in fresh, unprocessed surgical specimens from 22 neurosurgical patients. SRS detects tumor infiltration in near-perfect agreement with standard hematoxylin and eosin light microscopy (κ=0.86). The unique chemical contrast specific to SRS microscopy enables tumor detection by revealing quantifiable alterations in tissue cellularity, axonal density and protein:lipid ratio in tumor-infiltrated tissues. To ensure that SRS microscopic data can be easily used in brain tumor surgery, without the need for expert interpretation, we created a classifier based on cellularity, axonal density and protein:lipid ratio in SRS images capable of detecting tumor infiltration with 97.5% sensitivity and 98.5% specificity. Importantly, quantitative SRS microscopy detects the spread of tumor cells, even in brain tissue surrounding a tumor that appears grossly normal. By accurately revealing tumor infiltration, quantitative SRS microscopy holds potential for improving the accuracy of brain tumor surgery.

Quantitative comparison of 3D third harmonic generation and fluorescence microscopy images

“…The boundaries of detected bright objects (THG) and nuclei (fluorescence) are shown in green and red, respectively. helpful to figure out the best interpretation of other novel imaging modalities such as SRS microscopy [20,21], where such a quantitative comparison is currently lacking.…”

“…Such a qualitative link may verify that certain structures can be detected in both images, but remains indirect because it does not guarantee that each observed object or even the major part of it, indeed corresponds to, e. g., a brain cell. If the quantitative link between these novel images and fluorescence/H&E stained images can be established, it will facilitate and verify the interpretation of these novel images, and further strengthen their potential for real-time pathology of human brain tumors [6,21,24,25].…”

“…Like other multi-photon techniques [10][11][12][13][14][15][16][17][18], THG not only enables the recording of label-free images of unfixed 3D volumes of tissue [1][2][3][4][5][6][7][8][9][10] free from spatial distortion artifacts inherent to histopathology, but also potentially allows feedback on the nature of the tissue, i. e. whether it is healthy or tumorous, to the surgeon during surgery, as the relative speed of the imaging modalities approaches 'real' time, and no preparation steps of the tissue are required [5,6,[19][20][21][22][23]. THG was shown recently to yield label-free images of ex-vivo human tumor tissue of histopathological quality, in real-time [6].…”

“…THG images of structurally normal and tumor tissues have been compared, not one-toone, with histopathological images of the same samples stained with hematoxylin and eosin (H&E) [6]. Similarly, the interpretation of other label free imaging techniques, such as simulated Raman scattering (SRS) microscopy and optical coherence tomography (OCT), has been qualitatively linked to H&E stained images [20,21,24].…”

“…The active contour models [33][34][35][36][37][38][39][40][41][42][43] became popular for automatic cell/nuclei segmentation at the beginning of this century when the classical CV model [33] was proposed. THG images, with Raman and other nonlinear microscopy images, differ from labeled-fluorescence images in their complexity, inherent to their high information density [5,6,20,21,24,25,[44][45][46][47]62]. A limited number of segmentation methods has been explored on THG images of several tissue types.…”