摘要：The tumor microenvironment can be spatially heterogenous, which makes it challenging to fully characterize with standard 2D histology-based methods. In this study, we determined the feasibility of a CLARITY tissue-processing approach to analyze biopsies from breast cancer patients. Formalin-fixed human breast cancer core-needle biopsy specimens, were embedded, lipid-cleared, and multiplexed immunostained to identify key biomarkers (pan-cytokeratin, Ki67, CD3). Confocal microscopy was then used to image the specimens after refractive index matching. These data sets were then quantitatively compared to conventional slide-based FFPE histology. Using CLARITY, the gross and cellular morphology of the tissues were well preserved, and high optical transparency was achieved, with the exception of fibrotic regions. Specific staining of various cellular and nuclear markers was achieved using optimized antibody conditions. Manually determined composite Ki67 scores from the CLARITY datasets agreed with histology results. However, the CLARITY datasets (3D) revealed variation in the intra-tumoral Ki67 expression that was not evident in individual FFPE sections (2D). We further demonstrated that archived FFPE clinical specimens can be CLARITY-processed, immunostained, and imaged. In short, CLARITY-processed specimens may enable a more accurate, unbiased analysis of tumor samples in comparison to conventional slide-based histology, thus allowing for improved visualization of intra-tumoral heterogeneity.
