A subset of biopsies (n = 7) from inside the primary tumor border were processed by frozen section for fluorescence microscopy. Frozen biopsies were collected from n = 4 low-dose patient specimens and n = 3 high-dose patient specimens. Frozen biopsies were cut serially into 8 μm frozen tissue sections and placed on glass microscopy slides for confocal fluorescence microscopy before histological staining. Frozen sections were wrapped in tin foil to limit light exposure and stored at − 80 °C until fluorescence microscopy was performed. Serial sections were cut from each biopsy for confocal microscopy, H&E, and additional staining, as required. Slides were imaged with the Nikon A1R resonance scanning confocal fluorescence microscope (Nikon Instruments Inc. Melville, NY, U.S.A.) using 405 nm excitation, a 525/50-nm emission filter for background tissue autofluorescence (AF) and a 600/50 + 685/70-nm emission filter for PpIX fluorescence. Resonance mode was used to scan frozen tissue sections rapidly (30 or 420 frames/second), to optimize image acquisition settings and minimize photobleaching of PpIX fluorescence. Fluorescence images were acquired using Galvano scanning at 1024 × 1024 pixels.
Immediately following fluorescence microscopy, frozen sections were fixed in formalin and stained with H&E. Two additional serial sections of the same sample were stained using standard protocols with Masson’s trichrome (MT) to identify connective tissues and Oil Red O (ORO) to identify adipocytes. This enabled spatial correlation between fluorescent tissue features in the frozen section and tissue-specific staining for histology. Nikon NIS Elements C software (Nikon Instruments Inc. Melville, NY, USA) was used to convert fluorescence microscopy images to TIF format and Aperio ImageScope was used to export corresponding images of H&E, MT, and ORO-stained tissues to JPEG image format. Fluorescence H&E, MT, and ORO-stained tissue section images were examined by the study pathologist (S.J.D.).
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.