Optical diagnosis of gallbladder cancers via two-photon excited fluorescence imaging of unstained histological sections

Abstract

Two-photon excited fluorescence (TPEF) microscopy, based on signal from cells, can provide detailed information on tissue architecture and cellular morphology in unstained histological sections to generate subcellular-resolution images from tissue directly. In this paper, we used TPEF microscopy to image microstructure of human normal gallbladder and three types of differentiated carcinomas in order to investigate the morphological changes of tissue structure, cell, cytoplasm, and nucleus without hematoxylin and eosin (H&E) staining. It displayed that TPEF microscopy can well image the stratified normal gallbladder tissue, including the mucosa, the muscularis, and the serosa. The typical cancer cell, characterized by cellular and nuclear pleomorphism, enlarged nuclei, and augmented nucleolus, can be identified in histological sections without H-E staining as well. The quantitative results showed that the areas of the nucleus and the nucleolus in three types of cancerous cells were all significantly greater than those in normal gallbladder columnar epithelial cells derived from TPEF microscopic images. The studies demonstrated that TPEF microscopy has the ability to characterize tissue structures and cell morphology of gallbladder cancers differentiated from a normal gallbladder in a manner similar to traditional histological analysis. As a novel tool, it has the potential for future retrospective studies of tumor staging and migration by utilizing histological section specimens without H-E staining.

Comparison of optically‐derived biomarkers in healthy and brain tumor tissue under one‐ and two‐photon excitation

“…The added advantage of interrogating tissue non‐invasively, and in real‐time, without the need of exogenous, contrast further motivates the development of a non‐linear multimodal endomicroscope for intraoperative pathological assessment in lieu of repeat biopsy sampling. In fact, previous studies have reported on the ability of trained pathologists to identify gallbladder , breast , and liver tumors from TPEF images with an accuracy of ~ 90%, while being blind to the histopathological assessment. In neuro‐oncology, untrained pathologists correctly identified GBM and metastatic tissue with a sensitivity and specificity of 50% and 75%, respectively, before training the neuropathologists .…”

“…This study, among others , shows that qualitative examination of TPEF images not only mirrors the histological features observed by the gold standard (anaplasia, necrosis, and increased cellular density), but rather provides additional information related to tissue architecture such as the (dis)organization of blood vessels and collagen structures surrounding tumor cells (Figure C‐E). The added advantage of interrogating tissue non‐invasively, and in real‐time, without the need of exogenous, contrast further motivates the development of a non‐linear multimodal endomicroscope for intraoperative pathological assessment in lieu of repeat biopsy sampling.…”

Multimodal Analysis of Central Nervous System Tumor Tissue Endogenous Fluorescence With Multiscale Excitation

“…Training neuropathologists on TPEF images could improve these results since this trial was a first time experience using this imaging modality. Blind analysis of TPEF-SHG images have never been performed before on brain tumor, but on other types of cancer, such as gallblader cancer [21], breast masses [22], and liver cancer [23]. In these analysis, they found even higher results than us with accuracy of discrimination around 90%, due to training of the neuropathologist on bigger sets of data.…”

Label‐free detection of invasive micropapillary carcinoma of the breast using multiphoton microscopy

“…As far as we know, the histopathological staining procedure takes a long time and may cause environmental pollution [16]. Excessive or insufficient staining may cause some problems, for instance, chromatic aberration or location error of cytoplasm and nucleoli, which will affect the contrast of imaging as well as the resolution and thus affect the judgment of disease [20].…”

“…The interaction between laser and these fluorescent substances leads to fluorescence emission, and the non-centrosymmetric structure of collagen fibers can produce SHG signal. Therefore, MPM can produce subcellular resolution images to show detailed tissue microstructure and cell morphology [19,20]. Importantly, in the past few decades, the relationship between collagen changes in tumor microenvironment and tumor progression has been widely researched based on multiphoton imaging: Chen et al investigated the association of collagen signature in the tumor microenvironment with lymph node metastasis (LNM) in early gastric cancer (EGC), and the results suggested that the collagen signature is an independent indicator of LNM in EGC [21]; Burke et al have used SHG imaging to directly observe the changes of collagen fibers during the cancer development [22], Natal et al revealed that the histological special types of invasive breast cancer can be identified by collagen parameters in SHG images [23]; and Keely et al have defined three tumor associated collagen signatures (TACS1-3) of breast cancer by SHG images, and found that TACS-3 was associated with a poorer prognosis [24,25].…”