错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Clinical confocal laser endomicroscopy for imaging of autofluorescence signals of human brain tumors and non-tumor brain

  • Marlen Reichenbach,
  • Sven Richter,
  • Roberta Galli,
  • Matthias Meinhardt,
  • Katrin Kirsche,
  • Achim Temme,
  • Dimitrios Emmanouilidis,
  • Witold Polanski,
  • Insa Prilop,
  • Dietmar Krex,
  • Stephan B. Sobottka,
  • Tareq A. Juratli,
  • Ilker Y. Eyüpoglu,
  • Ortrud Uckermann

摘要

Purpose

Analysis of autofluorescence holds promise for brain tumor delineation and diagnosis. Therefore, we investigated the potential of a commercial confocal laser scanning endomicroscopy (CLE) system for clinical imaging of brain tumors.

Methods

A clinical CLE system with fiber probe and 488 nm laser excitation was used to acquire images of tissue autofluorescence. Fresh samples were obtained from routine surgeries (glioblastoma n = 6, meningioma n = 6, brain metastases n = 10, pituitary adenoma n = 2, non-tumor from surgery for the treatment of pharmacoresistant epilepsy n = 2). Additionally, in situ intraoperative label-free CLE was performed in three cases. The autofluorescence images were visually inspected for feature identification and quantification. For reference, tissue cryosections were prepared and further analyzed by label-free multiphoton microscopy and HE histology.

Results

Label-free CLE enabled the acquisition of autofluorescence images for all cases. Autofluorescent structures were assigned to the cytoplasmic compartment of cells, elastin fibers, psammoma bodies and blood vessels by comparison to references. Sparse punctuated autofluorescence was identified in most images across all cases, while dense punctuated autofluorescence was most frequent in glioblastomas. Autofluorescent cells were observed in higher abundancies in images of non-tumor samples. Diffuse autofluorescence, fibers and round fluorescent structures were predominantly found in tumor tissues.

Conclusion

Label-free CLE imaging through an approved clinical device was able to visualize the characteristic autofluorescence patterns of human brain tumors and non-tumor brain tissue ex vivo and in situ. Therefore, this approach offers the possibility to obtain intraoperative diagnostic information before resection, importantly independent of any kind of marker or label.