Studying the pathological and biochemical features in breast cancer progression by confocal Raman microspectral imaging of excised tissue samples

J Photochem Photobiol B. 2021 Sep:222:112280. doi: 10.1016/j.jphotobiol.2021.112280. Epub 2021 Aug 5.

Abstract

Confocal Raman microspectral imaging (CRMI) has been used to detect the spectra-pathological features of ductal carcinoma in situ (DCIS) and lobular hyperplasia (LH) compared with the heathy (H) breast tissue. A total of 15-20 spectra were measured from healthy tissue, LH tissue, and DCIS tissue. One-way ANOVA and Tukey's honest significant difference (HSD) post hoc multiple tests were used to evaluate the peak intensity variations in all three tissue types. Besides that, linear discrimination analysis (LDA) algorithm was adopted in combination with principal component analysis (PCA) to classify the spectral features from tissues at different stages along the continuum to breast cancer. Moreover, by using the point-by-point scanning methodology, spectral datasets were obtained and reconstructed for further pathologic visualization by multivariate imaging methods, including K-mean clustering analysis (KCA) and PCA. Univariate imaging of individual Raman bands was also used to describe the differences in the distribution of specific molecular components in the scanning area. After a detailed spectral feature analysis from 800 to 1800 cm-1 and 2800 to 3000 cm-1 for all the three tissue types, the histopathological features were visualized based on the content and structural variations of lipids, proteins, phenylalanine, carotenoids and collagen, as well as the calcification phenomena. The results obtained not only allowed a detailed Raman spectroscopy-based understanding of the malignant transformation process of breast cancer, but also provided a solid spectral data support for developing Raman based breast cancer clinical diagnostic techniques.

Keywords: Breast tissue; Cancer progression; Confocal Raman microspectral imaging; Ductal carcinoma in situ; Lobular hyperplasia.

MeSH terms

  • Breast Neoplasms / chemistry
  • Breast Neoplasms / metabolism
  • Breast Neoplasms / pathology*
  • Cluster Analysis
  • Discriminant Analysis
  • Disease Progression
  • Female
  • Humans
  • Microscopy, Confocal / methods*
  • Principal Component Analysis
  • Spectrum Analysis, Raman*