Vibrational Spectroscopy and Cancer - Vibrational Spectroscopy for Tissue Analysis - page 68

The main focus area of the comparative infrared spectroscopic study
of H. Fabian et al. was on human breast tumours, human breast tumour
cell lines, and xenografted human tumour cells. The results indicated that
substantial differences exist on a macroscopic level between the tumours,
tumour cell lines, and xenografted tumour cells, which are related to the
presence of a significant connective tissue matrix in the tumours. On a mac-
roscopic level tumour cell xenografts appear, in spectroscopic terms, to be
relatively homogenous with a relatively weak signature characteristic of
connective tissue. Differences on a microscopic level between adjacent small
(30-µm 2 ) areas of the same xenografted tumour could be detected, which
were due to local variations in collagen content. In addition to variations
in collagen content, variations in the deposition of microscopic fat droplets
throughout both human and xenografted tumours could be detected. The
results indicated the care with which infrared spectroscopic studies of tis-
sues must be carried out to avoid incorrect interpretation of results due to an
incomplete understanding of tissue pathology [79].
Tam et al. carried out a study on sample processing techniques for breast
cancer using Raman spectroscopy. Fifty breast biopsies were studied using
Raman spectroscopy prior to receipt of pathology reports. This was applied
to at least two of the three available tissue processing techniques using point
spectroscopy, mapping, and imaging. Differences in the spectra were related
to the various sample processing methods [80].
Kast et al. used Raman spectroscopy with near-infrared light excitation to
study normal breast tissue and tumours from 11 mice injected with a cancer
cell line [82]. Spectra were collected from 17 tumours, 18 samples of adjacent
breast tissue and lymph nodes, and 17 tissue samples from the contralateral
breast and its adjacent lymph nodes. Discriminant function analysis was
used for classification with principal component analysis scores as input
data. Discriminant function analysis and histology agreed on the diagno-
sis of all contralateral normal, tumour, and mastitis samples, except one
tumour that was found to be more similar to normal tissue. Normal tis-
sue adjacent to each tumour was examined as a separate data group and
classification of these tissues showed that some of them were diagnosti-
cally different from normal, tumour, and mastitis tissue. It was suggested
that this may reflect malignant molecular alterations prior to morphologic
changes, as expected in preneoplastic processes. Overall, it was concluded
that Raman spectroscopy not only distinguishes tumour from normal
breast tissue, but also defects early neoplastic changes prior to definite mor-
phologic alteration.
The main goal of the research carried out by Marzullo et al. was to inves-
tigate spectra of the borders of lesions of samples of infiltrating ductal car-
cinoma (IDC) and to determine the characteristics of these spectra [83]. For
this purpose, a total of 93 spectra were collected from five samples of healthy
tissues and from 13 samples of IDC breast tissues using FT-Raman spectros-
copy. Cluster analysis was used to separate the spectra into different groups.