Infrared spectra of basal cell carcinomas are distinct from non-tumor-bearing skin components

Citation
Lm. Mcintosh et al., Infrared spectra of basal cell carcinomas are distinct from non-tumor-bearing skin components, J INVES DER, 112(6), 1999, pp. 951-956
Citations number
23
Categorie Soggetti
Dermatology,"da verificare
Journal title
JOURNAL OF INVESTIGATIVE DERMATOLOGY
ISSN journal
0022202X → ACNP
Volume
112
Issue
6
Year of publication
1999
Pages
951 - 956
Database
ISI
SICI code
0022-202X(199906)112:6<951:ISOBCC>2.0.ZU;2-S
Abstract
Infrared spectroscopy, by probing the molecular vibration of chemical bonds , directly indicates tissue biochemistry, An expanding body of literature s uggests that infrared spectra distinguish diseased from normal tissue. The authors used infrared spectroscopy to examine basal cell carcinoma to explo re distinctive characteristics of basal cell carcinoma versus normal skin s amples and other skin neoplasms, Spectra of epidermis, tumor, follicle shea th, and dermis were acquired from unstained frozen sections, and analyzed q ualitatively, by t-tests and by linear discriminant analyses. Dermal spectr a were significantly different from the other skin components mainly due to absorptions from collagen in dermis, Spectra of normal epidermis and basal cell carcinoma were significantly different by virtue of subtle difference s in protein structure and nucleic acid content. Linear discriminant analys is characterized spectra as arising from basal cell carcinoma, epidermis, o r follicle sheath with 98.7% accuracy. Use of linear discriminant analysis accurately classified spectra as arising from epidermis overlying basal cel l carcinoma versus epidermis overlying non-tumor-bearing skin in 98.0% of c ases, Spectra of basal cell carcinoma, squamous cell carcinoma, nevi, and m alignant melanoma were qualitatively similar. Distinction of basal cell car cinoma, squamous cell carcinoma, and melanocytic lesions by linear discrimi nant analyses, however, was 93.5% accurate. Therefore, spectral separation of abnormal versus normal tissue was achieved with high sensitivity and spe cificity, which points to infrared spectroscopy as a potentially useful scr eening tool for cutaneous neoplasia.