Optical diagnosis of actinic cheilitis by infrared spectroscopy

• Autor(a) principal: das Chagas e Silva de Carvalho, Luis Felipe
• Data de Publicação: 2016
• Outros Autores: Pereira, Thiago Martini, Magrini, Taciana Depra, Cavalcante, Ana Sueli Rodrigues [UNESP], da Silva Martinho, Herculano, Almeida, Janete Dias [UNESP]
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1016/j.pdpdt.2016.07.013; http://hdl.handle.net/11449/178285
• topic: Actinic cheilitis; FT-IR spectroscopy; Optical biopsy; Oral cancer

Actinic cheilitis (AC) is considered a potentially malignant disorder of the lip. Biomolecular markers study is important to understand malignant transformation into squamous cell carcinoma. Fourier transform infra red (FT-IR) spectroscopy was used to analyze AC in this study. Objectives The aim of the study was to evaluate if FT-IR spectral regions of nucleic acids and collagen can help in early diagnosis of malignant transformation. Methods Tissues biopsies of 14 patients diagnosed with AC and 14 normal tissues were obtained. FT-IR spectra were measured at five different points resulting in 70 spectra of each. Analysis of Principal components analysis (PCA) and linear discrimination analysis (LDA) model were also used. In order to verify the statistical difference in the spectra, Mann-Whitney U test was performed in each variable (wavenumber) with p-value <0.05. Results After the Mann-Whitney U test the vibrational modes of CO (Collagen 1), PO2 (Nucleic Acids) and C[dbnd]O asymmetric (Triglycerides/Lipids) were observed as a possible spectral biomarker. These bands were chosen because they represent the vibrational modes related to collagen and DNA, which are supposed to be changed in AC samples. Based on the PCA-LDA results, the predictive model corresponding to the area under the curve was 0.91 for the fingerprint region and 0.83 for the high wavenumber region, showing the greater accuracy of the test. Conclusions FT-IR changes in collagen and nucleic acids could be used as molecular biomarkers for malignant transformation.