Identification and classification of prostate tumour tissues by FTIR

• Autor(a) principal: Fernandes, Inês Castro
• Data de Publicação: 2013
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/12648
• topic: Biomedicina; Cancro da próstata; Espectroscopia de infravermelhos

The development of prostate cancer (PCa) is not yet fully understood, but it is already recognized that it develops through pre malignant stages. Thus, the most important factor in the prevention of cancer is early detection, facilitating eradication of these pre-malignant cells prior to systemic invasion. There is a great clinical need for techniques that can detect these pre-malignant biochemical changes that can advance to cancer. The infrared (IR) spectroscopy is a powerful analytical tool to provide detailed information on the chemical composition of complex biological samples, such as tissues. This technique is based on the vibrations between the atoms of a molecule. The molecules absorb infrared light and give rise to distinctive spectral patterns in transmitted light. This is a non-destructive, highly sensitive and specific analytical technique that allows the identification of signals associated with prostate cancer at a biochemical level, providing an improved diagnosis of cancer and also facilitating the identification of tumors with aggressive biopotential. This dissertation is an exploratory study that intends to test the ability of the Fourier transform infrared spectroscopy (FTIR) to discriminate between control and PCa samples and to identify spectroscopic signals related to functional groups of biochemical compounds present in the malignant tissues. Prostate biopsy samples were collected from radical prostatectomy pieces of 8 patients characterized with elevated prostate specific antigen (PSA) and/or abnormal digital rectal examination (DRE). Also relevant histopathological information about the patients has been provided, including the Gleason grade and the TNM staging. PCa and control samples were analysed by FTIR and the resulting spectra were submitted to Principal Component Analysis (PCA) in the ranges 3000-2800 cm-1, 1200-900 cm-1 and 1800-1700 cm-1. The diagram of the factorial coordinates (scores) discriminated the spectra of PCa samples from the spectra of controls samples. The diagram of the factorial contributions (loadings) allowed the identification of spectroscopic signals associated with biochemical pathologic modifications expected in malignant tissues, mainly dysregulations in lipid metabolism, based on higher content in fatty acids and increased disorder of the chains of membrane lipids; also a lower polysaccharide (glucose) and glycogen content and an increase in the nucleic acids. In the future, further investigations would make possible the development of a universal FTIR multivariate classification model for prostate cancer.