Investigating the use of Raman spectroscopy as a histopathological tool to identify metastatic brain tumours and their sites of origin

Fullwood, Leanne (2013) Investigating the use of Raman spectroscopy as a histopathological tool to identify metastatic brain tumours and their sites of origin. Masters thesis, University of Central Lancashire.

[thumbnail of Thesis document]
Preview
PDF (Thesis document) - Accepted Version
Available under License Creative Commons Attribution Non-commercial Share Alike.

7MB

Abstract

It is reported that 13,000 people in the UK are diagnosed with tumours in the brain every year, of which 60% are metastatic. Current methods for diagnosing the disease can be subjective, invasive and have long diagnostic windows. Raman spectroscopy provides a non-destructive, non-invasive, rapid and economical method for diagnosis.

The aim of this study was to assess the use of Raman and immersion Raman spectroscopy for diagnosing metastatic brain and glioblastoma multiforme tumours and identifying primary sites of origin, and investigate the substrate effect on sample preparation and resultant spectra.

The tissue specimens used in this study were formalin fixed-paraffin preserved and were supported on spectroscopic substrates for analysis. Samples were dewaxed prior to analysis to reduce/eliminate the paraffin contributions in the Raman spectra. The substrate was shown to have a significant influence on this dewaxing procedure and thus resulting spectra. It was also observed that specimens on CaF2 and Spectrosil quartz retained paraffin after dewaxing, whereas specimens on Low-E substrates did not.

Through data examination, the 721 cm-1 and 782 cm-1 peaks were identified as being the most distinct peaks for discriminating between glioblastoma multiforme, metastatic and normal brain tissue spectra. A ratio score plot of these peaks determined classification sensitivities and specificities as 100% and 94.44% for glioblastoma multiforme, 96.55% and 100% for metastatic brain, and 85.71% and 100% for normal brain tissue respectively. Cancerous tissue was observed to retain more wax than normal tissue. This difference in dewaxing efficiency was attributed to alterations in tissue density between the histological types. Principle component-discriminant function analysis revealed separation between metastatic sites: breast, lung, melanoma, colon/rectum and oesophagus and stomach, showing the potential of Raman spectroscopy to identify primary sites of origin from metastatic brain tissue. Overall, this study demonstrated the diagnostic ability of Raman spectroscopy and the importance of substrate influence on tissue preparation and the quality of spectra.


Repository Staff Only: item control page