Abstract
INTRODUCTION: Raman spectroscopy is a low cost non-destructive and non-invasive technique showing great potential for the diagnosis of cancer. Biobanks are generally comprised of formalin fixed paraffin preserved (FFPP) tissue used in spectroscopic investigations. In order to carry out spectroscopic investigations the tissue needs to undergo a dewaxing process which removes and reduces paraffin contributions in the spectra. The ineffective removal of paraffin from the tissue and substrate contributions can affect the results and quality of the spectra. The aim of this study was to investigate the efficacy of tissue pre-processing steps used in spectral histopathology and the influence of commonly used substrates. METHOD: Nitrocellulose membrane filters of varying density, 8.0 µM, 5.0 µM and 1.2 µM were used to model human tissue. Filters underwent histological processing as clinical biopsies. Sections from each filter were cut at 10 microns and placed on three substrates, CaF2, Low-E and Glass which were subsequently dewaxed using two clearing agents, Histoclear and Xylene in three different bath times, 5 min, 10 min and 15 min. Spectra was collected from three biological replicates of each filter using Raman Spectroscopy with a 785 nm laser. RESULTS: CaF2 had the least background noise; however Glass and Low-E retained the least amount of paraffin whilst Xylene completely removed paraffin under the same bath conditions as Histoclear. CONCLUSION:Glass and Low-E are relatively cheaper and also provide reproducible results. These substrates can allow us to use standard clinical preparation procedures. Whilst Xylene is more effective in paraffin removal it is a hazardous chemical.