Derivation of a subtype-specific biochemical signature of endometrial carcinoma using synchrotron-based Fourier-transform infrared microspectroscopy.

Kerns, Jemma and Singh, Maneesh N. and Stringfellow, Helen F. and Walsh, Michael J. and Nicholson, James M. and Bahrami, Fariba and Ashton, Katherine M. and Pitt, Mark A. and Martin-Hirsch, Pierre L. and Martin, Frank L. (2009) Derivation of a subtype-specific biochemical signature of endometrial carcinoma using synchrotron-based Fourier-transform infrared microspectroscopy. Cancer Letters, 274 (2). pp. 208-217. ISSN 0304-3835

Abstract

Endometrial carcinoma consists of endometrioid (type I) and serous papillary (SP; type II) subtypes; a rarer form is malignant mixed müllerian tumours (MMMT; type II/mixed). We set out to determine whether one might be able to biochemically signature these subtypes using Fourier-transform infrared (FTIR) microspectroscopy and distinguish non-tamoxifen associated from tamoxifen-associated cases. Paraffin-embedded blocks were obtained from non-tamoxifen associated cases reported as endometrioid (n = 7), SP (n = 4) or MMMT (n = 4). From tamoxifen-associated cases, endometrioid (n = 1), SP (n = 3) and MMMT (n = 4) blocks were retrieved; benign tissues (n = 3) were also analysed. Exploiting synchrotron-based radiation, sections (10-μm thick) on BaF2 windows were interrogated through a 10 μm × 10 μm aperture. Point spectra were derived from 10 locations in each of six glandular elements per tissue; a further 20 stromal spectra were obtained. Following normalisation to Amide I, average spectra (1800–900 cm−1) per gland or stroma were analysed for variance using principal component analysis (PCA) and linear discriminant analysis (LDA). In scores plots, segregation of spectra from different subtypes or benign tissues was noted and it proved possible to distinguish tamoxifen-associated cases. In the PCA-LDA loadings plots, the wavenumbers that highlighted variance for benign or endometrioid carcinoma tissues were in the protein region (1800–1480 cm−1) whereas those contributing most to SP or MMMT segregation were primarily in the DNA/RNA region (1425–900 cm−1) of the vibrational spectrum. Our results suggest that the application of FTIR microspectroscopy is a powerful new approach in disease diagnosis and characterisation.