Raman spectroscopy for rapid diagnosis of lymphomas and metastatic lesions found in lymph nodes

Abstract

At least 50% of people will develop cancer at some point during their lifetime and half these will end in fatality. Improving patients’ prognosis relies on early and accurate diagnosis and treatment. Current diagnostic methods are based on histopathological analysis and are time-consuming, expensive and require biopsy. Raman spectroscopy can measure subtle biochemical changes and provides a rapid, non-destructive and objective technique that can be used in vivo for identifying pathological changes in tissue samples.

This study investigates both a standard Raman spectrometer system and also a Raman needle probe for their use as diagnostic techniques and clinical tools. Oesophageal, femoral and head and neck lymph nodes were analysed in this study. Metastatic lymph nodes from the three areas could be identified from the non-cancer lymph nodes with a sensitivity of 71% and specificity of 89%. Lymphoma was identified from non-cancer lymph nodes with a sensitivity of 64% and specificity of 86%. It was observed that oesophageal nodes often contained carbon particles, clinically diagnosed as anthracosis. These nodes were much harder to study than the femoral or head and neck, due to strong Raman signal detected from the carbon particles. Lymph nodes are embedded in adipose tissue and as a consequence, very strong lipid peaks were frequently observed in spectra. Spectral differences were exhibited in the measurements of the lymph nodes from the three different anatomical regions.

A comparison of the point measurements and mapped data showed no difference in classification. Therefore, indicating that just a few measurements can be sufficient enough sampling to represent a specimen, and demonstrates the practicability of Raman use in vivo for rapid analysis. The Raman needle probe feasibility study showed its potential for in vivo use for real-time diagnosis and as a surgical tool to support biopsy. A sensitivity and specificity of 80% and 79% for the identification of non-cancer head and neck lymph nodes from non-cancer provides similar accuracies to the standard Raman approach, therefore supports its viability for use as a diagnostic tool.