In 2016, there were an estimated 56 870 new cases of thyroid cancer (TC) in the USA. Fine needle aspiration cytology (FNAC) is the most safe, accurate and cost-effective method for the initial investigation of thyroid nodules. FNAC is limited by the inability to diagnose malignancy in follicular-patterned lesions accurately and, as a result, 20%-30% of cases under investigation for TC are classified as cytologically indeterminate, illustrating a problem with current FNAC procedure. Raman spectroscopy has shown promising results for the detection of many cancers; however, to date there has been no report on the performance of Raman spectroscopy on thyroid cytological samples. The aim of this study was to examine whether Raman spectroscopy could be used to correctly classify cell lines representing benign thyroid cells and various subtypes of TC.
A benign thyroid cell line and seven TC cell lines were prepared as ThinPrep® cytology slides and analysed with Raman spectroscopy. Principal components analysis and linear discriminant analysis were implemented to develop effective diagnostic algorithms for classification of Raman spectra of different TC subtypes.
The spectral differences separating benign and TC cell lines were assigned to differences in the composition of nucleic acids, lipids, carbohydrates and protein in the benign and cancer cells. Good sensitivities (74%-85%), specificities (65%-93%) and diagnostic accuracies (71%-88%) were achieved for the identification of TC.
These findings suggest that Raman spectroscopy has potential for preoperative TC diagnosis on FNAC samples.