RESUMO
PURPOSE: Current practice for diagnosing nasopharyngeal carcinoma (NPC) is based on invasive tissue biopsy. This study aims to explore the feasibility of using Raman spectroscopy to differentiate cancerous and non-cancerous nasopharyngeal tissue smears, expecting to realize minimal invasive diagnosis using smears from in vivo mucosa tissue by Raman spectroscopy. METHODS: Biopsy tissue smears were acquired from 74 patients with pathologically diagnosed nasopharyngeal diseases and measured using confocal Raman spectroscopy. RESULTS: Both fingerprint region and high wavenumber Raman spectra were acquired with distinguish features. Multivariate statistical analysis was used to differentiate cancerous and non-cancerous groups, achieving a diagnostic sensitivity of 87.2 and specificity of 85.7 % for differentiating NPC from nasopharyngeal non-cancerous smears. CONCLUSIONS: This work indicates that the method has a unique advantage in microanalysis for tissue smears which may provide a promising minimal invasive (or noninvasive) diagnosing tool for cancer diagnosis.