Abstract：
Several serum Raman spectroscopy (RS) studies have demonstrated its potential as an oral cancer screening tool. This study investigates influence of low tumour load (LTL) and high tumour load (HTL) on serum RS using hamster buccal pouch model of experimental oral carcinogenesis. Sera of untreated control, LTL, and HTL groups at week intervals during malignant transformation were employed. Serum Raman spectra were subjected to multivariate analyses-principal component analysis, principal component-based linear discriminant analysis (for stratification of study groups), and multivariate curve resolution-alternating least squares (MCR-ALS) (to comprehend biomolecular differences). Multivariate analysis revealed misclassifications between LTL and HTL at all week intervals. MCR-ALS components showed statistically significant abundances between control versus LTL and control versus HTL, but could not discern LTL and HTL. MCR-ALS components exhibited spectral mixtures of proteins, lipids, heme and nucleic acids. Thus, these findings support use of serum RS as a screening tool as varying tumour load is not a confounding factor influencing the technique.
摘要：
几项血清拉曼光谱（RS）研究已证明其作为口腔癌筛查工具的潜力。这项研究使用实验性口腔癌变的仓鼠颊袋模型研究了低肿瘤负荷（LTL）和高肿瘤负荷（HTL）对血清RS的影响。未经治疗的对照血清，LTL,在恶变期间每隔一周使用HTL组。血清拉曼光谱进行多变量分析-主成分分析,基于主成分的线性判别分析(用于研究组的分层)，和多变量曲线分辨率交替最小二乘（MCR-ALS）（以理解生物分子差异）。多变量分析显示在所有周间隔LTL和HTL之间存在错误分类。MCR-ALS组分在对照与LTL和对照与HTL之间显示出统计学上显著的丰度，但无法辨别LTL和HTL。MCR-ALS成分表现出蛋白质的光谱混合物，脂质，血红素和核酸。因此，这些发现支持使用血清RS作为筛查工具,因为不同的肿瘤负荷并不是影响该技术的混杂因素.
