Unité MéDIAN, UMR CNRS 6237-MEDyC, Université de Reims Champagne-Ardenne, IFR 53, UFR de Pharmacie, 51 rue Cognacq-Jay, 51096 Reims Cedex, France.
Anal Bioanal Chem. 2010 Sep;398(1):477-87. doi: 10.1007/s00216-010-3910-6. Epub 2010 Jun 26.
The potential of Raman spectroscopy for ex vivo and in vivo classification of normal and glioblastoma brain tumor development was investigated. High-quality spectra of normal and tumor tissues were obtained using a portable Raman spectrometer coupled to a microprobe with a signal integration time of 5 s. Ex vivo results demonstrated that by using the biochemical information contained in the spectra, we were able to distinguish between normal brain features (white and gray matter), invasion, and tumor tissues with a classification accuracy of 100%. Differences between these features resulted from variations in their lipid signal contributions, which probably reflect differences in the level of myelinization. This finding supports the ability of in vivo Raman spectroscopy to delineate tumor margins during surgery. After implanting C6 cells in rat brain, we monitored, in vivo, the development of glioblastoma tumor from days 0 to 20 post-implantation (PI). The classification exhibited a clear separation of the data into two clusters: one cluster was associated with normal brain tissues (cortex), and the second was related to data measured from tumor evolution. The second cluster could be divided into two subclusters, one associated with tumor tissue from 4 to 13 days PI and the second related to tumor tissue from 15 to 20 days PI. Histological analysis reveals that the differences between these two subclusters are: the presence of a massive infiltration zone in the brain tissue from 4 to 13 days PI, and; a maturation of the tumor characterized by the appearance of edematous and necrotic zones, as well as a diminution in the proliferative and invasive area, from 15 days. This work demonstrates the potential of Raman spectroscopy to provide diagnostic information for the early detection of tumors in vivo.
研究了拉曼光谱在离体和体内分类正常和胶质母细胞瘤脑肿瘤发展中的潜力。使用便携式拉曼光谱仪和带有 5 秒信号积分时间的微探针获得了正常和肿瘤组织的高质量光谱。离体结果表明，通过使用光谱中包含的生化信息，我们能够以 100%的分类准确率区分正常脑组织特征（白质和灰质）、浸润和肿瘤组织。这些特征之间的差异来自于其脂质信号贡献的变化，这可能反映了髓鞘化水平的差异。这一发现支持了在手术过程中使用体内拉曼光谱来描绘肿瘤边界的能力。在大鼠脑内植入 C6 细胞后，我们在植入后 0 至 20 天（PI）期间对胶质母细胞瘤肿瘤的发展进行了体内监测。分类结果清楚地将数据分为两个聚类：一个聚类与正常脑组织（皮层）相关，另一个聚类与从肿瘤演变中测量的数据相关。第二个聚类可以进一步分为两个子聚类，一个与植入后 4 至 13 天的肿瘤组织相关，另一个与植入后 15 至 20 天的肿瘤组织相关。组织学分析表明，这两个子聚类之间的差异在于：在植入后 4 至 13 天的脑组织中存在大量浸润区，以及；肿瘤的成熟特征为水肿和坏死区的出现，以及增殖和侵袭区的减少，从 15 天开始。这项工作表明了拉曼光谱在体内早期检测肿瘤方面提供诊断信息的潜力。
