Analysis of Collagen Presence in Brain Tumors by Raman Spectroscopy
Antipova E¹, Romanishkin I.D.², Savelieva T.A.¹,², Ryabova A.V.¹,²
¹ National Research Nuclear University "MEPhI", Moscow, Russia
² Prokhorov General Physics Institute of the Russian Academy of Sciences, Moscow, Russia
Abstract
During cancer progression, excess collagen increases the stiffness of the extracellular matrix (ECM), which promotes metastasis, impedes drug penetration and reduces treatment efficacy [1,2]. A detailed study of collagen as a promising biomarker of malignancy is highly relevant for complementing histopathology in accurate diagnosis and treatment planning.
In this study, it is proposed to employ Raman spectroscopy to analyze collagen in brain tumors and, for this purpose, a retrospective analysis of spectral data obtained ex vivo from six glioblastoma samples, five meningioma samples, and five healthy brain tissue samples is performed, each at 1-4 points, with 2-3 spectra at each point.
Raman scattering was excited using a Ramulaser-785 laser (StellarNet, USA) at a wavelength of 785 nm and a power of 150 mW, and was detected using a Raman-HR-TEC-785 fiber-optic spectrometer (StellarNet, USA). All spectra in the series were averaged, and the baseline was subtracted. The spectra were then smoothed using a Savitzky–Golay filter in the range of 800–1700 cm−1, and the background fluorescence signal was subtracted using the Vancouver algorithm. The spectra were processed using the MATLAB software package (MathWorks, R2023).
Using an automatic search algorithm, characteristic collagen peaks were identified in glioblastoma, meningioma, and normal tissues, and statistically significant differences were found between these data groups. The method involves approximating the spectra as a sum of Gaussian peaks, clustering the peak centers to account for their shifts, calculating the areas, and subsequent ANOVA analysis. The algorithm identified peaks with p < 0.05 and ranked them by significance.
As a result of the analysis, spectral peaks were found that provided statistically significant differences between groups corresponding to different diagnoses (p < 0.05). Among the protein peaks:
The peak at 1042.6 cm−1 (C-O and C-N bonds in the protein) in the meningiomas group showed high values (27.51 ± 1.28), in the glioblastoma and normal tissue groups the values were comparable (22.40 ± 1.09 and 22.7583 ± 0.72, respectively);
The peak at 1207.7 cm−1, corresponding to ν(C-C6H5) vibrations in tyrosine, hydroxyproline (collagen amino acids) and phenylalanine molecules, showed higher values in the meningiomas group (36.07 ± 8.00) compared to glioblastomas (22.44 ± 3.53) and normal tissue (22.37 ± 1.92);
The peak 1238.0 cm−1 (Amide III, reflecting the secondary structure of proteins) was significantly reduced in meningiomas (6.64 ± 0.98) relative to glioblastomas (19.70 ± 3.32) and normal (19.97 ± 4.89);
The peak at 1616.1 cm−1 (aromatic amino acids, protein) was highest in meningiomas (17.40±4.07), while in glioblastomas and normal its values were (11.56 ± 1.66 и 10.67±1.05)respectively.
The analysis revealed characteristic spectral differences between different types of tumors.
Funding
This work was supported by the Ministry of Science and Higher Education of the Russian Federation (Agreement No. 075-15-2025-559 dated June 11, 2025)
References
1. Mohiuddin E., Wakimoto H. Extracellular matrix in glioblastoma: opportunities for emerging therapeutic approaches // American Journal of Cancer Research. — 2021. — Vol. 11, No. 8. — P. 3742–3754. — PMCID: PMC8414390. — PMID: 34522446.
2. Das A., Tan W.-L., Smith D. R. Expression of extracellular matrix markers in benign meningiomas // Neuropathology. — 2003. — Vol. 23. — P. 275–281.
Speaker
Antipova E
National Research Nuclear University "MEPhI", Moscow, Russia
Russia
Discussion
Ask question
