Application of serum Raman spectroscopy in rapid and early discrimination of aplastic anemia and myelodysplastic syndrome
Haoyue Liang,Xiaodong Kong,Yansong Ren,Haoyu Wang,Ertao Liu,Fanfan Sun,Guoqing Zhu,Qiang Zhang,Yuan Zhou
DOI: https://doi.org/10.1016/j.saa.2023.123008
2023-12-05
Abstract:Background: Raman spectroscopy of hematological diseases has gained attention from various researchers. However, serum analysis of bone marrow failure (BMF), represented by aplastic anemia (AA) and myelodysplastic syndromes (MDS) has not been fully investigated. In this study, we aimed at establishing a simple, non-invasive serum detection method for AA and MDS. Method: Serum samples from 35 AA patients (N = 35), MDS patients (N = 25), and control volunteers (N = 23) were systematically analyzed via laser Raman spectroscopy, and orthogonal partial least squares discrimination analysis (OPLS-DA). Then, discrimination models between the BMFs and control were constructed and evaluated using the prediction set. Results: Compared to control volunteers, serum spectral data for BMF patients were specific. The intensities of Raman peaks representing nucleic acids (726, 781, 786, 1078, 1190, 1415 cm-1), proteins (1221 cm-1), phospholipid/cholesterol (1285 cm-1), and β-carotene (1162 cm-1) significantly decreased, while the intensity of lipids (1437 and 1446 cm-1) significantly increased. Intensities of Raman peaks representing nucleic acids (726 cm-1) and collagen (1344 cm-1) in the AA group were significantly lower than in the control group. Intensities of Raman peaks representing nucleic acids (726 and 786 cm-1), proteins (1003 cm-1), and collagen (1344 cm-1) in the MDS group were significantly lower than those of the control group. The intensity of Raman peaks representing lipids (1437 and 1443 cm-1) in the MDS group was significantly higher than in the control group. Patients with AA and MDS exhibited increased serum triglyceride levels and decreased high-density lipoprotein levels. Conclusions: The relationship between serological test data for patients and typing of AA and MDS provides essential information for rapid and early identification of BMF. This study shows the potential of Raman spectroscopy for non-invasive detection of different BMF types.