The Role of Raman Spectroscopy in Modern Cancer Research

Because Raman spectroscopy is a non-invasive technique that can detect molecular changes, it has found an important niche in cancer research and diagnostics. By measuring the inelastic scattering of light from tissues or cells, Raman spectroscopy provides detailed biochemical fingerprints that can distinguish between healthy and cancerous states with high specificity.

Below is a compilation of some recent studies that explored the application of Raman spectroscopy in cancer research, detection, and diagnosis.

Application of Surface-Enhanced Raman Spectroscopy in Head and Neck Cancer Diagnosis

This review article highlights the role of surface-enhanced Raman spectroscopy (SERS) in advancing head and neck cancer (HNC) diagnostics and treatment. SERS enables highly sensitive, molecular-level detection for applications such as gene expression analysis, tumor microenvironment profiling, and single-molecule assays targeting biomarkers like EGFR and PD-1/PD-L1 (1). This review touches upon several aspects to the applicability of SERS in HNC diagnostics, including its integration with liquid biopsy techniques such as ctDNA and salivary diagnostics (1). The article also explores SERS's potential in monitoring therapy response and its synergy with AI and multiomics, positioning it as a potential tool in advancing the future of precision oncology.

Infrared and Raman Spectroscopy of Blood Plasma for Rapid Endometrial Cancer Detection

A recent study explored the use of attenuated total reflectance Fourier transform infrared (FT-IR) and Raman spectroscopies for non-invasive detection of endometrial cancer (EC) using both “wet” and dry blood plasma samples. The results from the study showed that Raman analysis of wet plasma achieved 82% accuracy, ATR-FTIR reached 78%, and combining both methods improved accuracy to 86% (2). Meanwhile, the researchers found that dry plasma analysis with ATR-FTIR yielded 83% accuracy (2). The study also identified spectral similarities between EC and polycystic ovary syndrome (PCOS), indicating potential diagnostic overlap. These findings suggest that IR and Raman spectroscopies could offer fast, accurate screening tools for EC, though further research is needed to confirm their clinical utility (2).

Preliminary Study Demonstrating Cancer Cells Detection at the Margins of Whole Glioblastoma Specimens with Raman Spectroscopy Imaging

A recent study explored the applicability of a new, whole-specimen Raman spectroscopic imaging system designed to detect cancer cells at surgical margins in glioblastoma patients. Unlike traditional point-probe methods, the new system covers a 1 cm² area with high-resolution, sub-millimeter pixels, enabling rapid, label-free molecular imaging (3). Created using point-probe data from 24 patients, this tumor detection model achieved 90% sensitivity and 95% specificity (3). Applied to nine specimens from five patients, the system successfully generated cancer prediction maps validated by histopathology (3). Detection relied on Raman signals from amino acids like phenylalanine and tryptophan, as well as lipid and protein markers, showing promise for intraoperative guidance.

Multi-cancer Early Detection Based on Serum Surface-enhanced Raman Spectroscopy with Deep Learning: A Large-scale Case–Control Study

A recent study proposes the utility of a new serum-based platform for early multi-cancer detection by combining SERS with advanced data analysis methods, including deep learning and feature dimensionality enhancement. Analyzing serum samples from 1,655 early-stage cancer patients and 1,896 healthy controls, researchers applied techniques such as continuous wavelet transform, residual neural networks (ResNet), and class activation mapping (CAM) (4). The model achieved high diagnostic performance, with accuracy up to 94.75% and AUC values nearing or exceeding 0.99 for most cancer types (4). These findings of this study show that SERS, when integrated with AI, can be effective for non-invasive pan-cancer screening in clinical settings.

Raman Spectroscopy in Tandem with Machine Learning-based Decision Logic Methods for Characterization and Detection of Primary Precancerous and Cancerous Cells

A recent study looked at how Raman spectroscopy and machine learning (ML) can be used to differentiate between normal and cancerous cells, including the precancerous stage. In the study, researchers analyzed mouse fibroblast cells that were normal, precancerous, and cancerous, identifying spectral markers linked to cancer progression using ANOVA-based feature selection and log-likelihood decision logic for improved accuracy (5). The researchers achieved classification accuracies of 95.8% (normal vs. cancerous), 91% (normal vs. precancerous), and 86% (precancerous vs. cancerous) (5). These findings highlight the potential of Raman spectroscopy as a non-invasive, precise tool for early cancer detection and understanding carcinogenesis at the cellular level.

References

1. Yang, B.; Dai, X.; Chen, S.; Li, C.; Yan, B. Application of Surface-Enhanced Raman Spectroscopy in Head and Neck Cancer Diagnosis. Anal. Chem. 2025, 97 (7), 3781–3798. DOI: 10.1021/acs.analchem.4c02796
2. Schiemer, R.; Grant, J.; Shafiee, M. N.; et al. Infrared and Raman Spectroscopy of Blood Plasma for Rapid Endometrial Cancer Detection. Brit. J. Can. 2025, ASAP. DOI: 10.1038/s41416-025-03-050-0
3. Daoust, F.; Dallaire, F.; Tavera, H.; et al. Preliminary Study Demonstrating Cancer Cells Detection at the Margins of Whole Glioblastoma Specimens with Raman Spectroscopy Imaging. Sci. Rep. 2025, 15, 6453. DOI: 10.1038/s41598-025-87109-1
4. Lin, Y., Zhang, Q., Chen, H. et al. Multi-cancer Early Detection Based on Serum Surface-enhanced Raman Spectroscopy with Deep Learning: A Large-scale Case–Control Study. BMC Med. 2025, 23, 97. DOI: 10.1186/s12916-025-03887-5
5. Sharaha, U.; Hania, D.; Bykhovsky, D.; et al. Raman Spectroscopy in Tandem with Machine Learning-based Decision Logic Methods for Characterization and Detection of Primary Precancerous and Cancerous Cells. Analyst 2025, ASAP. DOI:10.1039/D5AN00360A