Raman Spectroscopy and Machine Learning Reveals Early Tumor Microenvironmental Changes Induced by Immunotherapy

Cancer immunotherapy provides durable clinical benefit in only a small fraction of patients, and identifying these patients is difficult due to a lack of reliable biomarkers for prediction and evaluation of treatment response. Here, we demonstrate the first application of label-free Raman spectroscopy for elucidating biomolecular changes induced by anti-CFLA4 and anti-PD-L1 immune checkpoint inhibitors (ICI) in the tumor microenvironment (TME) of colorectal tumor xenografts. Multivariate curve resolution-alternating least squares (MCR-AIS) decomposition of Raman spectral datasets revealed early changes in lipid, nucleic acid, and collagen content following therapy. Support vector machine classifiers and random forests analysis provided excellent prediction accuracies for response to both ICIs and delineated spectral markers specific to each therapy, consistent with their differential mechanisms of action. Corroborated by proteomics analysis, our observation of biomolecular changes in the TME should catalyze detailed investigations for translating such markers and label-free Raman spectroscopy for clinical monitoring of immunotherapy response in cancer patients. Significance: This study provides first-in-class evidence that optical spectroscopy allows sensitive detection of early changes in the biomolecular composition of tumors that predict response to immunotherapy with immune checkpoint inhibitors.

Automated summary

This study demonstrates the potential of label-free Raman spectroscopy for analyzing biomolecular changes induced by immune checkpoint inhibitors in the tumor microenvironment, providing a new approach for predicting response to immunotherapy.