Raman-spectroscopy: diagnostic value, risk stratification and future perspectives in HPV-associated cervical lesions: a review

Abstract

Cervical cancer continues to rank among the leading causes of cancer morbidity and mortality in women worldwide, underscoring the urgent need for innovative non-invasive, objective, and highly sensitive diagnostic methodologies, particularly for the early identification of lesions associated with the human papillomavirus (HPV). The advancement of such diagnostic techniques aligns with the World Health Organization's global strategy to eradicate cervical cancer as a public health concern. This review consolidates current findings regarding the application of Raman spectroscopy, including its surface-enhanced modality, in the diagnosis and risk stratification of cervical pathology. The analysis revealed that spectral examination of various biological samples – such as cytological preparations of the cervix, blood serum, and cervicovaginal fluid – facilitates the differentiation of benign, precancerous, and malignant conditions based on biochemical alterations in the composition of proteins, nucleic acids, lipids, glycogen, and other metabolites. Noteworthy emphasis is placed on the potential of identifying molecular abnormalities in morphologically unchanged cells, as well as the use of cervicovaginal fluid as a local substrate for liquid biopsy. This development presents promising opportunities for minimally invasive screening and monitoring. The integration of machine learning algorithms and multivariate statistical analysis significantly enhances the diagnostic accuracy of this methodology. However, it is important to note that limited sample sizes and protocol heterogeneity constrain existing studies. Consequently, Raman spectroscopy shows considerable promise for the early diagnosis and risk stratification of human papillomavirus-related cervical lesions. Nevertheless, its clinical implementation requires standardization of methodology, validation of diagnostic models in large independent cohorts, and further investigation into the predictive value of spectral markers.