Giovanna Melo Marques^1,2,3, Luana Quadros de Souza Leão^2,4, Adriano de Souza Carolino^2,5, Ana Carolina Rosa Silva⁶, Sofia Angiole Cavalcante³, Anderson Ferreira Gonçalves⁷, Marina Amaral Alves^8,9, Rafael Garrett da Costa^9,10, Priscila Ferreira de Aquino³

Introduction: The prevalence of cervical cancer in Brazil is an alarming public health issue, with high incidence and mortality rates among the female population, especially in the state of Amazonas, whose rates are higher than the national average. This carcinoma is preceded by cervical intraepithelial neoplasia (CIN), histopathologically classified by degree of dysplasia, with CIN 2 and CIN 3 considered high-grade cervical lesions. Although numerous prevention and treatment measures have been implemented to curb the spread of the disease, morbidity rates remain high in the state.  Thus, innovative approaches are required to understand CIN progression better and develop complementary early diagnosis strategies. The lipidomic approach may contribute to oncological studies by identifying potential lipid alterations during neoplasm development. Therefore, this study aims to investigate changes in the lipid profile present in the cervical tissue of women with high-grade precursor lesions of cervical cancer. Methodology: A total of 40 cervical tissue samples (lesion and adjacent margins) were collected from women diagnosed with high-grade cervical intraepithelial neoplasia who underwent surgical conization. The lipid profile was characterized by liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS). Data analysis was performed using multivariate statistical techniques developed in the R language, including Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA) and Heatmap Analysis. MS-DIAL software was used to identify differential lipids aided by the MS/MS Default library. Results and Discussion: The results revealed a similarity in the grouping of lesions and margins of both CINs, suggesting a possible recurrence of the disease even in histologically free margins. However, when grouping the lesion and margin samples by grade, a significant differentiation was found between the CIN 2 and CIN 3 tissue samples (AUC=0.99). This distinction allowed the identification of lipids with a greater impact on the differentiation of neoplasms. In addition, the heatmap graph detailed the levels of lipid expression, showing that lipids such as LPC 20:4 were abundant in CIN 2 samples, while PE 32:0 and TG 50:1 were more abundant in CIN 3. Triglycerides (TGs), the main energy reservoirs in eukaryotic cells, have been significantly associated with dyslipidemia — especially elevated levels — and the presence of high-grade intraepithelial lesions (HSIL). Phosphatidylethanolamines (PEs) and lysophosphatidylcholines (LPCs) are lipid groups that are fundamental for cell regulation and glycerophospholipid metabolism, which is considered the most relevant for tumorigenesis, as they act in the energy supply of atypical cells. Notably, PE 32:0 has been associated with early gastric cancer, and LPC 20:4 and TG 50:1 with lung cancer. Although direct reports linking altered levels of these specific lipids to CIN development are scarce, their established roles in cancer metabolism suggest their involvement in CIN progression. Conclusion: Therefore, this study demonstrated distinct lipid profiles between CIN 2 and CIN 3 tissues, identifying specific lipids for this differentiation. These findings may advance the understanding of the progression of high-grade intraepithelial neoplasia, highlighting the potential of combining mass spectrometry and bioinformatics to support monitoring and early diagnosis.

Agradecimentos: ILMD-Fiocruz Amazônia, UFAM, FCECON, LabMeta - (LADETEC-UFRJ), NANOPOL, PPSUS/FAPEAM/Secretary of State for Planning, Development, Science and Technology/Government of the State of Amazonas, CNPq and Capes.