Introduction Studies have shown that persistent infection by high risk human papillomavirus (HR-HPV), especially type 16, is responsible for cervical cancer development. For the development of a malignant phenotype, several mechanisms are involved, among them the process of epigenetic DNA methylation, which results in gene silencing, leading to abrogation of cell cycle control, escape from senescence, and induction of proliferation, that therefore can collaborate in carcinogenesis. Understanding how the epidemiological factors can influence the mechanism of methylation related to infection of the HR-HPV is an unclear situation to be unveiled. The objective of this study was to evaluate the infections caused by HPV16 in patients attended at Gynaecology Institute, of UFRJ, in the period between 2012 and 2013, considering aspects of methylation in host gene p^INK4a.

Methods The cervical smears were submitted to the detection of HPV DNA by Polimerase Chain Reaction (PCR) using primers MY09/11, genotyped with primers for the E6 gene of HPV16, evaluated the methylation of the host pINK4a gene by Nested-MSP technique, and collected epidemiological data, which included socio-demographic and behavioural factors of each patient; and, the data of the diagnostic test that served as reference, the cytopathology.

Results Our results showed nearly 76% of the studied samples presented some degree of gene p^INK4amethylated, while 24% were unmethylated. The results showed statistical significance for the correlation of the cytopathology with sexarca (p=0.05657), smoking (p=0.0317), pregnancies (p=5938e-05), parity (p=0.004425) and number of partners (p=0.0242). The p16 gene methylation correlations that showed statistical significance were pregnancy (p=0.02725), parity (p=0.01414),and typing (p=0.008121).

Conclusion We have found that the presence of HPV16 is associated with a greater chance of methylation of p16 in the lesions. Although, we also observed that the methylation of p16 gene is increasingly accompanying the severity of lesions, but without statistical relevance.