OBJECTIVES: Human papillomavirus (HPV) has been implicated as a major factor in cervical carcinogenesis. However, many pieces of evidence gathered over the last two decades suggest Epstein-Barr virus (EBV) plays a secondary role in this process. The purpose of the present meta-analysis was to determine whether the presence of EBV infection increases the risk of cervical carcinoma. METHODS: Based on 25 articles, the analysis yielded a 33.44% overall pooled prevalence of EBV. RESULTS: The pooled prevalence was higher in patients with carcinoma (43.63%) than in healthy patients (19.0%) or patients with cervical intraepithelial neoplasia 1 (CIN1) (27.34%) or CIN2/3 (34.67%). Co-infection with EBV and HPV displayed a similar pattern. EBV infection was significantly and positively associated with lesion grade in cervical epithelia and was more prevalent in malignant lesions. Moreover, cervical carcinoma occurred four times as often among EBV positive women as in women without EBV infection (OR=4.01 [1.87-8.58]; p<0.001). CONCLUSIONS: The existence of EBV(+)HPV(-) carcinomas, the confirmed expression of latent oncoproteins (EBNA1, EBNA2, LMP1) and EBERs in tumor cells, and the association of EBV with the integration of high-risk-HPV DNA in malignant specimens point to EBV as a co-factor (so far underestimated) in the genesis and/or progression of cervical carcinoma. However, further studies are necessary before the link between EBV and cervical carcinoma can be established.
OBJECTIVES:Human papillomavirus (HPV) has been implicated as a major factor in cervical carcinogenesis. However, many pieces of evidence gathered over the last two decades suggest Epstein-Barr virus (EBV) plays a secondary role in this process. The purpose of the present meta-analysis was to determine whether the presence of EBV infection increases the risk of cervical carcinoma. METHODS: Based on 25 articles, the analysis yielded a 33.44% overall pooled prevalence of EBV. RESULTS: The pooled prevalence was higher in patients with carcinoma (43.63%) than in healthy patients (19.0%) or patients with cervical intraepithelial neoplasia 1 (CIN1) (27.34%) or CIN2/3 (34.67%). Co-infection with EBV and HPV displayed a similar pattern. EBV infection was significantly and positively associated with lesion grade in cervical epithelia and was more prevalent in malignant lesions. Moreover, cervical carcinoma occurred four times as often among EBV positive women as in women without EBV infection (OR=4.01 [1.87-8.58]; p<0.001). CONCLUSIONS: The existence of EBV(+)HPV(-) carcinomas, the confirmed expression of latent oncoproteins (EBNA1, EBNA2, LMP1) and EBERs in tumor cells, and the association of EBV with the integration of high-risk-HPV DNA in malignant specimens point to EBV as a co-factor (so far underestimated) in the genesis and/or progression of cervical carcinoma. However, further studies are necessary before the link between EBV and cervical carcinoma can be established.