AIMS--To investigate the correlation between antibodies to the transforming protein E7 of human papillomavirus (HPV) type 16 and clinicopathological indices in women with cervical squamous carcinoma. METHODS--A synthetic peptide of the HPV type 16 E7 protein (amino acids 6 to 35) was used to screen sera from 29 children, 130 women with cervical intraepithelial neoplasia, 443 women with cervical cancer, and 222 controls, for antibodies against this viral antigen. Bivariate and multivariate analyses were used to investigate the correlation between the serological status in the pretreatment sera and clinicopathological indices (size of the lesions, histological grade, stomal infiltration, vascular invasion, and nodal spread). Survival analysis was done using the Cox regression model for all FIGO stages and stages IB and ILA. RESULTS--Cervical carcinoma patients had a significantly higher prevalence of antibodies to synthetic peptide E7/6-35 than women with cervical intraepithelial neoplasia (17.7% v 7%, p < 0.005) or controls (17.7% v 11%, p < 0.05). Bivariate analysis of the data on the presence of anti-E7/6-35 antibodies in the pretreatment sera from these patients and clinicopathological indices showed a significant correlation between the presence of anti-E7/6-35 antibodies and the size of the lesion (p = 0.0009), histological grade (p = 0.0031), and lymph node metastasis (p = 0.01). 0.011). In addition, the Cox regression model, analysing four risk factors which can be determined before treatment, showed a significant correlation between the presence of anti-E7/6-35 antibodies and a worse prognosis (p = 0.003). Survival analysis revealed that both for all FIGO stages (p = 0.0005) and for stages IB and IIA alone (p = 0.0021), anti-E7/6-35 positive patients before treatment had a significantly shorter life expectancy. CONCLUSIONS--The presence of antibodies against E7/6-35 in pretreatment sera from patients with cervical carcinoma correlates with the size of the lesions, lymph node involvement, and a worse prognosis.