This study is aimed at searching for an informative predictor of the clinical outcome of cervical cancer (CC) patients. The study included 135 patients with locally advanced cervical cancer (FIGO stage II-III) associated with human papillomavirus (HPV) 16/18 types or negative status of HPV infection. Using logistic regression, we analyzed the influence of the treatment method, clinical and morphological characteristics, and the molecular genetic parameters of HPV on the disease free survival (DFS) of patients treated with radiotherapy or chemoradiotherapy. Multivariate analysis revealed three factors that have prognostic significance for DFS, i.e., HPV-related biomarker (HPV-negativity or HPV DNA integration into the cell genome) (OR = 9.67, p = 1.2 × 10-4), stage of the disease (OR = 4.69, p = 0.001) and age (OR = 0.61, p = 0.025). The predictive model has a high statistical significance (p = 5.0 × 10-8; Nagelkirk's R2 = 0.336), as well as sensitivity (Se = 0.74) and specificity (Sp = 0.75). Thus, simultaneous accounting for the clinical and molecular genetic predictors (stage of the disease, patient age and HPV-related biomarker) makes it possible to effectively differentiate patients with prognostically favorable and unfavorable outcome of the disease.
Keywords: cervical cancer; chemoradiotherapy; human papillomavirus; prognosis; radiotherapy.