Integrated pre-surgical diagnostic algorithm to define extent of disease in cervical cance

Objectives Survival of patients with cervical cancer is strongly associated with the local extent of the primary disease. The aim of the study was to develop an integrated diagnostic algorithm, including ultrasonography (USG), magnetic resonance imaging (MRI), and examination under anesthesia, to define the local extent of disease in patients with newly diagnosed cervical cancer. Methods Patients with biopsy proven cervical cancer who underwent primary surgery from January 2013 to December 2018 in four participating centers were recruited. Patients who underwent USG, MRI, and examination under anesthesia prior to surgery were included in the study. Those for whom complete data were not available were excluded. Data regarding tumor size, parametrial invasion, and vaginal involvement obtained by USG, MRI, and examination under anesthesia were retrieved and compared with final histology. Specificity and sensitivity of the three methods were calculated for each parameter and the methods were compared with each other. An integrated pre-surgical algorithm was constructed considering the accuracy of each diagnostic method for each parameter. Results A total of 79 consecutive patients were included in the study. Median age was 53 years (range 28–87) and median body mass index was 24.6 kg/m2 (range 16–43). Fifty-five (69.6%) patients had squamous carcinoma, 18 (22.8%) patients had adenocarcinoma, and six (7.6%) patients had other histological subtypes. A statistically significant difference among the three methods was found for detecting tumor size (p=0.002 for tumors >2 cm and p=0.006 for tumors >4 cm) and vaginal involvement (p=0.01). There was no difference in detection of parametrial invasion between USG, MRI, and examination under anesthesia (p=0.26). Furthermore, regarding tumor size assessment, USG was found to be the significantly better method (p<0.01 for tumors >2 cm and p=0.02 for tumors >4 cm). Examination under anesthesia was the most accurate method for detection of vaginal involvement (p=0.01). Examination under anesthesia and MRI had higher accuracy than USG for identification of parametrial invasion. Application of the algorithm provided the correct definition of local extent of disease in 77.2% of patients (p=0.04). USG was the most accurate method to determine tumor size, while examination under anesthesia was found to be more accurate in prediction of vaginal involvement. Conclusion Our integrated diagnostic algorithm allows a higher accuracy in defining the local extent of disease and may be used as a tool to determine the therapeutic approach in women with cervical cancer.