Assessing changes to the brachytherapy target for cervical cancer using a single MRI and serial ultrasound

To assess changes to the brachytherapy target over the course of treatment and the impact of these changes on planning and resources.

Patients undergoing curative treatment with radiotherapy between January 2007 and March 2012 were included in the study. Intrauterine applicators were positioned in the uterine canal while patients were under anesthesia. Images were obtained by MRI and ultrasound at Fraction 1 and ultrasound alone at Fractions 2, 3, and 4. Cervix and uterine dimensions were measured on MRI and ultrasound and compared using Bland-Altman plots and repeated measures one-way analysis of variance.

Of 192 patients who underwent three fractions of brachytherapy, 141 of them received four fractions. Mean differences and standard error of differences between MRI at Fraction 1 and ultrasound at Fraction 4 for anterior cervix measurements were 2.9 (0.31), 3.5 (0.25), and 4.2 (0.27) mm and for posterior cervix 0.8 (0.3), 0.3 (0.3), and 0.9 (0.3) mm. All differences were within clinically acceptable limits. The mean differences in the cervix over the course of brachytherapy were less than 1 mm at all measurement points on the posterior surface. Replanning occurred in 11 of 192 (5.7%) patients, although changes to the cervix dimensions were not outside clinical limits.

There were small changes to the cervix and uterus over the course of brachytherapy that were not clinically significant. Use of intraoperative ultrasound as a verification aid accurately assesses the target at each insertion, reduces uncertainties in treatment delivery, and improves efficiency of the procedure benefiting both the patient and staff.