Changes In Tumor Anatomy And Dosimetry During Mr-Guided Pulsed Dose Rate Brachytherapy For Cervix Cancer

MRI guidance in intracavitary pulsed-dose rate brachytherapy (PDR-BT) for cervix cancer facilitates optimization of tumor coverage and sparing of organs at risk (OARs) through adaptation of radiation delivery to tumor and patient anatomy. Daily MR imaging during PDR-BT was performed to evaluate changes in the target volumes and OARs of interest, and the impact of these anatomic changes on dosimetry. Thirty-four patients undergoing PDR-BT for stage IB-IVA cervix cancer were analyzed following pelvic radiotherapy (45-50 Gy in 1.8-2 Gy fractions) and concurrent chemotherapy. An MR-compatible applicator without colpostats was inserted under general anesthesia. An optimized MR-guided treatment plan (35-40 Gy) was developed according to GEC-ESTRO guidelines. PDR-BT was delivered over 3 days. Repeat MR imaging was performed on day 2 (29 patients) and day 3 (16 patients) of treatment. New plans were generated using the day 2 and 3 images and compared to day 1. On Day 1, the mean D90 (EQD2 from EBRT and BT) was 95 Gy (SD 10) for the HRCTV. For the OARs, the mean EQD2 to 2cc of bladder wall was 80 Gy (SD 10), 70 Gy (SD 12) for the rectal wall and 74 Gy (SD 8) for the sigmoid. The contoured HRCTV was significantly larger on day 3 compared to day 1. The D90 for the HRCTV decreased from day 1 to day 2 (p < 0.00003), day 2 to day 3 (p = 0.02) and day 1 to 3 (p87 Gy (as suggested by Potter et al) in 31/34 (91%) of patients on day 1, 20/29 (69%) on day 2 and 5/16 (31%) on day 3. Among patients with acceptable day 1 plans, the day 2 and day 3 HRCTV D90s were within 5% of the 87 Gy threshold 67% of the time, and within 10% of this value 94% of the time. Sparing of OARs and tumor coverage can be optimized using MR-guided BT planning. Changes in tumor and uterine size may occur over a 3 day course of PDR-BT because of edema, inflammation and imaging artifact leading to contouring uncertainty. This translates to small changes in tumor dosimetry. Correlation with long-term outcome will be necessary to determine clinical impact.