Dose and Volume Parameters for MRI-Guided Cervix Cancer Brachytherapy in the Embrace Trial

Abstract

The international study on MRI-guided Brachytherapy in locally Advanced Cervical cancer (EMBRACE, www.embracestudy.dk) is a prospective observational study of treatment and clinical outcome data of 1416 patients with locally advanced cervical cancer from 23 centers, who were treated with radiochemotherapy (EBRT and image-guided brachytherapy (IGBT), between 2008 and 2016. EMBRACE data reflect the implementation of MRI-based IGBT through a variety of institutional protocols. The aim of this analysis was to report on the overall status of the implementation of this technique and to analyze dose-volume parameters according to GEC ESTRO recommendations. There were 1300 patients with complete treatment data records were included in this analysis. The distribution of implant types (intracavitary [ic] or intracavitary/interstitial [ic/is]), and dose rate schedules (high/pulsed dose rate [HDR/PDR]) was analyzed. Total treatment doses including EBRT and IGBT were converted to biologically equieffective dose in 2Gy fractions (EQD2) using the linear quadratic model, and dose reporting parameters for high-risk CTV (CTVHR) and organs at risk according to ICRU89 and GEC-ESTRO recommendations. A subgroup analysis was performed for different target volumes. Almost all patients were treated with EBRT of 45-50.4 Gy to the elective PTV, with 3D conformal (58%) or IMRT (42%) techniques. HDR BT (57%) was delivered mostly in 3 (20%), 4 (73%) or 5 (7%) fractions; PDR BT (42%) in 1 or 2 fractions (50%/50%). Most BT treatments were delivered with ring applicators (58%), followed by ovoids (27%), vaginal molds (12%) or cylinders (1%). IC applicators were exclusively used in 56% of patients, 44% were treated with ic/is only, or a combination of techniques. In ic/is implants the mean number of active needles per patient was 4.1±2.8. The mean overall treatment time for EBRT (including any sequential nodal boosts) and BT was 46.5±6.5 days. BT was administered after delivery of at least 45Gy of EBRT in 58% of patients, while 39% of BT treatments started after 30-45 Gy EBRT. The FIGO stage distribution was 19% 1B, 5% 2A, 52% 2B, 1% 3A, 15% 3B, 3% 4A, and 5% 4B. The median CTVHR volume was 28.7 cm3 (mean 34.5±21.5 cm3). The mean CTVHR D90 was 89.7±8.8 Gy EQD210. For ic treatments, the mean point A dose was 74.1±7.6 Gy EQD210. For ic and ic/is groups, with mean CTVHR volumes of 26.5 and 44.8 cm3, the mean CTVHR D90 were 90.5/88.7 Gy EQD210. Overall, for CTVHR volumes </> 30cm3, the mean D90 was 92.5/86.5 Gy EQD210, mean D2cm3 was 73.8/79.4 for bladder, and 61.1/64.3 Gy EQD23 for rectum. In total 76% of all patients received a CTVHR D90>85 Gy EQD210, 92% had a bladder D2cm3 <90 Gy EQD3, and 95% had a rectum D2cm3 <75Gy EQD3. The prospective EMBRACEII protocol will further promote the implementation of IGBT with ic/is technique. Further improvement of target coverage and organ sparing is expected.