Comparing Dosimetry of Locally Advanced Cervix Cancer Patients Treated with 3 vs. 4 Fractions of MRI-Guided Brachytherapy

Abstract

<h3>Purpose/Objective(s)</h3> MR-guided BT (MRgBT) with plan optimization enables dose escalation to the target and dose reduction to organs at risk (OAR), thereby improving patient outcomes. Worldwide, 28 Gy/4 fractions (F) is one of the most common MRgBT fractionations. There are limited data on a less resource-intensive 3-fraction regimen of 24 Gy/3F and its ability to meet EMBRACE II planning aims. This study compares target and OAR dosimetry of patients treated with 28 Gy/4F vs 24 Gy/3F MRgBT. <h3>Materials/Methods</h3> This was a retrospective review of 224 patients with stage IB-IVA cervix cancer treated with radical chemoradiation at a single center, where MRgBT fractionation transitioned from 28 Gy/4F (n = 91) to 24 Gy/3F (n = 133) gradually between 2016-2021. MR imaging and brachytherapy planning were performed for each fraction. The following were collected: stage, tumor size at diagnosis, dose fractionation, minimum equivalent dose in 2-Gy fractions to 98% of residual gross tumor volume (GTV_res D_98%), 90% of high-risk clinical target volume (CTV_HR D_90%), 98% of intermediate-risk (CTV_IR D_98%), 2 cm³ (D_2cm3) of sigmoid, rectum, small bowel, and bladder, and ICRU rectovaginal (RV) point. Continuous and categorical variables were compared using two-sided Wilcoxon's rank sum test and Fisher's exact test, respectively. Multivariable linear regression models were fitted to compare dosimetric parameters between the 4F and 3F group, adjusting for CTV_HR volume and T stage. <h3>Results</h3> The majority of patients had squamous cell carcinoma (86%), T2b disease (62%), and were treated with interstitial needles in addition to an intracavitary applicator (96%). There were no significant differences between those treated with 28 Gy/4F vs. 24 Gy/3F in histology, T stage, tumor size at diagnosis, number of interstitial needles, GTV_res D_98%, CTV_HR D_90%, sigmoid and small bowel D_2cm3. The 28 Gy/4F group had higher CTV_HR volume (median 28 vs 26 cm³, p = 0.04), CTV_IR D_98% (mean 65.5 vs 64.5 Gy, p = 0.03), rectum D_2cm3 (mean 61.7 vs 59.2 Gy, p = 0.04) and bladder D_2cm3 (81.3 vs 77.9 Gy, p = 0.03). There were no significant differences in the proportion of patients meeting the EMBRACE II dose limits between 28 Gy/4F and 24 Gy/3F respectively: GTV_res D_98% > 90 Gy (91 vs 97%); CTV_HR D_90% > 85 Gy (95 vs 98%); sigmoid D_2cm3 < 75 Gy (100 vs 99%); rectum D_2cm3 < 75 Gy, bladder D_2cm3 < 90 Gy, small bowel D_2cm3 < 75 Gy (100% for all); and ICRU RV < 75 Gy (95 vs 99%). There were also no significant differences in the proportion of patients meeting the EMBRACE II planning aims, except fewer patients treated with 28 Gy/4F met rectum D_2cm3 < 65 Gy (73 vs 85%, p = 0.027) and ICRU RV < 65 Gy (65 vs 84%, p = 0.005). <h3>Conclusion</h3> Cervix cancer patients treated with 24 Gy/3F MRgBT had comparable target doses and lower OAR doses compared to those treated with 28 Gy/4F. A less-resource intense fractionation schedule of 24Gy/ 3F is an alternative to 28 Gy/4F in cervix MRgBT.