GSOR04  Presentation Time: 12:15 PM: Evaluating Outcomes and Toxicities for a Newly Implemented MRI-Based Brachytherapy Program for Cervical Cancer

Background MRI-based brachytherapy for cervical cancer results in superior soft tissue delineation and unprecedented local control and toxicity rates. The vast majority of outcomes data are reported from European populations. Our institution initiated an MRI-based brachytherapy program in 2014, and this study updates our outcomes and toxicities at a U.S. academic center. Materials and Methods A retrospective review was performed on patients treated with MRI-based brachytherapy for cervical cancer from 2014-2022. A hybrid intracavitary/interstitial (IC/IS) applicator with capability for straight needles was introduced in 2016 and with oblique needles in 2022. EBRT was standardly 45Gy in 25 fractions. For brachytherapy, MRI was performed with the applicator in situ for all patients. Dose specification was 7Gy for 4 fractions with optimization aim of D90 HR-CTV EQD2 of 85-95Gy (α/β=10Gy) in 2 implants each delivering 2 fractions. We collected patient, tumor, and treatment characteristics, and disease and toxicity outcomes. Univariate-analysis was performed for disease control and toxicities. Kaplan-Meier method was used for survival estimates. Results 98 patients were included with median follow up of 24.5 months (IQR 11.9-39.8). Stage T3a-4a accounted for 29.2% of cases. Median tumor size on exam was 4.2cm (3-6). 93.6% of patients received cisplatin for a median of 5 cycles (4-6). Median treatment duration was 50 days (48-57). 25 patients were treated with a hybrid IC/IS applicator with a median of 6 needles (3-13). Dosimetry results include median GTV of 3.1cc (1.1-10.3), HR-CTV of 22.9cc (16.2-47.5), HR-CTV D98 of 80.3Gy (77.0-81.7) and D90 of 89Gy (86.1-90.6). Median bladder, rectum, sigmoid, and bowel doses were 82.1Gy (75.8-88.1), 65.9Gy (59.6-71.2), 65.1Gy (57.5-69.6), and 55Gy (48.9-61.3). Rectovaginal point dose was 65.2Gy (59.6-71.6). Chronic grade 2+ toxicities were seen in 11 (11.2%) patients for bladder, 13 (13.3%) rectosigmoid, 3 (3.1%) vaginal, and 1 (1.0%) patient for fatigue. Chronic grade 3+ toxicities were seen in the bladder (7 patients, 7.1%), rectosigmoid (4, 4.1%), and vagina (2, 2.1%). Local control was maintained at most recent follow up in 79 (83.2%) patients. Local control was favorably affected by smaller tumor size on exam (p=0.04) and smaller GTV volume at brachytherapy (p-0.02). PFS was negatively affected by adenocarcinoma histology (p=0.04), lager tumor size on CT (p=0.02), higher tumor stage (p=0.01), and larger volumes of GTV and HR-CTV (p=0.02, p=0.00). OS was negatively affected by larger volumes of GTV and HR-CTV (p=0.04, p=0.00). 3-year PFS and OS were estimated to be 60% and 75%. Conclusions MRI-based brachytherapy demonstrates excellent local control and low rates of G3 morbidity. These results are possible even in a population with relatively large volume primary tumors and extensive local disease compared to previously reported experiences from institutions. With the evolution of advanced hybrid applicators to include routine implementation of oblique needles, we anticipate even greater target coverage for patients with large targets, extensive parametrial involvement, or challenging topography. MRI-based brachytherapy for cervical cancer results in superior soft tissue delineation and unprecedented local control and toxicity rates. The vast majority of outcomes data are reported from European populations. Our institution initiated an MRI-based brachytherapy program in 2014, and this study updates our outcomes and toxicities at a U.S. academic center. A retrospective review was performed on patients treated with MRI-based brachytherapy for cervical cancer from 2014-2022. A hybrid intracavitary/interstitial (IC/IS) applicator with capability for straight needles was introduced in 2016 and with oblique needles in 2022. EBRT was standardly 45Gy in 25 fractions. For brachytherapy, MRI was performed with the applicator in situ for all patients. Dose specification was 7Gy for 4 fractions with optimization aim of D90 HR-CTV EQD2 of 85-95Gy (α/β=10Gy) in 2 implants each delivering 2 fractions. We collected patient, tumor, and treatment characteristics, and disease and toxicity outcomes. Univariate-analysis was performed for disease control and toxicities. Kaplan-Meier method was used for survival estimates. 98 patients were included with median follow up of 24.5 months (IQR 11.9-39.8). Stage T3a-4a accounted for 29.2% of cases. Median tumor size on exam was 4.2cm (3-6). 93.6% of patients received cisplatin for a median of 5 cycles (4-6). Median treatment duration was 50 days (48-57). 25 patients were treated with a hybrid IC/IS applicator with a median of 6 needles (3-13). Dosimetry results include median GTV of 3.1cc (1.1-10.3), HR-CTV of 22.9cc (16.2-47.5), HR-CTV D98 of 80.3Gy (77.0-81.7) and D90 of 89Gy (86.1-90.6). Median bladder, rectum, sigmoid, and bowel doses were 82.1Gy (75.8-88.1), 65.9Gy (59.6-71.2), 65.1Gy (57.5-69.6), and 55Gy (48.9-61.3). Rectovaginal point dose was 65.2Gy (59.6-71.6). Chronic grade 2+ toxicities were seen in 11 (11.2%) patients for bladder, 13 (13.3%) rectosigmoid, 3 (3.1%) vaginal, and 1 (1.0%) patient for fatigue. Chronic grade 3+ toxicities were seen in the bladder (7 patients, 7.1%), rectosigmoid (4, 4.1%), and vagina (2, 2.1%). Local control was maintained at most recent follow up in 79 (83.2%) patients. Local control was favorably affected by smaller tumor size on exam (p=0.04) and smaller GTV volume at brachytherapy (p-0.02). PFS was negatively affected by adenocarcinoma histology (p=0.04), lager tumor size on CT (p=0.02), higher tumor stage (p=0.01), and larger volumes of GTV and HR-CTV (p=0.02, p=0.00). OS was negatively affected by larger volumes of GTV and HR-CTV (p=0.04, p=0.00). 3-year PFS and OS were estimated to be 60% and 75%. MRI-based brachytherapy demonstrates excellent local control and low rates of G3 morbidity. These results are possible even in a population with relatively large volume primary tumors and extensive local disease compared to previously reported experiences from institutions. With the evolution of advanced hybrid applicators to include routine implementation of oblique needles, we anticipate even greater target coverage for patients with large targets, extensive parametrial involvement, or challenging topography.