Patients with medically inoperable endometrial cancer (MIEC) are curable with brachytherapy (BT)-based treatment yet have competing comorbidities making definitive treatment challenging. MRI demonstrates superior soft tissue anatomy and target volume delineation but with limited data in MIEC patients. We aim to report disease and toxicity outcomes with MRI-based BT and identify dose-volume relationships for toxicities in the treatment of MIEC patients treated with MRI-based BT. We conducted a retrospective multi-institutional analysis of MIEC patients undergoing definitive MRI-based BT (+/- EBRT). MRI-based BT was delivered with the applicator in situ or coregistered to a planning CT. We identified patient, tumor, and dosimetric factors associated with disease and toxicity outcomes. Kaplan-Meier method was used for survival estimates. Log rank test and Cox proportional hazards were used for univariate and multivariate analyses, respectively. T-test was used for dose-volume toxicity analysis. A total of 120 patients were included with a median follow up of 28.0 months. Median age was 68.5 years. ECOG PS was 0-1 in 70%. Clinical stage I was 83.3% and II-IV, 16.7%. Most patients (91.7%) were node negative. Endometrioid and high risk histologies comprised 83.3%, and 16.7%, respectively. EBRT + BT was delivered in 97 patients (80.8%) and BT alone in 23 patients (19.2%). Chemotherapy or hormonal therapy was delivered during treatment in 10 (8.3%) and 11 (9.2%) patients, respectively. Estimated 3-year freedom from local, nodal, and distant recurrence were 88.0%, 96.0%, and 89.1% respectively. Estimated 3-year PFS and OS were 60.9% and 62.9%, respectively. On UVA, older age, PS ≥2, high risk histology, higher grade, and larger GTV at BT were significant (p<0.1). On MVA, older age, higher grade, and larger GTV at BT (p<0.05) predicted for inferior PFS. Fifteen late grade ≥3 toxicities were experienced in 14 (11.6%) patients, 13 of whom received EBRT and BT and 1 who received BT alone. Grade ≥3 toxicities were rectal (2, 1.7%), sigmoid (8, 6.7%), bowel (1, 0.8%), bladder (3, 2.5%), and osseous (1, 0.8%). EBRT was delivered in 7 of 8 sigmoid toxicities. Median sigmoid doses (EQD2a/b = 3Gy) for patients with and without late grade ≥3 sigmoid toxicity were 69.6 Gy and 64.3 Gy, respectively (p = 0.009). MRI-based BT for MIEC patients results in high rates of local control and favorable rates of late grade ≥3 morbidity. Older age, higher grade, and larger GTV at BT predicted for poorer PFS. Sigmoid colon was the predominant organ at risk for grade ≥3 toxicity with a dose -volume relationship observed. Attention to the location of the sigmoid throughout the treatment course may add insight into its predilection for risk. Future work will include additional institutions and dose-volume relationships of target volumes and normal tissues for further disease control and toxicity analysis.
Read full abstract