Intra-fractional dosimetric analysis of image-guided adaptive intracavitary brachytherapy of cervical cancer

Abstract Background To assess the intra-fractional dosimetric variations of image-guided adaptive brachytherapy of cervical cancer. Methods A total of 38 fractions (9 patients) undergoing brachytherapy for cervical cancer underwent a CT scanning for treatment planning (planning CT) and a CBCT scanning immediately prior to treatment (pre-treatment CBCT). Variations between images acquired within one applicator insertions (intra-application) were evaluated. The DVH parameters including volume, D90 HRCTV and D2cc OARs were recorded. Results The variations (mean ± SD) of the volume and D90 HRCTV are − 2.00 ± 3.26% and − 1.18 ± 4.47%, respectively. The variations of D2cc for bladder, rectum, sigmoid and small intestine are − 0.6 ± 17.05%, 9.34 ± 14.59%, 7.15% ± 20.49% and 1.52 ± 12.57%, respectively. Most of them are statistically nonsignificant except the D2cc for rectum, which showed a significant increase (P = 0.001). Using 5% and 10% uncertainty of physical dose for HRCTV at a 6Gy ⋅ 5 HDR schedule, the possibility of total EQD2 dose lower than 85Gy is close to 0% and 3%, respectively. Performing similar simulation at 15% and 20% uncertainty of a 4Gy physical dose for OARs, the possibility of total EQD2 dose exceeding 75Gy is about 70%. Less than 1% of the total EQD2 of OARs would exceed 80Gy. Conclusions Average intra-fractional dosimetric variation of HRCTV was small in a interval of less than 1 hour, and the possibility of total EQD2 exceeding 85Gy is higher than 97%. The intra-fractional dosimetric variations of OARs might result in an overdose for OARs in a single fraction or the whole treatment. It is necessary to detect unfavorable anatomical changes by re-imaging and take interventions to minimize applied doses and reduce the risk of complications.