Interfractional Variation of Treatment Delivery for High-Dose-Rate Interstitial Brachytherapy for Primary or Recurrent Gynecologic Malignancies: A Single-Institution Experience

Abstract

For women who require high-dose-rate interstitial brachytherapy (HDR-ISBT) for primary or recurrent gynecologic malignancy, there can be significant interfractional variation from patient motion, post-implant swelling and organ filling. The purpose of this study was to evaluate the interfractional variation of dose distribution and displacement of catheters during HDR-ISBT in a single implant, consecutive day regimen. A total of 15 patients with primary or recurrent gynecologic malignancies who underwent HDR-ISBT were reviewed. Interstitial catheters and fiducials were placed under fluoroscopy and ultrasound guidance to confirm desired placement for clinical target volume (CTV) coverage and reproducibility. Patients underwent 4-5 treatments over 2-3 consecutive days. CT imaging was acquired on day of implant and days 2 and/or 3. Displacement of fiducials, their center of mass (CM) and self-consistency was calculated for each fraction. Additionally, the position of catheter tips and CTV’s relative to CM of fiducials were measured. The effects on dosimetric parameters such as dose covering 90% of high-risk CTV (D90), equivalent uniform BED (EUBED) and generalized biologically equivalent uniform dose (gBEUD) were compared for each fraction. 50 fiducials were tracked over 70 fractions. For a majority of patients (10 out of 15), the fiducial markers CM moved less than 0.2 cm. The positions of 278 catheters over the same fractions were measured relative to the fiducials CM and found to have an average interfractional movement of -0.06 +/- 0.31 cm. The CTV to needles interfractional shift was -0.09 +/- 0.36 cm. The average D90 for all fractions was 92.5 +/- 6.5% of the prescription dose (PD). The average gEUD across all patients was 1.57 +/- 0.19 and 18.44+/- 15.08 Gy/PD (a = -5 and 2, respectively). The average EUBED across all patients was 1.98+/- 0.16 and 2.43+/- 0.23 Gy/PD (α = 0.3 and 0.1, respectively). Volume integrated radiobiological metrics, gBEUD and EUBED showed very consistent effects for all implants. Center of mass for 3 or more markers seems a clinically acceptable reference for ensuring catheter tip and CTV positional interfractional reproducibility. Given the dose inhomogeneities accepted with HDR-ISBT, D90 is less sensitive than volume integrated metrics for describing the dose delivered. Combining the effects of a reproducibly accurate implant with unique identification of the importance of ‘hot’ sub-volumes is a promising method for improving patient outcomes with HDR-ISBT.