Interobserver variability in rectum contouring in high-dose-rate brachytherapy for prostate cancer: A multi-institutional prospective analysis.

ABSTRACT

PURPOSE: The aim of this study was to evaluate the interobserver variability (IOV) of rectum contouring, and its dosimetric consequences, for high-dose-rate brachytherapy in patients with prostate cancer across multiple institutions. METHODS AND MATERIALS Five radiation oncologists contoured rectums in 10 patients on transperineal ultrasound image sets after establishing a delineation consensus. The D0.1cc, D1cc, and D2cc rectum volume parameters were determined. The mean, standard deviation, and range of each dose-volume histogram parameter were evaluated for each patient. The IOV was determined using the coefficient of variation, and the dosimetric impacts on the total dose were analyzed by estimating the biologically equivalent dose (EQD2α/ß = 3). RESULTS: The interobserver coefficients of variation (±standard deviation) for the reported D0.1cc, D1cc, and D2cc were 5 ± 1.84%, 4 ± 1.26%, and 4 ± 1.33%, respectively. As for the impact on the total dose, the mean dose differences for D0.1cc, D1cc, and D2cc were 10 Gy, 7.3 Gy, and 6.6 Gy, respectively. CONCLUSIONS: The D2cc is robust as evident by the low IOV (<5%). However, some variability ranges almost overlap with the clinical threshold level, which may present dosimetric and clinical complications. General rectal contouring guidelines for prostate high-dose-rate brachytherapy are desirable to reduce discrepancies in delineation.