2094 - Optimizing PTV Generation and Comparison of OAR Dosimetry with Different Rectal Spacers in Patients Treated with HDR-Like Prostate SBRT

Purpose/Objective(s): Prostate SBRT utilizing a highly inhomogeneous planning approach (HDR-like SBRT) has been associated with lower post-treatment PSA nadirs than more conventional, relatively homogeneous SBRT planning. However, this approach may lead to underdosing of the posterior prostate due to PTV cropping from the anterior rectum. Randomized studies have demonstrated dosimetric and clinical benefits to the use of rectal spacers in patients receiving prostate radiotherapy. Theoretically, these devices may yield improved disease control due to improved coverage of the posterior peripheral zone in patients treated with HDR-like SBRT. We aimed to determine if the use of different rectal spacers can improve PTV generation and coverage in patients treated with this technique.

Materials/Methods: We retrospectively identified 62 patients who underwent SBRT for clinically localized prostate cancer using a frameless robotic radiosurgery system. All patients were treated with HDR-like SBRT (3625 cGy in 5 fractions with minimum hotspot of 5500 cGy). 22 patients had SpaceOAR Vue (SOV), 20 patients had Barrigel (BG), and 20 patients did not have a rectal spacer (NonSpacer, NS). Spacers were chosen at the discretion of individual providers, and thus patients may have been prior selected for a given spacer due to anatomic and/or strategic coverage considerations not accounted for in these analyses. Institutional PTV margins involve 2 mm isometric expansions (5 mm on any side(s) with Gleason grade group 2+ disease) cropped out of the rectum. For comparison purposes, a secondary, idealized research PTV (PTVres) was generated in which expansions were made without rectal cropping. The treated PTV coverage (PTV Cov) was compared to PTVres coverage (PTVres Cov), as were other planning metrics including contoured spacer volume, maximum & mean rectal dose, maximum rectal mucosa dose, maximum urethra dose, and mean bladder dose. 1-way ANOVA with Tukey’s HSD testing was completed to compare the effect of rectal spacers across the 3 groups.

Results: Mean treated PTV Cov did not differ by use of rectal spacer (95.3% NS; 95.5% SOV; 95.0% BG, p = 0.187) but there was a significant difference in the PTVres Cov (92.5% for NS patients; 94.6% for SOV patients; 94.3% for BG patients, p < 0.001). Mean reduction in PTVres Cov from PTV Cov was not different between patients treated with SOV or BG (p = 0.800). Use of any rectal spacer was associated with reductions in mean rectal dose (1411.7 cGy NS; 1213.6 cGy SOV; 1198.9 cGy BG; p = 0.012) and maximum urethral dose (3890.6 cGy NS; 3742.3 cGy SOV; 3737.9 cGy BG; p = 0.000). Mean spacer volume did not differ between the SOV or BG groups (11.19 mL versus 9.78 mL, p = 0.107).

Conclusion: Our results support the use of rectal spacers in conjunction with HDR-like prostate SBRT, as they were associated with improved target volume generation and coverage, as well as reductions in multiple OAR metrics. No significant dosimetric differences were identified when comparing between different types of rectal spacer.