Characterization of a Novel Radiopaque Perirectal Hydrogel Spacer for Prostate Cancer Radiotherapy

Purpose/Objective(s) Polyethylene glycol (PEG) hydrogel perirectal spacers can reduce prostate cancer (PCa) radiotherapy (RT)-related gastrointestinal (GI) toxicity. Accurate hydrogel contouring requires T2 MRI simulation (sim), as hydrogels have Hounsfield units (HU) comparable to soft tissue on computerized tomography (CT). A PEG hydrogel containing covalently bound iodine was recently developed and approved for clinical use. However, its characteristics in routine clinical practice have yet to be reported. Here we evaluated iodinated hydrogel volume, perirectal spacing, and HU from treatment images for PCa patients receiving RT at our institution. Materials/Methods Patients with biopsy-proven, clinically localized PCa (cT1c-cT2N0M0) who had same-day CT/MRI sim after fiducial markers and iodinated hydrogel placement were identified. Hydrogel was contoured on CT sim, MRI, and cone beam CT (CBCT) at treatment start (CBCT#1) and completion (CBCT#2). Prostate and hydrogel volumes and anterior-posterior mid-gland perirectal spacing at midline were measured on CT/MRI and CBCTs, and HU was measured on CT sim and CBCTs using treatment planning software. Patient differences were evaluated using paired, two-tailed Student's t-tests, with significance at P Results Twenty-three patients were identified (mean age 71.1 ± 7.8 years; mean prostate volume 61.9 ± 29.2 cm3; mean initial PSA 7.36 ± 1.9), treated with IMRT: 79.2 Gy/44 fx (n = 3), 70 Gy/28 fx (n = 15), or 36.25 Gy/5 fx (n = 5). Hydrogel was identified on all sim scans and was larger on T2 MRI vs CT (MRI mean volume 9.9 ± 1.7 cm3 vs CT 9.3 ± 1.6 cm3 vs; P = 0.03), and comparable mean perirectal distance at mid-gland/midline (MRI 1.02 ± 0.55 cm vs CT 1.03 ± 0.53 cm; P = 0.65). Perirectal spacing was maintained between CT sim and CBCT#1 (1.03 ± 0.51 cm; P = 0.97) done an average of 16.7 ± 7.1 days post sim and CT sim and CBCT#2 (1.03 ± 0.53 cm vs 1.02 ± 0.54; P = 0.74) done an average of 53.8 ± 19 days post sim. No significant change in hydrogel volume between CBCT#1 and CBCT#2 for 44 fx (10.9 ± 1.7 cm vs 10.4 ± 1.3 cm; P = 0.17), 28 fx (10.2 ± 1.4 vs 9.83 ± 1.7 cm; P = 0.11), or 5 fx (9.9 ± 2.1 vs 9.54 ± 1.9; P = 0.45) patients was observed. Hydrogel mean HU on CT sim was not significantly changed at CBCT#1 (CT sim 152.1 ± 30.5 vs CBCT#1 129.1 ± 51.6; P = 0.08), but mean HU at CBCT#2 (109.3 ± 47.9 cm) was lower vs CT sim (P = 0.001). For 44 fx treatments hydrogel HU significantly decreased between CBCT#1 and CBCT#2 (140.7 ± 10.0 vs 94.2 ± 80.4; P = 0.01), but not for 28 fx (123.5 ± 54.4 vs 103.4 ± 44.0; P = 0.13) or 5 fx (138.4 ± 38.5 vs 136 ± 37.2; P = 0.60) regimens. Conclusion PEG-based iodinated hydrogel can be delineated on CT sim comparable to MRI sim, and maintains visibility and perirectal spacing through 5, 28, and 44 fx RT courses as assessed by CBCT. Hydrogel HU decreased over 44 fx relative to CT sim, with no significant difference in HU seen for 28 fx and 5 fx courses. Use of perirectal hydrogel containing covalently bound iodine may obviate reliance on T2 MRI for accurate hydrogel contouring during RT treatment planning.