A novel model to correlate hydrogel spacer placement, perirectal space creation, and rectum dosimetry in prostate stereotactic body radiotherapy

Abstract

Background: The SpaceOAR hydrogel is employed to limit rectal radiation dose during prostate radiotherapy. We identified a novel parameter - the product of angle θ and hydrogel volume - to quantify hydrogel placement. This parameter predicted rectum dosimetry and acute rectal toxicity in prostate cancer patients treated with stereotactic body radiotherapy to 36.25 Gy in 5 fractions.

Methods: Twenty men with low- and intermediate-risk prostate cancer underwent hydrogel placement from 2015 to 2017. Hydrogel symmetry was assessed on the CT simulation scan in 3 axial slices (midgland, 1 cm above midgland, 1 cm below midgland). Two novel parameters quantifying hydrogel placement - hydrogel volume and angle θ formed by the prostate, hydrogel, and rectum - were measured, and the normalized product of θ and hydrogel volume calculated. These were then correlated with perirectal distance, rectum maximum 1-3 cc point doses (rD_max 1-3 cc), and rectum volumes receiving 80-95% of the prescription dose (rV80-95%). Acute rectal toxicity was recorded per RTOG criteria.

Results: In 50% of patients, hydrogel placement was symmetric bilaterally to within 1 cm of midline in all three CT simulation scan axial slices. Lateral hydrogel asymmetry < 2 cm in any one axial slice did not affect rectum dosimetry, but absence of hydrogel in the inferior axial slice resulted in a mean increase of 171 cGy in the rD_max 1 cc (p < 0.005). The perirectal distance measured at prostate midgland, midline (mean 9.1 ± 4.3 mm) correlated strongly with rV95 (R² 0.6, p < 0.001). The mean hydrogel volume and θ were 10.3 ± 4.5 cc and 70 ± 49°, respectively. Perirectal distance, rV95 and rD_max 1 cc correlated with hydrogel angle θ (p < 0.01), and yet more strongly with the novel metric θ*hydrogel volume (p < 0.001). With a median follow up of 14 months, no rectal toxicity >grade 2 was observed. Low grade rectal toxicity was observed in a third of men and resolved within 1 month of SBRT. Men who had these symptoms had higher rD_max 1 cc and smaller θ*hydrogel volume measurements.

Conclusions: Optimal hydrogel placement occurs at prostate midgland, midline. The novel parameter θ*hydrogel volume describes a large proportion of rectum dosimetric benefit derived from hydrogel placement, and can be used to assess the learning curve phenomenon for hydrogel placement.

Keywords: Dosimetry; Prostate cancer; Rectal toxicity; SpaceOAR hydrogel; Stereotactic body radiotherapy.

Fig. 1

Measurement of a perirectal distance at CTV center on axial CT image (a), and θ in 3-dimensional space on axial and sagittal CT images (a, b). The perirectal distance is measured at each vertex in a 4 × 4 cm coronal grid centered on the prostate midgland, at midline to generate a perirectal space map comprising up to 25 perirectal distance measurements for each patient (c)

Fig. 2

Rectum maximum 1 cc point dose (rD_max 1 cc) by hydrogel symmetry (SYM) score. T-test comparison of means revealed a statistically significant difference between groups SYM1 and Missing Inf Slice, denoted by * = p < 0.01

Fig. 3

Perirectal distance at CTV center versus rV95%, R² 0.60, p < 0.01, (a). Normalized θ*hydrogel volume versus: rV95%, R² 0.60, p < 0.01, (b); rD_max 1 cc, R² 0.64, p < 0.01, (c); and perirectal distance at CTV center, R² 0.66, p < 0.01, (d)

Fig. 4

Normalized θ*hydrogel volume versus rectum maximum 1 cc point dose (rD_max 1 cc), showing acute low grade rectal toxicity (n = 6/20)