Song et al. (2013) (1) conducted a prospective pilot clinical trial with 52 patients at four institutions. Spacer hydrogel was injected after CT and MRI planning scans with repeat scans after the injection. IMRT plans were composed using each set of CT and MRI scans. A prostate rectal separation of >7.5 mm was achieved in 95.8% of the patients. A decreased rectal V70 >25% occurred in 95.7%. No significant differences were found in prostate, planning treatment volume (PTV), rectal, and bladder volumes. Four of the 52 patients were not successfully injected and a separate publication was planned detailing these occurrences. Acute toxicity was not addressed in this report. The author concluded statistically significant rectal dose reductions across the entire dose range occurred in >90% of the patients.
Uhl et al. (2014) (3) reported on the 12 month toxicity of 52 patients who received IMRT (78 Gy) for localized prostate cancer along with a prostate rectal PEG spacer. Injection was not successful in four and in the rectal wall in one patient, leaving 47 patients in the study. In addition to toxicity data at 3, 6, and 12 months, proctoscopy was performed at 12 months. Grade 1 acute rectal toxicity was noted in 19 (39.6%) and grade 2 in 6 (12.5%). No patients had grade 3 or 4 acute toxicity. Late grade 1 toxicity was experienced by 2 (4.3%), but none had grades 2, 3, or 4 toxicity. Grades 1, 2, and 3 genitourinary (GU) toxicity occurred in 20 (41.7%), 17 (35.4%), and 1 (2.1%), respectively. Forty-five of the 47 patients had proctoscopy at 12 months after IMRT treatment. Using the Vienna Rectoscopy Scale (VRS), 32 (71%) had a score of zero. Grade 2 congested mucosa was noted in 1 (3%) and telangiectasias were found in 28%: grade 1 - 13%, grade 2 – 13%, and grade 3 – 2%. Ulceration, stricture, or necrosis were not found.
Hamstra et al. (2017) (4) studied the extended three year follow-up data of the above subjects. Patient participation was voluntary and included 63% of both spacer (n=94) and control (n=46) patients. A comparison of the group volunteer members was not presented, except that there was no difference between groups regarding participation. A median follow-up of 37 months (range 26-46) occurred for the controls and 37.1 (range 32-47) months for the spacer subjects.
Data showed the spacer group had a smaller volume of rectal radiation for all volumes from V50 to V80 (p<.0001). Relative reductions were 54% for V50, 79% for V70, and 96% for V80. Grade >1 rectal toxicity at three years was decreased by 75% in the spacer patients (spacer 2%, 95% CI 1%- 6%) and (control 9%, 95% CI 4% - 20%), p<.03. No grade >2 rectal toxicity was seen in the spacer patients with 6% in the control group and one case of grade 3. It was noted that the toxicity in the control group was less than usually reported and no explanation was available. There were no differences in grades 1 or 2 urinary toxicities between the groups at three years with the exception of urinary incontinence in 15% of the controls and 4% in the spacer group (p = .046).
Susil et al. (2010) (13) injected 20 cadavers with hydrogel using a transperineal approach achieving an average of 12.5 mm of separation. Average rectal volume receiving 70 Gy decreased from 19.9% to 4.5%. The authors discussed the potential advantages and possible risks as well as various substances that could be used.
Mariados et al. (2015) (14) reported on a prospective, randomized, controlled, multicenter trial with 222 patients with stage T1 or T2 prostate cancer. Computed tomography (CT) and magnetic resonance imaging (MRI) scans were performed for treatment planning, followed by placement of fiducial markers. Patients were then randomized 2:1 to receive a polyethylene glycol (PEG) prostate rectal absorbable spacer (SpaceOAR® system) injection or no injection. The planning CTs and MRIs were repeated prior to image-guided intensity modulated radiation therapy (79.2 Gy in 1.8-Gy fractions). A primary endpoint of a >25% reduction in the rectal volume (rV70) was achieved in 97.3% of the spacer patients. The primary safety endpoint was the proportion of patients experiencing grade 1 or greater rectal or procedural adverse events (AEs) in the first six months. The treatment group had a reduction in pain during RT (spacer 2.7% and control 11.8%, p = 0.022) but overall there were no statistically significant differences in acute AEs between the treatment (34.2%) and control (31.5%) groups. There were also no differences between groups in urinary toxicity. Late rectal toxicity (3 – 15 months) was seen in 2.0% of the spacer patients and 7.0% of the controls which was statistically significant (p = 0.044). No differences were found in bowel and urinary quality-of-life (QOL) at three months and both groups had 5- and 10-point bowel QOL declines at 6, 12, and 15 months. There was a statistically significant difference (p = 0.003) in urinary QOL at six months between groups favoring the treatment group, but there was no difference at 15 months.
Pieczonka et al. (2016) (15) also reported on the same group immediately above. Insertion of the spacer was described as “very easy” in 98.7% and successful in 99.3%. It was noted that the mean perirectal space was 12.6 mm after implant and 10.9 mm at 12.4 weeks, with absorption at 12 months.
The Expanded Prostate Cancer Index Bowel Composite (EPIC) (16) quality of life (QOL) and minimally important difference (MID) (17) tools were used to assess patient opinions of treatment. Bowel QOL declined in both groups in the first three months with return to baseline at six months. At three years the spacer group was near or greater than baseline, but the control group had decreased (p=.002). Differences were at the 5 point level of MID but not at the 10 point level. A correlation between an increasing rectal V50 to V80 and a decline in bowel QOL was found. Urinary QOL also declined in both groups in the first three months with return to baseline at six months. At three years, there was a statistical difference between the two groups favoring the spacer group, but it did not meet the MID level. However, it was also stated that there was a statistically significant difference between the groups regarding urinary frequency favoring the spacer arm (5%) versus the controls (18%) p = .05. No differences were found in the sexual QOL or vitality/hormonal QOL.
Hatiboglu et al. (2012) (18) reported on 29 of the patients in the study immediately above. The method of selecting the 29 out of 52 patients is not described, and the study is described as prospective, single-arm, open-label performed at four institutions. Safety evaluation and performance of the spacer were the main objectives. Scans (CT and MRI) were performed prior to and after spacer injection and after IMRT at 3 and 6 months. An independent reviewer measured the distance between the prostate and rectum. “Functional” (7.5 mm space after spacer injection) and “clinical success” (> 25% reduction in rectal V70) occurred in 28/29 (96.6%) and 26/27 (96.3%) of patients. Two patients were excluded due to technical difficulties loading their data for review.
Whalley et al. (2016) (19) studied 30 patients with T1-T3 prostate cancer for whom dose-escalated radiation therapy was considered appropriate and who were enrolled in a Phase I/II trial. A contemporary control group of 110 patients receiving the same dose was identified for comparison. Primary endpoints were comparison of the rectal volume receiving 30 – 82 Gy and post-operative toxicity. Secondary endpoints were acute and late toxicity. Hydrogel (spacer) was successfully injected into 29 patients with injection into the rectal lumen in one. Mean difference of rectal- prostate separation was 10.5 mm. Toxicity related to the injection occurred in five but resolved within a week. Acute radiation gastrointestinal (GI) toxicity occurred in 13 (43%), which was primarily increased stool frequency. It was noted that stool softeners had been prescribed. Two patients had grade 1 rectal bleeding. There was no >2 stage acute gastrointestinal toxicity. Late grade 1 GI toxicity of increased stool frequency occurred in five (16.6%) spacer patients. One patient received laser coagulation at 13 months and at 18 months the bleeding had not recurred. In the control group, acute grade 1 toxicity occurred in 56 (50.6%) and grade 2 in five (4.5%). Late grade 1 GI toxicity in the control patients was noted in 46 (41.8%) occurring at a median of 11.5 months after radiation with a range of 6 – 43 months.
Symptoms were increased stool frequency and rectal bleeding not requiring intervention. Late grade 2 toxicity was seen in four (3.6%) patients and occurred at a median of 20 months. There was no grade 3 toxicity in either group. Late grade 1 toxicity was reduced in the spacer group compared to the control (p = 0.04) but there was no difference in late grade 2 toxicity. Median follow-up in this study was greater than two years. The authors noted that GI toxicity occurs at a median of 17 months with peaks at 1.5 and 4.5 years [Zelefsky et al. (2008) (20) and Odrazka et al. (2010) (21)].
Habl et al. (2016) (22) stopped using the spacer gel due to the development of rectal fistulae in two patients and the fistulae were presumed to be due to the gradual accumulation of gel within the anterior rectal wall. Correspondence from Fagundes et al. (2016) (23) has questioned the fistulae being due to the use of the spacer.
The ECRI Institute which performs technology assessments recently evaluated the SpaceOAR® System. (24) A literature search of articles published between January 1, 2000 and May 17, 2017 was performed. The full text of seven articles and seven abstracts were reviewed. It was concluded from one RCT and three non-RCTs that the hydrogel was well tolerated; worked as intended to reduce rectal irradiation and long term but not acute toxicity, and improved bowel quality of life. Acute rectal toxicity reduction was not found. The report stated studies with longer term follow-up of greater than five years comparing different spacers are needed.
Hypofractionation (HFX) EBRT
Moderate HPX has become an accepted alternative to conventionally-fractioned EBRT, despite lack of standardized dose constraints and follow-up beyond 5 years (7, 25-27). However, guideline support for ultra-HFX (sometimes referred to as stereotactic body radiation therapy or SBRT) as a treatment option, even in very low- to favorable intermediate-risk prostate cancer, is mixed. NCCN guidelines include ultra-HPX treatment options (7), ASTRO/ASCO/AUA guidelines (25) grade their ultra-HFX recommendation “conditional” (“remaining uncertainty in the balance between benefit and risk”), and the European Association of Urology (EAU) recommends restricting ultra-HFX to prospective clinical trials (26).
Studies of HFX using the PEG spacer are lacking. A small, single institution, retrospective chart review of 50 men with low- or intermediate-risk prostate cancer treated with ultra-HFX after PEG placement claimed to show that “lower rates of acute rectal toxicity were observed compared with previous, similarly-fractionated SBRT reports that were performed without spacer placement” (28).
NCCN Guidelines recommend selective use of the PEG spacer (standard EBRT or HFX) when modern EBRT localization techniques are “insufficient to improve oncologic cure rates and/or reduce side effects due to anatomic geometry or other patient related factors, such as medication usage and/or comorbid conditions” (7). ASTRO/ASCO/AUA HPX specific guidelines mention prostate--rectal spacers as one of several techniques “to facilitate the meeting of rectal and bladder dose-volume constraints” (25).