Background

External beam radiotherapy is one of the standard treatments recommended for the management of localised low- or intermediate-risk prostate adenocarcinoma according to the classification of D’Amico [1,2,3]. It gives similar results to surgery and therefore represents a therapeutic alternative for patients who are inoperable or do not want to undergo surgery [4]. Although dose escalation to ≥ 76 Gy improves treatment efficacy, this increase in dose may be associated with increased genitourinary and gastrointestinal toxicity [5,6,7]. Modern radiotherapy in prostate cancer is based on intensity-modulated radiotherapy (IMRT) and image-guided radiotherapy (IGRT) techniques [8,9,10,11,12]. However, due to the close proximity of the prostate and rectum, rectal toxicity (such as rectitis, rectal bleeding…) remains a major problem in patient management.

One method of increasing the distance between the prostate and the rectum is to use a spacer implanted into the rectoprostatic space. Recent studies have shown the possibility of reducing the radiation doses received by the rectum when spacers are used [13,14,15]. To date, the most studied devices are gels containing polyethylene glycol and hyaluronic acid. These devices create a space between the rectum and the prostate, ranging from 7‒20 mm depending on the implantation protocol [14,15,16,17]. However, there has been no long-term evaluation of their use.

Another medical device that can be used is the BioProtect balloon (BioProtect Ltd, Tzur Yigal, Israel). This is a balloon made from a biodegradable polymer, which is filled with saline or iodine solution. It is inserted using a transperineal technique, after hydrodissection of the rectoprostatic space. Its efficacy and safety have been demonstrated in animal studies [18].

After obtaining a favourable opinion from the SFRO (French Society of Oncological Radiotherapy) in 2013, a pilot study was set up to evaluate the use of this balloon spacer. This report describes the long-term outcomes obtained in the first treated patients.

Methods

Aim, design and setting

This study was carried out in the Pasteur Clinic, Toulouse, France, between October 2013 and March 2015. The aim was to evaluate the long-term outcomes and safety of the BioProtect balloon spacer in patients treated with radiotherapy for low- or intermediate-risk prostate adenocarcinoma.

Study participants

All patients treated with curative radiotherapy for low- or intermediate-risk prostate adenocarcinoma, who underwent insertion of the ProSpace® rectal-prostate balloon spacer before treatment initiation, were included. This study was approved by national institutional review boards of the SFRO and all patients obtained written information about the study objectives and gave their written informed consent prior to balloon placement.

The rectal-prostate balloon spacer device

ProSpace® is a sterile medical device consisting of a biodegradable inflatable balloon (mounted on a deployer) and a delivery kit (echogenic needles, a dilator and an introducer sheath). After antibiotic prophylaxis (fluoroquinolone) and rectal preparation, implantation of the device was performed under local anaesthesia or neuroleptanalgesia by our experienced interventional radiologist previously trained and certified by BioProtect. It is recommended to use a urinary catheter to empty the bladder at the start of the procedure which helps with balloon positioning. In the case of difficulty when penetrating the skin of the perineum, a small incision (3‒5 mm) can be made with a scalpel at the insertion site. Hydrodissection of the rectoprostatic space is performed, guided by endorectal ultrasound, before inserting the device. Once the balloon is positioned, it is inflated with sterile saline (15‒20 ml) using a plastic Luer-Lok syringe (20 ml or 50 ml). It is possible to add 1‒2 ml of an iodinated contrast agent to the saline solution to better visualise the balloon on the dosimetric scan. This procedure has been previously described by Vanneste et al. [19]. There is no formal contraindication apart from the risk of capsular penetration or invasion of the seminal vesicles.

Radiotherapy planning

All patients underwent a dosimetric scan before implantation of the ProSpace® balloon and 1 week after implantation. Patients were required to have an empty rectum and a full bladder, according to the department's usual protocol, during pre-treatment magnetic resonance imaging (MRI), dosimetric scans and all radiotherapy sessions. Rigid fusion was performed between the MRI and dosimetric scans, to assist with prostate delineation. Clinical target volume 1 (CTV1) corresponded to the prostate and seminal vesicles and CTV2 corresponded to the prostate. If necessary (based on Roach Formula), a pelvic lymph node CTV could also be delineated. Planning target volume (PTV) corresponded to a homogeneous geometric extension of 10 mm in all directions and 5 mm posteriorly. The prescribed doses for PTV1 and PTV2 were 46 Gy and 78 Gy, respectively. If necessary, pelvic lymph node PTV was planned to receive a dose of 46 Gy. In accordance with International Commission on Radiation Units and Measurements (ICRU) reports 50 and 83, the coverage objectives for PTVs were consistent with current recommendations for each PTV with: V95% > 95%, Dmin > 90%, Dmax < 107% [1, 20, 21]. Organs at risk (OARs) were delineated according to international recommendations and dose constraints corresponded to international standards [22, 23]. When necessary, posterior PTV uncovering was allowed so as not to exceed the rectal constraints [1]. Pre- and post-implantation dosimetry was performed by TPS Eclipse (Varian, Palo Alto, CA). Treatment was performed by IMRT with daily IGRT using cone beam computed tomography (CBCT).

Patient follow-up

All patients underwent clinical examination that included digital rectal examination, laboratory tests that included prostate-specific antigen (PSA), pelvic MRI and prostate biopsies prior to the approval of radiotherapy at a multidisciplinary meeting. An assessment of quality of life (QoL) was also performed, with the use of validated questionnaires (IPSS: International Prostate Symptom Score, QLQ-C30, PR25), during the initial consultation before ProSpace® balloon placement and after spacer placement just before starting radiotherapy.

Patients were then followed weekly during radiotherapy, every 3 months for 6 months, then every 6 months for 5 years, and subsequently every year. At each follow-up, a physical examination and a PSA assay were performed along with completion of QoL questionnaires (before balloon insertion, before the start of radiotherapy, at mid-treatment, at the end of treatment and at each visit to the end of the 2-year follow-up period). Toxicity was assessed according to the NCI Common Terminology Criteria for Adverse Events (CTCAE v4.0). Late toxicity was defined as any toxicity occurring > 6 months after the end of radiotherapy. Targeted scans were only performed when clinically required or after PSA elevations, in accordance with the usual recommendations [2]. No additional imaging was performed to observe the resorption of the balloon.

Study objectives

The main objective of the study was to evaluate the dosimetric benefit of the ProSpace® rectal-prostate spacer for the main OARs (rectum, bladder, etc.). The secondary objectives were to evaluate the feasibility and tolerability of ProSpace® balloon placement and to evaluate its long-term therapeutic efficacy and tolerance according to CTCAE v4.0 as well as by patients through the IPSS, QLQ-C30 and PR25 QoL questionnaires.

Statistical analysis

Quantitative values are expressed as mean or median (as specified) and (range: min‒max). Qualitative data are expressed as n (%). Time to event was calculated from the first day of radiotherapy to the occurrence of the event. Dosimetric values were compared using the Wilcoxon test for paired data. A p value of < 0.05 was used to indicate statistical significance.

Results

Study population

Sixteen patients were enrolled between October 2013 and March 2015. Median age was 73 years (range: 65–81 years), median PSA at diagnosis was 9.5 ng/ml (range: 5.6–17.9 ng/ml) and median prostate volume was 40.5 cm3 (range: 20–84 cm3). Clinically, the lesions were stage T1c (n = 7), T2a (n = 4) and T2b (n = 5). No patient had a T3 lesion on clinical examination or MRI. Lesions with a Gleason score 6 (3 + 3) were found in six patients and lesions with a Gleason score 7 (3 + 4) were found in 10 patients. No patient had locoregional or distant lymph node involvement in the standard extension assessment. All patients received prostate radiotherapy at a dose of 78 Gy in 39 sessions, one session a day, 5 days a week. Two patients received pelvic radiotherapy at a dose of 46 Gy, according to a normofractionated regimen. All radiotherapy sessions were performed using the IMRT technique with daily IGRT using CBCT. Hormone therapy with an LHRH analogue was prescribed in five patients for 6 months (Table 1).

Table 1 Characteristics of the study population
Full size table

Implantation of the ProSpace® balloon

The median time between the pre- and post-implantation dosimetric CT scan was 28 days. The placement of the ProSpace® balloon was considered easy by the interventional radiologist who performed the entire procedure. There was no implantation failure or immediate or late complication related to the procedure for all 16 balloons. Simple step-1 analgesia was necessary in most cases, which enabled outpatient performance of the procedure. The mean volume of natural saline injected into the balloon was 17.21 ml (range: 16.5–17.5 ml). The mean coverage of the balloon was 49.3 mm (range: 44–58 mm) in length, 32.1 mm (range: 28–35.3 mm) in width and 16.7 mm (range: 12–20 mm) in thickness. From top to bottom, the space created was a mean of 16.3 mm (range: 11–20.5 mm) at the base of the prostate, 12.1 mm (range: 4–16 mm) at the middle and 8.9 mm at the apex (range: 5–15 mm). Laterally, the space created at the middle of the prostate was a mean of 13.4 mm (range: 5–22 mm) on the right and 12.1 mm (range: 2–19 mm) on the left. The mean distance between the apex and the lower pole of the balloon was 7.9 mm (range: 0–23 mm) (Table 2). During the last radiation session, a CBCT imaging control was performed, confirming the presence of the balloon. No significant changes in the size or shape of the balloon were observed at the end of the treatment.

Table 2 Characteristics of the implanted ProSpace® balloons
Full size table

Dosimetry

Dosimetric data were collected from the dosimetric scan before and after implantation of the ProSpace® balloon in all 16 patients.

Regarding the high doses received by the rectum, there was a mean decrease of 9.73 Gy at the D2.5 cc level (63.5 vs. 73.3; p < 0.0005), a decrease of 8.72 Gy at the D10cc level (54.3 vs. 63; p < 0.0005), a decrease of 6.34 Gy at the D15cc level (50.4 vs. 56.7; p = 0.001) and a decrease of 4.54 Gy at the D20cc level (47.1 vs. 51.7; p = 0.018). For D25%, D50%, D60%, D70% and D75%, the doses received in the rectum were statistically lower after the balloon was placed. On average, rectal volumes receiving a dose of 70 Gy, 60 Gy and 50 Gy were significantly lower after balloon implantation: -4.81 cc (1.5 vs. 6.3; p < 0.0005), -8.08 cc (6.4 vs. 14.5; p = 0.002) and -9.06 cc (16.7 vs. 25.7; p = 0.003), respectively. There was no statistically significant differences in Dmax, V39Gy, mean dose or Dmin (Table 3). For the bladder doses, there was no significant difference in any of the dosimetric points of interest.

Table 3 Dosimetric parameters of the rectum before and after implantation of the ProSpace® balloon
Full size table

Regarding PTV coverage, this was systematically covered according to the recommendation of V95% > 95%. Dmin > 90% was respected after balloon implantation but was frequently (11/16 pts) reduced before implantation to protect the rectum and respect the usual constraints. However, there were significant differences in coverage after balloon implantation. Mean V95% was 1.61% higher after balloon implantation (99.7 vs. 98.1; p < 0.0005), mean Dmin was 4.04% higher after balloon implantation (90.5 vs. 85.3; p = 0.01) and mean V98% was 3.14% higher (96.1 vs. 93; p < 0.001) (Table 4). Figure 1 shows the dosimetric gains obtained after placement of the ProSpace® balloon in one of the patients of the study cohort. All the patient experiments a dosimetric gain after the placement of the spacer, regardless the volume of the prostate.

Table 4 PTV coverage before and after implantation of the ProSpace® balloon
Full size table
Fig. 1
figure 1

Dose-Volume Histogram. Exemple of dosimetric gain after ProSpace® Ballon implantation. Before(◼▢) and After (▲) ballon implantation

Full size image

Efficacy

The median duration of follow-up was 85.5 months (range: 27–98 months) for the entire study population. One patient was lost to follow-up approximately 2 years after treatment completion, following a move to a different district, and one patient died of heart failure approximately 5 years after radiotherapy. Among the surviving patients, only one patient experienced a biochemical recurrence 70 months after radiotherapy. This was a T2a N0 adenocarcinoma, Gleason 7 (3 + 4), initial PSA of 9.22 ng/ml.

Toxicity

Cumulative urinary toxicity at 5 years was grade 1 in 32% of patients (5/16) and grade 2 in 6% (1/16 patients). No grade 3 or 4 toxicity was found. Cumulative gastrointestinal toxicity at 5 years was grade 1 in 6% (1/16 patients). No toxicity of grade 2 or higher was found. Cumulative sexual toxicity at 5 years was grade 1 in 6% (1/16 patients) and grade 2 in 19% of patients (3/16 patients). Median IPSS score was 2 before balloon implantation, 4 after balloon implantation and 5 at the end of radiotherapy. Three months after the end of radiotherapy, the IPPS score was 3 and remained stable until 24 months post-radiotherapy (Fig. 2). Balloon resorption was confirmed during a follow-up MRI 6 months after the end of radiotherapy.

Fig. 2
figure 2

Late Toxicites. A Bowel Toxicty. B Urinary Toxicity. C Sexual Toxicity. D Evolution of IPSS score

Full size image

Quality of life

Before implantation of the ProSpace® balloon, “functional” domain scores were > 85 and “symptom” domain scores were < 10 for QLQ-C30 and PR25. The mean PR25 sexual activity score was 58 before balloon implantation. After implantation of the spacer, there were no significant changes in the various QoL indicators, which indicated good acceptability of the ProSpace® balloon by the patients. At the end of radiotherapy, a slight decrease in the overall health score and a slight increase in fatigue, pain, loss of appetite, diarrhoea and urinary symptom scores were noted. All scores returned to baseline within 6 months after the end of treatment, with the exception of fatigue which returned to baseline 12 months after radiotherapy. Regarding the PR25 questionnaire, there was an increase in the urinary symptom score at the end of radiotherapy before returning to pre-radiotherapy values 3 months after the end of treatment. In contrast, there was no significant modification of the score for digestive symptoms: mean score was 2.6% before treatment, 5.7% after treatment and 5% 3 months after radiotherapy, 3.8% 12 months after radiotherapy and 3.3% 18 months after radiotherapy (Fig. 3).

Fig. 3
figure 3

Evaluation of the quality of life

Full size image

Discussion

This is the first prospective study with long term toxicity datas, evaluating the practical usefulness of a new rectal-prostate spacer, the ProSpace® balloon, as an additional tool for modern radiotherapy by IMRT with daily IGRT using CBCT. The main limitation of our study is the small number of patients enrolled. Our results suggest that the use of a balloon is simple and does not delay the initiation of treatment in a clinically meaningful way. The placement of the balloon by an experienced interventional radiologist proved to be easy or very easy, without immediate or delayed complications. We didn’t report any acute toxicity related to the ballon placement. Because of the hydrodissection of the rectoprostatic space is performed, we should keep in mind the rectum perforation risk [24]. In another report, acute grade 3 toxicity (rectum perforation and urethral damage) directly related to the implantation procedure occurred in 1.49% [25]. No radiotherapy treatment was postponed due to any issues arising as a result of the placement of the ProSpace® balloon. Patient satisfaction after ProSpace® balloon placement was high and patients reported a high QoL before and after balloon implantation. The use of the balloon allowed a significant dosimetric gain, both in terms of protection and sparing of the rectum, especially in comparison with high doses, but also in terms of coverage of the PTV, since no uncovering of the PTV was necessary in order to respect the rectum constraints and the Dmin constraint was still satisfied. In modern IMRT, the V98% > 95% target set is a knew objective in the treatment planning. By the addition of the ballon, The V98% > 95% target set in our study was also achieved, with a reduction of the dose received by the rectum. No dosimetric impact on the bladder was found. Filling the balloon with saline solution containing a iodinated contrast agent made it easier to visualise the device on the dosimetric scan [19]. It was also an aid to the daily repositioning of the patient during treatment sessions. The use of spacers can also reduce prostate gland and the seminal vesicles intra-fraction motion [26, 27].

In the long-term, rectal toxicity remains particularly low, with no grade ≥ 2 toxicity identified and QoL scores remain very high, especially those based on digestive criteria. Our results are consistent with previous studies that have identified a strong link between the radiation dose received by the rectum, especially in its lower part, and the digestive QoL [28, 29]. In the context of modern radiotherapy with IMRT and IGRT, the expected rates of grade 2 and grade 3 rectal toxicities are approximately 20% and 3%, respectively [6]. Several rectal dose constraints have been published to reduce rectal toxicity, but there is no strong consensus on the doses to be respected [1]. Therefore, long-term follow-up is essential to ensure the correlation between dosimetric gain and the absence of long-term toxicity. On the other hand, there does not seem to be any positive correlation between the use of a rectal-prostate spacer and genitourinary toxicity. In a phase III study evaluating the utility of the SpaceOAR® rectal-prostate spacer, a hydrogel, Hamstra et al. explored different urinary structures such as the bladder, bladder wall, bladder neck, prostate or penile bulb, and failed to demonstrate a clear relationship between dosimetric characteristics and urinary toxicity [15]. Over a 3-year follow-up, the use of this spacer decreased grade ≥ 1 (9.2% vs. 2.0%, p = 0.028) and grade ≥ 2 (5.7% vs. 0%, p = 0.012) rectal toxicity.

The development of radiotherapy for prostate cancer is limited by its genitourinary toxicity and especially by its long-term gastrointestinal toxicity, which is directly related to the doses received by the rectum. In the GETUG 06 trial, a randomised phase III trial that compared a standard radiotherapy at a dose of 70 Gy vs. 80 Gy dose escalation for intermediate- to high-risk prostate cancer, Beckendorf et al. reported, in the 80 Gy experimental arm, grade ≥ 2 delayed gastrointestinal and genitourinary toxicity rates of 19.5% and 17.5%, respectively [6]. Because of this significant increase in toxicity it was not possible to create a new standard of care despite an improvement in biochemical relapse-free survival at 5 years [6]. The randomised phase III trial FLAME, which evaluated the utility of an intralesional boost of 95 Gy associated with prostate irradiation with 77 Gy, demonstrated a benefit in term of biochemical control but no benefit in terms of specific survival of prostate cancer and overall survival. There was no difference in long-term gastrointestinal toxicity, with grade ≥ 2 delayed gastrointestinal toxicity of 12% and 13%, equivalent in both arms. In that trial, priority was given to respecting the constraints for organs at risk over PTV boost coverage, which may explain the negative survival results [30].

Other radiotherapy techniques for prostate cancer with dose escalation have been evaluated, including moderate hypofractionated radiotherapy, severe hypofractionated radiotherapy under stereotactic conditions, exclusive or boost brachytherapy and proton therapy [31,32,33,34,35,36,37,38,39,40]. For each of these therapeutic modalities, gastrointestinal toxicity and long-term digestive QoL were correlated with the doses received by the rectum. In a phase I/II dose escalation study of radiotherapy under stereotactic conditions, nearly 10% of patients in the highest dose group (50 Gy in five fractions) required a colostomy[41]. A rectal-prostate spacer, shown to be useful in conventional fractionation radiotherapy, could also bring a significant benefit when used to enable therapy intensification, regardless of the technical modality chosen by the clinician. The use of a rectal-prostate spacer could improve the therapeutic index by allowing higher doses to be delivered to the prostate without increasing the doses received by the rectum.

That pilot study was followed by a prospective trial, the BioPro-RCMI study, which confirmed the results obtained in our cohort [42]. The use of a rectal-prostate spacer is in the process of being approved by the competent health authorities for reimbursement in patients with prostate cancer. The selection of patients is important. In these studies, placement of a rectal-prostate spacer was contraindicated in patients at risk of capsular penetration or seminal vesicle invasion [14, 15, 42]. No data exist at this time to support the safety of spacer use in patients with prostate cancer with capsular penetration or seminal vesicle invasion. A theoretical risk of dissemination of tumour cells during hydrodissection of the rectoprostatic space exists and should not be overlooked.

Conclusions

The use of a rectal-prostate spacer of the ProSpace® balloon type seems to be well accepted by patients with intermediate- or low-risk prostate cancer, allowing a double dosimetric gain: a significant decrease in doses received by the rectum and an improvement in the coverage of the high-risk PTV. The early results were confirmed long-term with low gastrointestinal toxicity (0% grade ≥ 2) and preserved QoL in all treated patients.