Journal Pre-proof

PROSTATE BED DELINEATION GUIDELINES FOR POSTOPERATIVE

RADIOTHERAPY, ON BEHALF OF THE GFRU (FRANCOPHONE GROUP OF
UROLOGICAL RADIOTHERAPY)

Sophie ROBIN, MD, Marjory JOLICOEUR, MD, Samuel PALUMBO, MD, Thomas
ZILLI, MD, Gilles CREHANGE, MD PhD, Olivier DE. HERTOGH, MD, Talar
DERASHODIAN, MD, Paul SARGOS, MD, Carl SALEMBIER, MD, Stéphane
SUPIOT, MD, PhD, Corina UDRESCU, PhD, Olivier CHAPET, MD, PhD.
PII: S0360-3016(20)34498-9
DOI: https://doi.org/10.1016/j.ijrobp.2020.11.010
Reference: ROB 26703

To appear in: International Journal of Radiation Oncology • Biology • Physics

Received Date: 19 July 2020

Revised Date: 14 October 2020
Accepted Date: 2 November 2020

Please cite this article as: ROBIN S, JOLICOEUR M, PALUMBO S, ZILLI T, CREHANGE G,
HERTOGH OD, DERASHODIAN T, SARGOS P, SALEMBIER C, SUPIOT S, UDRESCU C, CHAPET
O, PROSTATE BED DELINEATION GUIDELINES FOR POSTOPERATIVE RADIOTHERAPY,
ON BEHALF OF THE GFRU (FRANCOPHONE GROUP OF UROLOGICAL RADIOTHERAPY),
International Journal of Radiation Oncology • Biology • Physics (2020), doi: https://doi.org/10.1016/
j.ijrobp.2020.11.010.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier Inc.

 PROSTATE BED DELINEATION GUIDELINES FOR

POSTOPERATIVE RADIOTHERAPY, ON BEHALF OF THE GFRU

(FRANCOPHONE GROUP OF UROLOGICAL RADIOTHERAPY)

Sophie ROBIN, MD1, Marjory JOLICOEUR, MD2, Samuel PALUMBO, MD3, Thomas ZILLI,

MD4, Gilles CREHANGE, MD PhD5, Olivier DE HERTOGH, MD6, Talar DERASHODIAN,

of
MD2, Paul SARGOS, MD7, Carl SALEMBIER, MD8, Stéphane SUPIOT, MD, PhD9,

Corina UDRESCU, PhD1, Olivier CHAPET1 MD, PhD.

ro
-p
1. Radiation Oncology Department, Centre Hospitalier Lyon Sud, Pierre Benite, France
re
2. Radiation Oncology Department, Charles LeMoyne Hospital, CISSS Montérégie-
centre, Montréal, Canada
lP

3. Radiation Oncology Department, CHU UCL Namur – Sainte Elisabeth, Namur,

na

Belgium.
4. Radiation Oncology Department, Geneva University Hospital, Geneva, Switzerland
ur

and Faculty of Medicine, Geneva, Switzerland

5. Radiation Oncology Department, Institut Curie, Saint-Cloud, France.
Jo

6. Radiation Oncology Department, CHR Verviers East Belgium, Verviers, Belgium

7. Radiation Oncology Department, Jewish General Hospital, McGill, Montreal, Canada
8. Department of Radiotherapy, Europe Hospitals Brussels, Belgium.
9. Radiation Oncology Department, Institut de Cancérologie de l’Ouest, Nantes Saint-
Herblain, France and CRCINA CNRS Inserm, University of Nantes and Angers, Nantes,
France

1
Corresponding Author:

Pr. Olivier CHAPET

Service de Radiothérapie Oncologie
Centre Hospitalier Lyon Sud
165 Chemin du Grand Revoyet
69495 PIERRE BENITE, FRANCE
Phone number: +33 4 78 86 42 51
Fax: +33 4 78 86 42 65

of
E-mail: [email protected]

ro
Short Running Title: GFRU prostate bed radiotherapy guidelines
-p
re
Number of pages:
lP

Title page and additional information: 2 pages

Abstract: 1 page
Manuscript: 24 pages
na

Number of tables: 1
Number of figures: 4
ur

Word count
Abstract: 267 words
Jo

Manuscript: 3845 words

Conflict of interest: The authors report no conflict of interest.

Funding: None

Acknowledgements: we would like to express our gratitude to all the colleagues and
members of the GFRU, the Francophone Group of Urological Radiotherapy.

2
 PROSTATE BED DELINEATION GUIDELINES FOR

POSTOPERATIVE RADIOTHERAPY, ON BEHALF OF THE GFRU

(FRANCOPHONE GROUP OF UROLOGICAL RADIOTHERAPY)

of
ro
-p
Short Running Title: GFRU prostate bed radiotherapy guidelines
re
lP
na
ur
Jo

1
ABSTRACT

Purpose: Prostate bed (PB) irradiation is considered the standard post-operative treatment after

radical prostatectomy (RP) for tumors with high-risk features and/or persistant PSA, or for

salvage treatment in case of biological relapse. Four consensus guidelines have been published to

standardize practices and reduce the inter-observer variability in PB delineation, however with

discordant recommendations. In order to improve the reproducibility in the PB delineation, the

Francophone Group of Urological Radiotherapy (GFRU – Groupe Francophone de

of
Radiothérapie Urologique) worked to propose a new and more reproducible consensus guideline

ro
for PB clinical target volume (CTV) definition.

-p
Methods and Materials: A four-step procedure was used. First, a group of 10 GFRU prostate
re
experts evaluated the four existing delineation guidelines for post-operative radiotherapy
lP

(EORTC, FROGG, RTOG, and PMH) in order to identify divergent issues. Second, datasets of

50 magnetic resonance imaging (MRI) studies (25 after RP and 25 with an intact prostate gland)
na

were analyzed to identify the relevant anatomical boundaries of the PB. Third, a literature review

of surgical, anatomical, histological, and imaging data was performed to identify the relevant PB
ur

boundaries. Fourth, a final consensus on PB-CTV definition was reached among experts.
Jo

Results: Definitive limits of the PB-CTV delineation were defined, using easily visible

landmarks on computed tomography scans (CT). The purpose was to ensure a better

reproducibility of PB definition for any radiation oncologist even without experience in post-

operative radiotherapy.

Conclusions: New recommendations for PB delineation based on simple anatomical boundaries

and available as a CT image atlas are proposed by the GFRU. Improvement in uniformity in PB-

CTV definition and treatment homogeneity in the context of clinical trials are expected.

2
INTRODUCTION

Radical prostatectomy (RP) is one of the standard treatments for localized prostate cancer

[1]. However, approximately one third of patients experience a biological recurrence

within the decade following surgery [2,3].

Salvage radiotherapy is recommended for the management of biochemical relapse

after RP [2-5]. The clinical target volume (CTV) for post-operative radiotherapy is the

of
prostate bed (PB), sometimes extended to the pelvic lymph nodes [6]. However, after

ro
surgery, the delineation of the CTV is complex and subject to large intra- and inter-
-p
observer variations [7]. Four guidelines are already available, in order to assist the
re
radiation oncologist with the delineation of the PB [8-11]. Nevertheless, these guidelines
lP

differ in several major points, such as the borders of the PB at the apex or at the base,
na

limiting therefore an overall contouring agreement among the radiation oncology

community. Moreover, use of modern imaging techniques like multiparametric magnetic

ur

resonance imaging (mpMRI) and more recently prostate-specific membrane antigen

Jo

(PSMA) PET/CT have been implemented in the restaging workflow of relapsing prostate

cancer, with a clear impact on treatment volume delineation [12-14].

The GFRU identified the need to generate new recommendations for PB

delineation and a consensual atlas based on simple and reproducible anatomical

landmarks, easy to be identified on planning computed tomography (CT) datasets.

3
MATERIALS AND METHODS

A group of 10 GFRU prostate radiation oncologists’ experts (4 from France, 3

from Belgium, 1 from Switzerland, and 2 from Quebec-Canada) worked together for the

definition of the PB, following a four-step procedure:

Step 1: PB delineation guidelines review

of
Four existing guidelines for PB-CTV delineation were analyzed: the European

ro
Organization for Research and Treatment of Cancer (EORTC) [9]; the Faculty of
-p
Radiation Oncology Genito-Urinary Group (FROGG) [10]; the Radiation Therapy
re
Oncology Group (RTOG) [8]; and the Princess Margaret Hospital (PMH) [11] consensus
lP

guidelines. For the definition of each anatomical boundary of the PB, the four guidelines
na

were compared and variations in PB definition were identified (Table 1).

ur

Step 2: Literature review

Jo

The methodology used in this study was based on a critical, not systematic,

review of the literature of the last three decades up to March 2020, on PubMed, to collect

surgical, radiological, anatomical and/or histological information likely to help finding an

accurate and converging definition of the delineation limits of the PB differing between

the four guidelines.

4
Step 3: Magnetic resonance imaging (MRI)-based analysis of PB boundaries

For each boundary of the PB delineation, the GFRU group performed an analysis

of 50 prostate T2-MRI series. Twenty-five patients have already had a prostatectomy and

25 different patients had an intact prostate gland.

Step 4: Consensus on PB definition and CT image

of
Based on the analysis of these four guidelines, the review of the literature and the

ro
analysis of the MRI acquisitions, a final consensus on limits for PB-CTV definition was
-p
reached among the 10 GFRU experts and a CT image atlas was proposed (maximum
re
thickness of 3 mm for the continuous CT scan slices with an injection of contrast agent).
lP

The final consensus on these guidelines was established among the panelists after several
na

meetings conducted from 2016 to prepare the GFRU contouring workshops.

ur

RESULTS
Jo

INFERIOR LIMIT

Existing guidelines

Two landmark structures are commonly used to define the inferior limit of the

PB-CTV delineation: the vesico-urethral anastomosis (VUA) with an additional margin

below it [8,10,11] and the penile bulb [9-11]. The recommended limit to define the PB-

CTV apex ranges on the different guidelines between 5 and 12 mm below the VUA. The

distance from the cranial part of the penile bulb to the inferior limit of the PB-CTV also

ranges from a minimum of one CT slice (thickness not defined) up to15 mm.

5
Analysis of the literature

Urethrography has been used to define the prostate apex by providing a clear

visualization of the penile urethra to the point where it enters the urogenital diaphragm

[15]. The penile bulb, an easily identifiable soft tissue structure, lying immediately below

the urogenital diaphragm of the pelvic floor, can be used as a surrogate landmark for the

prostate apex [16,17]. Studies correlating the penile bulb location with the prostate apex,

of
suggest an average distance between the two structures of 15 mm based on the MRI

ro
imaging [17]. Incorporation of this average distance into treatment planning has been
-p
associated with satisfactory target coverage of the apical region of the prostate [18].
re
Lock et al. compared on 10 patients the relative accuracy of urethrogram or
lP

penile bulb delineation as surrogate markers for the prostate apex [19]. The authors
na

showed that the penile bulb can be used to identify the prostate apex, and that the

measurements between the penile bulb and the apex are consistent between patients and
ur

through the course of treatment. Penile bulb can be reliably contoured between observers,
Jo

ensuring a localization of the prostate apex comparable to urethrography [19].

MRI analysis

Apex-penile bulb distance

The distance between the prostate apex and the penile bulb measured on the 25

prostate T2-MRI acquisitions was on average 6.7 mm (range, 4.7-11 mm) (Figure 1). All

the measurements were inferior to 15 mm in contrast with the results of the literature

[17]. Using the EORTC definition (where the apex was localized at 15 mm from the

6
penile bulb) [9], the PB-CTV would not be correctly covered at the apex for all the 25

patients. For the three other guidelines [8,10,11], the first slice of PB-CTV corresponds to

the first slice above the penile bulb. Assuming that the slice thickness of the planning CT

does not exceed 5 mm, the PB-CTV apex would be correctly covered for all cases

analyzed.

VUA-penile bulb distance

of
The distance between the VUA and the penile bulb was measured on the 25 post-

ro
operative T2-MRI acquisitions (Figure 1). This distance ranged from 10.3 mm to 27 mm,
-p
with an average of 18.4 mm. In the PMH, FROGG, and RTOG guidelines, the inferior
re
limit of the PB-CTV is defined at 8 mm, 5-6 mm, and 8-12 mm below the VUA,
lP

respectively [8,10,11]. According to these three guidelines, on the 25 post-operative T2-

na

MRI series, the most inferior slice of the PB-CTV delineation would be on average 9.6

mm (range, 2.3-19 mm), 12.1 mm (range, 4.8-21.5 mm), and 7.6 mm (range, 0.3-17 mm)
ur

above the penile bulb, respectively (Figure 2). In the present analysis on the 25 post-
Jo

operative MRI acquisitions, major discrepancies compared to existing guidelines were

demonstrated.

GFRU analysis

Identification of the VUA is not easy on CT imaging because of the postsurgical

rearrangements and requires the use of an intravenous injection of contrast. Moreover,

according with the measures above, there is some variability in the distance between the

VUA and the penile bulb on post-prostatectomy MRI. Consequently, the current

7
definitions of the apex based on VUA (PMH, FROGG and RTOG) could be

inappropriate to systematically cover the inferior limit of the PB-CTV as it has already

been shown by Manji et al. [20].

The penile bulb is a structure easily identified on the CT imaging even without

contrast injection and its position remains stable after RP. By starting the delineation of

PB-CTV 5 mm above the penile bulb, the apical part of the PB-CTV was correctly

covered on all 25 analyzed prostate MRI acquisitions.

of
ro
GFRU definition
-p
At its most inferior part the PB_CTV lies between the inferior limit located
re
5mm above the penile bulb. The posterior limit is represented by the anterior wall
lP

of the rectum or of the anal canal. The lateral and anterior limits are the pelvic
na

muscles or the insertion of the corpora cavernosa (Figure 4A). These limits are in

correlation with the other guidelines.

ur
Jo

MIDDLE SECTION

Existing guidelines

The four guidelines propose very similar limits to delineate the middle section of

the PB-CTV: the pubic symphysis anteriorly, the levator ani or the obturator internus

muscles laterally, and the anterior rectal wall posteriorly [8-11]. A small variation is

proposed by the RTOG and FROGG guidelines [8,10], which suggest that the posterior

limit of the PB-CTV needs to be concave on both side of the rectum to better include the

rectoprostatic angles [8,10].

8
Analysis of the literature

Nevoux et al. performed a quantitative tissue analysis of prostate cancer foci in an

unselected series of 96 cysto-prostatectomy specimens [21]. They showed that 75% of

the tumors are located in the peripheral zone [21]. In the middle part of the prostate, the

of
tumors of more than 0.5 cc are mainly located in the peripheral zone and more

ro
specifically in the two posterolateral areas [21]. When target volumes were delineated
-p
using the RTOG guideline, the CTV coverage was marginal in the posterolateral regions
re
near to the rectum and the mesorectal fascia [8]. In another series analyzing 121 surgical
lP

specimens, the extracapsular extension occurred postero-laterally along the neurovascular

na

bundle in all analyzed cases [22]. These results clearly support the need to have a

concave delineation of the posterior limit of the PB-CTV on both sides of the rectum.
ur
Jo

MRI analysis

A specific analysis of the available T2-MRI series was not found to be relevant

for the delineation of the middle section of the PB-CTV.

GFRU analysis

Anteriorly and laterally, the four guidelines converge on similar recommendations

based on the anatomical definition of the structures surrounding the prostate. Posteriorly,

9
the anterior rectal wall represents the limit. Based on the review of the literature, the two

posterolateral recto-prostatic angles need to be included in the PB-CTV volume.

GFRU definition

In the middle section, the posterior limit of the PB-CTV is the anterior

border of the rectum including the posterolateral angles on both sides of the rectum

of
of 5 mm. The experts considered that it is a reasonable compromise to cover the risk

ro
of relapse and to limit the irradiation of the rectal wall. The lateral limits are the
-p
internal borders of the levator ani or of the obturator internus muscles. The muscles
re
should not be included in the PB-CTV. The anterior limit is represented by the
lP

posterior border of the pubic symphysis (Figure 4B, C).

na

ANTERIOR UPPER LIMIT

ur

Existing guidelines
Jo

Based on the EORTC guidelines [9], the anterior upper limit of the PB-CTV

should include “the VUA and the urethral axis”. In the PMH and RTOG guidelines, this

limit is represented by the top of the edge of the pubic bone [8,11]. In the FROGG

guidelines [10], from the lower border of the PB-CTV to 3cm superior, the anterior

border of the PB-CTV is the posterior aspect of the symphysis pubis. In these last three

guidelines [8,10,11], at least 1.5 cm of the bladder neck must be included in the

delineation (up to 2 cm in the RTOG guideline) [8].

10
Analysis of the literature

Based on the Nevoux et al. study [21], no significant tumor (> 0.1cc) is generally

found on the pathological RP specimen in the upper anterior third of the prostate.

MRI analysis data based on the guidelines

On the 25 prostate MRI acquisitions, the length of contact between the prostate

and the pubic bone was measured and the ratio between the length of this contact and the

of
total length of the pubic bone was calculated. The ratio varied from one patient to another

ro
from 17% to 90%. However, in 80% of cases, this percentage was inferior to 66%

(Figure 3).
-p
re
lP

GFRU analysis
na

In a study by Freitag et al., 119 patients with biochemical recurrence after RP

were restaged both with hybrid 68Ga-PSMA-11-PET/CTlow-dose and PET/MRI including a

ur

multiparametric MRI (mpMRI) protocol of the PB [23]. The authors observed that the
Jo

detection rate of local recurrences using the PET-component was significantly influenced
68
by the proximity to the bladder, with the risk to miss relapses due to the Ga-PSMA

residual urinary radioactivity [23]. These findings were confirmed by another study by

Achard et al. suggesting the added value of mpMRI imaging for the detection of PB
18
recurrences compared to standard whole body hybrid F-choline PET/MRI protocols

[24]. Compared to PET, mpMRI was able to detect more local relapses (17 vs 14 patients

over 58 analyzed), mostly located in the anastomotic region, the bladder neck and the SV

bed [24].

11
 In a systematic literature review concerning the current role of mpMRI in the

detection of locoregional recurrence, Barchetti et al. reported that after RP, the most

common site of local recurrence is the vesico-urethral anastomosis around the urinary

bladder and/or membranous urethra [25]. Other common sites of local recurrence are

retrovesical (between the urinary bladder and rectum), within retained SVs, at the anterior

or lateral surgical margins of the prostatectomy bed (e.g., abutting the levator ani

muscles) and at the resection site of the vas deferens [25].

of
Similar results were observed by Zilli et al. in a series of 171 prostate cancer

ro
patients relapsing after RP and restaged with an endorectal MRI before salvage
-p
radiotherapy [26]. Among the 131 patients with a positive MRI imaging, the peri-
re
anastomotic (35.9%) and the bladder neck region (33.6%) were the most common sites of
lP

local recurrence, followed by penile bulb (19%) and the SV bed (3.8%) [26].
na

Based on the above studies, on the Nevoux et al. study [21] and the analysis on

the 25 MRI acquisitions, the use of the top of the edge of the pubic bone as upper anterior
ur

limit of the PB-CTV seems to be a quite generous landmark. An upper limit located at
Jo

2/3 of the pelvic bone (closer to the FROGG definition [10]) appears a reasonable

solution to cover the area at risk of relapse and to limit the volume of bladder included in

the high-dose volume. In the three guidelines [8,10,11], a length of 1.5 cm of the bladder

neck has to be included in the PB-CTV. This rule is necessary to cover the VUA and the

interface between the prostate and the bladder. In the FROGG guideline [10], the PB-

CTV must be extended by at least 3 cm from the lower slice of delineation. This minimal

length is reasonable according to the size of the prostate and the necessity to cover at

least 1.5 cm of bladder neck.

12
GFRU definition

Delineation of the upper anterior limit of the PB-CTV must fulfill three

criteria (Figure 4D, E):

Criteria n°1: At least 1.5 cm of the bladder neck must be included in the PB-

CTV.

Criteria n°2: The PB-CTV must cover the posterior border of the pubic bone

of
on at least 2/3 of its length.

ro
Criteria n°3: At least 3 cm are necessary between the lower and upper slices
-p
of delineation of the PB-CTV along the pubic bone.
re
When the three criteria are fulfilled, the anterior delineation of the CTV
lP

along the pubic bone is discontinued.

na

SEMINAL VESICLES BED

ur

Existing guidelines
Jo

In the EORTC guidelines [9], delineation of the seminal vesicles (SV) bed is

recommended only in case of SV invasion on the surgical specimen. The PB-CTV

includes the original location of the SV. In the PMH guidelines [11], the PB-CTV is

delineated up to the vas deferens (5 mm above the inferior border of the vas deferens) and

must include all the surgical clips. The FROGG guidelines use the same limits but

specify that residual SV must be included in the volume [10]. Lastly, in the RTOG

guidelines the PB is delineated up to the vas deferens, or 3 to 4 cm above the top of the

pubic symphysis and includes SV remnants if pathologically involved [8].

13
Analysis of the literature

In several studies on MRI analysis, the rate of involvement of the bottom of the

SV is very low, ranging between 0% and 13%. In the Samaratunga et al. study, 16% of

the SV invasions were located in the distal third of the SV [27]. Kestin et al. measured

the length of cancer involvement from the prostate to the SV junction [28]. On the 81

pathologic specimens analyzed in this study, the risk of SV involvement beyond 2 cm

of
was approximately 1% [28]. In another study on 71 patients treated with RP, 12 patients

ro
(17%) had a SV involvement but none of them had a pathological involvement of the last

1 cm of the SV [29].
-p
re
lP

MRI analysis
na

The distance between the distal part of the SV and the top of the pubic bone is

used in the RTOG guidelines [8]. This definition is by far the easiest to apply. In the 25
ur

prostate MRI acquisitions, the distance between the extremity of the SV and the top of
Jo

the pubic symphysis was measured on average at 26.4 mm [range, 3-56.2 mm]. This

distance was less than 4 cm and 3 cm in 84% and 76% of the cases, respectively.

GFRU analysis

The vas deferens arises from the testicle, following the epididymal canal and it

ends at the confluence of the SV and the ejaculatory duct. The union between the vas

deferens and the neck of the SV forms the ejaculatory duct at the base of the prostate.

Using the vas deferens to define the upper border of the SV bed might present some

14
limitations:

1- The vas deferens is not always visible on the planning CT.

2- The surgical section of vas deferens may vary from a surgery (and surgeon) to

another.

3- The vas deferens may retract upward and backward after RP.

For these reasons, vas deferens may not be the most appropriate anatomical landmark to

define the upper limit of the delineation of the SV bed. According to the analysis of the

of
25 MRI acquisitions and the review of the literature, the RTOG definition [8] based on

ro
the top of pubic bone (+ 3cm) seems accurate, and highly reproducible.
-p
re
GFRU definition
lP

1. If SV are pathologically involved (Figure 4F, I, J):

na

The superior limit of delineation of the SV bed is defined at 3 cm above the top of

the pubic symphysis. This limit can be extended up to 4 cm in case of involvement of

ur

the last third of the SV on the histopathological specimen. The posterior limit is the
Jo

anterior border of the mesorectum. The GFRU recommends the inclusion of the

posterior third of the bladder wall (with a thickness of 1cm) to better encompass the

the SV bed. The lateral limits are the internal obturator muscles.

2. If SV are not pathologically involved (Figure 4E, G, H):

In order to cover the prostate-SV junction, the superior border of the SV bed is

maintained to the first 1 cm above the pubic symphysis, keeping the same anterior,

posterior and lateral limits used in case of SV involvement.

15
Considering the minimal risk of pathological involvement of the bottom part of the

SV, the superior border limit of the SV bed can be reduced in order to respect the

dose constraints to the rectum and the bladder.

DISCUSSION

Based on the existing literature, postoperative RT for prostate cancer is associated

of
with a large interobserver variability in the PB-CTV contouring [30]. Systematic errors in

ro
PB-CTV definition may impact the final dosimetry and treatment delivery by translating
-p
into possible underdosage of the target and/or overdosage of the healthy tissues [30].
re
International guidelines have been developed to assist radiation oncologists in
lP

standardizing the contouring process and potentially reducing its variability [8-11].
na

Why is a new guideline necessary?

ur

The present GFRU analysis of the existing guidelines shows some large variations
Jo

in the limits of delineation of the PB-CTV which may induce significant variations in the

doses delivered to the target and to the organs at risk [31]. Differences in methodology

used for defining the PB-CTV in the four guidelines can explain this variability. The

EORTC guidelines do not provide a precise description of the methodology used [9]. The

PMH guidelines have been generated evaluating data based on the topography of the

post-RP relapses, as well as based on radiological anatomy and surgical findings [11].

The FROGG guidelines are the result of an expert’s debate on the PMH contouring atlas

[10]. The RTOG atlas uses an algorithm to determine the PB-CTV borders taking into

16
consideration the site of post-RP relapse combined with surgical and anatomical data [8].

In the Malone et al. study [32], the four consensus guidelines were compared in

20 patients in terms of treatment volumes and organs at risk irradiation. The PB-CTV

differed significantly between the four guidelines, allowing a potential impact on long-

term clinical outcome and treatment-related toxicity [32]. The PB-CTV volume defined

using the EORTC guidelines was significantly smaller than the CTVs defined using the

other recommendations, with a more limited coverage of the PB in the anterior and

of
superior directions [32]. In another study, Ost et al. analyzed the inter-observer

ro
variability in contouring the PB-CTV according to the EORTC guidelines [33]. They
-p
showed only a moderate observer agreement for both the PB-CTV (mean kappa, 0.49;
re
range, 0.35–0.62) and the SV bed (mean kappa, 0.42; range, 0.22–0.59) [33].
lP

Taking into consideration limitations and divergences of the existing guidelines,

na

the GFRU defined the need for a new guideline and atlas, able to limit the inter-observer

variations with well-defined anatomical limits easily identified by any physician.

ur
Jo

Why a guideline based on CT imaging?

Matching postoperative MRI and CT may be challenging due to anatomical

variations in the PB shape between the two exams. Based on these considerations, the

GFRU experts’ panel estimates that a guideline based on only one single imaging

modality is more adapted to homogenize contouring of the PB-CTV among the radiation

oncology community. As observed by Barkati et al. using the RTOG guidelines, defining

the PB-CTV based on CT imaging resulted in a statistically significant lower inter-

observer variability (mean dice similarity coefficient: 0.76) compared to a MRI-based

17
contouring (mean dice similarity coefficient: 0.66) [34]. The increasing use of mpMRI

for restaging and radiotherapy planning, along with the diffusion among the radiation

oncology community of educational platforms for contouring, are both expected to

reduce this interobserver variability in PB definition.

What are the interests of this new guideline?

We applied a three-step methodology was applied based on the analysis of the

of
existing guidelines and their discordances, on a review of the literature, and on the study

ro
of 50 MRI datasets. This guideline, written by experts in the field of prostate
-p
radiotherapy, is based on simple anatomical structures easy to be identified on a planning
re
CT: the penile bulb, the bladder, the rectum, the mesorectum, the pubic symphysis, and
lP

the pelvic muscles. Anatomical boundaries more difficult to be identified on CT imaging

na

or associated with a variable position in the pelvis (such vas deferens or the VUA) were

avoided.
ur

We acknowledge that a stronger methodology based on a systematic review of the

Jo

literature, a larger panel of experts including specialists from other disciplines, and

integration of agreement measures among the panelists, would have provided more

robustness to our guidelines. However, a validation analysis of our consensus guidelines

is actually ongoing using the available datasets from three contouring workshops

organized annually from 2017 by the GFRU.

CONCLUSIONS

New recommendations for PB-CTV delineation based on simple anatomical

boundaries and available as a CT image atlas are proposed by the GFRU for

18
postoperative prostate radiotherapy. Improvement in uniformity in PB-CTV definition

and treatment homogeneity in the context of clinical trials are expected. Further

validation of these consensus guidelines is ongoing based on the data of several

contouring workshops organized by the GFRU with radiation oncologists from France,

Belgium, Switzerland, and Morocco.

References

of
1. Hamdy FC, Donovan JL, Lane JA, et al. 10-year outcomes after monitoring, surgery,

or radiotherapy for localized prostate cancer. N Engl J Med 2016;375:1415-1424.

ro
2. Cornford P, Bellmunt J, Bolla M, et al. Eau-estro-siog guidelines on prostate cancer.
-p
Part ii: Treatment of relapsing, metastatic, and castration-resistant prostate cancer.
re
Eur Urol 2017;71:630-642.
lP

3. Pisansky TM, Thompson IM, Valicenti RK, et al. Adjuvant and salvage radiation
na

therapy after prostatectomy: Astro/aua guideline amendment, executive summary

2018. Pract Radiat Oncol 2019;9:208-213.

ur

4. Carrie C, Magne N, Burban-Provost P, et al. Short-term androgen deprivation therapy

Jo

combined with radiotherapy as salvage treatment after radical prostatectomy for

prostate cancer (getug-afu 16): A 112-month follow-up of a phase 3, randomised

trial. Lancet Oncol 2019;20:1740-1749.

5. Shipley WU, Seiferheld W, Lukka HR, et al. Radiation with or without antiandrogen

therapy in recurrent prostate cancer. N Engl J Med 2017;376:417-428.

6. Pollack A, Karrison TG, Balogh AG, et al. Short term androgen deprivation therapy

without or with pelvic lymph node treatment added to prostate bed only salvage

radiotherapy: The nrg oncology/rtog 0534 spport trial 2018;102:1605.

19
7. Latorzeff I, Sargos P, Loos G, et al. Delineation of the prostate bed: The "invisible

target" is still an issue? Front Oncol 2017;7:108.

8. Michalski JM, Lawton C, El Naqa I, et al. Development of rtog consensus guidelines

for the definition of the clinical target volume for postoperative conformal

radiation therapy for prostate cancer. Int J Radiat Oncol Biol Phys 2010;76:361-

368.

9. Poortmans P, Bossi A, Vandeputte K, et al. Guidelines for target volume definition in

of
post-operative radiotherapy for prostate cancer, on behalf of the eortc radiation

ro
oncology group. Radiother Oncol 2007;84:121-127.
-p
10. Sidhom MA, Kneebone AB, Lehman M, et al. Post-prostatectomy radiation therapy:
re
Consensus guidelines of the australian and new zealand radiation oncology
lP

genito-urinary group. Radiother Oncol 2008;88:10-19.

na

11. Wiltshire KL, Brock KK, Haider MA, et al. Anatomic boundaries of the clinical

target volume (prostate bed) after radical prostatectomy. Int J Radiat Oncol Biol
ur

Phys 2007;69:1090-1099.
Jo

12. Amzalag G, Rager O, Tabouret-Viaud C, et al. Target definition in salvage

radiotherapy for recurrent prostate cancer: The role of advanced molecular

imaging. Front Oncol 2016;6:73

13. Mottet 2020 : http://uroweb.org/guidelines/compilations-of-all-guidelines/

14. Bhargava P, Ravizzini G, Chapin BF and Kundra V. Imaging biochemical recurrence

after prostatectomy: Where are we headed? AJR 2020;214:1248-1258.

15. Schild SE, Wong W. The role of retrograde urethrography in the planning of prostate

cancer radiotherapy. Med Dosim 1997;22:83-86.

20
16. Li XM, Gao XS, Guo XM, et al. Using ct imaging to delineate the prostatic apex for

radiation treatment planning. Chin J Cancer 2010;29:914-922.

17. Plants BA, Chen DT, Fiveash JB, et al. Bulb of penis as a marker for prostatic apex in

external beam radiotherapy of prostate cancer. Int J Radiat Oncol Biol Phys

2003;56:1079-1084.

18. Milosevic M, Voruganti S, Blend R, et al. Magnetic resonance imaging (mri) for

localization of the prostatic apex: Comparison to computed tomography (ct) and

of
urethrography. Radiother Oncol 1998;47:277-284.

ro
19. Lock MI, Heinrichs A, Bhattacharya G, et al. The utility of penile bulb contouring to
-p
localise the prostate apex as compared to urethrography. J Med Imaging Radiat
re
Sci 2018;49:76-83.
lP

20. Manji M, Crook J, Schmid M, et al. Target volume definition for post prostatectomy
na

radiotherapy: Do the consensus guidelines correctly define the inferior border of

the ctv? Rep Pract Oncol Radiother 2016;21:525-531.

ur

21. Nevoux P, Ouzzane A, Ahmed HU, et al. Quantitative tissue analyses of prostate
Jo

cancer foci in an unselected cystoprostatectomy series. BJU Int 2012;110:517-

523.

22. Chao KK, Goldstein NS, Yan D, et al. Clinicopathologic analysis of extracapsular

extension in prostate cancer: Should the clinical target volume be expanded

posterolaterally to account for microscopic extension? Int J Radiat Oncol Biol

Phys 2006;65:999-1007.

23. Freitag MT, Radtke JP, Afshar-Oromieh A, et al. Local recurrence of prostate cancer
68
after radical prostatectomy is at risk to be missed in Ga-PSMA-11-PET of

21
 PET/CT and PET/MRI: comparison with mpMRI integrated in simultaneous

PET/MRI. Eur J Nucl Med Mol Imaging 2017;44(5):776-87.

24. Achard V, Lamanna G, Denis A, et al. Recurrent prostate cancer after radical

prostatectomy: restaging performance of 18F-choline hybrid PET/MRI. Med

Oncol 2019;36:67.

25. Barchetti F and Panebianco V. Multiparametric MRI for recurrent prostate cancer

post radical prostatectomy and postradiation therapy. BioMed Res Int

of
2014:316272.

ro
26. Zilli T, Jorcano S, Peguret N, et al. Results of dose-adapted salvage radiotherapy after
-p
radical prostatectomy based on an endorectal MRI target definition model. Am J
re
Clin Oncol 2017;40:194-9.
lP

27. Samaratunga H, Samaratunga D, Perry-Keene J, et al. Distal seminal vesicle invasion

na

by prostate adenocarcinoma does not occur in isolation of proximal seminal

vesicle invasion or lymphovascular infiltration. Pathology 2010;42:330-333.

ur

28. Kestin L, Goldstein N, Vicini F, et al. Treatment of prostate cancer with radiotherapy:
Jo

Should the entire seminal vesicles be included in the clinical target volume? Int J

Radiat Oncol Biol Phys 2002;54:686-697.

29. Korman HJ, Watson RB, Civantos F, et al. Radical prostatectomy: Is complete

resection of the seminal vesicles really necessary? J Urol 1996;156:1081-1083.

30. Foppiano F, Franzone P, Bertoni F, et al. Airo pros 01– 04 multi-centric study in

contouring ctv for conformal radiotherapy after radicalprostatectomy. Int J Radiat

Oncol Biol Phys 2006;66:S358.

22
31. Croke J, Malone S, Roustan Delatour N, et al. Postoperative radiotherapy in prostate

cancer: The case of the missing target. Int J Radiat Oncol Biol Phys

2012;83:1160-1168.

32. Malone S, Croke J, Roustan-Delatour N, et al. Postoperative radiotherapy for prostate

cancer: A comparison of four consensus guidelines and dosimetric evaluation of

3d-crt versus tomotherapy imrt. Int J Radiat Oncol Biol Phys 2012;84:725-732.

33. Ost P, De Meerleer G, Vercauteren T, et al. Delineation of the postprostatectomy

of
prostate bed using computed tomography: Interobserver variability following the

ro
eortc delineation guidelines. Int J Radiat Oncol Biol Phys 2011;81:e143-149.
-p
34. Barkati M, Simard D, Taussky D, et al. Magnetic resonance imaging for prostate bed
re
radiotherapy planning: An inter- and intra-observer variability study. J Med
lP

Imaging Radiat Oncol 2016;60:255-259.

na
ur
Jo

23
Figure legends

of
Figure 1. Average distance with range (in mm) between the penile bulb and the apex on

ro
25 prostate magnetic resonance imaging (MRI) studies (left), and between the penile bulb
-p
and the vesico-urethral anastomosis on 25 post-operative MRI (right).
re
lP

Figure 2. Average distance with range (in mm) between the inferior border of the
na

prostate bed and the penile bulb on 25 post-operative MRI studies as defined by the

Princess Margaret Hospital (PMH) [11], Faculty of Radiation Oncology Genito-Urinary

ur

Group (FROGG) [10], and Radiation Therapy Oncology Group (RTOG) [8] guidelines.
Jo

Figure 3. Ratio of the length (in mm) of the prostate in contact with the pubic bone (PBo)

and the total length (in mm) of the PBo as measured on 25 prostate MRI studies.

Figure 4. Computed tomography-based atlas for prostate bed clinical target delineation

based on the GFRU consensus guidelines.

24
Table 1. Consensus guidelines for post-operative prostate bed clinical target delineation.

Inferior border Anterior border Lateral border Posterior border Superior border
EORTC Include the bladder neck.
Up to but not including
[9] Up to the neurovascular For patients with invasion of
Including the apex. Including the the outer rectal wall.
bundles (if removed : up the seminal vesicles, the
15 mm cranially from the anastomosis and the Cranially including the
to the ilio-obturatic prostate bed including the
penile bulb. urethral axis. most posterior part of
muscles). apex and the original location
the bladder neck
of the seminal vesicles.
PMH -Cranial boundary :
[11] The sacro- recto-genito-

f
oo
-Caudal boundary : pubic fascia, lateral to
Posterior edge of the the neurovascular The superior surgical clips (if
-Cranial boundary :
symphysis pubis up to structures. present) or 5mm above the
the mesorectal fascia.

pr
the top of the At the cranial aspect of inferior border of the vas
8mm below the
symphysis pubis. the CTV, it is not deferens.
vesicourethral anastomosis -Caudal boundary :

e-
necessary to extend to Retained seminal vesicles
or the top of the penile bulb. the anterior border of
-Cranial boundary : the obturator muscle. were included when
the rectal wall and

Pr
The posterior 1.5 cm pathologically involved.
levator ani.
of the bladder wall. -Caudal boundary :
the medial border of the

al
levator ani and obturator
internus.

rn
FROGG The space delineated by
[10] the levator ani and

u anterior rectal wall is at

Jo
risk of recurrence and The superior border should
-From the lower
should be encompassed encompass all of the vesicle
border of the CTV to
in the CTV if rectal seminal bed as defined by
5-6mm below the 3cm superior, the
dose constraints allow. non-vascular clips and should
vesicourethral anastomosis anterior border of the
Ensure a minimum 2 include the distal portion of
but should extended lower CTV is the posterior
The medial border of the cm margin from the the vas deferens (usually
to include all surgical clips aspect of the
levator ani muscle or posterior extent of the visualised superiorly as thin,
inferiorly. symphysis pubis.
obturator internus CTV to the posterior horizontal cylindrical
When the anastomosis is
muscle. rectal wall to prevent structures).
not clearly defined, the -More superiorly, the
the entire If the seminal vesicles are
inferior border will be the anterior border of the
circumference of pathologically involved by
slice above the penile bulb. CTV encompasses the
rectum receiving the tumour, ensure any residual
posterior 1.5cm of the
full radiation dose. vesicles are also included in
bladder.
More superiorly, the CTV.
posterior border of the
CTV is the anterior
mesorectal fascia.
RTOG -Above the superior
[8] edge of the symphysis
pubis :
-Below the superior -Above the superior
Sacrorectogenitopubic
edge of the symphysis edge of the symphysis
fascia.
pubis : pubis : Level of cut end of vas
If concern about
8-12 mm below Posterior edge of Mesorectal fascia. deferens or 3-4 cm above top
extraprostatic disease at
vesicourethral anastomosis. pubic bone. of symphysis.
base may extend to
May include more if -Below the superior Vas may retract
obturator internus.
concern for apical margins -Above the superior edge of the symphysis postoperatively.
(respecting penile bulb). edge of the symphysis pubis : Include seminal vesicle
-Below the superior
pubis : Anterior rectal wall. remnants if pathologically
edge of the symphysis
Posterior 1-2cm of May need to be involved.
pubis :

f
bladder wall. concave around lateral

oo
Levator ani muscles,
aspects.
obturator internus
muscles.

pr
GFRU -If SV are pathologically

e-
Delineation of the
involved :
upper anterior limit of
The superior limit of

Pr
the PB-CTV must
delineation of the SV bed is
fulfill three criteria :
defined at 3 cm above the top
-At least 1.5 cm of the
of the pubic symphysis. This

al
bladder neck must be
limit can be extended up to 4
included in the PB-
cm in case of involvement of

rn
CTV.
the last third of the SV on the
-The PB-CTV must
5 mm above the PB. histopathological specimen.

u
The posterior limit is the
cover the posterior
Internal borders of the The GFRU recommends the
Jo
border of the pubic The anterior border of
anterior wall of the rectum levator ani or obturator inclusion of the posterior third
bone on at least 2/3 of the rectum including
or of the anal canal. internus muscles. of the bladder wall (with a
its length. the posterolateral
The lateral and anterior The muscles are not thickness of 1cm) to better
-At least 3 cm are angles in both sides of
limits are the pelvic muscles included in the volume encompass the the SV bed.
necessary between the the rectum in 5 mm.
or the insertion of the of delineation.
lower and upper slices
corpus cavernosum. -If SV are not pathologically
of delineation of the
involved :
PB-CTV along the
The superior border of the SV
pubic bone.
bed is reduced to the first 1
When the three rules
cm above the pubic
criteria are fulfilled,
symphysis, keeping the same
the anterior
anterior, posterior and lateral
delineation of the
limits used in case of SV
CTV along the pubic
involvement.
bone is discontinued.
Jo
ur
na
lP
re
-p
ro
of
Jo
ur
na
lP
re
-p
ro
of
Jo
ur
na
lP
re
-p
ro
of
Jo
ur
na
lP
re
-p
ro
of