www.redjournal.

org

Clinical Investigation

NRG Oncology Updated International Consensus

Atlas on Pelvic Lymph Node Volumes for Intact
and Postoperative Prostate Cancer
William A. Hall, MD,* Eric Paulson, PhD,* Brian J. Davis, MD, PhD,y
Daniel E. Spratt, MD,z Todd M. Morgan, MD,x David Dearnaley, FRCR, MD,k
Alison C. Tree, FRCR, MD,k Jason A. Efstathiou, MD, DPhil, FACRO, FASTRO,{
Mukesh Harisinghani, MD,# Ashesh B. Jani, MD, MSEE, FASTRO,**
Mark K. Buyyounouski, MD, MS,yy Thomas M. Pisansky, MD,y
Phuoc T. Tran, MD, PhD,zz R. Jeffrey Karnes, MD,xx
Ronald C. Chen, MD, MPH, FASCO, FASTRO,kk Fabio L. Cury, MD,{{
Jeff M. Michalski, MD, MBA, FASTRO,##
Seth A. Rosenthal, MD, FACR, FASTRO,*** Bridget F. Koontz, MD,yyy
Anthony C. Wong, MD, PhD,zzz Paul L. Nguyen, MD,xxx
Thomas A. Hope, MD,kkk Felix Feng, MD,zzz
Corresponding author: William A. Hall, MD; E-mail: [email protected] submitted work and is a member of the ASTRO Board of Directors.
Disclosures: R.C. reports personal fees from Abbvie and Accuray, D.E.S. reports personal fees from Janssen, Blue Earth, and AstraZenica,
outside the submitted work. F.L.C. reports grants and nonfinancial sup- outside the submitted work. P.T.T. reports grants from Astellas Pharm
port from Boston Scientific, personal fees from Varian Medical Systems, and Bayer Healthcare; grants, personal fees, and other from RefleXion
and grants from Sanofi, outside the submitted work. B.J.D. reports per- Medical Inc; and personal fees from Noxopharm, outside the submitted
sonal fees from Boston Scientific, Inc, outside the submitted work. D.D. work; in addition, P.T.T. has a patent Compounds and Methods of Use in
reports personal fees from The Institute of Cancer Research, during the Ablative Radiotherapy (#9114158) licensed to Natsar Pharm. A.C.T.
conduct of the study; in addition, D.D. has a patent EP1933709B1 is- reports grants and personal fees from Elekta; grants from Accuray and
sued. J.A.E. reports personal fees from Blue Earth Diagnostics, Boston Varian; personal fees from Janssen, Genesis healthcare, and Ferring; and
Scientific, AstraZeneca, Taris Biomedical, Janssen, Bayer Healthcare, nonfinancial support from Astellas, outside the submitted work. All other
and Roivant Pharma, outside the submitted work. F.F. reports personal authors have nothing to disclose.
fees from Dendreon, EMD Serono, Janssen Oncology, Ferring, Sanofi, Data Sharing: Research data are stored in an institutional repository
Bayer, Blue Earth Diagnostics, Celgene, Medivation/Astellas, Clovis and will be shared upon request to the corresponding author.
Oncology, and Genentech and other from PFS Genomics and Nutcracker Supplementary material for this article can be found at https://doi.org/
Therapeutics, outside the submitted work; in addition, F.F. has a patent 10.1016/j.ijrobp.2020.08.034.
EP3047037 A4 issued. W.A.H. reports technical support from MIM AcknowledgmentsdMedical College of Wisconsin Libraries and MIM
Software Inc, during the conduct of the study, and other from Elekta AB, provided technical support (MIM Inc, Beachwood, OH). Katie Kruger
outside the submitted work. T.A.H. reports grants from Philips Health- provided meeting coordination and manuscript formatting. This work
care and Advanced Accelerator Applications and personal fees from represents independent research supported by the National Institute for
Curium and Ipsen, outside the submitted work. B.F.K. reports grants Health Research Biomedical Research Centre at The Royal Marsden NHS
from Janssen Scientific Affairs, grants and personal fees from Blue Earth Foundation Trust and the Institute of Cancer Research, London. The views
Diagnostics, grants from Merck, and personal fees from Demos Pub- expressed are those of the authors and not necessarily those of the NIHR or
lishing, outside the submitted work. P.L.N. reports personal fees from the Department for Health and Social Care. The project described was
COTA, Ferring, Astellas, Dendreon, Blue Earth Diagnostics, and Boston supported by the National Center for Advancing Translational Sciences,
Scientific; grants and personal fees from Astellas, Bayer, and Janssen; National Institutes of Health, Award Number KL2TR001438. The content
and personal fees and other from Augmenix, outside the submitted work. is solely the responsibility of the author(s) and does not necessarily
E.P. reports other from Elekta AB, outside the submitted work. H.M.S. represent the official views of the NIH.
reports personal fees from Janssen and other from Radiogel, outside the

Int J Radiation Oncol Biol Phys, Vol. 109, No. 1, pp. 174e185, 2021
0360-3016/$ - see front matter Ó 2020 Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.ijrobp.2020.08.034
Volume 109  Number 1  2021 NRG Oncology pelvic nodal contouring atlas 175

Howard M. Sandler, MD, FASCO, FASTRO,### and

Colleen A.F. Lawton, MD, FACR, FASTRO*
*Medical College of Wisconsin, Department of Radiation Oncology, Milwaukee, Wisconsin; yMayo Clinic,
Department of Radiation Oncology, Rochester, Minnesota; zDepartment of Radiation Oncology,
University of Michigan, Ann Arbor, Michigan; xDepartment of Urology, University of Michigan, Ann
Arbor, Michigan; kThe Royal Marsden NHS Foundation Trust and The Institute of Cancer Research,
London, UK; {Department of Radiation Oncology, Massachusetts General Hospital, Boston,
Massachusetts; #Department of Radiology, Massachusetts General Hospital, Boston, Massachusetts;
**Department of Radiation Oncology, Emory University, Atlanta, Georgia; yyDepartment of Radiation
Oncology, Stanford University, Stanford, California; zzDepartment of Radiation Oncology, Johns
Hopkins, Baltimore, Maryland; xx Department of Urology, Mayo Clinic, Rochester, Minnesota;
kk
Department of Radiation Oncology, University of Kansas, Kansas City, Kansas; {{Department of
Radiation Oncology, McGill University, Montreal, Canada; ##Department of Radiation Oncology,
Washington University, St. Louis, Missouri; ***Department of Radiation Oncology, Sutter Medical
Group, Roseville, California; yyyDepartment of Radiation Oncology, Duke Cancer Institute, Durham,
North Carolina; zzz Department of Radiation Oncology, University of California San Francisco, San
Francisco, California; xxxDepartment of Radiation Oncology, Dana Farber Harvard Cancer Center,
Boston, Massachusetts; kkkDepartment of Radiology and Biomedical Imaging, University of California
San Francisco, San Francisco, California; and ###Department of Radiation Oncology, Cedars-Sinai
Medical Center, Los Angeles, California

Received Jun 3, 2020. Accepted for publication Aug 7, 2020.

Purpose: In 2009, the Radiation Therapy Oncology Group (RTOG) genitourinary members published a consensus atlas for
contouring prostate pelvic nodal clinical target volumes (CTVs). Data have emerged further informing nodal recurrence pat-
terns. The objective of this study is to provide an updated prostate pelvic nodal consensus atlas.
Methods and Materials: A literature review was performed abstracting data on nodal recurrence patterns. Data were pre-
sented to a panel of international experts, including radiation oncologists, radiologists, and urologists. After data review, par-
ticipants contoured nodal CTVs on 3 cases: postoperative, intact node positive, and intact node negative. Radiation oncologist
contours were analyzed qualitatively using count maps, which provided a visual assessment of controversial regions, and
quantitatively analyzed using Sorensen-Dice similarity coefficients and Hausdorff distances compared with the 2009 RTOG
atlas. Diagnostic radiologists generated a reference table outlining considerations for determining clinical node positivity.
Results: Eighteen radiation oncologists’ contours (54 CTVs) were included. Two urologists’ volumes were examined in a
separate analysis. The mean CTV for the postoperative case was 302 cm3, intact node positive case was 409 cm3, and intact
node negative case was 342 cm3. Compared with the original RTOG consensus, the mean Sorensen-Dice similarity coeffi-
cient for the postoperative case was 0.63 (standard deviation [SD] 0.13), the intact node positive case was 0.68 (SD 0.13),
and the intact node negative case was 0.66 (SD 0.18). The mean Hausdorff distance (in cm) for the postoperative case
was 0.24 (SD 0.13), the intact node positive case was 0.23 (SD 0.09), and intact node negative case was 0.33 (SD 0.24). Four
regions of CTV controversy were identified, and consensus for each of these areas was reached.
Conclusions: Discordance with the 2009 RTOG consensus atlas was seen in a group of experienced NRG Oncology and in-
ternational genitourinary radiation oncologists. To address areas of variability and account for new data, an updated NRG
Oncology consensus contour atlas was developed. Ó 2020 Elsevier Inc. All rights reserved.

Introduction contouring and identification of pelvic nodal regions

considered to be “at risk.” Treated volumes also have been
historically correlated with clinical outcomes for prostate
The treatment of pelvic lymph nodes with external beam
radiation therapy (RT) is a frequent component of the patients.5 The Radiation Therapy Oncology Group (RTOG)
developed a consensus-based contouring atlas in 2009 that
management of patients with prostate cancer.1 Pelvic lymph
has served as a foundation for nodal contouring on several
node irradiation is a common practice for men receiving
prospective clinical trials.6 This guideline has also been
prostate RT with high-risk disease, clinically lymph node-
used in standard clinical practice. A consensus atlas en-
positive disease, and in the postprostatectomy setting.2-4
courages a consistent application of nodal treatments across
There exists a wide range of approaches to pelvic nodal
176 Hall et al. International Journal of Radiation Oncology  Biology  Physics

providers and institutions to allow additional understanding group. Figures were reviewed with the group, including
of the effects of this component of treatment. locations of failure patterns. Surgeons and radiologists
Since publication of the original RTOG atlas, additional participated in these calls and were available for com-
data on patterns of tumor recurrence have emerged through mentary and questions. After the video conference pre-
both retrospective and prospective imaging studies. Multi- sentations, slides (with notes from the video conferencing)
ple publications have presented data to support a change in were circulated to all participants for additional individual
recommendations for pelvic nodal contouring from the review.
original RTOG consensus atlas.7-11 Given these data, the After this data presentation, radiation oncologists were
NRG Oncology genitourinary (GU) core committee asked to contour the nodal CTV. A total of 3 cases formed
thought it was appropriate to update the consensus atlas for the primary contouring subjects. These cases were selected
pelvic nodal contouring and to expand the existing atlas to by the first and senior authors (W.A.H. and CA.F.L.). Case
address the postoperative and clinically node-positive set- 1 was a 58-year-old man with a history of unfavorable
tings. The objective of this study was to both expand and intermediate-risk adenocarcinoma of the prostate, clinical
refine the existing consensus nodal atlas to account for stage T1cN0M0, grade group 3, Gleason score 4 þ 3, and
contemporary research findings. initial serum prostate-specific antigen (PSA) of 5.92 ng/
mL, who underwent surgical resection. Final pathology
showed grade group 3, Gleason score 4 þ 3 adenocarci-
Methods and Materials noma, positive margins, extensive seminal vesicle
involvement, and 1 of 8 nodes positive in a right obturator
The first and senior authors (W.A.H. and CA.F.L.) along node (pT3bN1M0). Case 2 was a 66-year-old man with
with the NRG Oncology GU core committee recruited an high-risk adenocarcinoma of the prostate who underwent a
international panel of physicians including radiation on- biopsy due to a PSA rising to 13.7 ng/mL. Biopsy showed
cologists, diagnostic radiologists (with expertise in nuclear grade group 4, Gleason score 4 þ 4, with clinical stage of
medicine and magnetic resonance imaging [MRI]), and T2bN1M0. He was clinically node positive, with 2 enlarged
urologists. The study was approved by the Institutional regional nodes on his diagnostic pelvic computed tomog-
Review Board at the Medical College of Wisconsin before raphy (CT). Case 3 was a 65-year-old man with high-risk
initiating research activities. All participants in the con- adenocarcinoma of the prostate, clinical stage T2aN0M0,
touring effort were informed via e-mail correspondence and grade group 5, Gleason score 4 þ 5, and PSA 38.2 ng/mL.
verbal review at the start of the video conferencing of their Urologists (T.M.M. and R.J.K.) were also asked to
rights as participants in this nodal contouring effort. Care contour “dissection” regions using their anticipated
was taken to anonymize individual observer contour con- dissection templates using case 3. These surgical contours
tributions within the group. were not included in the primary nodal contouring analysis.
The first step in the update was a review of the literature Contours were completed using MIM cloud (MIM Soft-
on pelvic nodal recurrence patterns published since 2007. ware Inc, Cleveland, OH). Contouring physician observers
This literature search was performed in collaboration with were blinded to other participants’ contour results during
the Medical College of Wisconsin Libraries. Primary this process of contouring. Only the first, second, and senior
search sources included: (1) PubMed (((pelvic AND author (W.A.H., EP, and CA.F.L.) had access to all contour
(lymph node drainage OR lymphatic drainage))) AND results collectively. Observers were required to contour a
prostate cancer) and (2) Google Scholar (terms: prostate nodal CTV and, if so inclined, to contour a nodal gross
cancer nodal drainage, prostate cancer nodal radiation, tumor volume.
prostate cancer nodal failure patterns, post-operative pros- Contour analysis was performed using the Sorensen-
tate cancer nodal failure, prostate-specific membrane anti- Dice similarity coefficient and Hausdorff distance.12,13
gen [PSMA] nodal failure, fluciclovine F-18 nodal failure These metrics were calculated and compared with a base-
patterns, and C-11 Choline PET prostate lymph nodes). line contour that was created by the first (W.A.H.) and
Along with the primary search terms, several additional senior (CA.F.L.) authors following the 2009 RTOG nodal
“similar publication” links from the references were used. contouring atlas.6 The contour volumes were statistically
Finally, all participants were asked to send relevant litera- compared using a Mann-Whitney test. The CTV contours
ture and references to the first author (W.A.H.) for review, of all individual observers were used to create a count map
organization, and presentation. Publications selected by the having the same resolution as the underlying image mo-
group were considered representative of the most recent dality. Within such a count map, each voxel value is
and relevant data in 4 different categories: (1) existing determined by the superposition of observers who included
updated nodal consensus atlases, (2) modern surgical/intact the corresponding image voxel within their CTV. For 18
disease lymphatic drainage patterns, (3) postoperative observers, the maximum count is 18. If all image voxels
recurrence patterns, and (4) novel molecular positron were included in a contour, they would present as a solid
emission tomography (PET)-based recurrence patterns. single color. If some of the voxels were not included in a
Publications were presented in detail via video conference contour set, they would present as a different color, based
for discussion and commentary from all members in the on the number of observers who included those voxels.
Volume 109  Number 1  2021 NRG Oncology pelvic nodal contouring atlas 177

Within a count map, different iso-surfaces with different Results

colors were created. A total of 18 colors would be available
with 18 observers. This enabled very careful “qualitative” Eighteen radiation oncologists finished 3 full contour sets for
observation of specific regions that were controversial and a total of 54 volumes, all of which were included in the final
presented a method to highlight specific areas of contro- contour analysis. The urologists’ contours were not included
versy for focused discussion and arbitration. The spread in in the final consensus contour analysis but instead were used
volume over these percentile surfaces provided an indica- for observation and consideration only. Observers practiced
tion of the CTV similarities within the observers and in the United States, Canada, and the United Kingdom with a
highlighted controversial regions. This method also pro- median of more than 15 years of practice.
vided a means by which to visually highlight particular The mean CTV for the postoperative case was 302 cm3,
areas of disagreement that were present in contoured vol- the intact node positive case was 409 cm3, and the intact node
umes among the observers. Diagnostic radiologists (T.A.H. negative case was 342 cm3. Compared with the original
and M.H.) presented a summary of criteria for node posi- RTOG consensus atlas contour (created by authors W.A.H.
tivity in the pelvis using a variety of imaging modalities and C.A.F.L.), the mean Sorensen-Dice similarity coefficient
(Fig. 1). for the postoperative case was 0.63 (SD 0.13), the intact node
Results of the consensus contouring exercise were sub- positive case was 0.68 (SD 0.13), and the intact node negative
sequently reviewed at the January 2020 NRG Oncology case was 0.66 (SD 0.18). The mean Hausdorff distance (in
meeting in person for those attending and were simulta- centimeters) for the postoperative case was 0.24 (SD 0.13),
neously presented via video conference for those unavai- the intact node positive case was 0.23 (SD 0.09), and the
lable to attend. Finally, areas of controversy identified in intact node negative case was 0.33 (SD 0.24). These values
the contour analytics were adjudicated via an anonymous represented the “quantitative” contour results.
online survey. The new step-by-step contour recommen- Several “qualitative” variations were identified when
dations were reviewed and circulated to the group. Com- using the count maps. Taken collectively, these variations
mon dose and fractionation schedules and corresponding provided a visual representation of consensus (“warmer”
constraints were included for group review and comment. colors, e.g., yellow, green) and controversial (“cooler”
Community radiation oncology feedback on these updates colors, e.g., magenta) areas. The 4 areas of greatest vari-
was solicited from the Michigan Radiation Oncology ability consisted of (1) the superior-most aspect of the
Quality Consortium via video conference and e-mail. common iliac nodes, (2) the transition from the external

CT/MRI-based PSMA PET-based Fluciclovine PET-based Example of positive Example of positive Example of positive
Anatomic Location CT/MRI-based Size
Morphology Criteria Criteria node on CT node on MR node on PET

> 1 cm: Uptake

Irregular Border and/or greater than BM
Short axis heterogenous Uptake greater than
Mesorectal, Presacral < 1 cm: Uptake
> 4 mm morphology (only for LN blood pool
> 3mm on MRI) greater than blood
pool

> 1 cm: Uptake

Irregular Border and/or greater than BM
Internal Iliac, Short axis Uptake greater than
heterogenous < 1 cm: Uptake
Obturator > 7mm blood pool
morphology greater than blood
pool

> 1 cm: Uptake

Irregular Border and/or greater than BM
Common IIiac and Short axis Uptake greater than
heterogenous < 1 cm: Uptake
External IIiac > 8 mm blood pool
morphology greater than blood
pool

Irregular Border and/or Asymmetric uptake

Short axis Asymmetric uptake
Inguinal heterogenous that is greater than
> 8 mm greater than BM
morphology liver

Fig. 1. Summary criteria for clinical node positivity. Abbreviations: CT Z computed tomography; LN Z lymph node;
MRI Z magnetic resonance imaging; PET Z positron emission tomography; PSMA Z prostate-specific membrane antigen.
178 Hall et al. International Journal of Radiation Oncology  Biology  Physics

iliac to the inguinal nodes, (3) the inclusion of the peri- vascular structures into the inguinal canal (Fig. 3i), often
prostatic nodes, and (4) the inclusion of perirectal nodes best seen on the coronal images (Fig. 3j).
(Fig. 2a-d). Contours of clinically positive nodes were also 7. The external iliac contours should typically end when the
controversial. These areas were discussed in detail via an vessels are completely lateral to the most medial aspect of
in-person meeting and a video conference and were the the acetabulum (near the mid-femoral head and fovea). At
subject of specific questions in the anonymous survey. The that point, the contours should be tapered off (Fig. 3k-l).
results of the survey formed the consensus steps (1-10).
Consensus on final borders for each of these areas was 8. The obturator nodes can be between 1 and 2 cm in width
reached via a written survey specifically addressing po- and should extend to the posterior edge of the obturator
tential changes to these areas. The refined steps to contour internus muscle (Fig. 3k).
the nodal CTV can be seen in Figures 3 and 4. 9. Begin to taper the obturator nodes at the top of the
Prophylactic nodal contouring steps for clinically node- seminal vesicles (or the top of the postoperative bed),
negative patients including both intact and postoperative extending approximately 1 cm anterior to the anterior
cases (Fig. 3a-m and Fig. 4a-g): edge of the obturator internus muscle (Fig. 3k-l; MRI
1. Commence contours at the bifurcation of the aorta into registration can be useful in this area).
the common iliac arteries or the proximal inferior vena
cava to the common iliac veins, whichever occurs more 10. The obturator nodes should end where the seminal ves-
superiorly (typically at the level of L4-L5; Fig. 3a-b). icles join the prostate, or approximately the midportion
of the contoured postoperative CTV bed (Fig. 3m).
2. Contour approximately 5 to 7 mm around each iliac
vessel, including the entire circumference of both the Modifications when treating clinically node positive
iliac artery and vein. Bone, bowel, bladder, and muscle cases:
should be excluded from the nodal CTV contour. Where 1. Steps 1 to 10 should be followed for prophylactic
clinically indicated, CTV margins can be more generous, regions.
particularly anterior to vessels (10 mm). Ensure coverage
posteriorly in the area formed between the psoas major 2. Figure 1 should be referenced to help identify suspicious
and the vertebral body (Fig. 3c-d). nodes; all suspicious nodes should be considered for review
with diagnostic radiology and contoured as appropriate.
3. The width of the interspace between the external and
internal iliac contours should be approximately 1.5 to 3
3. Prophylactic nodal volumes should extend approxi-
cm. This will vary depending on patient anatomy
mately 5 to 7 mm around clinically suspicious nodes;
(Fig. 3e).
this may alter the prophylactic nodal volumes in steps 1
4. Include the prevertebral, presacral, and posterior meso- to 10.
rectal nodes to the bottom of S3 (Fig. 3f).
4. Residual (shrunken) gross nodes, posteandrogen depri-
5. The posterior border of the CTV coming off the internal vation therapy (ADT), should form the primary boost
iliac vessels should extend to the anterior edge of the volume (additional information in dosing section below).
piriformis muscle after the course of the pudendal artery
and inferior gluteal artery (Fig. 3g-h). Radiation dosing to pelvic nodes:
6. The transition from the external iliac to the inguinal nodes  Prophylactic nodes: A dose range of 45 to 50.4 Gy is
occurs when the external iliac vessels cross beneath the acceptable when using conventional fractionation. The
inguinal ligament into the inguinal canal. Examine for this majority of participants do not change their prophylactic
transition and begin tapering off external iliac nodes at that nodal dose whether treating an intact prostate case or
point. This should correspond to the entrance of the postoperative.

Fig. 2. Count maps showing controversial regions identified.

Volume 109  Number 1  2021 NRG Oncology pelvic nodal contouring atlas 179

Fig. 3. (A-M) New consensus contours on computed tomography. Abbreviation: CTV Z clinical target volumes.

 Gross nodes: Should be treated as high as clinically

feasible (up to the dose being delivered to the primary  All available/relevant scans (eg, PET and MRI) should be
tumor) while respecting normal organ tolerances. Nodal carefully considered by the radiation oncologist when
volumes should be examined pre- and post-ADT, and the delineating nodal coverage.
post-ADT tumor volume should serve as the high dose
boost volume.  In general, the CTV should exclude bone, bladder,
muscle, and bowel.
Overarching points for consideration when contouring  Simulation images that are suggestive of clinically suspi-
pelvic nodes with the new guidelines: cious nodes (criteria in Fig. 1) should be reviewed with a
180 Hall et al. International Journal of Radiation Oncology  Biology  Physics

Fig. 3. (continued).

diagnostic radiologist and may be included in boost vol- exhibit extranodal tumor extension may have more
umes at the clinical discretion of the radiation oncologist. generous CTVs. Surgical clips should be identified and
 In some circumstances, small portions of bowel may abut potentially included at the discretion of the radiation
vascular structures or large portions of small bowel may oncologist. Close collaboration with colleagues having
be in the pelvis. As mentioned earlier (in step 2), the expertise in urology and diagnostic radiology is rec-
CTV should exclude bowel (including both small and ommended. Altered lymph node spread is common,14
large bowel). Rarely, bowel may be included in the CTV and larger volume expansions, including postoperative
at the discretion of the radiation oncologist secondary to changes of uncertain significance, may also be neces-
extenuating clinical circumstances (eg, adjacent involved sary. PET scans or other advanced imaging acquired
node or tumor extension). Normal tissue constraints should be registered and included in the treatment
should be prioritized by the radiation oncologist when planning process.
treating pelvic nodes. Clinical review and discretion on  Consideration should be given to the comorbidities and
the part of the radiation oncologist is needed in each of medical history of each individual patient.
these circumstances. The results of areas that urologic surgeons identified as
 For postoperative cases: Pathology and operative re- part of their dissection template are presented in Figure E1.
ports should be carefully considered in treatment vol- Finally, given the wide range of contour volumes, an
umes. Regions with pathologically involved nodes that example of a larger contour set, including perirectal nodes,
Volume 109  Number 1  2021 NRG Oncology pelvic nodal contouring atlas 181

Fig. 4. (A-G): New consensus contours on magnetic resonance.

182 Hall et al. International Journal of Radiation Oncology  Biology  Physics

Table 1 Constraints for consideration when treating pelvic Table 1 (continued )

nodes
60 Gy in 20 fractions
75.6-79.2 Gy in 42-44 fractions, (treating nodes to 44-47 Gy
treating nodes to 45-50.4 Gy over 20 fractions*) (8)
with a sequential boost
Femur_R V (3500 cGy) 5%
Rectum (24) V (4500 cGy) 50% Dmax 3700 cGy
V (7000 cGy) 15% Colon Dmax 5000 cGy
V (>7200 cGy) <10 cm3 Small bowel (bowel loops) Dmax 4000 cGy
Bladder V (4500 cGy) 50% V (3700 cGy) 90 cm3
V (7000 cGy) 15% V (3300 cGy) 130 cm3
Femur_L V (5000 cGy) 2% Pubic bone V (5700 cGy) 20%
Dmax 5250 cGy Penile bulb (25) V (2200 cGy) 50%
Femur_R V (5000 cGy) 2%
Dmax 5250 cGy Abbreviation: PTV Z planning target volume.
* Safety and efficacy of hypofractionation to pelvic nodes is
Colon V (6000 cGy) 2%
currently the subject of ongoing investigation and has not been
Dmax 6250 cGy established.
Small bowel (bowel loops) V (5000 cGy) 10% y
Group consensus constraint.
Dmax 5200 cGy z
Patient reported quality of life data for the bladder constraints is
Pubic bone V (7000 cGy) 25% the subject of ongoing investigation.
Penile bulb (should not V (5000 cGy) 50%
sacrifice PTV coverage)
70 Gy in 28 fractions, treating nodes can be seen in Figure E2. Such expanded volumes may be
to 45-50.4 Gy with a simultaneous rarely considered for highly select and advanced T4 lesions
integrated boost at the discretion of the radiation oncologist.15 Considerable
Rectum (24) V (4500 cGy) 45% discretion is needed when including mesorectal nodes in
V (5500 cGy) 25% the treatment volume, and normal tissue constraints should
V ( 6500 cGy) 15% be prioritized.
V (>6500 cGy) <10 cm3 Figure 1 was created by the diagnostic radiologists
Bladder V (4500 cGy) 45% (T.A.H., M.H.) and nuclear medicine expert (T.A.H.)
V (5500 cGy) 25% to include criteria for clinical node positive prostate
V (6500 cGy) 15% lesions.16-21 These criteria are helpful for radiation oncolo-
Femur_L V (5000 cGy) 1% gists to be aware of and most importantly discuss with their
Dmax 5250 cGy
diagnostic radiology and nuclear medicine colleagues. In
Femur_R V (5000 cGy) 1%
Dmax 5250 cGy
addition, commonly used dose constraints were collated for
Colon Dmax 5500 cGy different dose and fractionation schedules and are displayed
Small bowel (bowel loops) V (4650 cGy) 2 cm3 in Table 1.22-25 These may be helpful for radiation oncolo-
Dmax 5200 cGy gists to consider when treating pelvic nodes.
Pubic bone V (6000 cGy) 30%
Penile bulb (should not Make dose as low as
sacrifice PTV coverage) reasonably achievable
60 Gy in 20 fractions Discussion
(treating nodes to 44-47 Gy
over 20 fractions*) (8) Prophylactic treatment of pelvic lymph nodes in the man-
Rectum (22) V (2000 cGy) 85% agement of prostate cancer remains an active area of clin-
(no circumferential dose) ical inquiry and investigation that presently lacks
V (3000 cGy) 57% consensus. Data are emerging suggesting some efficacy to
V (4000 cGy) 38% pelvic nodal treatment.1 In the context of this ongoing in-
V (5000 cGy) 22% quiry, expert consensusebased guidelines consider its use
V (6000 cGy) 1%y an acceptable management option.3,4 Constant evaluation
Bladderz V (4000 cGy) 50% and evidence-based updating of available consensus
V (4800 cGy) 25% guidelines are imperative. Careful examination of the
V (5680 cGy) 5% evolution of guidelines over time is essential to ensure
V (6000 cGy) 3 % evidence-based improvement. The overarching goal of our
Femur_L V (3500 cGy) 5%
process was to perform a timely evaluation and update the
Dmax 3700 cGy
2009 RTOG consensus guidelines. We did not seek to
(continued) reinvent the atlas, but rather sought to update and refine it.
Volume 109  Number 1  2021 NRG Oncology pelvic nodal contouring atlas 183

Our study shows the 2009 RTOG pelvic lymph node considered carefully by the panel; it should be noted that
consensus guidelines no longer accurately reflect the inclusion of these more generous nodal volumes should be
practice patterns of prostate cancer experts from around the highly selected.
world or the state-of-the-art assessment of lymph node re- The third general category of data considered included
gions at risk for prostate cancer metastasis. Furthermore, novel MRI techniques and newly published clinical patterns
we developed a guideline process to develop treatment of recurrence data. Several series directly compared the
volume contouring standards that could be used as a tem- anatomic distribution of nodal metastases with the pub-
plate for other disease sites, and for research or clinical lished RTOG contouring guideline. Meijer et al examined
collaboratives. magnetic resonance lymphography in a modern cohort of
These guidelines were updated using an evidence-based intact intermediate- and high-risk patients and noted that
process. Several categories of updated data were considered more than 50% of patients had positive nodes outside of the
in detail by the group of observers who participated in this RTOG nodal atlasecontoured volumes. Common sites
contouring effort. These publications fell into 4 broad were in the high common iliac, perirectal, and paraortic
categories: (1) existing updates to contouring guidelines, regions.9 It was also noted that a high percentage of patients
(2) surgical mapping and lymphatic drainage series, (3) in the postoperative setting had aberrant nodal spread, with
clinical recurrence series, and (4) PET/postoperative a particularly large percentage of patients exhibiting nodal
recurrence series. International groups have proposed a few spread in the perirectal area.14 Data on patterns of recur-
modifications to the existing RTOG nodal contouring atlas rence have also been published directly comparing failure
that were considered in detail by the authors. The first was patterns to the existing RTOG atlas. Spratt et al conducted a
an updated atlas produced by the PIVOTAL trialists group,8 retrospective series of pelvic nodal failures and mapped
of which one author (D.D.) also participated as an inter- those in relation to the existing RTOG nodal atlas.10 This
national representative in this NRG Oncology contouring series concluded that an increase in the superior border of
activity. The PIVOTAL atlas recommended modifications the pelvic nodal treatment volume to cover the common
to the existing RTOG contouring recommendations but did iliac stations to L4/L5 would cover more than 90% of first
not include node-positive, PET, MRI, or postoperative nodal recurrences.10 Such findings regarding the common
nodal contouring recommendations. The second recently iliac nodal stations have been supported by other publica-
updated consensus atlas that specifically focused on pros- tions, demonstrating that a number of recurrences were
tate nodal treatment was from the Groupe d’Etude des located outside of the standard RTOG atlas treatment
Tumeurs Uro-Génitales.7 This atlas incorporated some volumes.32,33
novel PET recurrence pattern data available at that time. The final category of contemporary data considered was
The Groupe d’Etude des Tumeurs Uro-Génitales atlas does novel prostate-specific PET data. More specifically, how
not include specific contouring recommendations for node- prostate PET scans might influence nodal volumes in both
positive or postoperative patients. The NRG Oncology the intact treatment naı̈ve setting and the postoperative,
group provides the current updated consensus atlas with 3 biochemically recurrent setting. Series including PSMA,
overarching goals: (1) refining the current RTOG intact Fluciclovine F18, and C-11 choline PET were considered
prophylactic atlas recommendations, (2) addressing clini- and reviewed. Several of the published PSMA PET series
cally node-positive disease, and (3) addressing contouring mapped areas of nodal recurrence that were outside of the
in the postoperative setting. existing RTOG template. These recurrence locations were
The second broad category of data considered was presented and reviewed by the observers for consideration
newly available surgical data. Much of this focused on as to how this might influence the existing nodal treatment
novel sentinel node data and other surgical nodal mapping volumes.11,34-38 Several of these series visually mapped
techniques. Current surgical methods of addressing pelvic PET recurrence locations in relation to the existing RTOG
nodes were considered. Most contemporary surgical consensus atlas.39
guidelines recommend an extended pelvic lymph node After the literature review, a comprehensive contouring
dissection when a nodal dissection is performed.4,26,27 exercise took place. There were both quantitative and
Surgical dissection and nodal mapping data provided qualitative assessments of these contour results. The
valuable insight into common sites of nodal drainage. quantitative results of the contouring exercise yielded
These data partially informed the updated nodal atlas rec- Sorensen-Dice coefficients reflecting poor agreement.40
ommendations. It is notable that the internal iliac, external These findings were consistent within the postoperative
iliac, and obturator nodes comprise the vast majority of contours, intact node positive, and intact node negative
nodal drainage sites of the prostate. However, the common contour sets. Qualitatively, a total of 4 areas were visually
iliac, presacral, and paraortic/caval nodes can also represent identified as controversial using the count map strategy.
10% or more of nodal drainage sites mapped.26,28-31 Other The count map strategy was believed to be very helpful to
drainage sites, such as perirectal nodes, have also repre- recognize areas needing focused discussion as compared
sented more than 10% of nodal drainage sites in some with just the numerical metrics. Considered collectively,
sentinel node mapping series, but this is highly variable and these metrics were supportive of the need for an updated
inconsistent.31 Appropriate applications of the data were consensus contouring atlas. Several areas of this updated
184 Hall et al. International Journal of Radiation Oncology  Biology  Physics

atlas differ from the existing 2009 RTOG atlas, including published PET- and MRI-based nodal recurrence data,
the superior, vascular margins, and inferior boundary which support a prudent expansion of target volumes. In
recommendations. addition, we have presented higher resolution CT and MRI
A few important points must be considered when sets, with annotations that may assist in educating and
examining the new contouring steps presented. These are obtaining uniformity of practice. Full DICOM image files,
intended to provide approximate guidelines, not to rigidly with contoured structure sets, are available as supplements
constrain the radiation oncologist from exercising clinical to provide greater detail for practitioners.
judgment in an individual case. Radiation oncologists
should carefully examine and incorporate all oncologic and
diagnostic scan information into their treatment plans. Conclusions
Some clinical circumstances may warrant more generous
treatment volumes or more constrained treatment volumes. A new NRG Oncology consensus nodal contouring atlas is
Factors specific to the comorbidities and individual presented, with several changes to the existing RTOG
patient’s medical history should also be considered. We consensus atlas. Extensive imaging data and studies pro-
have presented variations for consideration, along with vided a basis for the CTVs that radiation oncologists should
step-by-step guidelines to ensure an overarching consensus consider when targeting pelvic nodal tissues. The included
recommendation. guidelines are intended to provide greater detail and ac-
As novel PET-based imaging continues to develop, count for recently published nodal failure pattern data.
this additional information may help individualize RT Moreover, variations in contouring strategies are presented,
planning. Many published series highlight apparently along with dosimetric constraints for consideration when
atypical anatomic sites of nodal recurrence, such as in treating the pelvic lymph nodes.
perirectal or periaortic nodes.39 Perirectal nodes in
particular were a source of significant discussion, espe-
cially for T4 tumors.15 Routinely including areas such as References
the perirectal region was thought by the majority of the
group to create an unnecessarily large treatment volume. 1. Pollack A, Karrison T, Balogh JA, Low D, Bruner D. Short term
However, a variation in contours is also presented for androgen deprivation therapy without or with pelvic lymph node
consideration (Fig. E2) when considered clinically indi- treatment added to prostate bed only salvage radiotherapy: The NRG
cated by the radiation oncologist. Other studies have Oncology/RTOG 0534 SPPORT trial. Int J Radiat Oncol Biol Phys
2018;102:1605.
recently addressed considerably more generous treatment 2. Lieng H, Kneebone A, Hayden AJ, et al. Radiotherapy for node-
volumes and the tolerance of such an approach.41 As positive prostate cancer: 2019 recommendations of the Australian
mentioned, advanced molecular imaging studies should and New Zealand Radiation Oncology Genito-Urinary group. Radio-
be reviewed by radiation oncologists, in collaboration ther Oncol 2019;140:68-75.
with nuclear medicine, whenever available. 3. Heidenreich A, Bastian PJ, Bellmunt J, et al. EAU guidelines on
prostate cancer. Part 1: screening, diagnosis, and local treatment with
There are limitations to this activity that merit consid- curative intent-update 2013. Eur Urol 2014;65:124-137.
eration. We do not address the controversial topic of “in- 4. Mohler JL, Antonarakis ES, Armstrong AJ, et al. Prostate cancer,
dications” for pelvic nodal RT. That is currently the subject version 2.2019, NCCN clinical practice guidelines in oncology. J Natl
of multiple trials (NCT01368588, NCT01952223, Compr Canc Netw 2019;17:479-505.
ISRCTN80146950) and is considered beyond the scope of 5. Roach M 3rd, DeSilvio M, Valicenti R, et al. Whole-pelvis, “mini-
pelvis,” or prostate-only external beam radiotherapy after neoadjuvant
the present study. This study does not aggregate or meta- and concurrent hormonal therapy in patients treated in the Radiation
analyze formally all reported PET-based patterns of failure; Therapy Oncology Group 9413 trial. Int J Radiat Oncol Biol Phys
this was also considered beyond the scope of the present 2006;66:647-653.
study. We also did not address the ideal planning target 6. Lawton CA, Michalski J, El-Naqa I, et al. RTOG GU Radiation
volume definition. This will depend on target proximity to oncology specialists reach consensus on pelvic lymph node volumes
for high-risk prostate cancer. Int J Radiat Oncol Biol Phys 2009;74:
organs at risk and image guidance methods. This is a 383-387.
consensus atlas that went through extensive revision, 7. Sargos P, Guerif S, Latorzeff I, et al. Definition of lymph node areas
refinement, and peer review, but prospective validation of for radiotherapy of prostate cancer: A critical literature review by the
the atlas was not formally conducted. Dosimetric con- French Genito-Urinary Group and the French Association of Urology
straints are presented for consideration; however, optimal (GETUG-AFU). Cancer Treat Rev 2015;41:814-820.
8. Harris VA, Staffurth J, Naismith O, et al. Consensus guidelines and
dose constraints were not the primary focus of the analysis, contouring atlas for pelvic node delineation in prostate and pelvic
and these should be interpreted accordingly. Finally, we node intensity modulated radiation therapy. Int J Radiat Oncol Biol
acknowledge that any guideline is a work in progress and Phys 2015;92:874-883.
that refinement and enhancement is expected as the science 9. Meijer HJ, Fortuin AS, van Lin EN, et al. Geographical distribution of
that forms its basis advances. lymph node metastases on MR lymphography in prostate cancer pa-
tients. Radiother Oncol 2013;106:59-63.
The objective and results of this study serve as a 10. Spratt DE, Vargas HA, Zumsteg ZS, et al. Patterns of lymph node
refinement and evidence-based update to the existing failure after dose-escalated radiotherapy: Implications for extended
RTOG atlas. Our aspiration was to account for recently pelvic lymph node coverage. Eur Urol 2017;71:37-43.
Volume 109  Number 1  2021 NRG Oncology pelvic nodal contouring atlas 185

11. Schiller K, Devecka M, Maurer T, et al. Impact of 68Ga-PSMA-PET 28. de Korne CM, Wit EM, de Jong J, et al. Anatomical localization of
imaging on target volume definition and guidelines in radiation radiocolloid tracer deposition affects outcome of sentinel node pro-
oncology: Patterns of failure analysis in patients with primary diag- cedures in prostate cancer. Eur J Nucl Med Mol Imaging 2019;46:
nosis of prostate cancer. Radiat Oncol 2018;13:36. 2558-2568.
12. Taha AA, Hanbury A. An efficient algorithm for calculating the exact 29. Harke NN, Godes M, Wagner C, et al. Fluorescence-supported
Hausdorff distance. IEEE Trans Pattern Anal Mach Intell 2015;37: lymphography and extended pelvic lymph node dissection in
2153-2163. robot-assisted radical prostatectomy: A prospective, randomized trial.
13. Sorensen-Dice. MathWorks. Available at: https://www.mathworks. World J Urol 2018;36:1817-1823.
com/help/images/ref/dice.html. Accessed January, 3, 2020. 30. Nguyen DP, Huber PM, Metzger TA, Genitsch V, Schudel HH,
14. Meijer HJ, van Lin EN, Debats OA, et al. High occurrence of aberrant Thalmann GN. A specific mapping study using fluorescence sentinel
lymph node spread on magnetic resonance lymphography in prostate lymph node detection in patients with intermediate- and high-risk
cancer patients with a biochemical recurrence after radical prosta- prostate cancer undergoing extended pelvic lymph node dissection.
tectomy. Int J Radiat Oncol Biol Phys 2012;82:1405-1410. Eur Urol 2016;70:734-737.
15. Abu-Gheida I, Bathala TK, Maldonado JA, et al. Increased frequency 31. Winter A, Kowald T, Paulo TS, et al. Magnetic resonance sentinel
of mesorectal and perirectal LN involvement in T4 prostate cancers. lymph node imaging and magnetometer-guided intraoperative detec-
Int J Radiat Oncol Biol Phys 2020. tion in prostate cancer using superparamagnetic iron oxide nano-
16. Ramirez M, Ingrand P, Richer JP, et al. What is the pelvic lymph node particles. Int J Nanomedicine 2018;13:6689-6698.
normal size? Determination from normal MRI examinations. Surg 32. De Bruycker A, De Bleser E, Decaestecker K, et al. Nodal oligor-
Radiol Anat 2016;38:425-431. ecurrent prostate cancer: anatomic pattern of possible treatment failure
17. Maurer T, Eiber M, Schwaiger M, Gschwend JE. Current use of in relation to elective surgical and radiotherapy treatment templates.
PSMA-PET in prostate cancer management. Nat Rev Urol 2016;13: Eur Urol 2019;75:826-833.
226-235. 33. Parker WP, Evans JD, Stish BJ, et al. Patterns of recurrence after
18. Hofman MS, Lawrentschuk N, Francis RJ, et al. Prostate-specific postprostatectomy fossa radiation therapy identified by C-11 choline
membrane antigen PET-CT in patients with high-risk prostate cancer positron emission tomography/computed tomography. Int J Radiat
before curative-intent surgery or radiotherapy (proPSMA): A pro- Oncol Biol Phys 2017;97:526-535.
spective, randomised, multi-centre study. Lancet 2020;395. 34. Schiller K, Sauter K, Dewes S, et al. Patterns of failure after radical
19. Songmen S, Nepal P, Olsavsky T, Sapire J. Axumin positron emission prostatectomy in prostate cancer: Implications for radiation therapy
tomography: Novel agent for prostate cancer biochemical recurrence. planning after 68Ga-PSMA-PET imaging. Eur J Nucl Med Mol Im-
J Clin Imaging Sci 2019;9:49. aging 2017;44:1656-1662.
20. Eiber M, Herrmann K, Calais J, et al. Prostate Cancer Molecular 35. Calais J, Kishan AU, Cao M, et al. Potential impact of 68Ga-PSMA-11
Imaging Standardized Evaluation (PROMISE): Proposed miTNM PET/CT on the planning of definitive radiation therapy for prostate
classification for the interpretation of PSMA-ligand PET/CT. J Nucl cancer. J Nucl Med 2018;59:1714-1721.
Med 2018;59:469-478. 36. Habl G, Sauter K, Schiller K, et al. 68Ga-PSMA-PET for radiation
21. Savir-Baruch B, Banks KP, McConathy JE, et al. ACR-ACNM prac- treatment planning in prostate cancer recurrences after surgery: Indi-
tice parameter for the performance of fluorine-18 fluciclovine-PET/CT vidualized medicine or new standard in salvage treatment. Prostate
for recurrent prostate cancer. Clin Nucl Med 2018;43:909-917. 2017;77:920-927.
22. Wilkins A, Naismith O, Brand D, et al. Derivation of dose/volume 37. Byrne K, Eade T, Kneebone A, et al. Delineating sites of failure
constraints for the anorectum from clinician- and patient-reported following post-prostatectomy radiation treatment using 68Ga-PSMA-
outcomes in the CHHiP trial of radiation therapy fractionation. Int J PET. Radiother Oncol 2018;126:244-248.
Radiat Oncol Biol Phys 2020;106:928-938. 38. Sinha S, Witztum A, Vasudevan H, et al. 68Ga-PSMA-11 PET-based
23. Ferreira MR, Thomas K, Truelove L, et al. Dosimetry and gastroin- prostate cancer lymph node atlas reveals patterns of potential
testinal toxicity relationships in a phase II trial of pelvic lymph node geographic miss in consensus pelvic nodal contours. Int J Radiat
radiotherapy in advanced localised prostate cancer. Clin Oncol (R Coll Oncol Biol Phys 2019;105:S136.
Radiol) 2019;31:374-384. 39. Schiller K, Stöhrer L, Düsberg M, et al. PSMA-PET/CT-based lymph
24. Michalski JM, Gay H, Jackson A, Tucker SL, Deasy JO. Radiation node atlas for prostate cancer patients recurring after primary treat-
dose-volume effects in radiation-induced rectal injury. Int J Radiat ment: Clinical implications for salvage radiation therapy [e-pub ahead
Oncol Biol Phys 2010;76(3 Suppl):S123-S129. of print]. Eur Urol Oncol. https://doi.org/10.1016/j.euo.2020.04.004.
25. Murray J, Gulliford S, Griffin C, et al. Evaluation of erectile potency Accessed May 23, 2020.
and radiation dose to the penile bulb using image guided radiotherapy 40. Delpon G, Escande A, Ruef T, et al. Comparison of automated atlas-
in the CHHiP trial. Clin Transl Radiat Oncol 2020;21:77-84. based segmentation software for postoperative prostate cancer radio-
26. Mattei A, Fuechsel FG, Bhatta Dhar N, et al. The template of the therapy. Front Oncol 2016;6:178.
primary lymphatic landing sites of the prostate should be revisited: 41. Jethwa KR, Hellekson CD, Evans JD, et al. 11C-Choline PET guided
Results of a multimodality mapping study. Eur Urol 2008;53:118-125. salvage radiation therapy for isolated pelvic and paraortic nodal
27. Mottet N, Bellmunt J, Bolla M, et al. EAU-ESTRO-SIOG guidelines recurrence of prostate cancer after radical prostatectomy: Rationale
on prostate cancer. Part 1: Screening, diagnosis, and local treatment and early genitourinary or gastrointestinal toxicities. Adv Radiat Oncol
with curative intent. Eur Urol 2017;71:618-629. 2019;4:659-667.