Volumetric tumor burden and its effect on brachial plexus dosimetry in head and neck intensity-modulated radiotherapy
To determine the effect of gross tumor volume of the primary (GTV-P) and nodal (GTV-N) disease on planned radiation dose to the brachial plexus (BP) in head and neck intensity-modulated radiotherapy (IMRT). Overall, 75 patients underwent definitive IMRT to a median total dose of 69.96Gy in 33 fractions. The right BP and left BP were prospectively contoured as separate organs at risk. The GTV was related to BP dose using the unpaired t-test. Receiver operating characteristics curves were constructed to determine optimized volumetric thresholds of GTV-P and GTV-N corresponding to a maximum BP dose cutoff of > 66Gy. Multivariate analyses were performed to account for factors associated with a higher maximal BP dose. A higher maximum BP dose (> 66 vs ≤ 66Gy) correlated with a greater mean GTV-P (79.5 vs 30.8cc; p = 0.001) and ipsilateral GTV-N (60.6 vs 19.8cc; p = 0.014). When dichotomized by the optimized nodal volume, patients with an ipsilateral GTV-N ≥ 4.9 vs < 4.9cc had a significant difference in maximum BP dose (64.2 vs 59.4Gy; p = 0.001). Multivariate analysis confirmed that an ipsilateral GTV-N ≥ 4.9cc was an independent predictor for the BP to receive a maximal dose of > 66Gy when adjusted individually for BP volume, GTV-P, the use of a low anterior neck field technique, total planned radiation dose, and tumor category. Although both the primary and the nodal tumor volumes affected the BP maximal dose, the ipsilateral nodal tumor volume (GTV-N ≥ 4.9cc) was an independent predictor for high maximal BP dose constraints in head and neck IMRT.
Volumetric tumor burden and its effect on brachial plexus dosimetry in
head and neck intensity-modulated radiotherapy
Paul B. Romesser, M.D.,
Muhammad M. Qureshi, M.B.B.S., Nataliya Kovalchuk, Ph.D., and
Minh Tam Truong, M.D.
Department of Radiation Oncology, Boston Medical Center, Boston University School of Medicine, Boston, MA
Received 28 March 2013
Accepted 5 December 2013
Head and neck cancer
Gross tumor volume
To determine the effect of gross tumor volume of the primary (GTV-P) and nodal (GTV-N) disease on
planned radiation dose to the brachial plexus (BP) in head and neck intensity-modulated radiotherapy
(IMRT). Overall, 75 patients underwent deﬁnitive IMRT to a median total dose of 69.96 Gy in 33 fractions.
The right BP and left BP were prospectively contoured as separate organs at risk. The GTV was related to
BP dose using the unpaired t-test. Receiver operating characteristics curves were constructed to
determine optimized volumetric thresholds of GTV-P and GTV-N corresponding to a maximum BP dose
cutoff of 466 Gy. Multivariate analyses were performed to account for factors associated with a higher
maximal BP dose. A higher maximum BP dose (466 vs r66 Gy) correlated with a greater mean GTV-P
(79.5 vs 30.8 cc; p¼0.001) and ipsilateral GTV-N (60.6 vs 19.8 cc; p¼0.014). When dichotomized by the
optimized nodal volume, patients with an ipsilateral GTV-N Z4.9 vs o4.9 cc had a signiﬁcant difference
in maximum BP dose (64.2 vs 59.4 Gy; p¼0.001). Multivariate analysis conﬁrmed that an ipsilateral
GTV-N Z4.9 cc was an independent predictor for the BP to receive a maximal dose of 466 Gy when
adjusted individually for BP volume, GTV-P, the use of a low anterior neck ﬁeld technique, total planned
radiation dose, and tumor category. Although both the primary and the nodal tumor volumes affected
the BP maximal dose, the ipsilateral nodal tumor volume (GTV-N Z4.9 cc) was an independent predictor
for high maximal BP dose constraints in head and neck IMRT.
&2014 Published by Elsevier Inc. on behalf of American Association of Medical Dosimetrists.
Brachial plexopathy after head and neck cancer (HNC) radio-
therapy is a rare but potentially devastating complication without an
A recent report has suggested that brachial plexop-
athy symptoms maybe underreported in the HNC population, with a
risk approaching 12% correlating with a dose-response relationship
for the development of brachial plexus (BP)–related neuropathies
among patients treated with radiotherapy for HNC.
Within the past
10 years, intensity-modulated radiotherapy (IMRT) has emerged as
the standard radiotherapy approach for treating HNC. The radiation
dose to the BP is signiﬁcantly increased among patients undergoing
IMRT compared with conventional radiotherapy for the treatment of
IMRT optimization without consideration of the BP can lead to
signiﬁcant dose inhomogeneity within the BP, which may lead to an
increased risk of long-term toxicity.
The Radiation Therapy Oncology Group recommends limiting
the BP maximum dose to 60 to 66 Gy, depending on the study
Original data from Emami et al.
suggest that the
tolerance dose before the conformal radiotherapy era was 60 Gy.
Radiologic atlases have been published to improve accuracy and
reduce interobserver variability in contouring the BP.
institution, the BP has been routinely contoured as an organ at risk
(OAR) since 2004 with the intent to limit the BP maximum dose
less than 60 Gy, while achieving tumor coverage with the pre-
scription dose. A recent report of 114 patients with HNC treated
with IMRT reported that a signiﬁcantly higher planned radiation
dose was delivered to the BP in patients with laryngeal, hypophar-
yngeal, and oropharyngeal cancer with locally advanced disease
(63.4 vs 58.4 Gy; p¼0.002) as compared with more distant sites
such as the nasopharynx.
Similarly, it was reported that advanced
nodal disease (N2/3) correlated with a higher maximum BP dose
than N0/1 disease (60.9 vs 52.8 Gy; po0.0001).
dose to the BP in patients receiving IMRT has been shown to be a
signiﬁcant contributing factor in the development of symptomatic
Exploring dosimetric factors that may
journal homepage: www.meddos.org
0958-3947/$ –see front matter Copyright Ó2014 Published by Elsevier Inc. on behalf of American Association of Medical Dosimetrists
Meeting presentation: Presented in part at the 54th Annual Meeting of the
American Society for Radiation Oncology, October 28 to 31, 2012, Boston, MA.
Reprint requests to: Minh Tam Truong, M.D., Department of Radiation Oncol-
ogy, Boston Medical Center, 830 Harrison Avenue, Moakley Building LL 238, Boston,
E-mails: firstname.lastname@example.org, email@example.com
Current address: Department of Radiation Oncology, Memorial Sloan-
Kettering Cancer Center, New York, NY.
Medical Dosimetry 39 (2014) 169–173
contribute to a greater risk of brachial plexopathy will help deﬁne
and determine attainable dose limits to the BP during IMRT
optimization, which may subsequently reduce the risk of BP-
related complications. Hence, the purpose of this study was to
evaluate the inﬂuence of volumetric tumor burden on planned BP
dose in head and neck IMRT, when the BP is routinely contoured as
an avoidance structure for IMRT optimization.
Methods and Materials
Between August 2005 and May 2011, 75 patients with HNC were treated with
deﬁnitive IMRT. Patients undergoing primary surgery and adjuvant postoperative
radiation were excluded from this study. All patients were staged according to the
2002 American Joint Committee on Cancer.
The study was conducted as a
retrospective review and approved with a waiver of informed consent by the
institutional review board.
Patient simulation and immobilization technique
Before receiving radiation therapy (RT), computed tomography (CT) simulation
was performed with 2- to 3-mm slice thickness, extending from the vertex of the
scalp to at least 5 cm below the clavicle. The patients were immobilized on a carbon
ﬁber Civco “S-frame”with a type S thermoplastic head and neck board.
IMRT planning technique
The following structures were contoured by the physician on the planning CT:
gross tumor volume (GTV), clinical target volume (CTV), planning target volume
(PTV), and OAR including the bilateral BPs (each contoured as its own separate
OAR). Margins of 7 to 15 mm were added to GTV to generate the CTV, followed by 3
to 5 mm expansion to PTV. GTVs were contoured incorporating diagnostic CT,
positron emission tomographic, and magnetic resonance images when available
from pretreatment scans.
Treatment planning was performed using Pinnacle treatment planning software,
version 6.0 to 8.0m (Philips Medical Systems, Fitchburg, WI). The IMRT optimization
objectives constrained the BP maximum dose to o60 Gy if adjacent nodal disease
was present or o56 Gy for patients receiving elective nodal irradiation based on the
gradient method. In the cases when the BP overlapped with PTV, priority was given
to PTV coverage while keeping hot spots outside of the BP. IMRT plans were
normalized such that 95% of PTV was covered with the prescription dose (66 to
69.96 Gy) and no more than 1% of PTV received less than 93% of prescription dose,
and no more than 1% or 1 cc of PTV received more than 110% of prescription dose.
All patients were treated with 7 to nine 6-MV photon beam step-and-shoot
technique. In 9 patients (12.0%), an upper IMRT plan was matched to a low anterior
neck (LAN) ﬁeld; the remaining patients were treated with full-length IMRT ﬁelds.
Elective nodal areas and regions at risk for subclinical disease were treated to 54 to
60 Gy using a dose painting technique.
The volume, mean, minimum, and maximum planned doses to the BP were
recorded in the dose-volume statistics report generated by the treatment planning
system at the time of treatment and retrospectively collected. The volumes of the
primary tumor (GTV-P) and nodal disease (GTV-N) were retrospectively collected
from the original treatment contours.
Brachial plexus contouring technique
The right BP and left BP were contoured as separate structures in all 75
patients. The contouring methodology has been previously described.
superior and lower limits of the BP, between the C4-C5 and T1-T2 neural foramina,
were identiﬁed on a sagittal CT view. The ventral rami of C5-T1 exiting through the
intervertebral neural foramina were contoured on the axial CT. The BP trunks were
contoured between the anterior and middle scalene muscles to the insertion of the
scalene muscles into the ﬁrst rib.
Descriptive statistics were calculated for tumor characteristics and dose-
volume histogram parameters obtained from the radiation plans. For the analysis
of GTV-P with planned radiation dose parameters (maximum and mean dose to
BP), the BP (left or right) receiving the higher dosage was used. For the analysis of
GTV-N with planned BP radiation dose parameters, dose parameters from the
ipsilateral BP were used. Cutoff values of 60, 66, and 70 Gy for maximum dose to BP
were used to categorize the cohort into 2 dose groups. Comparisons between the
2 groups were made with the unpaired t-test.
Receiver operating characteristics (ROC) curves were constructed, and opti-
mized sensitivity- and speciﬁcity-deﬁned volumetric thresholds of GTV-P and GTV-
N were identiﬁed as the cutoff values that best predicted delivery of maximum
radiation dose of 466 Gy to either plexus (for GTV-P) and to ipsilateral BP (for
GTV-N), respectively. ROC curve analyses were performed to demonstrate overall
discriminatory power of a predictive model over the whole range of GTV values.
The area under the ROC curve (AUC) was used to assess the predicted validity of
GTV-P and GTV-N.
The closer the AUC value is to 1.0, the more predictive the GTV
volume parameters were with respect to delivery of maximum radiation dose of
466 Gy to BP. Based on these cutoff values, patients were dichotomized into
2 groups (GTV ovs Zcutoff value) and compared in terms of mean and maximum
dose delivered to the BP.
Univariate and multivariate analyses were performed using the general linear
model (Proc GLM) of SAS 9.1 system (SAS Institute, Cary, NC), and crude and
adjusted maximum and mean dose to BP (with standard errors) were calculated for
ROC cutoff-based GTV groups. The following potential confounding variables were
explored in these analyses: BP volume (continuous variable), GTV (continuous
variable), use of a LAN ﬁeld (categorical), Tumor category (categorical: T0-T3 vs T4),
Nodal category (categorical: N0-N1 vs N2-N3), and total radiation dose (continuous
variable). Finally, patients were evaluated for local control, nodal control, and
overall survival from conclusion of RT until last available follow-up or death.
Patients with a follow-up of less than 3 months were excluded from treatment
outcome analysis unless there was a disease recurrence during that period.
Actuarial control rates at 2 years were estimated using the Kaplan-Meier prod-
uct-limit method. A 2-sided hypothesis was used for all tests, and a probability
value of less than 0.05 was considered statistically signiﬁcant.
Patient and treatment characteristics
The median age of study population was 58.0 years (range: 31
to 86 years). Stage III to stage IV disease was present in 65 (86.7%)
patients. Four patients had an unknown primary and 3 were
treated for nodal disease recurrence, hence there were 7 patients
with T0 category disease, without a GTV-P. Complete tumor
characteristics are described in Table 1.
The median follow-up for the entire patient cohort (n¼75)
was 24.2 months, (range: 0 to 72.1 months). There were 6 patients
who died from unknown or noncancer-related causes within
3 months (range: 0 to 2.4 months) of completing RT, that were
excluded from the disease control analysis.
Local and nodal recurrences occurred in 12 (17.4%) and 7 (10.1%)
patients, respectively. The median time to recurrence was
3.4 months (range: 1.9 to 19.3 months) and 2.1 months (range:
0.3 to 3.5 months) for local and nodal failure, respectively. The
estimated 2-year actuarial local control, nodal control, and overall
survival were 80.8%, 89.8%, and 77.9%, respectively. To date, we
have not detected any symptomatic brachial plexopathies, though
our median follow-up is short to reliably detect such a potential
Dose and volume statistics for 150 BPs
The mean (⫾standard deviation, SD) BP volume, BP mean dose,
and BP maximum dose were 8.5 ⫾4.5 cc, 43.7 ⫾10.0 Gy, and 59.6
⫾11.5 Gy, respectively (Table 2). There were no statistically
signiﬁcant differences between mean right and left BP volume
(p¼0.958), mean dose (p¼0.604), or maximum dose (p¼0.646).
Gross tumor volume
The GTV-P was contoured in 68 patients (mean ¼40.8 cc; SD ¼
51.1 cc). Of 150 BP, 84 BP were adjacent to GTV-N (mean ¼30.5 cc;
SD ¼67.8 cc), with no statistically signiﬁcant difference between
right and left sides (p¼0.624). In 66 BP, no adjacent nodes were
P.B. Romesser et al. / Medical Dosimetry 39 (2014) 169–173170
Correlating maximum BP dose with primary GTV
Overall, 40 (58.8%), 14 (20.6%), and 11 (16.2%) patients received
a maximum dose of 460, 466, and 470 Gy, respectively, to at
least 1 BP. A higher maximum BP radiation dose correlated with a
greater mean GTV-P: 79.5 vs 30.8 cc (p¼0.001) and 98.6 vs 29.7 cc
(po0.0001), for 466 vs r66 Gy and 470 vs r70 Gy,
respectively (Table 3). When dichotomized by proximity of the
primary site to the BP: oropharynx, hypopharynx, and larynx
tumors (in closer proximity to the BP) were compared with
nasopharynx, oral cavity, and others (more distant from the BP),
a signiﬁcant difference in mean GTV-P was noted for the 66 and
70 Gy cutoffs (Fig.).
Correlating maximum BP dose with nodal GTV
A total of 55 (65.5%), 22 (26.2%), and 13 (15.5%) individual BP
received a maximum dose of 460, 466, and 470 Gy, respec-
tively. A higher maximum BP radiation dose correlated with a
greater mean ipsilateral GTV-nodal; 60.6 vs 19.8 cc (p¼0.014) and
79.0 vs 21.6 cc (p¼0.004) for 466 vs r66 Gy and 470 vs r
70 Gy, respectively (Table 3).
ROC curves analysis
ROC curve analysis identiﬁed 79.6 cc (AUC ¼64.2%; CI: 51.6% to
75.4%) and 4.9 cc (AUC ¼65.6%; CI: 54.4% to 75.6%) as the
optimized cutoff values of GTV-P and ipsilateral GTV-N, respec-
tively, which best predicted a maximum BP radiation dose of
466 Gy. When dichotomized by the ROC threshold, a larger
ipsilateral GTV-N volume (Z4.9 vs o4.9 cc) signiﬁcantly corre-
lated with a higher maximum ipsilateral BP dose (64.2 vs 59.4 Gy;
p¼0.001) but no difference in maximum BP dose was noted for
larger GTV-P volume (63.0 vs 60.8 Gy for Z79.6 vs o79.6 cc; p¼
0.518) (Table 4).
Tumor characteristics of 75 patients with head and neck cancer treated from 2005
Hypopharynx 10 (13.3)
Larynx 20 (26.7)
Nasophayrnx 4 (5.3)
Oral cavity 7 (9.3)
Oropharynx 28 (37.3)
Unknown primary 4 (5.3)
I 4 (5.3)
II 6 (8.0)
III 9 (12.0)
IV 56 (74.7)
T0 7 (9.3)
T1 10 (13.3)
T2 18 (24.0)
T3 20 (26.7)
T4 20 (26.7)
N0 18 (24.0)
N1 6 (8.0)
N2 41 (54.7)
N3 10 (13.3)
n¼number of patients; AJCC ¼American Joint Committee on Cancer.
Note: 3 patients were treated for recurrent disease in nodes (tumor category, T0).
Other sites include sinonasal and external auditory canal tumor.
Gross tumor volume and dose-volume histogram statistics of 75 patients with head
and neck cancer with 150 brachial plexuses
nMean (SD) Median (range)
Gross tumor volume (cc)
Primary 68 40.8 (51.1) 18.3 (2.9 to 214.4)
Nodal 84 30.5 (67.8) 9.0 (0.37 to 512.1)
BP volume (cc) 150 8.5 (4.5) 7.0 (1.8 to 29.6)
BP minimum dose (Gy) 150 26.2 (10.6) 29.6 (0.96 to 40.2)
BP maximum dose (Gy) 150 59.6 (11.5) 60.5 (1.9 to 78.4)
BP mean dose (Gy) 150 43.7 (10.0) 45.8 (0.46 to 61.6)
Gross tumor volume of primary tumor was contoured for 68 patients
while of 150 brachial plexuses (75 on each side), 84 were associated with an
ipsilateral nodal disease (gross tumor volume nodal) and 66 were with no nodal
Correlating various thresholds of maximum radiation dose to brachial plexus with
gross tumor volume
Value Mean SD
r60 28 27.2 37.8 ref 29 19.7 32.8 ref
460 40 50.3 57.1 0.066 55 36.1 80.1 0.292
r66 54 30.8 34.6 ref 62 19.8 33.5 ref
466 14 79.5 80.8 0.001 22 60.6 116.7 0.014
r70 57 29.7 34.0 ref 71 21.6 40.1 ref
470 11 98.6 81.4 o
13 79.0 139.3 0.004
For the analysis of GTV-Primary, the maximum dose is from the brachial
plexus (left or right) receiving the greater amount of radiation dosage; for analysis
of GTV-nodal, maximum dose is from the brachial plexus corresponding to the side
with nodal disease.
Fig. Correlating maximum dose to brachial plexus (left or right) with primary gross
tumor volume devised by tumor subsite (oropharynx, hypopharynx, and larynx
compared with nasopharynx, oral cavity, and others).
P.B. Romesser et al. / Medical Dosimetry 39 (2014) 169–173 171
The difference in maximum ipsilateral BP dose, when dicho-
tomized by GTV-N of 4.9 cc, retained signiﬁcance after individually
adjusting for BP volume, GTV-P, the use of a LAN ﬁeld technique,
total planned radiation dose, and tumor category. After adjusting
for GTV-P, the maximum BP dose was 64.1 and 59.4 Gy (p¼0.001)
for GTV-N category o4.9 and Z4.9 cc, respectively.
Tumor volume of the primary head and neck disease site and
nodal disease in head and neck IMRT inﬂuences the BP maximum
dose. Our data suggest that ipsilateral GTV-N and rather than
the GTV-P predominantly drives the maximum BP dose. Patients
with larger nodal GTVs were at signiﬁcant risk of receiving a
greater radiation dose to the BP, with an ipsilateral nodal volume
of 4.9 cc identiﬁed as the optimized threshold of GTV-N above
which the maximum ipsilateral BP dose was more likely to exceed
66 Gy. For the primary head and neck disease site, when dichot-
omizing by proximity to the BP, oropharynx, hypopharynx, and
larynx tumors (which are anatomically in closer proximity to the
BP) compared with nasopharynx, oral cavity, and others, have a
lower primary tumor volumetric thresholds when using the 60,
66, and 70 Gy BP dose cut points. In our study, we found that the
relationship between the GTV-P and BP dose constraints was more
signiﬁcant for patients with oropharynx, hypopharynx, and larynx
Chen et al.
recently reported the ﬁrst study suggesting a dose-
response relationship for the development of brachial plexopathies
with a 1.39 times greater odds of developing symptoms with each
1 Gy increase in the maximum BP dose. On multivariate analysis,
neck dissection (hazard ratio ¼9.55; po0.001) and maximum BP
dose (hazard ratio ¼1.87; po0.001) were independent predictors
of brachial plexopathy. This study reported a brachial plexopathy
rate of 12%, which was signiﬁcantly higher than that of previously
published reports. This increased rate was likely because of a
prospective questionnaire used to identify patients with subjective
symptoms suggestive of brachial plexopathy.
was adopted from studies on breast cancer and has not been
validated in the HNC population. Therefore its sensitivity to detect
brachial plexopathy in contrast to other treatment-related upper
extremity neuropathies related to head and neck treatment has
not yet been determined. The symptoms on the questionnaire may
detect the sequelae of head and neck surgery, particularly a neck
dissection with or without sacriﬁce of the spinal accessory nerve,
chemotherapy-related peripheral neuropathy, or other treatment-
related neuropathies, and may not be purely attributed to the
effect of IMRT on the BP. For example, tingling in the hands and
ﬁngers could be related to peripheral neuropathy from cisplatin-
based chemotherapy regimens, whereas pain in the arm or
shoulder may also be related to effect of neck dissection on the
spinal accessory nerve or from postoperative ﬁbrosis and not
exclusively owing to radiation-induced neuropathy. Despite these
limitations, Chen et al. have demonstrated the importance of
patient-reported outcomes as a measure of toxicity from the
patient’s perspective with dose correlation of the BP dose in head
and neck IMRT.
Multiple studies have demonstrated that a larger GTV-P corre-
lates with increased rates of locoregional recurrence, development
of distant metastasis, and mortality.
Given the higher risk of
locoregional failure in patients with larger GTVs, a more aggressive
approach is justiﬁed in giving priority to GTV and tolerating higher
levels of BP maximal dose, especially given the potential for cure.
As such, it is our preference that priority is given to tumor
coverage, even if this results in exceeding guidelines for BP dose
constraints. This approach has maintained high local and regional
control rates in this patient population, for whom 87% of patients
have locally advanced HNC without brachial plexopathies reported
to date in our series. Arguably, we did not perform patient-
reported outcomes to determine more subtle patient symptoms
relating to upper extremity dysfunction and plan to evaluate this
in future studies. Future studies could also examine the inﬂuence
of primary site on brachial plexopathy risk, as our study demon-
strated that tumors in closer proximity to the BP such as the
oropharynx, larynx, and hypopharynx generally have lower tumor
volume thresholds when correlating to BP dose compared with
HNCs of the nasopharynx and oral cavity.
The BP roots and trunks are located between the anterior and
middle scalene muscles, which is medial and adjacent to the nodal
levels II through IV. In the setting of gross nodal disease, treatment
of the involved nodes to 66 to 70 Gy results in overlap of the nodal
PTVs on the BP roots and trunks, hence the increased likelihood of
exceeding BP dose constraints even when the volume of nodal
disease is low. It has been suggested that the BP roots, trunk,
divisions, cords, and branches may have differences in radiation
dose tolerance given the differences in reported BP radiation dose
tolerance in the breast cancer literature,
as compared with the
Furthermore, BP contouring guidelines in HNC
usually do not deﬁne the BP outside the IMRT ﬁeld, and hence the
BP traversing the axilla (within the ﬁeld of breast cancer radiation)
is usually outside the region of head and neck IMRT optimization.
However, there have been no convincing studies evaluating the
differential sensitivity along the length of the BP. This observation
warrants further evaluation as it could potentially identify areas at
In this study, we deﬁned an ipsilateral nodal volume threshold
of threshold 4.9 cc above which there is a signiﬁcantly greater risk
of prescribing a maximal BP dose of greater than 66 Gy. Methods
to reduce BP dose include reducing PTV margins on elective nodal
volumes or by reducing CTV and PTV margins on gross nodal
disease. Reduction of PTV expansions to 2.5-mm CTVs may be
permitted in the context of onboard imaging with cone-beam CT
or daily kilovoltage imaging, which allows for improvement in
treatment setup reproducibility. In head and neck IMRT, maintain-
ing maximum BP dose constraints within tolerance guidelines can
Maximum and mean dose to brachial plexus volume by primary and nodal gross
Maximum dose (Gy)*Mean dose (Gy)*
Mean SD Mean SD
o79.6 cc 57 60.8 6.5 44.5 7.5
Z79.6 cc 11 63.0 21.4 46.3 16.3
pValue 0.518 0.559
No nodal disease 66 56.0 14.9 40.8 12.3
Nodal disease 84 62.5 6.7 46.0 7.0
pValue 0.0004 0.002
o4.9 cc 29 59.4 4.4 44.1 7.8
Z4.9 cc 55 64.2 7.1 47.0 6.4
pValue 0.001 0.075
For the analysis of GTV-primary, the maximum and mean dose is from the
brachial plexus (left or right) receiving the greater amount of radiation dosage; for
analysis of GTV-nodal, radiation dose is from the brachial plexus corresponding to
the side with nodal disease.
GTV-primary and GTV-nodal volumes of 79.6 and 4.9 cc are the optimized
cutoff values which best predicted maximum radiation dose of 466 Gy to either
plexus (for GTV-P) and to corresponding brachial plexus (for GTV-N).
P.B. Romesser et al. / Medical Dosimetry 39 (2014) 169–173172
be challenging in patients with gross nodal disease Z5 cc.
Although our objective of head and neck IMRT is to achieve 95%
of PTV coverage with the prescription dose (66 to 69.96 Gy) as a
priority over BP dose constraints, there is the potential that a
reduction in nodal CTV and PTV expansions may improve our
ability to limit BP dose. Although different optimization algorithms
are currently available and clinically implemented, the principles
of head and neck IMRT optimization remain similar. Most
approaches are based on the gradient method and simulated
annealing, which sample and explore the possible solution space
with differing strategies. Hence, with the same input parameters,
there will be different outputs after a single iteration. Although the
treatment planning process is inherently iterative and continu-
ously adapting, the radiation dosimetrist/planner may add addi-
tional optimization contours to further shape the dose locally, and
subsequently modify the optimization parameters multiple times
to achieve the desired dose constraints. In our experience, the
results are isodose lines that are sculpted to conform to the target
and avoid adjacent normal tissue, according to the original
optimization goals taking into account the treatment delivery
The results of this study assist in informing the radiation
oncologist of the expected BP dosimetry in head and neck IMRT
planning and its relationship to the tumor burden, when the BP is
routinely contoured as an avoidance structure. If the BP is not
avoided during the IMRT optimization process, the doses to the BP
are expected to be higher than the BP dose ﬁndings of this study.
To understand the clinical implications and true tolerance of the
BP in the HNC population in the context of volumetric tumor
burden, future studies that enable correlation of tumor and BP
dosimetric data with nerve conduction studies or electromyogra-
phy in treated patients and using a validated patient-reported
symptom module would be required.
1. Schierle, C.; Winograd, J.M. Radiation-induced brachial plexopathy: Review.
Complication without a cure. J. Reconstr. Microsurg. 20:149–52; 2004.
2. Chen, A.M.; Hall, W.H.; Li, J.; et al. Brachial plexus–associated neuropathy after
high-dose radiation therapy for head-and-neck cancer. Int. J. Radiat. Oncol. Biol.
Phys. 84:165–9; 2012.
3. Chen, A.M.; Hall, W.H.; Li, B.Q.; et al. Intensity-modulated radiotherapy
increases dose to the brachial plexus compared with conventional radiotherapy
for head and neck cancer. Br. J. Radiol. 84:58–63; 2011.
4. Truong, M.T.; Romesser, P.B.; Qureshi, M.M.; et al. Radiation dose to the brachial
plexus in head-and-neck intensity-modulated radiation therapy and its relation-
ship to tumor and nodal stage. Int. J. Radiat. Oncol. Biol. Phys. 84:158–64; 2012.
5. Hall, W.H.; Guiou, M.; Lee, N.Y.; et al. Development and validation of a
standardized method for contouring the brachial plexus: Preliminary dosimet-
ric analysis among patients treated with IMRT for head-and-neck cancer. Int. J.
Radiat. Oncol. Biol. Phys. 72:1362–7; 2008.
6. Emami, B.; Lyman, J.; Brown, A.; et al. Tolerance of normal tissue to therapeutic
irradiation. Int. J. Radiat. Oncol. Biol. Phys. 21:109–22; 1991.
7. Truong, M.T.; Nadgir, R.N.; Hirsch, A.E.; et al. Brachial plexus contouring with CT
and MR imaging in radiation therapy planning for head and neck cancer.
Radiographics 30:1095–103; 2010.
8. Yi, S.K.; Hall, W.H.; Mathai, M.; et al. Validating the RTOG-endorsed brachial
plexus contouring atlas: An evaluation of reproducibility among patients
treated by intensity-modulated radiotherapy for head-and-neck cancer. Int. J.
Radiat. Oncol. Biol. Phys. 82:1060–4; 2012.
9. Edge, S.B., Byrd, D.R., Compton, C.C., editors. AJCC Cancer Staging Manual. 7th
ed., Chicago, IL: Springer. 2010.
10. Metz, C.E. Basic principles of ROC analysis. Semin. Nucl. Med. 8:283–98; 1978.
11. Hanley, J.A.; McNeil, B.J. The meaning and use of the area under a receiver
operating characteristic (ROC) curve. Radiology 143:29–36; 1982.
12. Hoeller, U.; Rolofs, K.; Bajrovic, A.; et al. A patient questionnaire for radiation-
induced brachial plexopathy. Am. J. Clin. Oncol. 27:1–7; 2004.
13. Romesser, P.B.; Qureshi, M.M.; Shah, B.A.; et al. Superior prognostic utility of
gross and metabolic tumor volume compared to standardized uptake value
using PET/CT in head and neck squamous cell carcinoma patients treated with
intensity-modulated radiotherapy. Ann. Nucl. Med. 26:527–34; 2012.
14. Powell, S.; Cooke, J.; Parsons, C. Radiation-induced brachial plexus injury: Follow-
up of two different fractionation schedules. Radiother. Oncol. 18:213–20; 1990.
15. Cooke, J.; Powell, S.; Parsons, C. The diagnosis by computed tomography of
brachial plexus lesions following radiotherapy for carcinoma of the breast. Clin.
Radiol. 39:602–6; 1988.
P.B. Romesser et al. / Medical Dosimetry 39 (2014) 169–173 173