Int. J. Radiation Oncology Biol. Phys., Vol. 81, No. 3, pp.

e111–e118, 2011
Copyright Ó 2011 Elsevier Inc.
Printed in the USA. All rights reserved
0360-3016/$ - see front matter

doi:10.1016/j.ijrobp.2011.01.004

CLINICAL INVESTIGATION Head and Neck

REIRRADIATION FOR HEAD-AND-NECK CANCER: DELICATE BALANCE BETWEEN

EFFECTIVENESS AND TOXICITY

FRANK HOEBERS, M.D., PH.D.,*# WILMA HEEMSBERGEN, PH.D.,*y SUZANNE MOOR, R.T.T.,*
MARTA LOPEZ, PH.D.,y MARTIN KLOP, M.D., PH.D.,z MARGOT TESSELAAR, M.D., PH.D.,x
AND COEN RASCH, M.D., PH.D.*

Departments of *Radiation Oncology, yBioinformatics and Statistics, zHead and Neck Oncology and Surgery, and xMedical Oncology,
The Netherlands Cancer Institute / Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands, #Maastricht University Medical
Center, Department of Radiation Oncology (MAASTRO clinic), GROW School for Oncology and Developmental Biology, Maastricht,
The Netherlands

Purpose: To analyze the effectiveness and toxicity of reirradiation (re-RT) for head-and-neck cancer.
Methods and Materials: A retrospective data analysis was performed of 58 patients who underwent re-RT with
curative intent. Re-RT was given as definitive treatment in 53% of patients, whereas salvage surgery preceded re-
irradiation in 47%. The median cumulative RT dose was 119 Gy (range, 76–140). Concurrent chemotherapy was
administered with re-RT (CRT) in 57% of patients. Event-free survival was defined as survival without recurrence
and without serious toxicity ($Grade 3).
Results: Median follow-up was 57 months (range, 9–140). Locoregional (LR) control was 50% at 2 and 5 years. The
2-year and 5-year overall survival (OS) was 42% and 34%. The following factors were associated with improved OS:
postoperative re-RT (vs. primary re-RT), treatment with RT only (vs. CRT) and interval >3 years between previous
RT and re-RT. For patients treated with postoperative re-RT and definitive re-RT, the 5-year OS was 49% and 20%,
respectively. Patients treated with CRT had a 5-year OS of 13%. Serious (late) toxicity $Grade 3 was observed in 20
of 47 evaluable patients (43%). Three cases of treatment-related death were recorded. The 2- and 5-year serious
toxicity-free interval was 59% and 55%, respectively. Associated with increased risk of serious toxicity were CRT
and higher re-RT dose. The event-free survival rates at 2 and 5 years were 34% and 31%, respectively.
Conclusions: Re-RT in head-and-neck cancer is associated with poor survival rates of 13–20% in patients with
inoperable disease treated with primary (chemo-) re-RT. For this subgroup, however, no other curative options
are available. Long-term disease control and survival can be achieved in patients who receive re-RT as an
adjunct to surgical resection. The rates of serious toxicity after re-RT are high, with an incidence of approxi-
mately 45% at 5 years. Approximately 1 in 3 patients survived re-RT without recurrence and severe
complications. Ó 2011 Elsevier Inc.

Reirradiation, Head-and-neck cancer, Toxicity, Complications.

INTRODUCTION rates (20–35%) and a median survival of less than 9 months

(4). The possibilities for full-dose reirradiation after previous
In head-and-neck cancer the incidence of recurrent disease
radiation are limited by potential normal tissue complications
may be as high as 30–50% after radiation treatment (RT)
caused by cumulative radiation doses exceeding the normal
(1, 2). There is also a risk of approximately 20% for second
tissue tolerance. However, over the past two decades, experi-
primary tumors (3), based on the continued exposure to risk
ence with reirradiation has been described by several groups
factors associated with malignancies (mainly nicotine abuse
(see review by Kasperts et al). (5). The general conclusion
and alcohol consumption). Recurrent and/or second primary
from most of these noncontrolled studies is that in this popu-
tumors after previous RT can be regarded as a challenging
lation of patients with poor prognosis, reirradiation may be
problem for the multidisciplinary head-and-neck cancer
associated with long-term survival in a small proportion of
team. The treatment of choice after previous RT is salvage
patients and that the incidence of severe toxicity is substan-
surgery. In the case of locally advanced or inoperable disease,
tial. Although scientific evidence is still limited, with mainly
chemotherapy can be offered, with relatively low response
retrospective or Phase I–II studies and only one randomized

Reprint requests to: Frank Hoebers, M.D., Ph.D., Maastricht
University Medical Center, Department of Radiation Oncology
(MAASTRO clinic), GROW School for Oncology and Develop-
mental Biology, Dr. Tanslaan 12, 6229 ET Maastricht, The Nether-
e111
lands. Tel (+31) 88 4455666; Fax (+31) 88 4455667; E-mail: frank.
hoebers@maastro.nl
Conflict of interest: none.
Received Aug 12, 2010, and in revised form Dec 29, 2010.
Accepted for publication Jan 3, 2011.
e112 I. J. Radiation Oncology d Biology d Physics Volume 81, Number 3, 2011

Table 1. Patient, tumor, and treatment characteristics of 58 with RT. The following selection criteria were applied: histologic
patients proof of recurrent disease or second primary tumor after previous
radiation and no evidence of distant metastases. The diagnostic
No. %
evaluation included physical examination, radiologic evaluation
Patient of the head and neck by computed tomography (CT) and/or mag-
Sex netic resonance imaging, panendoscopy with biopsies and screen-
M 42 72 ing for distant metastases using CT and/or positron emission
F 16 28 tomography—(CT). For previously irradiated patients presenting
Mean age (y, range) 60 (34–81) with recurrent or second primary tumors, surgical salvage has re-
Tumor mained the standard of care in our clinical practice. In cases of ir-
Primary tumor site resectability, primary reirradiation (with or without concurrent
Oropharynx 18 31 chemotherapy) was to be discussed in the multidisciplinary tumor
Larynx 12 21
board meeting and, if deemed appropriate, presented to the patient
Oral cavity 5 9
Nasopharynx 4 7 as a therapeutic option. In general, for patients with recurrent dis-
Hypopharynx 4 7 ease the minimum time interval between previous RT and reirradia-
Esophagus 3 5 tion was 2 years, unless the recurrence occurred in a nonirradiated
Nasal cavity/sinus 3 5 area (e.g., the opposite side of the neck). Postoperative reirradiation
Unknown primary 1 2 was considered only if the pathologic features of the surgical spec-
Other* 8 14 imen indicated a high risk of subsequent recurrence: positive mar-
Histology gins, lymph node metastasis with extracapsular extension (ECE),
Squamous cell carcinoma 51 88 and/or multiple lymph node metastases.
Salivary gland carcinoma 4 7 Between 1998 and 2009, 58 patients with head-and-neck cancer
Nasopharyngeal carcinoma 1 2
underwent reirradiation with curative intent. In 88% of cases, the
Othery 2 3
Treatment histology of the retreated tumor was squamous cell carcinoma.
Indication for reirradiation Head-and-neck tumor sites included oropharynx, nasopharynx,
Second primary tumor 21 36 hypopharynx, larynx, or oral cavity in 75% of cases (Table 1).
recurrence 37 64
Reirradiation setting
Primary/definitive 31 53 Treatment
Postoperative 27 47 Radiotherapy was generally given with 4- to 6-MV photon linear
Indication for postoperative re-RT accelerators using an immobilization mask. Treatment was given
R1 resection 12 44 according to the standard of practice at the time of reirradiation
R2 resection 1 4 using either standard parallel opposing fields, with or without an
Nodal ECE 10 37 adjacent anteroposterior supraclavicular beam, or an intensity-
pN2 (a/b/c) 4 15 modulated (IM) RT technique, depending on resources. From
Concurrent chemotherapy: 2001 to 2002 on, three-dimensional (3D) RT and IMRT were intro-
Cisplatin-based 33 57
duced into the routine clinical practice, and 30 patients were treated
None 25 43
with these techniques. The remaining 28 patients received standard
Abbreviations: ECE = extracapsular extension; RT = radiation techniques with parallel opposing beams. The spinal cord was con-
therapy. sidered the most important organ at risk during reirradiation.
* Other sites included salivary gland (3), trachea (1), vestibulum Depending on the interval between the date of first RT and the re-
nasi (1), and skin (3). irradiation, some degree of repair of sublethal damage to the spinal
y
Other histologies included basal cell carcinoma (1) and Merkel cord was taken into account. The cumulative RT dose of the first
cell carcinoma (1). treatment (including the calculated repair) and the retreatment
was kept below 50 Gy for the spinal cord. Because a substantial pro-
trial available (6), to our knowledge, this treatment approach portion of patients received irradiation and/or reirradiation with
standard parallel opposing fields, no 3D dose information was
has gained acceptance over time.
available for other organs at risk except for the spinal cord. In ad-
The aim of this retrospective single-center study was to
dition, some patients received their first radiation treatment outside
analyze the effectiveness of reirradiation for head-and-neck our center, which again prohibited calculation of cumulative doses
cancer with regard to toxicity, locoregional (LR) control, to organs at risk (e.g., mandible, large vessels, nerves, vertebrae).
and survival. All patients who received reirradiation with Cumulative radiation doses to these structures could therefore not
curative intent with overlapping fields, after a previous radia- be calculated, and no dose restrictions for them were used.
tion treatment to the head-and-neck area for a recurrence or In the majority of cases, target volume definition for reirradiation
secondary primary tumor, were included in this study. included the gross tumor volume (GTV) for both the primary tumor
and the lymph nodes. Given the differences in treatment techniques,
a uniform description of the GTV to clinical target volume (CTV)
METHODS AND MATERIALS expansion could not be given, but a maximum margin of 1.5 cm was
applied to the GTV to define the CTV. The CTV to planning target
Patients volume margin was 5 mm in 3D RT and IMRT patients. In addition
Patients were selected who received external beam reirradiation to this CTV, in most cases elective areas were also included of re-
to the head-and-neck area with curative intent (intended RT dose gions that were considered to be at risk for microscopic disease.
$50Gy) to overlapping areas that had previously been treated This included other neck levels of the ipsilateral or contralateral
 Reirradiation in head-and-neck cancer d F. HOEBERS et al.
Fig. 1. Overlap between initial fields and reirradiation fields.
BoS = base of skull.
side of the neck. In a postoperative setting, the localization of the
tumor was reconstructed on the simulator films or planning CT,
and the target volume included the (resected) GTV (with a maxi-
mum 1.5-cm margin for CTV) and the entire surgical bed including
the neck. Because some of the patients were treated with conven-
tional parallel opposing fields and others with IMRT, a semiquanti-
tative descriptive method was chosen to assess the degree of
overlap. This was based on the amount of the head-and-neck area
being reirradiated (Fig. 1).
In general, the radiation dose to the elective areas was 46 Gy,
followed by a boost of 14 to 20 Gy to the high-risk area in a postop-
erative setting or a boost of 24 Gy in a definitive RT setting.
The median dose of the first treatment was 67 Gy (range, 35–70).
The median delivered dose of the reirradiation was 66 Gy (range,
16–70), resulting in a median cumulative dose of 119 Gy (range,
76–140). The median interval between first RT and reirradiation
was 3.0 years (range, 0.3–43).
Concurrent chemotherapy with reirradiation was given in 57% of
cases. This included daily low-dose intravenous cisplatin at a dosage
of 6 mg/m2 in 24 patients according to our previously reported
protocol (7). Seven patients received standard high-dose intravenous
cisplatin (100 mg/m2) in weeks 1, 4, and 7. In 1 patient, cisplatin was
administered intra-arterially at a dosage of 150 mg/m2 in weeks 1, 2,
3, and 4, according to the RADPLAT protocol (RADiotherapy and
concurrent intra-arterial cisPLATin) (8). One patient was treated
with cisplatin and 5-fluorouracil concurrently with radiation. The
planned treatment was completed in 95% of all patients.
Endpoints of survival, tumor control, and toxicity
Acute and late toxicity was scored retrospectively and converted
to the Common Terminology Criteria for Adverse Events (CTC),
version 3.0. For late Grade $3 toxicity, the first date of occurrence
was recorded for actuarial analysis. Severe toxicity that could be at-
tributed to the reirradiation but that occurred within 6 months was
also regarded as an event. Patients with less than 6 months follow-
up and no evidence of severe toxicity at that time were censored
from the severe toxicity analysis. Follow-up was calculated from
the start of reirradiation. LR control was measured from start of
retreatment until LR recurrence. Disease-free survival (DFS) was
defined as the time from start of retreatment until LR recurrence,
distant metastasis, or death, whichever came first. For serious
toxicity-free interval, patients were at risk from start of treatment
until serious toxicity (CTC $Grade 3, event). To evaluate the num-
e113
Fig. 2. Locoregional control (solid line) and overall survival
(dashed line) for all patients undergoing reirradiation.
ber of patients who had benefited from reirradiation, we defined
a composite endpoint: event-free survival (EFS), which was defined
as survival without disease recurrence and without serious toxicity
(CTC $Grade 3).
Statistical analysis
Kaplan-Meier plots and log-rank testing were used to describe
survival. Univariate Cox proportional hazard regression analysis
was performed to establish factors that were associated with LR
control. Given the small sample size (n = 58), no multivariate anal-
ysis was performed. We considered p values <0.05 to be statistically
significant.
RESULTS
The median follow-up was 17 months (range, 0.5–140) of
all 58 patients and 57 months (range, 9–140) among 20 sur-
viving patients. At the time of last follow-up, 20 patients had
no evidence of disease (35%), 1 patient was alive with dis-
ease (2%), and 37 patients were deceased (63%). The cause
of death in these patients was LR tumor in 65% of cases, dis-
tant metastases in 9%, both LR disease and metastases in
9%, treatment related in 9% (n = 3), and unrelated causes
(e.g., pulmonary or cardiac disease) in 8%. For all patients,
the estimated LR control was 50% at 2 and 5 years. The DFS
was 36% at 2 years and 34% at 5 years. Overall survival (OS)
was 42% at 2 years and 34% at 5 years (Fig. 2).
Toxicity and complications after reirradiation
The incidence and worst grades of acute toxicity were
Grade 3 mucositis in 49% of patients (n = 26), Grade 3
skin in 22% (n = 12), tube feeding (nasogastric or by percu-
taneous radiologic gastrostomy) in 48% (n = 27, mainly in
patients treated with concurrent chemoreirradiation: 67%
vs. 20% in patients treated with RT alone), Grade 2 renal
toxicity in 13% (n = 4, in CRT patients only). One patient
required tracheotomy during treatment because of airway
obstruction due to tumor and/or edema.
Late toxicity data are presented in Table 2. The most
frequent late toxicities were stricture/stenosis of the pharynx,
vascular (transient ischemic attack, cerebrovascular accident,
e114 I. J. Radiation Oncology d Biology d Physics
Table 2. Late toxicity scores according to the Common
Criteria Terminology for Adverse Events after reirradiation
Toxicity scores Grade No.
Skin ulceration 0–2 37
3 1
4 0
5 0
Too short FU 20
Fibrosis of deep 0–2 38
connective tissue 3 0
4 1
5 0
Too short FU 19
Late mucosal damage 0–2 37
3 0
4 0
5 0
Too short FU 21
Stricture/stenosis 0–2 36
of pharynx 3 6
4 1
5 0
Too short FU 20
Edema of larynx 0–2 35
3 0
4 1
5 0
Too short FU 22
Osteoradionecrosis 0–2 35
3 6
4 0
5 0
Too short FU 17
Large vessels/vascular 0–2 35
damage 3 1
4 4
5 2
Too short FU 16
Salivary gland 0–2 35
changes: saliva 3 0
4 0
5 0
Too short FU 23
Cumulative serious Yes 20
toxicity ($Grade 3) None 27
Too short FU 11
Abbreviation: FU = follow-up.
carotid blowout), and osteoradionecrosis. Other complica-
tions included severe trismus in 1 patient, and both severe tris-
mus and cranial nerve palsy in another. In total, 25 events of
serious toxicity ($Grade 3) occurred in 20 (43%) of 47 evalu-
able patients. No cases of radiation-associated myelopathy
were seen during follow-up.
In 3 patients, the cause of death was treatment related.
These included two cases of carotid blowout, which oc-
curred within a few months after the end of reirradiation.
At the time of death no evidence of recurrent disease was
present. One patient developed an aspiration pneumonia dur-
ing RT, which was complicated by respiratory failure and ad-
mission to the intensive care unit for mechanical ventilation.
This patient did not recover and died.
Volume 81, Number 3, 2011
%
97
3
0
0
97
0
3
0
100
0
0
0 Fig. 3. Serious toxicity-free interval ($Grade 3, solid line) and
event-free survival, defined as survival without serious toxicity
82 and without disease recurrence, (dashed line) for all patients.
15
3
0 Serious toxicity-free interval and EFS
The serious toxicity-free interval was 59% at 2 years and
97
0 55% at 5 years (Fig. 3). The EFS (i.e., survival without serious
3 toxicity and without disease recurrence) was 34% at 2 years
0 and 31% at 5 years (Fig. 3), indicating that about one third of
the patients survived reirradiation without severe toxicity.
85
15
0 Prognostic factors and univariate analysis
0 A correlation between several treatment-related factors
was observed. Concurrent chemoreirradiation was associ-
83
ated with definitive reirradiation: in 81% of cases, concur-
2
10 rent chemoreirradiation was given as definitive treatment,
5 whereas in 76% of cases treated with RT without chemother-
apy, reirradiation was given postoperatively (chi-square test,
100 p < 0.001). Based on hospital treatment policy guidelines,
0
0
the prescribed radiation dose for postoperative reirradiation
0 was lower (usually 60 Gy) than in cases of definitive chemo-
irradiation (usually 70 Gy), resulting in a higher cumulative
43 dose for the latter group: mean of 123 Gy vs. 113 Gy. A large
57 overlap between previous RT and reirradiation, defined as
overlap to the bilateral neck or unilateral neck, was present
in 47% of cases (Fig. 1). However, in patients treated with
concurrent chemoreirradiation, this was 54% compared to
36% in patients treated with reirradiation alone (chi-square
test, p = n.s.). In general, therefore, concurrent chemoreirra-
diation was associated with definitive reirradiation to
a higher dose and more overlap, whereas reirradiation alone
usually was given as postoperative reirradiation to a lower
dose and less overlap.
The results of univariate analyses for LR control are given
in Table 3. Factors tested were sex, age, primary tumor site,
histology, indication for reirradiation, reirradiation setting,
indication for postoperative reirradiation, reirradiation
dose, dose of first course of radiation, concurrent chemother-
apy, cumulative radiation dose, and interval between first
radiation and reirradiation. Factors associated with better
LR control were second primary tumor (vs. recurrence),
postoperative reirradiation (vs. definitive reirradiation),
 Reirradiation in head-and-neck cancer d F. HOEBERS et al. e115

Table 3. Univariate analysis for locoregional control in 55 patients by Cox regression

Parameter Freq. Events HR 95% CI p value

Indication for re-RT

Second PT 21 6 0.38 0.15 0.94 0.036
Recurrence 34 20 Reference
Re-RT setting
Definitive re-RT 29 20 Reference
Postoperative re-RT 26 6 0.24 0.10 0.61 0.003
Concurrent chemo-RT
No, RT alone 24 3 0.10 0.03 0.34 < 0.001
Yes 31 23 Reference
Interval
#3 years 26 16 Reference
>3 years 29 10 0.43 0.19 0.95 0.036
Cumulative RT dose
Continuous 1.04 1.01 1.07 0.011

Abbreviations: HR = hazard ratio; CI = confidence interval; RT = radiation therapy; PT = primary tumor.

treatment with reirradiation alone (vs. concurrent chemo- Location of LR recurrence relative to radiation dose levels
reirradiation), longer interval between first course and reirra- For patients with LR recurrence and for whom treatment
diation, and lower cumulative RT dose. plans were available, the radiation portals or IMRT plans
were reviewed to locate the recurrence relative to the deliv-
Prognostic factors for serious toxicity ered radiation dose. In 18 cases (82%) the recurrence oc-
Univariate analysis for serious toxicity ($Grade 3) was curred locally within the high-dose area ($60 Gy), in 4%
performed for the following factors (Table 4): indication within the intermediate-dose area (46–59 Gy), none within
for reirradiation, reirradiation setting, reirradiation dose, the area that was treated electively for suspected micro-
dose of first radiation, concurrent chemotherapy, cumulative scopic disease (46 Gy), and 14% out of the RT field.
radiation dose, degree of overlap between first radiation and
reirradiation, and interval. Increased risk for serious toxicity DISCUSSION
was associated with concurrent chemoreirradiation and The current study represents a single-center data analysis
a higher reirradiation dose. Definitive reirradiation and of experience with reirradiation in patients with head-and-
higher cumulative dose were of borderline significance. neck cancer with reasonable long-term disease control and
When serious toxicity endpoints were evaluated sepa-
rately, stricture or stenosis of the pharynx was associated Table 4. Univariate analysis for serious toxicity ($Grade 3)
in 47 patients by Cox regression
with chemoradiation, treatment for recurrence, larger over-
lapping areas, more dose, and longer intervals. Osteoradio- Parameter HR p value
necrosis was associated with chemoradiation, primary RT,
more dose, and shorter intervals. Vascular damage was asso- Indication for re-RT
Second PT 0.55 0.19
ciated with treatment for recurrence (data not shown). Recurrence Reference
Concurrent chemo-RT
No, RT alone Reference
Subgroups analysis
Yes 2.84 0.032
As clearly demonstrated from the univariate analyses pre- Re-RT setting
sented in Tables 3 and 4, patients treated by concurrent Definitive re-RT Reference
chemoreirradiation had the worst outcome: OS was only Postoperative re-RT 0.39 0.05
20% and 13% at 2 and 5 years, respectively (Fig. 4), and Interval
#3 years 0.99 0.66
LR control was only 22% at 2 and 5 years. The serious >3 years Reference
toxicity-free interval for these patients was 32% at 5 years, Overlap
and EFS was 11% at 5 years. Bilateral 1.69 0.30
For patients in whom salvage surgery could be performed, Unilateral 1.07 0.92
followed by postoperative reirradiation, a more favorable Partial Reference
Re-RT dose
outcome was observed: LR control was 76% at 5 years, Continuous 1.11 0.036
OS was 49% at 5 years, serious toxicity-free interval was Dose first RT
72% at 5 years, and EFS was 53% at 5 years. Patients who Continuous 1.04 0.58
received postoperative reirradiation for recurrence in the Cumulative RT dose
neck had better outcomes than did those who were retreated Continuous 1.06 0.06
because of recurrent primary tumor: LR control at 2 years Abbreviations: HR = hazard ratio; RT = radiation therapy; PT =
was 84% vs. 35%. primary tumor.
e116 I. J. Radiation Oncology d Biology d Physics
Fig. 4. Overall survival by treatment modality: reirradiation only
(solid line) vs. concurrent chemoreirradiation (dashed line). RT =
radiation therapy.
survival in a subset of patients. LR control for all patients was
50% at 2 and 5 years, with corresponding OS rates of 42%
and 34%. At 2 and 5 years after reirradiation, 59 and 55%
of patients still alive were free of severe toxicity. For the com-
bined endpoint of EFS (no recurrence and no severe toxicity),
the 2- and 5-year rates were 34% and 31%, indicating that ap-
proximately one third of patients survived reirradiation with-
out severe toxicity. These are the patients who benefited most
from reirradiation: tumor cure, without toxicity.
The subgroup of patients treated with definitive concur-
rent chemoreirradiation had the worst outcome: LR control
and OS were only 22% and 20% at 2 years. These patients
were also at higher risk for severe toxicity: 68% incidence
at 5 years. These results are probably related to a negative
selection of patients with poor prognosis: locally advanced
and inoperable disease treated with definitive CRT. In these
cases, the reirradiation dose was higher and the overlap be-
tween initial RT and reirradiation was more extensive, which
may have caused the increased risk of severe toxicity. How-
ever, these assumptions are based on subgroup analyses from
this relatively small patient series. Other series of definitive
(chemo-) reirradiation have reported 2-year survival rates
ranging from 21% to 35% (9–11) and comparable LR
control (12). Although the figures seem to be poor, it has
to be kept in mind that for this patient group, no other cura-
tive treatment options are available. Chemotherapy alone in
the recurrent or metastatic setting results in median survival
rates of 5 to 9 months (13). A low chance of cancer cure,
therefore, has to be weighed against the high risk of severe
toxicity. In our opinion, offering this treatment option of
concurrent chemoreirradiation to a motivated patient is indi-
cated, assuming that all its limitations are discussed openly.
The subgroup of patients who received postoperative reir-
radiation as an adjunct to salvage surgery seemed to have
a more favorable outcome. LR control and OS were 76%
and 49% at 5 years. The results seem a little better than those
Volume 81, Number 3, 2011
obtained by Janot et al. (6) in their randomized Phase III
study of postoperative concurrent chemoreirradiation vs.
wait-and-see after salvage surgery: In the group that received
retreatment, the LR control was 60% and OS was 25% at 5
years. The entry criteria for that study were more liberal than
the indications for postoperative reirradiation in our series.
In the current series, all patients who received postoperative
reirradiation had either positive margins, ECE, or multiple
nodes in the pathology specimen, compared to only 49%
of cases in the study by Janot et al. (6). The remaining pa-
tients were included because of other factors, including
a neck dissection pathology specimen with a single positive
node (without ECE), or a deep infiltrating tumor (but nega-
tive margin). In addition, in our group of patients receiving
postoperative reirradiation, the severe toxicity rate of 28%
at 5 years seems favorable compared with 39% in the ran-
domized trial. Inasmuch as this trial did not show any benefit
in terms of OS between the retreatment arm and the wait-
and-see arm, it was argued (14) that postoperative reirradia-
tion could be withheld from these patients and that it might
be better to reserve the retreatment for patients who do expe-
rience recurrence after salvage surgery. The real question
then would be one of delivering immediate postoperative
reirradiation vs. delayed reirradiation (‘‘pay now or pay
later’’). However, in the wait-and-see arm of the trial by
Janot et al. (6), half of the patients experienced recurrence,
and only half of these received delayed definitive CRT,
with only 20% of these achieving complete remission, indi-
cating that the results of definitive reirradiation compare
poorly with the results of reirradiation in the postoperative
setting. In a previous study by Jones et al. (15), it was shown
that in a population of patients who underwent salvage
surgery after previous radiotherapy, the risk of subsequent
recurrence was very high and was associated with margin
status: 66% of patients with positive margins experienced
recurrence, compared with 47% of patients with negative
margins. None of the patients in this study underwent
reirradiation.
Other noncontrolled reports on outcome after postoperative
reirradiation have shown similar 2- or 3-year LR control rates
between 40% and 74% (12, 16–18). We propose that patients
be selected for postoperative reirradiation who seem at very
high risk of untreatable recurrence. These indications
include at least positive margins and/or ECE (16, 19–21).
Although there is no true evidence for the addition of
concurrent chemotherapy to postoperative reirradiation,
extrapolation of data from irradiation in the setting of
primary radiation (22–24) suggests that this may increase
the therapeutic effect, and also toxicity, in high-risk patients.
The incidence and grades of late toxicity were consider-
able, with a cumulative rate of 43% of patients experiencing
$Grade 3 toxicity. Actuarial analysis at 2 and 5 years re-
vealed that 59% and 55% of patients were without severe
($Grade 3) toxicity. These results are comparable with those
of other studies that have reported incidence rates of $Grade
3 toxicity ranging from 11% to 57% (9, 12, 17, 25–28).
Toxicity in our study was seen mainly in patients who
 Reirradiation in head-and-neck cancer d F. HOEBERS et al.
were treated with definitive chemoreirradiation to a higher
dose (Table 3). It could be possible that the higher rate of
toxicity might be explained by the fact that these patients
had inoperable disease, indicating larger radiation fields,
larger overlap, and higher dose.
Several limitations of our study have to be acknowledged.
The retrospective design resulted in a heterogeneous study
population in terms of tumor site, histology, and treatment.
Especially with respect to treatment, selection bias was pres-
ent: patients with inoperable disease were obviously not can-
didates for surgical salvage and were therefore treated with
definitive chemoreirradiation. The scoring of toxicity from
retrospective data is difficult by nature, with a risk of missing
data, at least for minor toxicities and complications (Grade
1–2), which may be underestimated. For severe complica-
tions ($Grade 3), this risk is probably low because these
complications need medical attention and/or treatment and
will therefore be documented in the medical charts.
In our series of reirradiation, the target included the gross
tumor (or high-risk volume in the postoperative setting) and
elective areas of the neck in most cases. This resulted in
overlapping volumes encompassing the bilateral neck or
unilateral neck in almost 50% of cases. The need for elective
treatment of nodal echelons in the setting of reirradiation can
be questioned, first, because it was noted in the reirradiation
series from the University of Michigan that in cases with re-
current disease, 96% recurrences were within the GTV (28).
No reirradiation to areas of potential subclinical disease was
given. In our series, also, more than 80% of recurrences de-
veloped within the high-dose region. Elective nodal treat-
ment will result in a larger volume being retreated, and
this in turn may be associated with increased toxicity from
reirradiation. Secondly, drainage patterns after previous ra-
diation (and/or surgery) may be altered (29), and thereby un-
expected nodal metastases could occur despite coverage of
REFERENCES
1. Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemo-
therapy and radiotherapy for organ preservation in advanced la-
ryngeal cancer. N Engl J Med 2003;349:2091–2098.
2. Overgaard J, Hansen HS, Specht L, et al. Five compared with
six fractions per week of conventional radiotherapy of
squamous-cell carcinoma of head and neck: DAHANCA 6
and 7 randomised controlled trial. Lancet 2003;362:933–940.
3. Schwartz LH, Ozsahin M, Zhang GN, et al. Synchronous and
metachronous head and neck carcinomas. Cancer 1994;74:
1933–1938.
4. Wong SJ, Machtay M, Li Y. Locally recurrent, previously irra-
diated head and neck cancer: Concurrent re-irradiation and che-
motherapy, or chemotherapy alone? J Clin Oncol 2006;24:
2653–2658.
5. Kasperts N, Slotman B, Leemans CR, et al. A review on
re-irradiation for recurrent and second primary head and neck
cancer. Oral Oncol 2005;41:225–243.
6. Janot F, de Raucourt D, Benhamou E, et al. Randomized trial
of postoperative reirradiation combined with chemotherapy
after salvage surgery compared with salvage surgery alone
in head and neck carcinoma. J Clin Oncol 2008;26:5518–
5523.
e117
the ‘‘standard’’ echelons. Other series on reirradiation have
also included nodal areas of subclinical disease within the re-
irradiation volume (18, 25, 27), whereas most have not (10–
12, 16, 17, 19, 26, 28) or have included only the first nodal
area (6). Based on the above, we have decided to adopt
our treatment policy not to include all areas of potential sub-
clinical disease within the target volume. It can be antici-
pated that this reduction in retreated volume may result in
fewer complications.
Based on the available literature and our own experience,
the following recommendations can be made for patients
with recurrent disease or a second primary tumor after previ-
ous irradiation: If the tumor is resectable, salvage surgery is
the mainstay of treatment. Postoperative reirradiation should
be considered in patients at high risk of subsequent recurrence
(e.g., in cases of positive resection margins and/or extranodal
spread). If the disease is irresectable, definitive reirradiation
with concurrent chemotherapy can be offered to the patient.
Even though the likelihood of a positive outcome is limited
in these cases, it should be realized that no other curative
treatment options are available. Treatment within the context
of clinical trials remains warranted to establish evidence-
based treatment policies. This may include alterations in
radiation fractionation, more effective chemotherapy, and/or
other modalities like photodynamic therapy (30).
In conclusion, reirradiation in head-and-neck cancer is as-
sociated with poor survival rates in patients with inoperable
disease who are given definitive (chemo-) reirradiation. For
this subgroup, however, no other curative options are avail-
able. Long-term disease control and survival can be achieved
in patients who receive reirradiation as an adjunct to surgical
resection. Rates of serious toxicity ($Grade 3) after reirra-
diation are high, with an incidence of approximately 45%
at 5 years. Approximately 1 in 3 patients survived re-RT
without recurrence and severe complications.
7. Hoebers FJ, Heemsbergen W, Balm AJ, et al. Concurrent che-
moradiation with daily low dose cisplatin for advanced stage
head and neck carcinoma. Radiother Oncol 2007;85:42–47.
8. Rasch CR, Hauptmann M, Schornagel J, et al. Intra-arterial ver-
sus intravenous chemoradiation for advanced head and neck
cancer: Results of a randomized phase 3 trial. Cancer 2010;
116:2159–2165.
9. Langer CJ, Harris J, Horwitz EM, et al. Phase II study of low-
dose paclitaxel and cisplatin in combination with split-course
concomitant twice-daily reirradiation in recurrent squamous
cell carcinoma of the head and neck: Results of Radiation Ther-
apy Oncology Group Protocol 9911. J Clin Oncol 2007;25:
4800–4805.
10. Kramer NM, Horwitz EM, Cheng J, et al. Toxicity and outcome
analysis of patients with recurrent head and neck cancer treated
with hyperfractionated split-course reirradiation and concur-
rent cisplatin and paclitaxel chemotherapy from two prospec-
tive phase I and II studies. Head Neck 2005;27:406–414.
11. de Crevoisier R, Bourhis J, Domenge C, et al. Full-dose reirra-
diation for unresectable head and neck carcinoma: Experience
at the Gustave-Roussy institute in a series of 169 patients. J
Clin Oncol 1998;16:3556–3562.
e118 I. J. Radiation Oncology d Biology d Physics
12. Lee N, Chan K, Bekelman JE, et al. Salvage re-irradiation for
recurrent head and neck cancer. Int J Radiat Oncol Biol Phys
2007;68:731–740.
13. Wong S, Machtay M, Li Y. Locally recurrent, previously irra-
diated head and neck cancer: Concurrent re-irradiation and
chemotherapy, or chemotherapy alone? J Clin Oncol 2006;
24:2653–2658.
14. Wong SJ, Spencer S. Reirradiation and concurrent chemother-
apy after salvage surgery: Pay now or pay later. J Clin Oncol
2008;26:5500–5501.
15. Jones AS, Bin Hanafi Z, Nadapalan V, et al. Do positive resec-
tion margins after ablative surgery for head and neck cancer ad-
versely affect prognosis? A study of 352 patients with recurrent
carcinoma following radiotherapy treated by salvage surgery.
Br J Cancer 1996;74:128–132.
16. Kasperts N, Slotman BJ, Leemans CR, et al. Results of postop-
erative reirradiation for recurrent or second primary head and
neck carcinoma. Cancer 2006;106:1536–1547.
17. Sulman E, Schwartz D, Le T, et al. IMRT reirradiation of head
and neck cancer: Disease control and morbidity outcomes. Int J
Radiat Oncol Biol Phys 2009;73:399–409.
18. Salama JK, Vokes EE, Chmura SJ, et al. Long-term outcome of
concurrent chemotherapy and reirradiation for recurrent and
second primary head-and-neck squamous cell carcinoma. Int
J Radiat Oncol Biol Phys 2006;64:382–391.
19. de Crevoisier R, Domenge C, Wibault P, et al. Full dose reirra-
diation combined with chemotherapy after salvage surgery in
head and neck carcinoma. Cancer 2001;91:2071–2076.
20. Bartelink H, Breur K, Hart G, et al. The value of postoperative
radiotherapy as an adjuvant to radical neck dissection. Cancer
1983;52:1008–1013.
21. Bernier J, Cooper JS, Pajak TF, et al. Defining risk levels in lo-
cally advanced head and neck cancers: A comparative analysis
of concurrent postoperative radiation plus chemotherapy trials
of the EORTC (#22931) and RTOG (# 9501). Head Neck 2005;
27:843–850.
Volume 81, Number 3, 2011
22. Pignon JP, Bourhis J, Domenge C, et al. Chemotherapy
added to locoregional treatment for head and neck
squamous-cell carcinoma: Three meta-analyses of updated
individual data. MACH-NC Collaborative Group. Meta-
Analysis of Chemotherapy on Head and Neck Cancer.
Lancet 2000;356:767–768.
23. Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradi-
ation with or without concomitant chemotherapy for locally
advanced head and neck cancer. N Engl J Med 2004;350:
1945–1952.
24. Cooper J, Pajak T, Forastiere A, et al. Postoperative concurrent
radiotherapy and chemotherapy for high-risk squamous-cell
carcinoma of the head and neck. N Engl J Med 2004;350:
1937–1944.
25. Duprez F, Madani I, Bonte K, et al. Intensity-modulated radio-
therapy for recurrent and second primary head and neck cancer
in previously irradiated territory. Radiother Oncol 2009;93:
563–569.
26. Dawson LA, Myers LL, Bradford CR, et al. Conformal re-
irradiation of recurrent and new primary head-and-neck cancer.
Int J Radiat Oncol Biol Phys 2001;50:377–385.
27. Goldstein DP, Karnell LH, Yao M, et al. Outcomes following
reirradiation of patients with head and neck cancer. Head
Neck 2008;30:765–770.
28. Popovtzer A, Gluck I, Chepeha DB, et al. The pattern of failure
after reirradiation of recurrent squamous cell head and neck
cancer: Implications for defining the targets. Int J Radiat Oncol
Biol Phys 2009;74:1342–1347.
29. Wagner A, Kermer C, Zettinig G, et al. Validity of sentinel
lymph node (SLN) detection following adjuvant radioche-
motherapy (RCT) in head and neck squamous cell
carcinoma (HNSCC). Technol Cancer Res Treat 2007;6:
655–660.
30. Lou PJ, Jager HR, Jones L, et al. Interstitial photodynamic ther-
apy as salvage treatment for recurrent head and neck cancer. Br
J Cancer 2004;91:441–446.