Radiotherapy and Oncology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Radiotherapy and Oncology

journal homepage: www.thegreenjournal.com

Original article

IAEA-HypoX. A randomized multicenter study of the hypoxic

radiosensitizer nimorazole concomitant with accelerated radiotherapy
in head and neck squamous cell carcinoma
Mohamed A. Hassan Metwally a,⇑, Rubina Ali b, Maire Kuddu c, Tarek Shouman d, Primoz Strojan e,
Kashif Iqbal b, Rajiv Prasad f, Cai Grau g, Jens Overgaard a
a
Department of Experimental Clinical Oncology, Aarhus University Hospital, Denmark; b Oncology Department, Nuclear Medicine, Oncology & Radiotherapy Institute, Islamabad,
Pakistan; c Radiation Oncology Center, North Estonia Medical Center, Tallinn, Estonia; d Radiation Oncology Department, National Cancer Institute, Cairo, Egypt; e Department of
Radiation Oncology, Institute of Oncology, Ljubljana, Slovenia; f Applied Radiation Biology and Radiotherapy Section, International Atomic Energy Agency, Vienna, Austria;
g
Department of Oncology, Aarhus University Hospital, Denmark

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: To test the hypothesis that radiotherapy (RT) of head and neck squamous cell carcinoma
Received 8 January 2015 (HNSCC) can be improved by hypoxic modification using nimorazole (NIM) in association with acceler-
Received in revised form 10 April 2015 ated fractionation.
Accepted 10 April 2015
Materials and methods: The protocol was activated in March 2012 as an international multicenter ran-
Available online xxxx
domized trial in patients with HNSCC. Tumors were treated to a dose of 66–70 Gy, 33–35 fractions, 6 frac-
tions per week. NIM was administered in a dose of 1.2 g per m2, 90 min before the first daily RT fraction.
Keywords:
The primary endpoint was loco-regional failure. The trial was closed prematurely by June 2014 due to
Clinical trials
Accelerated radiotherapy
poor recruitment. An associated quality assurance program was performed to ensure the consistency
Hypoxic radiosensitization of RT with the protocol guidelines.
HNSCC Results: The trial was dimensioned to include 600 patients in 3 years, but only 104 patients were ran-
Hypoxia domized between March 2012 and May 2014 due to the inability to involve three major centers and
Nimorazole the insufficient recruitment rate from the other participating centers. Twenty patients from two centers
had to be excluded from the analysis due to the unavailability of the follow-up data. Among the remain-
ing 84 patients, 82 patients were evaluable (39 and 43 patients in the RT + NIM and the RT-alone arms,
respectively). The treatment compliance was good with only six patients not completing the full planned
RT course, and 31 patients (79%) out of 39 allocated for NIM, achieving at least 90% of the prescribed drug
dose. At the time of evaluation, 40 patients had failed to achieve persistent loco-regional tumor control,
and a total of 45 patients had died. The use of NIM improved the loco-regional tumor control with an
18 month post-randomization cumulative failure rate of 33% versus 51% in the control arm, yielding a risk
difference of 18% (CI 3% to 39%; P = 0.10). The corresponding values for overall death was 43% versus
62%, yielding a risk difference of 19% (CI 3% to 42%; P = 0.10). Sixteen patients, out of 55 patients ana-
lyzed for hypoxic gene expression, were classified as having more hypoxic tumors. Such patients, if trea-
ted with RT alone, had a higher loco-regional tumor failure rate as compared to the rest of the patients
with known hypoxic status (P = 0.05).
Conclusion: Although the trial was incomplete and suffered from a small number of patients, the results
suggested an improvement in loco-regional tumor control and overall survival in patients with advanced
HNSCC given the hypoxic modifier NIM in addition to accelerated fractionation RT. However, the trial also
revealed that conducting multicenter and multinational study combining drug and RT in developing
countries may suffer from uncontrolled and unsolvable problems.
Ó 2015 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology xxx (2015) xxx–xxx

Advanced head and neck squamous cell carcinoma (HNSCC) is a
loco-regional disease that is frequently treated with radiotherapy
⇑ Corresponding author at: Department of Experimental Clinical Oncology,
Aarhus University Hospital, Nørrebrogade 44, Building 5, 8000 Aarhus C, Denmark.
E-mail address: hassan@oncology.dk (M.A. Hassan Metwally).
(RT). Enhancing the effect of radiation treatment, using different
treatment strategies, is an area of extensive research. One of these
strategies considered the reduction of the overall treatment
time of the RT course by a so called "accelerated fractionation"
schedule. The aim of acceleration is to overcome the effect of

http://dx.doi.org/10.1016/j.radonc.2015.04.005
0167-8140/Ó 2015 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Hassan Metwally MA et al. IAEA-HypoX. A randomized multicenter study of the hypoxic radiosensitizer nimorazole con-
comitant with accelerated radiotherapy in head and neck squamous cell carcinoma. Radiother Oncol (2015), http://dx.doi.org/10.1016/
j.radonc.2015.04.005
2 IAEA-HypoX trial of HNSCC
radiation-induced repopulation of tumor cells during treatment
[1]. Accelerated fractionation has demonstrated an improvement
in the loco-regional tumor control and disease specific survival in
patients with HNSCC in some clinical trials, as well as in meta-
analysis, using different strategies to reduce the overall treatment
time [2–4]. One of these trials was the IAEA-ACC trial, where
acceleration of RT was performed by adding an extra fraction in
each week of treatment (going from 5 fractions/week to 6 frac-
tions/week), thus allowing the delivery of the same dose in a
shorter overall treatment time. The accelerated fractionation
schedule in this trial improved the 5-year actuarial rate of loco-
regional tumor control by 12% relative to the conventional
fractionation arm [4].
Another factor that plays an important role in the tumor
response to radiation, particularly in patients with HNSCC, is the
absence of oxygen (hypoxia) in tumor tissue [5,6]. Different meth-
ods have been used to improve the radiation treatment effect
through the modification of hypoxia, resulting in a significantly
better loco-regional tumor control, disease-specific and overall
survival in different randomized trials [7–10] as well as demon-
strated in meta-analyses [6,11,12]. One of the modifications was
the concomitant administration of hypoxic radiosensitizers with
the radiation treatment [11–13]. Among these, nimorazole (NIM)
has demonstrated a significant improvement in loco-regional con-
trol in HNSCC [8], and since it has neither serious nor long-lasting
side effects, it has been adopted for routine clinical use in some
places [8,14,15].
Repopulation and hypoxia are two independent factors, and it is
thus to be expected that combining accelerated fractionation and
hypoxic modification will result in an improvement in the treat-
ment effect without overlapping toxicity. Such a principle has been
adopted by the DAHANCA group [3,14,16], but the true benefit (in
the form of tumor control and morbidity) has not been evaluated
in a controlled clinical trial. Furthermore, this treatment principle
does not require additional RT resources, and is therefore applicable
in a resource limited environment. Therefore, the aim of the present
study was to determine the possible therapeutic gain of using NIM
as a hypoxic radiosensitizer in conjunction with accelerated frac-
tionated RT for invasive HNSCC, and evaluate the tolerance, compli-
ance and toxicity of using the drug, in a randomized trial.
So far, treatment with hypoxic modification have mostly been
performed in Western Europe and North America, and it is unclear
to which extent such treatment principle can be generalized to RT
practice in the developing world, where the therapeutic resources
are less and the patients have often a more heavy tumor burden.
Therefore, the current study was attempted to test the applicability
of such treatment strategy in a global setting. The study is there-
fore in line with the strategy developed by the IAEA aimed at
improving RT of HNSCC through an optimization of the radiobio-
logical properties [4,13,17].
A recent study has identified a gene profile, consisting of 15
genes, that was able to distinguish between more-hypoxic and
less-hypoxic tumors in vivo, based on gene expression [18]. The
gene-profile was also able to predict which patients will benefit
from the hypoxic modification using NIM [19]. Tumors categorized
as more-hypoxic based on this gene profile were associated with a
significantly poorer clinical outcome than less-hypoxic tumors. A
prospective study by EORTC has been activated to investigate the
benefit of adding NIM to accelerated concomitant chemoradiation
in patients with HNSCC [20], where the indication of NIM in the
patients is evaluated based on the status of the hypoxic gene-
profile. The classification of tumors in more or less hypoxic accord-
ing to the 15-gene profile is expected to offer a good prediction for
treatment individualization, where NIM is only given if hypoxic
radiosensitization is warranted, thus avoiding unnecessary side
effects [15].
Please cite this article in press as: Hassan Metwally MA et al. IAEA-HypoX. A randomized multicenter study of the hypoxic radiosensitizer nimorazole con-
comitant with accelerated radiotherapy in head and neck squamous cell carcinoma. Radiother Oncol (2015), http://dx.doi.org/10.1016/
j.radonc.2015.04.005
The present report has been performed according to the
CONSORT guidelines for reporting clinical trials [21].
Patients and methods
The IAEA-HypoX study protocol was activated in March 2012 as
a multicenter randomized double arm trial. The trial was activated
in six centers situated in four different countries.
The study design was a stratified, balanced, and randomized
phase III study in patients with HNSCC. The criteria for eligibility
were invasive squamous cell carcinoma of the larynx (stage I–IV,
except stage I–II glottis cancer), pharynx (stage I–IV, except
nasopharynx), or oral cavity (stage I–IV) according to the TNM clas-
sification of malignant tumors, 7th ed. [22]. Patients had to be
P18 years old, with performance status 0–2 according to WHO cri-
teria, normal liver and kidney functions and without neurological
disorders as assessed by clinical examination. Patients had not to
be planned for surgical excision (except biopsy) before inclusion
in the trial (including elective neck dissection). Patients had not
to be pregnant or with distant metastases at the time of inclusion
in the trial. The trial was designed according to the Helsinki
declaration and was approved by the relevant ethics committees
and health authorities. The trial was registered under the
ClinicalTrials.gov with the following identifier: NCT01507467.
The full trial protocol is provided as a Supplementary material.
Prior to randomization, patients were stratified according to pri-
mary tumor and nodal stage, primary tumor localization (pharynx,
larynx, and oral cavity), performance status, and center. Patients
were randomized centrally to accelerated fractionated RT with or
without NIM.
Patients were treated with external beam RT based on Co-60 or
linear accelerator machines. The treatment principles and dose
specifications were prescribed according to the guidelines given
in the ICRU-50 and ICRU-83 reports. Treatment was planned with
either a conventional 2D technique on a simulator or CT-based 3D
conformal RT (3D-CRT) or Intensity Modulated Radiation Therapy
(IMRT) techniques; however, the chosen technique had to be used
for the entire course of treatment.
The treatment was given in 6 fractions per week, to a centrally
absorbed target dose of 2 Gy per fraction. The macroscopic tumor
was treated to a dose of 66–70 Gy in 33–35 fractions in both arms.
Treatment had to be given with 1 fraction per day through the first
five weekdays; the sixth fraction was given on either a weekend
day or as an extra fraction on one of the first five weekdays, but
always allowing at least 6 h interval between fractions. Patients
treated with IMRT are allowed to receive the sixth fraction as
simultaneously integrated boost (SIB) technique. Concomitant
boost using separate IMRT plans was not allowed. With IMRT
treatment, the primary tumor and involved nodes (PTV1) were
prescribed a total dose 66–70 Gy (2 Gy/fraction), high-risk
sub-clinical disease sites (PTV2) had to receive at least 60 Gy
(1.7–1.8 Gy/fraction), and lower-risk targets (PTV3) were pre-
scribed at least 50 Gy (1.4–1.5 Gy/fraction).
NIM was supplied by Azanta A/S, Denmark, in the form of 500 mg
oral tablets. The drug was administered in doses of approximately
1.2 g per m2 body surface area (BSA), 90 min before each first daily
radiation treatment as follows: 1.5 g (3 tablets) for patients with
BSA 61.6 m2, 2 g (4 tablets) for patients with BSA >1.6 to 61.9 m2,
and 2.5 g (5 tablets) for patients with BSA >1.9 m2. Total dose over
the entire radiation period (33–35 fractions) should be approxi-
mately 36 g/m2 and must not exceed 40 g/m2 or a total of 75 g.
A RT quality assurance (RTQA) program was associated with the
trial from its beginning. The participating centers were asked to
send the details of the RT planning for the first five recruited
patients with the documentation of the received doses according
to the ICRU-50 and ICRU-83 guidelines. The plans, reported doses
 M.A. Hassan Metwally et al. / Radiotherapy and Oncology xxx (2015) xxx–xxx
for different volumes, and the overall treatment times were
independently checked by the quality coordinator for any devia-
tions from the protocol guidelines. A list of the reported deviations
according to predefined criteria was sent to the centers immedi-
ately afterward.
Patients were evaluated weekly during treatment. The common
radiation-related adverse events (dysphagia, mucosal edema,
mucositis, and skin reaction) were graded and documented. NIM
related adverse events (nausea, vomiting, flushing, and skin rash)
were also graded according to the CTCAE v3.0 [23]. Patients were seen
two months after the end of treatment to record persistent acute toxi-
city and early tumor response. Afterward, the patients were seen
every three months. The study design, stratification, randomization,
treatment and follow up, are detailed in the trial protocol.
The trial was designed to include 600 patients. This number was
expected to be recruited over a 3-year period. Assuming a true
improvement of the loco-regional tumor control rate of 12% in the
RT + NIM arm; then the probability that such an event would be
detected at a significant level of P = 0.05 was greater than 90%. The
primary endpoint was time to loco-regional tumor failure (loco-
regional tumor control). The definition of this end-point was com-
plete and persistent disappearance of the disease in the primary site
(T-site) and regional lymph nodes (N-site) after RT. Failure was
recorded in the event of a recurrent tumor, or if the primary tumor
and/or nodal disease never completely disappeared. In the latter sit-
uation the tumor was then assumed to have failed at the time of the
end of RT course. All time estimates were done using the date of ran-
domization as the initial value. The primary end-point does not
include the effect of a successful procedure with salvage surgery.
The Kaplan–Meier method was used to estimate time-to-event
curves. Patterns of first failure were analyzed using a "competing-
risks" analysis for the loco-regional failure. The treatment groups
were compared with respect to the time to loco-regional failure,
and overall survival using the Cox proportional hazards model.
The trial was closed prematurely after 27 months by June 2014
due to the poor recruitment rate. The total recruitment was 104
patients compared to an expected recruitment of more than 400
patients at the same time point. Only 6 out of 9 centers have suc-
ceeded to obtain the required approvals for conducting the trial in
their countries. The present analysis was performed for the random-
ized patients on an intention to treat basis, and patients were
included in their randomization group irrespective of whether or
not they had completed the planned treatment. Frequency tables
with counts and percentages were used to describe the patients and
their distribution in the two treatment arms. Patients were followed
up in all centers until the time of evaluation on March 30, 2015.
An associated protocol for translational research was included
in this trial. The aim was to identify biomarkers that can predict
which patients will benefit from the use of NIM. The biological
analysis was performed retrospectively. Biological materials were
collected from the participating centers as formalin fixed and
paraffin embedded samples. Hypoxia status was determined
through gene expression analysis of hypoxia-induced 15-gene pro-
file on RNA extracted from the samples. Gene expression was
determined by quantitative PCR analysis. Immunohistochemical
staining for the HPV-associated p16 expression was performed
on 49 tumor samples and tumors were classified as being HPV
positive or negative according to a standardized DAHANCA proto-
col [24]. The results of the HPV status determination for further
four patients from one of the centers were included in the analysis.
Results
From March 2012 to May 2014, 104 patients were included.
However, all included patients from two centers were excluded
from the final analysis due to major deficiency of data and the lack
Please cite this article in press as: Hassan Metwally MA et al. IAEA-HypoX. A randomized multicenter study of the hypoxic radiosensitizer nimorazole con-
comitant with accelerated radiotherapy in head and neck squamous cell carcinoma. Radiother Oncol (2015), http://dx.doi.org/10.1016/
j.radonc.2015.04.005
3
of communication from the local investigators. These centers have
randomized a total of 20 patients (9 patients were in the RT-alone
arm, and 11 patients were in the RT + NIM arm). Among the
remaining 84 eligible patients, two patients (one in each treatment
arm) did not start the allocated treatment after randomization.
One patient was lost to follow up immediately after randomization
due to unknown reasons and the other patient was diagnosed with
claustrophobia during preparation procedures for RT. The remain-
ing 82 patients were evaluable (39 and 43 patients in the RT + NIM
and the RT-alone arms, respectively). The trial flow chart is shown
in (Fig. 1).
There were 30 females and 52 males with a median age at ran-
domization of 56 years (range 28–75 years). Most of patients (78%)
presented with well/moderate differentiated tumors, 9% had poor/
undifferentiated tumors, and 13% had unknown differentiation
with equal distribution between both treatment arms. Only 5
patients, out of 53 patients with known HPV/p16 status, were
HPV/p16 positive (3 and 2 patients in RT-alone and RT + NIM arms,
respectively). The patient- and treatment-related characteristics in
both randomization arms are shown in (Table 1). The majority of
patients have received radiation treatment using the 2D planning
technique (56 patients). Most of patients received 66 Gy in 33 frac-
tions to the gross tumor volume, and the vast majority completed
the full prescribed radiation dose (66–70 Gy). Only 6 patients could
not complete the full RT course: four were due to severe treat-
ment-related toxicity; one was due to non-compliance, and one
died during treatment due to complications of jejunostomy.
As a whole, 31 (79%) patients received at least 90% of the pre-
scribed NIM dose and 19 (49%) patients completed the full pre-
scribed dose (Table 1). The cause of failure to fulfill the planned
treatment was mainly due to the treatment-related acute toxicity,
with the majority of patients complaining of mild to moderate
nausea and vomiting. A few patients suffered from flushing and
skin rash (Table 2).
At the time of evaluation, 40 patients had failed to achieve per-
sistent loco-regional tumor control, and a total of 45 patients had
died. The median follow-up time of patients who were still alive
at the time of evaluation was 19 months (range 1–34 months),
and the median follow-up time of patients who died was 7 months
(range 1–29 months). From a total of 40 patients with loco-regional
Fig. 1. Trial profile and outcome.
4 IAEA-HypoX trial of HNSCC

Table 1 Table 2
Patient and treatment characteristics as a function of the randomization arm.§ Treatment related morbidity in 82 treated patients as a function of the randomization
arm.§
RT-alone (43) RT + NIM (39) All patients (82)
Age (years) RT-alone (43) RT + NIM (39)
Median (range) 56 (31–75) 54 (28–72) 56 (28–75) RT related morbidity
Gender Dysphagia (Gr 1–2) 25 (58%) 20 (51%)
Female 15 (35%) 15 (38%) 30 (37%) (Gr 3–4) 18 (42%) 18 (46%)
Male 28 (65%) 24 (62%) 52 (63%) Mucosal edema (Gr 1–2) 37 (86%) 30 (77%)
Performance status (Gr 3) 4 (9%) 6 (15%)
0–1 40 (93%) 37 (95%) 77 (94%) Mucositis (Gr 1–2) 23 (53%) 19 (49%)
2–3 3 (7%) 2 (5%) 5 (6%) (Gr 3–4) 20 (47%) 20 (51%)
Tumor site Skin reaction (Gr 1–2) 22 (51%) 22 (56%)
Larynx 9 (21%) 9 (23%) 18 (22%) (Gr 3–4) 21 (49%) 16 (41%)
Pharynx 20 (46%) 17 (44%) 37 (45%)
Interventions for morbidity
Oral cavity 14 (33%) 13 (33%) 27 (33%)
Tube feeding 7 (16%) 9 (23%)
T-classification
Tracheostomy 7 (16%) 7 (18%)
T1–2 14 (33%) 17 (44%) 31 (38%)
T3–4 29 (67%) 22 (56%) 51 (62%) Nimorazole-related morbidity*
N-classification Nausea (%) (Gr 1–2) 26 (67%)
N0 13 (30%) 11 (28%) 24 (29%) (Gr 3–4) 2 (5%)
N1–3 30 (70%) 28 (72%) 58 (71%) Vomiting (%) (Gr 1–2) 17 (44%)
Stage (Gr 3–4) 0
I–II 7 (16%) 4 (10%) 11 (13%) Flushing (%) (Gr 1) 5 (13%)
III–IV 36 (84%) 35 (90%) 71 (87%) (Gr 2) 6 (15%)
HPV/p16 status Rash (%) (Gr 1) 1 (3%)
HPV/p16 +ve 3 (7%) 2 (5%) 5 (6%) (Gr 2–3) 2 (5%)
HPV/p16 ve 24 (56%) 24 (62%) 48 (59%)
*
Unknown 16 (37%) 13 (33%) 29 (35%) Incidence is expressed as an approximate percentage in the 39 patients treated
Hypoxic status with nimorazole, and side effects were graded according to the CTCAE v3.0.
§
More hypoxic 8 (19%) 8 (20%) 16 (19%) No statistically significant differences in the distribution of characteristics
Less hypoxic 18 (42%) 21 (54%) 39 (48%) between the two arms.
Unknown 17 (39%) 10 (26%) 27 (33%)
RT technique
2D 30 (70%) 26 (67%) 56 (68%) tumors treated with RT + NIM whether less or more hypoxic) were
3DCRT 1 (2%) 0 1 (1%) ranging between 33% and 44%. The distribution of the hypoxic sta-
IMRT 12 (28%) 13 (33%) 25 (31%)
tus between the two treatment arms is shown in (Table 1).
RT Dose
<66 Gy 1 5 6 Among the first five recruited patients from each center, the
66 to <68 Gy (OTT)* 26 (37 days) 19 (37 days) 45 (37 days) quality assurance check did not find any major deviations from
68 to <70 Gy (OTT) 5 (37 days) 5 (38 days) 10 (37 days) the RTQA protocol as regards treatment planning. All deviations
P70 Gy (OTT) 11 (40 days) 10 (42 days) 21(40 days)
were found in the overall treatment time and were mainly due
Nimorazole Dose
P90% of dose received 31 (79%)
to patients’ reasons or machine breakdown. A report with the
100% of dose received 19 (49%) observed major and minor deviations, following the quality assur-
*
ance check, was sent to each investigator for future considerations.
OTT, Overall Treatment Time (expressed as median value).
§
No statistically significant differences in the distribution of characteristics
between the two arms. Discussion

The current IAEA-HypoX study has utilized the experience from

failure, 18 patients had failure in the T-site, 8 patients had failure in
the N-site, and 14 patients had failure in both the T- and N-sites.
Two patients had distant metastasis as the first failure event.
Salvage surgery was successfully performed in 6 patients with
loco-regional failure (4 and 2 patients in RT-alone and RT + NIM
arms, respectively) (Fig. 1).
The use of NIM improved the loco-regional tumor control with
an 18 month post-randomization cumulative failure rate of 33%
versus 51% in the control group, yielding a risk difference of 18%
(CI 3% to 39%; P = 0.10). The corresponding values for overall
death were 43% versus 62%, yielding a risk difference of 19% (CI
3% to 42%; P = 0.10) (Fig. 2). The hazard of the loco-regional fail-
ure in the RT + NIM arm was 0.72 (CI 0.38–1.38; P = 0.31) com-
pared to the control arm, with corresponding hazard of death
being 0.65 (CI 0.35–1.19; P = 0.16).
Histopathological tissue materials from 55 tumors were ana-
lyzed for gene expression. According to the hypoxic 15-gene pro-
file, 16 patients were classified as having more hypoxic tumors.
Fig. 3 shows the distribution of patients as a function of the loco-
regional failure rate, treatment arm and the hypoxic status. The
highest loco-regional failure rate (75%) was in the group of patients
with more hypoxic tumors who were treated with accelerated RT
alone (P = 0.05). While the loco-regional failure rates in the other
3 groups (i.e. less hypoxic tumors treated with RT alone, and
Please cite this article in press as: Hassan Metwally MA et al. IAEA-HypoX. A randomized multicenter study of the hypoxic radiosensitizer nimorazole con-
comitant with accelerated radiotherapy in head and neck squamous cell carcinoma. Radiother Oncol (2015), http://dx.doi.org/10.1016/
j.radonc.2015.04.005
the previous clinical trials, that investigated the effect of reducing
the overall treatment time [4] to overcome the problem of tumor
repopulation during RT, and the use of radiosensitizers in conjunc-
tion with radiation to overcome the problem of hypoxia [8]. The
use of hypoxic modification concomitantly with the accelerated
fractionation schedule was expected to offer a reasonable improve-
ment in the treatment that does not require additional RT
resources, and is therefore applicable in a resource limited
environment. The initial steps of the RTQA program showed a clear
understanding of the protocol guidelines by the centers with full
compliance with its specifications. Together with the results of
RTQA program, the good compliance with the accelerated frac-
tionation schedule and the NIM treatment proved the applicability
of such treatment strategy in the countries where the trial was
conducted. The inability to recruit the pre-defined number of
patients was due to logistic, legislative, and political reasons in
both the active participating countries and the countries that could
not participate.
The treatment was well tolerated and the majority of patients
completed the accelerated RT course and the prescribed dose of
NIM. No important or long-lasting toxicity has been noted with
the use of NIM and the toxicity was in agreement with the previous
phase I and II studies [25,26], and the compliance was in agree-
ment with that documented from other studies of the drug
 M.A. Hassan Metwally et al. / Radiotherapy and Oncology xxx (2015) xxx–xxx 5

A B

100

100
75
Cumm. risk difference at 18 months:

75
18% (-3% to 39%) P=0.10 RT+NIM

Loco−regional failure

Overall survival (%)

56% 51%
RT−alone

50

50
RT−alone
52%

RT+NIM
25

25
29%

Events All Events All

RT+NIM 17 39 RT+NIM 18 39
RT−alone 23 43 RT−alone 27 43 p=0.16
0

0
0 6 12 18 24 0 6 12 18 24

At risk At risk
RT+NIM 39 (12) 20 (1) 18 (0) 13 (4) 6 RT+NIM 39 (8) 30 (6) 22 (2) 15 (1) 9
RT−alone 43 (20) 19 (2) 11 (0) 8 (1) 5 RT−alone 43 (9) 34 (11) 18 (5) 10 (2) 5

Fig. 2. (A) Cumulative incidence curve comparing the two arms regarding cumulative loco-regional failure rates. (B) Kaplan–Meier curves comparing survival in both
treatment arms.

100 The proportion of HPV-related tumors in the present cohort was

found low; however, due to a traditionally high rate of tobacco use
in the countries where the study was conducted, this observation
80 75 % meets the expectations [28,29]. Together with uncertain predictive
More hypoxic
value of hypoxia in HPV/p16 positive patients [30], HPV positivity
Loco-regional failures (%)

Less hypoxic had a marginal role in the present trial.

60
44 %
38 %
40
20
0 loco-regional tumor control by about 10% over the conventional
RT-alone
Fig. 3. Frequency of loco-regional failure in 55 patients with known hypoxic status
(according to the 15-gene hypoxia profile) as a function of treatment arm (higher
loco-regional failure rates in patients with more hypoxic tumors treated with RT-
alone as compared to the remaining patients with known hypoxic status, P = 0.05).
[8,14,15,27]. The compliance of the 39 patients who were treated
with NIM was in agreement with the compliance seen in a larger
cohort of HNSCC patients (653 patients) who were treated with
accelerated RT + NIM in a previous study (Supplementary Fig. 1)
[15]. The vast majority of patients have complained of mild to
moderate nausea and vomiting that did not interfere with comple-
tion of the full course of treatment, and 82% of the patients
received 75% of the prescribed drug dose.
The analysis of the loco-regional failure rate in relation to the
treatment arm and the hypoxic status according to the 15-gene
profile (Fig. 3) showed that patients treated with NIM had a similar
rate of loco-regional failure as those who did not receive NIM, if
they were having less-hypoxic tumors, while the loco-regional fail-
ure rate differs obviously in patients having more-hypoxic tumors.
This confirms the findings reported by Toustrup et al. [19] who
tested the predictive value of the 15-gene hypoxia profile in
allocating patients with HNSCC for treatment with NIM. It also
underlines the relevance of giving NIM with RT in treating
HNSCC patients having more-hypoxic tumors.
Please cite this article in press as: Hassan Metwally MA et al. IAEA-HypoX. A randomized multicenter study of the hypoxic radiosensitizer nimorazole con-
comitant with accelerated radiotherapy in head and neck squamous cell carcinoma. Radiother Oncol (2015), http://dx.doi.org/10.1016/
j.radonc.2015.04.005
The trial did not compare the investigational arm (RT + NIM)
with the current practice of giving chemotherapy concomitantly
with radiation treatment to patients with advanced HNSCC [31–
33], although this may have both a biological [34] and a clinical
33 % rationale [35]. The main reason behind the trial design was the
assumption made by the investigators that the poor tolerability
of HNSCC patients in their countries to chemotherapy will result
in poor compliance with treatment, which in turn will mask the
potential benefit of such combined approach. Furthermore, the
accelerated fractionation schedule was expected to improve the
RT+NIM fractionation schedule that is commonly used with the concomi-
tant chemotherapy treatment [4]. Thus, the currently investigated
strategy was expected to be more tolerable by the patients in the
participating countries than the concomitant chemoradiation regi-
men, with less exhaustion of resources and similar added clinical
benefit. The trial, also, tested the applicability and the tolerability
of the investigated treatment strategy in a global setting. The same
concept is used in the NIMRAD randomized phase III study recently
activated in UK [36]. The study investigates the benefit of adding
NIM to RT alone in patients with advanced HNSCC not suitable
for synchronous chemotherapy or cetuximab.
The observed loco-regional tumor control rate in the acceler-
ated RT-alone arm in the current study (around 50% at 18 months)
is coinciding with the results from the same arm in the IAEA-ACC
trial, that investigated the benefit of accelerated fractionation over
the conventional fractionation in a similar patient population [4].
This confirms the level of loco-regional tumor control in this study,
and indicates the potential benefit of adding NIM to accelerated RT
in such patient population, which is not proven due to the incom-
plete nature of the trial.
Although the current study shows a statistically non-significant
improvement in the tumor control in patients treated with NIM
concomitantly with accelerated RT, it should be emphasized that
the trial was closed early with an insufficient number of patients
and relatively short observation time. Neither the small number
6 IAEA-HypoX trial of HNSCC
of patients allowed more elaborate analyses of this study.
Furthermore, a longer observation time and more patients would
be needed to reach a final conclusion regarding the magnitude of
the effect of adding NIM to the radiation treatment. However, the
likely importance of hypoxic modification seems apparent and fur-
ther attempts toward overcoming this problem appear to be
relevant.
Authors’ disclosures of potential conflicts of interest
None to declare.
Acknowledgments
This study was supported by funding from IAEA – the
International Atomic Energy Agency; the Faculty of Health
Sciences, Aarhus University; CIRRO – The Lundbeck Foundation
Centre for Interventional Research in Radiation Oncology; the
Danish Council for Strategic Research, and DAHANCA – Danish
Head and Neck Cancer Society, Denmark.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.radonc.2015.04.
005.
References
[1] Withers HR, Taylor JM, Maciejewski B. The hazard of accelerated tumor
clonogen repopulation during radiotherapy. Acta Oncol 1988;27:131–46.
[2] Bourhis J, Overgaard J, Audry H, et al. Hyperfractionated or accelerated
radiotherapy in head and neck cancer: a meta-analysis. Lancet
2006;368:843–54.
[3] 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;362:933–40.
[4] Overgaard J, Mohanti BK, Begum N, et al. Five versus six fractions of
radiotherapy per week for squamous-cell carcinoma of the head and neck
(IAEA-ACC study): a randomised, multicentre trial. Lancet Oncol
2010;11:553–60.
[5] Overgaard J, Horsman MR. Modification of hypoxia-induced radioresistance in
tumors by the use of oxygen and sensitizers. Semin Radiat Oncol
1996;6:10–21.
[6] Overgaard J. Hypoxic radiosensitization: adored and ignored. J Clin Oncol
2007;25:4066–74.
[7] Overgaard J, Hansen HS, Andersen AP, et al. Misonidazole combined with split-
course radiotherapy in the treatment of invasive carcinoma of larynx and
pharynx: report from the DAHANCA 2 study. Int J Radiat Oncol Biol Phys
1989;16:1065–8.
[8] Overgaard J, Hansen HS, Overgaard M, et al. A randomized double-blind phase
III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in
supraglottic larynx and pharynx carcinoma. Results of the Danish Head and
Neck Cancer Study (DAHANCA) Protocol 5–85. Radiother Oncol
1998;46:135–46.
[9] Saunders M, Dische S. Clinical results of hypoxic cell radiosensitisation from
hyperbaric oxygen to accelerated radiotherapy, carbogen and nicotinamide. Br
J Cancer Suppl 1996;27:S271–8.
[10] Kaanders JH, Bussink J, van der Kogel AJ. Clinical studies of hypoxia
modification in radiotherapy. Semin Radiat Oncol 2004;14:233–40.
[11] Overgaard J. Clinical evaluation of nitroimidazoles as modifiers of hypoxia in
solid tumors. Oncol Res 1994;6:509–18.
[12] Overgaard J. Hypoxic modification of radiotherapy in squamous cell carcinoma
of the head and neck–a systematic review and meta-analysis. Radiother Oncol
2011;100:22–32.
Please cite this article in press as: Hassan Metwally MA et al. IAEA-HypoX. A randomized multicenter study of the hypoxic radiosensitizer nimorazole con-
comitant with accelerated radiotherapy in head and neck squamous cell carcinoma. Radiother Oncol (2015), http://dx.doi.org/10.1016/
j.radonc.2015.04.005
[13] Wardman P. Chemical radiosensitizers for use in radiotherapy. Clin Oncol (R
Coll Radiol) 2007;19:397–417.
[14] Overgaard J, Sand HH, Lindelov B, et al. Nimorazole as a hypoxic
radiosensitizer in the treatment of supraglottic larynx and pharynx
carcinoma. First report from the Danish Head and Neck Cancer Study
(DAHANCA) protocol 5–85. Radiother Oncol 1991;20:143–9.
[15] Metwally MA, Frederiksen KD, Overgaard J. Compliance and toxicity of the
hypoxic radiosensitizer nimorazole in the treatment of patients with head and
neck squamous cell carcinoma (HNSCC). Acta Oncol 2014;53:654–61.
[16] DAHANCA. DAHANCA. 2-4-2015. <http://www.dahanca.dk/>.
[17] Grau C, Prakash AJ, Jabeen K, et al. Radiotherapy with or without mitomycin c
in the treatment of locally advanced head and neck cancer: results of the IAEA
multicentre randomised trial. Radiother Oncol 2003;67:17–26.
[18] Toustrup K, Sorensen BS, Nordsmark M, et al. Development of a hypoxia gene
expression classifier with predictive impact for hypoxic modification of
radiotherapy in head and neck cancer. Cancer Res 2011;71:5923–31.
[19] Toustrup K, Sorensen BS, Lassen P, Wiuf C, Alsner J, Overgaard J. Gene
expression classifier predicts for hypoxic modification of radiotherapy with
nimorazole in squamous cell carcinomas of the head and neck. Radiother
Oncol 2012;102:122–9.
[20] Intergroup Study (EORTC-1219-ROG-HNCG – DAHANCA-29):. A blind
randomized multicenter study of accelerated fractionated chemo-
radiotherapy with or without the hypoxic cell radiosensitizer Nimorazole
(Nimoral), using a 15-gene Signature for Hypoxia in the Treatment of HPV/p16
Negative Squamous Cell Carcinoma of the Head and Neck. Head Neck 2014.
<https://clinicaltrials.gov/ct/show/NCT01880359>.
[21] Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated
guidelines for reporting parallel group randomized trials. Ann Intern Med
2010;152:726–32.
[22] Sobin Leslie H, Gospodarowicz Mary K, Wittekind Christian. TNM classification
of malignant tumours. Wiley-Blackwell; 2009.
[23] Common Terminology Criteria for Adverse Events (CTCAE) v3.0. NCI. 3-10-
2014. <http://ctep.cancer.gov/protocolDevelopment/electronic_applications>.
[24] Lassen P, Overgaard J. Scoring and classification of oropharyngeal carcinoma
based on HPV-related p16-expression. Radiother Oncol 2012;105:269–70.
[25] Overgaard J, Overgaard M, Timothy AR. Studies of the pharmacokinetic
properties of nimorazole. Br J Cancer 1983;48:27–34.
[26] Timothy AR, Overgaard J, Overgaard M. A phase I clinical study of Nimorazole
as a hypoxic radiosensitizer. Int J Radiat Oncol Biol Phys 1984;10:1765–8.
[27] Hassan Metwally MA, Jansen JA, Overgaard J. Study of the population
pharmacokinetic characteristics of Nimorazole in head and neck cancer
patients treated in the DAHANCA-5 Trial. Clin Oncol (R Coll Radiol)
2015;27:168–75.
[28] de MC, Ferlay J, Franceschi S. Global burden of cancers attributable to
infections in 2008: a review and synthetic analysis. Lancet Oncol
2012;13:607–15.
[29] Strojan P, Zadnik V, Sifrer R. Incidence trends in head and neck squamous cell
carcinoma in Slovenia, 1983–2009: role of human papillomavirus infection.
Eur Arch Otorhinolaryngol 2014.
[30] Lassen P, Eriksen JG, Hamilton-Dutoit S, Tramm T, Alsner J, Overgaard J. HPV-
associated p16-expression and response to hypoxic modification of
radiotherapy in head and neck cancer. Radiother Oncol 2010;94:30–5.
[31] Pignon JP, Le MA, Maillard E, Bourhis J. Meta-analysis of chemotherapy in head
and neck cancer (MACH-NC): an update on 93 randomised trials and 17,346
patients. Radiother Oncol 2009;92:4–14.
[32] Marcu L, van DT, Olver I. Cisplatin and radiotherapy in the treatment of locally
advanced head and neck cancer–a review of their cooperation. Acta Oncol
2003;42:315–25.
[33] Homma A, Inamura N, Oridate N, et al. Concomitant weekly cisplatin and
radiotherapy for head and neck cancer. Jpn J Clin Oncol 2011;41:980–6.
[34] Grau C, Overgaard J. Effect of cancer chemotherapy on the hypoxic fraction of a
solid tumor measured using a local tumor control assay. Radiother Oncol
1988;13:301–9.
[35] Bentzen J, Toustrup K, Eriksen JG, Primdahl H, Andersen LJ, Overgaard J. Locally
advanced head and neck cancer treated with accelerated radiotherapy, the
hypoxic modifier nimorazole and weekly cisplatin. Results from the DAHANCA
18 phase II study. Acta Oncol 2015:1–7.
[36] Thomson D, Yang H, Baines H, et al. NIMRAD – a phase III trial to investigate
the use of nimorazole hypoxia modification with intensity-modulated
radiotherapy in head and neck cancer. Clin Oncol (R Coll Radiol)
2014;26:344–7.