Clinical Therapeutics/Volume 37, Number 3, 2015

Efcacy and Tolerability of Intranasal Fentanyl Spray in

Cancer Patients With Breakthrough Pain
Morten Throns, MD1,2; Lars Popper, MD, PhD3; Martin Eeg, MSc4;
Ellen Jaatun, MD1,5; Magnar Kvitberg, MD6; and Stein Kaasa, MD, PhD1,2
1

Norwegian University of Science and Technology, Faculty of Medicine, Department of Cancer Research
and Molecular Medicine, European Palliative Care Research Centre, Trondheim, Norway; 2St Olavs
Hospital, Trondheim University Hospital, Cancer Clinic, Trondheim, Norway; 3Takeda Pharma A/S,
Clinical Science, Roskilde, Denmark; 4Takeda Pharma A/S, Analytical Science, Roskilde, Denmark;
5
St Olavs Hospital, Trondheim University Hospital, Department of Oto-Rhino-Laryngology, Trondheim,
Norway; and 6Drammen Hospital, Department of Oncology, Palliative Unit, Drammen, Norway
ABSTRACT
Purpose: The aims of this study were to explore the
efcacy of intranasal fentanyl spray* (INFS) 400 g
to evaluate 12-week tolerability of the nasal mucosa
and to explore safety data for all dose strengths of
INFS in patients with cancer-related breakthrough
pain (BTP).
Methods: Patients received a test dose of INFS 50
g, followed by a titration phase. Those patients with
doses titrated to 200 or 400 g entered a randomized,
double-blind, cross-over efcacy phase, in which 8
episodes of BTP were randomly treated with INFS 400 g
(6 episodes) and placebo (2 episodes), followed by a
tolerability phase. Patients with doses titrated to 50 or
100 g entered the tolerability phase directly. Primary
outcome was measured by pain intensity difference at
10 minutes, analyzed using ANCOVA, and presented
as least square mean difference. Examination of the
nasal cavity was conducted at inclusion and after 12
weeks of treatment by an otorhinolaryngologist.
Findings: Forty-six patients were included. Thirtyeight patients doses were titrated to an effective dose of
INFS; 50 g (n 8), 100 g (n 9), 200 g (n 9), and
400 g (n 12); 15 patients entered the efcacy phase
and 31 entered the tolerability phase. In the efcacy
phase, 88 and 29 episodes of BTP were treated with INFS
400 g and placebo, respectively. Pain intensity difference
at 10 minutes least square mean for INFS 400 g was 2.5
(95% CI, 1.423.49) (P o 0.001) and least square
mean difference between INFS 400 g and placebo was
*

Trademark: Instanyls (Takeda Pharmaceutical Company

Limited, Zurich, Switzerland).

1.1 (95% CI, 0.411.79) (P 0.002). Runny nose

(10%) and change in color of the mucosa (9%) were the
most frequent ndings of nasal examination, and nausea
and dizziness were the most frequent treatment-related
adverse events. One serious adverse event (ie, respiratory
depression) was considered related to INFS.
Implications: INFS 400 g is effective and nasal
tolerability and overall safety prole is acceptable
during 12 weeks of use. ClinicalTrials.gov identier:
NCT01429051. (Clin Ther. 2015;37:585596) & 2015
Elsevier HS Journals, Inc. All rights reserved.
Key words: breakthrough pain, cross-over efcacy
phase, Instanyl, intranasal fentanyl spray, pain intensity difference.

INTRODUCTION
More than 50% of cancer patients suffer from cancerrelated pain,1 despite improved guidelines and
treatment alternatives.2
The term breakthrough pain (BTP) was introduced by
Portenoy and Hagen in 1989.3 Prevalence varies from
40% to 80%, according to setting and denition,4 and
characteristics are typically described as a median of 3
BTP episodes per day, time to peak intensity of 5 to 10
minutes, and duration of untreated episodes of 45 to 60
minutes. The majority of patients report their BTP as
severe.5
Accepted for publication December 12, 2014.
http://dx.doi.org/10.1016/j.clinthera.2014.12.010
0149-2918/$ - see front matter
& 2015 Elsevier HS Journals, Inc. All rights reserved.

March 2015

585
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

Clinical Therapeutics
Immediate-released (IR) morphine taken orally on
demand in addition to slow-released morphine is
commonly used and recommended in the treatment
of BTP.2 This is despite IR morphines delayed onset
of analgesia (3060 minutes).6,7 Rapid-acting fentanyl preparations have been introduced as an alternative approach to IR morphine for the treatment of
BTP.8 Rapid-acting fentanyl exists in a variety of
administration forms.9 Most studies of these
products are placebo controlled and show favorable
analgesic effect in the treatment of BTP compared
with placebo. Analgesic effect is typically achieved
within 10 to 30 minutes and adverse effects are similar
to other opioids.9
Pharmacokinetic studies of nasal administration of
opioids show rapid uptake and action.10,11 Two drugs
for treating BTP are available for nasal administration; intranasal fentanyl spray* (INFS) and fentanyl
pectin nasal spray. INFS is available as 50, 100, and
200 g/dose, and pharmacokinetic studies have found
a median Tmax value between 12 and 15 minutes12
and a bioavailability of 89%.13 The efcacy of INFS
has been reported in several studies, showing rapid
time to analgesic effect,14 superior pain intensity
difference at 10 minutes (PID10) compared with
placebo,15 and that onset of meaningful pain relief is
experienced earlier compared with oral transmucosal
fentanyl citrate.16 Indirectly compared with other fastacting fentanyl components, INFS has provided the
greatest PID compared with placebo at 15 minutes
and 30 minutes post baseline.9 Adverse effects of INFS
are typically opioid related, such as nausea, vertigo,
dizziness, and myoclonus,15,16 but nasal ulcers have
been observed,16 and a systematic evaluation of the
nasal cavity in long-term use is required.
Optimal treatment of BTP is an unmet need among
cancer patients. Clinical experience has found that 10%
to 20% of patients are in need of a higher dose of INFS
than that approved and on the market today. INFS 400
g has therefore been developed. Pharmacokinetic studies
in healthy subjects have indicated dose linear increase in
the exposure in the dose range of 200 g to 2  400 g.17
In the present trial, the efcacy of INFS 400 g is
compared with placebo, and the nasal tolerability and
general adverse effects during 12 weeks of follow-up
for all dose strengths of INFS are evaluated.

Lazandas (Archimedes Development Limited, Bedminster,

New Jersey).

586

METHODS
The present placebo-controlled, double-blind, randomized, cross-over efcacy study was conducted
from August 2011 until January 2013 in 11 European
centers (Norway, 2 sites; Hungary, 7 sites; and
Russia, 2 sites). After a test dose of INFS 50 g, the
patients doses were titrated to an effective dose of
INFS for their BTP. Patients with doses titrated to 50
g and 100 g proceeded directly to a tolerability
phase lasting for 3 months. Patients with doses
titrated to 200 or 400 g INFS were included in the
efcacy phase, in which 8 episodes of BTP were
treated with INFS 400 g (6 episodes) and placebo
(2 episodes). Having completed the efcacy phase,
these patients continued into the tolerability phase.
An examination of the nasal cavity was conducted by
an otorhinolaryngologist at baseline and after 12
weeks of INFS treatment, and patient-assessed tolerability was evaluated during all phases of the study
(Figure 1).

Patients
Cancer patients aged Z18 years with BTP episodes
between 3 times per week and 4 times per day, and life
expectancy of more than 3 months were eligible for this
study. Both inpatients and outpatients were included.
Use of oral opioids or transdermal fentanyl (morphine
equivalent doses of 601000 mg/24 h) for treatment of
stable and controlled background pain (BGP), dened
as a mean r4 on an 11-point Numeric Rating Scale
(NRS), were required. The exclusion criteria were
history of abuse, severe hepatic impairment (alanine
aminotransferase or aspartate aminotransferase 43 
upper normal range), severe renal impairment (serum
creatinine Z3.0 mg/dL [265 mol/L]), and severe
impairment of respiratory function, which might increase the risk of relevant respiratory depression. In
addition, patients having concomitant conditions resulting in runny nose (eg, rhinitis), patients who had
been at some point treated with facial radiotherapy,
and patients treated with nasal surgery within the last
30 days before screening were excluded. Patients with
head injury, primary brain tumor, or other pathologic
conditions that could signicantly increase the risk of
elevated intracranial pressure or impaired consciousness were also excluded. Any kind of drugs for intranasal administration or the use of a nasopharyngeal
probe were not allowed, and patients having recurrent
episodes of epistaxis were excluded. Intake of

Volume 37 Number 3
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

M. Throns et al.

Screening
BGP 4
BTP episodes
< 3/week
4/day
Nose examination

Titration
test dose 50 g

Tolerability phase

Titrated to 50 or
100 g

End of treatment
visit
Nose examination

Titrated to 200
or 400 g

Efficacy phase
Randomly placed with a doses
of INPS 400 g and 2 doses
placebo

710 days

12 weeks

Figure 1. Flowchart study schedule. BGP background pain; BTP breakthrough pain.

monoamine oxidase inhibitors was contraindicated due

to reports of severe and unpredictable potentiation by
monoamine oxidase inhibitors. Finally, patients known
to be hypersensitive to fentanyl or other opioids and
patients participating in another interventional trial
were not included.

Assessments
The participants recorded information about pain,
adverse effects, and general impression of the efcacy
of the treatment in a diary every day during the trial.
Pain intensity (PI) was assessed by the patient on
an 11-point NRS in the efcacy phase. Patients
assessed current PI at baseline (0 minutes) and at 5,
10, 30, and 60 minutes after intake of INFS or
placebo (0 no pain and 10 worst possible pain).
General impression (GI) of the overall efcacy of the
treatment was assessed by the patients 60 minutes
after intake of INFS (INFS or placebo in the efcacy
phase) in all phases of the study. A 5-point Verbal
Rating Scale (0 poor, 1 fair, 2 good, 3 very
good, and 4 excellent) was applied for this
assessment.
A successfully treated episode of BTP was dened as
no need for rescue analgesia within 60 minutes after
rst intake of INFS or placebo, and score of Z2 for
GI. In addition to this, no occurrence of severe adverse
drug reactions, such as pronounced hypoventilation,

severe respiratory depression, unacceptable sedation or

drowsiness, any local tolerance problems, or any
intolerable side effects were accepted.
Local nasal tolerability was assessed by an otorhinolaryngologist at baseline and at the end of the study
(3 days from the end of treatment [EOT]). Clinical
signs like epistaxis and runny nose were scored on a
4-point rating scale (0 not present, 1 present in a
mild degree, 2 present in a moderate degree, and
3 present in a severe degree.) At the EOT visit, the
otorhinolaryngologist rated the ndings of the examination as no change, improvement from baseline, or
worsening compared with baseline. In addition,
changes were rated as related or not related to the
use of INFS. Side effects were additionally recorded by
the patient in a diary.

Medication
For the BTP episodes, INFS was used. This nasal
spray contains a phosphate-buffered solution of fentanyl
citrate and was available in the following strengths:
0.5 mg/mL, 1 mg/mL, 2 mg/mL, and 4 mg/mL in
multiple-dose sprays. The corresponding doses were
50, 100, 200, and 400 g INFS/puff. One puff
denes and equals one dose and was applied in
one nostril.
If the patient experienced insufcient pain relief
with one dose of INFS, an additional dose was allowed

March 2015

587
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

Clinical Therapeutics
for the same BTP episode at a minimum of 10 minutes
after the rst administration. A maximum of 4 BTP
episodes was allowed to treat with INFS each day and
each episode had to be separated by a minimum of 4
hours. The patients were prompted to use both nostrils
during the trial and to switch nostril from one
administration to the next. If BTP episodes occurred
more frequent, or if INFS of placebo did not lead to
sufcient pain relief, rescue medication was allowed to
administer in all phases of the trial 20 minutes after the
rst administration INFS or placebo. There were no
restrictions in the choice of rescue medication so that
the participants could use their usual rescue medication, with which they were familiar.
In the efcacy phase, 2 of 8 BTP episodes were
treated with placebo. The placebo spray contained a
phosphate-buffered solution of sodium citrate and
was visually identical to the INFS 400 g spray.

Study Design
Screening Period
To evaluate eligibility, each patients BGP and BTP
were assessed in a screening period lasting for 7 to
10 days. Before entering the titration phase, an
examination of the nasal cavity was conducted by
an otorhinolaryngologist.

Titration Phase
Patient doses were stepwise titrated to an effective
dose of INFS (50, 100, 200, or 400 mg) after a test
dose of INFS 50 g. One or two doses of the same
dose strength of INFS, separated by at least 10
minutes from the rst administration, was accepted.
If BTP was successfully treated in 3 of 4 episodes, the
titration was considered as successful.

Efficacy Phase
Patients with doses titrated to 200 or 400 g
entered a randomized, double-blind, cross-over efcacy phase directly, in which 8 episodes of BTP were
treated. Patients with doses were titrated to 200 g
and 400 g received INFS 400 g as the active drug
during the efcacy phase. In this phase, the BTP
episodes were treated randomly with INFS 400 g
(6 episodes) and placebo (2 episodes). Due to safety
considerations, patients with doses titrated to 50 and
100 g were not included in the efcacy phase, as
undesirable adverse effects might occur if these patients were to be treated with 400 g INFS.

588

Tolerability Phase
Patients with doses titrated to 50 or 100 g INFS
entered the tolerability phase directly, and patients
with doses titrated to 200 or 400 g entered the
tolerability phase after completing the treatment of 8
BTP episodes in the efcacy phase. The tolerability
phase was terminated 12 weeks after start of the
titration phase for all patients. Adjustments of doses
were allowed during the tolerability phase.

Study Visits and End of Treatment Visit

In all phases of the trial, one or more study visits
were conducted. At the EOT visit, patients delivered
their diary, and a physical examination, including an
examination of the nasal cavity, was conducted, the
latter by an otorhinolaryngologist (3 days to EOT).

Statistical Analysis
All patients that received one or more doses of
INFS during the study were included in the safety
analysis. Patients in the efcacy phase that treated at
least one episode of BTP were included in the efcacy
analysis.
Categorical variables were summarized by number
and percentage of patients in each category, and
continuous variables were summarized by number of
patients, mean, SD, median, minimum, and maximum
values. All tests were 2-sided and conducted at 5%
signicance level. Patients were presented according to
dose level after the titration phase, regardless of additional dose adjustments during the tolerability phase.

Sample Size
In accordance with previous INFS studies, sample
size estimations were calculated to show a signicant
difference for PI at 10 minutes between INFS 400 g
and placebo. With at least 14 patients in the primary
analysis set, the trial had a power of at least 90% to
detect a difference of 1.24 in PID10 between 400 g
and placebo with a signicance level of 5%.

Missing Values
A last observation carried forward rule was applied
for missing data and for efcacy data obtained after
rescue medication.

Efficacy
The null hypothesis was that there was no difference
in PID10 between BTP episodes treated with INFS and

Volume 37 Number 3
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

M. Throns et al.
those treated with placebo. The primary objective was
met if the difference in PID10 between 400 g INFS and
placebo was statistically signicant (P o 0.05).
For each episode, PID10 was calculated as the
difference between PI at time 0 and 10 minutes,
PID10  (PID10  PID0), where PID0 was PID at
time 0. The reverse scale was applied so that a positive
value was a decrease (improvement) in the pain.
The analysis of the primary end point, that is, the 8
repeated measurements of PID10 for each patient, was
based on a linear mixed effects model, including the
episode baseline PI as covariate, the treatment as xed
effect, and the patient as a random effect. The treatment
effect and treatment difference to placebo were presented
as least squares (LS) means. Within-patient and betweenpatient variance estimates were presented as SD.
The PID5, PID30, and PID60 were analyzed using
the same methodology.
The sum of pain intensity difference at 30 minutes
(SPID30) and at 60 minutes (SPID60) represented the mean
improvement in PI derived from PI scores in the efcacy
phase during 30- and 60-minute intervals. The SPIDs at
30 and 60 minutes were derived from the imputed PI
scores as described for the primary efcacy end point. The
calculation of the SPID at time t (SPIDt) was calculated
using the trapezoidal rule, as the area under the curve for
PID during the time intervals 0 to 30 minutes (SPID30)
and 0 to 60 minutes (SPID60), divided by the length of the
respective time intervals, 30 minutes and 60 minutes, and
analyzed similarly to the primary end point.
Overall responder rates of PI in the efcacy phase
were presented in two ways; as a reduction in PI of
33% and 50% at 5, 10, 30, and 60 minutes and as a
reduction in PI of 1, 2, and 3 (010 NRS) at the same
time points. The responder rates were analyzed in a
logistic regression model, including treatment and
period as xed effects and patient as a random effect.
The GI score at 60 minutes after the rst dose of
INFS (INFS or placebo in the efcacy phase) was
analyzed in a mixed effect linear model, including
treatment as a xed effect and patient as a random
effect in all phases of the study.

Tolerability
According to the nose tolerability, the incidence of
changes that worsened from baseline and were rated
as related to the use of INFS by the otorhinolaryngologist was presented as incidence with corresponding

95% CI. The total number of nostrils examined was

used in the calculations.

Ethical Considerations
This study was conducted in accordance with the
principles of Good Clinical Practice and the Declaration of Helsinki. Local ethical committees and medical
agencies in the participating countries approved the
nal protocol and written informed consent was
obtained from each patient included.

RESULTS
Forty-six patients, 31 females and 15 males, with a
mean age of 61 years (range, 3879 years) were
included. The most frequent cancer diagnosis were
breast (30%), urothelial (22%), and gastrointestinal
(20%), and most frequent site of metastasis was bone
(54%) (Table I). Most patients used fentanyl patch as
BGP medication (54%), and the most common rescue
medications were diclofenac (33%), metamizole
sodium (33%), tramadol (13%), oxycodone (13%),
and morphine (9%)
Forty-one patients received a test dose of INFS of
50 g; 4 patients started the titration phase without
taking the test dose. Thirty-eight patients doses were
titrated to effective doses of INFS as follows: 50 g,
n 8; 100 g, n 9; 200 g, n 9; and 400 g,
n 12. Of the 21 patients with doses titrated to 200
or 400 g INFS, 6 patients with doses all titrated to
200 g did not enter the efcacy phase. This was due
to patient withdrawal (n 3), death (n 1),
investigators judgment (n 1), and 1 patient who
refused to increase the dose from 200 g to 400 g to
participate in the efcacy phase. All patients with
doses titrated to 400 g continued to the efcacy
phase. In the efcacy phase, 2 of 15 patients were
withdrawn; 1 patient entered the efcacy phase by
mistake after their dose was titrated to 50 g only and
another patient died the day after completing the
efcacy phase and the randomization was unblended,
leading to withdrawal. The death was not related to
the study drug. These 2 patients were included in the
full analysis set that was used for efcacy analysis.
A total of 31 patients entered, and 16 patients
completed the tolerability phase. Withdrawal (n 5),
death (n 5), and adverse effects (n 2) were the
most common reasons for discontinuation in
this phase (Figure 2).

March 2015

589
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

Clinical Therapeutics

Table I. Demographic characteristics of patients.

All Patients
(n 46)

Variable
Sex, n (%)
Female
Male
Age, y
Mean (SD)
Median (minimummaximum)
Diagnosis (primary tumor), n
Breast
Urothelial or prostate
Gastrointestinal
Pulmonal
Female genitalia
Others
Metastases, n
Bone
Lung
Liver
Lymph nodes
Peritoneum
Adrenals
BGP medication, n
Hydromorfon
Morphine sulfate
Oxycodone
Targiniq (oxycodone or nalaxon)
Tramadol
Tramadol hydrochloride
Fentanyl
Country, n
Hungary
Norway
Russia

31 (67.4)
15 (32.6)
61.0 (90.9)
61 (3879)
14
10
10
5
3
4
25
7
12
7
3
3
6
6
4
3
9
4
25
34
10
2

BGP background pain.

Efficacy
In the efcacy phase, 88 and 29 BTP episodes were
treated with INFS 400 g and placebo, respectively.
Mean PID10 was 2.4 (SD 2.29) and 1.5 (SD 1.52) for
INFS 400 g and placebo, respectively, and the PID LS
mean was 2.5 (95% CI, 1.423.49) and 1.4 (95% CI,
0.232.48) for INFS 400 g and placebo. LS mean

590

difference (INFS  placebo) was 1.1 (95% CI,

0.411.79) (P 0.002) (Table II).
The PID LS mean treatment differences between
INFS and placebo at 30 and 60 minutes were statistically signicant: 1.1 (95% CI, 0.36 to 1.87; P
0.004) and 1.0 (95% CI, 0.24 to 1.84; P 0.012),
respectively. The PID LS mean treatment difference at
5 minutes was 0.5 (95% CI, 0.04 to 0.95; P 0.07),
however, it was not statistically signicant.
SPID30 and SPID60 were higher for INFS 400 g
compared with placebo, showing model adjusted LS
mean treatment difference of 0.3 (95% CI, 0.060.45;
P 0.009) and 0.3 (95% CI, 0.080.51; P 0.008),
respectively.
The responder rates for PID were numerically
higher for INFS 400 g compared with placebo at
all time points. A statistically signicant difference
was seen for PID5 Z1, PID10 >33%, and PID10 Z2
with responder rates of 61% (P 0.046), 44% (P
0.037), and 51% (P 0.041), respectively. The
corresponding odds ratios compared with placebo
were 2.1 (95% CI, 1.04.2), 2.6 (95% CI,
1.16.6), and 2.5 (95% CI, 1.05.8), respectively.
Rescue medication for BTP was used in 29.5% of the
BTP episodes treated with INFS and in 51.7% of BTP
episodes treated with placebo.
The GI score at 60 minutes was favorable for INFS
compared with placebo; LS mean GI scores were 1.9
(95% CI, 1.412.47) vs 1.1 (95% CI, 0.521.71)
and model-adjusted LS mean treatment difference was
0.8 (95% CI, 0.421.23; P r 0.001).

Tolerability
Twenty-seven patients (54 nostrils) were examined
by an otorhinolaryngologist to evaluate the symptoms
of the nasal cavity at baseline and at the EOT visit to
assess tolerability of INFS. Nineteen patients were not
examined, this was due to unsuccessful titration (n
7), death (n 6), withdrawn by clinician (n 3),
withdrawal by patient (n 2), and other reasons (n
1). The most common adverse events (AEs) considered
by the otorhinolaryngologists to be related to INFS
were change in color of the mucosa (5 nostrils), runny
nose (5 nostrils), inammation (4 nostrils), stuffed nose
(4 nostrils), and edema (4 nostrils). None of the AEs
related to the nasal cavity was considered as serious
AE (SAE) (Table III).
Thirteen SAEs and 246 AEs were reported from 12
and 29 patients, respectively, during the trial. One SAE

Volume 37 Number 3
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

M. Throns et al.

Included n = 46
Test dose n = 42
Not successfully titrated
n=7

Titration started n = 45
Titration succeded n = 38
50 g
n=8

100 g
n=9

200 g
n=9

400 g
n = 12

Efficacy phase started n = 15

Not entering the efficacy

phase*:
Withdrawal by patient
n=3
Deaths
n=1
Did not want to increase dose n = 1
Investigators judement
n=1

Efficacy phase completed n = 13

Tolerability phase started

Tolerability phase completed

Not completing the

tolerability phase:

n = 31

n = 16

Withdrawal by patient
Deaths
Discontinued due to AEs
Investigators judgement
Other:

n=3
n=5
n=2
n=1
n=2

Figure 2. Flowchart patients. AE adverse event. *All of these patients doses were titrated to 200 g
intranasal fentanyl spray (Instanyls [Takeda Pharmaceutical Company Limited, Zurich,
Switzerland]).

(respiratory depression) experienced by a 68-year-old

woman with metastatic breast cancer was considered
related to INFS. This patient also had hypertension, type
2 diabetes, and hypothyroidism after a thyroidectomy.
She had been treated with transdermal fentanyl (100 g/h)
as BGP medication, in addition to oral tramadol for

more than 1 year. During 11 days (titration and efcacy

phase), this patient received 400 g INFS for 4 BTP
episodes and 400 400 g INFS for 7 BTP episodes. At
the actual episode, she had received INFS 400 400 g
(at time 0 and time 10 minutes) for treating her BTP
episode. She developed gasping breathing and cyanosis

Table II. Pain intensity difference at 10 minutes for intranasal fentanyl spray* and placebo.
PID calculations
n
Mean (SD)
Median (minimummaximum)
LS mean (95% CI)
P
LS mean, difference INFS  placebo (95% CI)
P

INFS

Placebo

15
2.4 (2.29)
1.5 (07)
2.5 (1.423.49)
o0.001
1.1 (0.411.79)
0.002

15
1.5 (1.52)
0.5 (07)
1.4 (0.232.48)
0.020

INFS intranasal fentanyl spray; LS least square.

*
Trademark: Instanyls (Takeda Pharmaceutical Company Limited, Zurich, Switzerland).

March 2015

591
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

Clinical Therapeutics

Table III. Local nasal tolerability of intranasal

fentanyl spray.*
Worsening Related to
INFS (n 54 nostrils
[27 patients])
Symptoms

n Incidence

95% CI

Change of color
(palered)
Inammation
Sore nose
Ulceration
Dry nose
Runny nose (increases
secretion)
Stuffed nose (swollen)
Edema
Epistaxis

0.09

0.020.17

4
2
2
0
5

0.07
0.04
0.04

0.000.14
0.000.09
0.000.09

0.10

0.020.18

4
4
1

0.08
0.08
0.02

0.000.15
0.000.15
0.000.05

INFS intranasal fentanyl spray.

*
Trademark: Instanyls (Takeda Pharmaceutical
Company Limited, Zurich, Switzerland).

25 minutes after the rst administration of INFS,

decreased respiratory frequency, and thereafter she became unconscious. Nalorphine was given intravenously
twice and she resolved without sequelae but was discontinued from the trial. Other SAEs were not considered to be related to the use of INFS. Eight patients died
during the trial, 1 due to cardiovascular insufciency and
7 due to the underlying cancer diagnosis.
The most common AEs reported during 12 weeks of
treatment were dizziness (52 episodes), nausea (34
episodes), fatigue (17 episodes), vomiting (12 episodes),
and abdominal pain (11 episodes) (Table IV). About
half of the AEs were considered to be treatment related
(146 of 259); nausea, dizziness, and nasal discomfort
were most frequent, reported by 10 (22%), 9 (20%),
and 4 (9%) of the patients, respectively. The AE prole
was similar for all 4 dose strengths. A total of 91 AEs
had an onset within 30 minutes after intake of the rst
dose of the study drug. Dizziness and nausea were the
most frequent early-onset AEs, reported by 8 and 7
patients, respectively. Three patients experienced a total
of 6 AEs that led to discontinuation of the study. These
AEs were nausea, nasopharyngitis, dizziness, azotemia,
respiratory depression, and urticaria.

592

DISCUSSION
The present study found that INFS 400 g is statistically and clinically signicantly more effective than
placebo for pain relief at 10 minutes after administration. The study presents long-term follow-up data,
including a thorough medical examination of the nose
cavity, concluding that INFS is well tolerated.

Effect
This study reported the superiority of INFS 400 g
compared with placebo and the primary end point
proving that INFS 400 g was statistically more
effective than placebo using ANCOVA analysis
was met. When considering a statistically signicant
difference, it is necessary to evaluate whether a difference in PI is of clinical signicance. Studies of pain
management have suggested that PID Z2 on a 0 to 10
NRS or of at least a 30% reduction compared with
baseline should be considered as a clinically important
difference.18,19 In the present study, the mean difference in PI was 2.4 for BTP episodes treated with INFS
400 g at 10 minutes. A reduction in PI of 33% at 10
minutes was seen in 44% and 24% of the pain
episodes treated with INFS 400 g and placebo,
respectively (P = 0.037). This suggests that pain relief
achieved with INFS 400 g is clinically signicant.
These results are comparable with an earlier study in
which Kress et al15 reported clinically meaningful pain
relief at 10 minutes for other dose strengths of INFS
(50200 g).15
These efcacy results are also comparable with
other placebo-controlled studies of rapid-acting fentanyl formulations that report pain relief compared
with placebo in 10 to 15 minutes for fentanyl pectin
nasal spray,2022 fentanyl buccal tablets,23 fentanyl
buccal soluble lm,24 fentanyl sublingual tablet,25 and
fentanyl sublingual spray.26 Primary end points and
statistical methods used in the different studies vary,
and the results are, consequently, challenging to
compare. However, a recent network meta-analysis,
indirectly comparing INFS, fentanyl buccal tablets,
fentanyl soluble lm, and oral transmucosal fentanyl
citrate reported Z33% reduction in PI at 15 minutes
for all of these drugs compared with placebo, with
odds ratios of 2.9, 2.1, 1.3, and 1.2, respectively.9 In
the present study, the corresponding odds ratio was
2.6 compared with placebo at 10 minutes (P 0.037).
Although the present study was not powered to
show a difference in PID5 between INFS 400 g and

Volume 37 Number 3
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

M. Throns et al.

Table IV. Most common adverse events in total and related to intranasal fentanyl spray* and number of
patient-reported adverse events.
Adverse Events
Any event
Dizziness
Nausea
Fatigue
Vomiting
Abdominal pain
Hot ush
Hypoesthesia
Headache
Somnolence
Hyperhidrosis
Nasal discomfort
Nasal edema
Vertigo
Malignant neoplasm
progression
*

No. of
Events

No. of Patients With

Any Event

No. of Related
Events

No. of Patients With

Related Events

259
52
34
17
12
11
11
7
6
6
6
5
5
5
5

35
10
15
3
6
3
1
2
4
2
3
4
3
1
5

146
47
28
17
6
0
11
1
2
5
3
5
3
5
0

18
9
10
3
3
0
1
1
1
1
2
4
1
1
0

Trademark: Instanyls (Takeda Pharmaceutical Company Limited, Zurich, Switzerland).

placebo, a trend was reported (P 0.07), as well as a

PID Z1 (010 NRS) in a higher proportion of the
patients for INFS 400 g compared with placebo (P
0.046). This indicates an early onset of action of this
potent analgesic drug.
Early onset of pain relief has also been found in the
use of fentanyl sublingual spray. Rauck et al26
reported a statistically signicant improvement in
mean SPID from 5 minutes of this drug compared
with placebo. In studies with fentanyl pectin nasal
spray, a PID5 Z1 (NRS 010) compared with
placebo is reported.21,22 To our knowledge, these
are the only studies that indicate such an early onset
of analgesia as found for INFS 400 g in the
present study.
However, it is difcult to interpret if PID of 1 at 0
to 10 NRS is clinically meaningful to patients with
cancer pain at onset of pain relief. Therefore, additional studies are needed to evaluate this aspect.
Although the present study found that INFS is
among the most rapid-acting fentanyl formulations, a
study directly comparing the rapid-acting fentanyl

formulations should be performed to nd which

pharmacologic agent offers the most rapid pain relief.
A comparative study should not only consider efcacy
and adverse effects, but also explore the convenience
of use. The ability to eat and drink after intake of the
drugs might be an important aspect to the patients
that could favor the nasal fentanyl preparations. In
addition, nausea and dry mouth, which are commonly
encountered symptoms in cancer and well-known side
effects of opioids, could be a challenge in the oral
treatment of pain. This problem is avoided by using
nasal spray and studies have shown that patients nd
this approach convenient.16,22
Patient satisfaction was not addressed directly in
the present study, but rescue medication was used in
only 30% of pain episodes in the INFS group
compared with 52% in the placebo group, corresponding to a higher GI score in the INFS group
compared with placebo. The study also reports that as
many as 82% of the patients doses were titrated to an
effective dose of INFS, demonstrating the convenient
use of this drug.

March 2015

593
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

Clinical Therapeutics

Tolerability
The adverse effects of INFS were similar through all
4 doses (50, 100, 200, and 400 g) and seem to be
comparable with other opioids, both rapid-acting
fentanyl compounds and traditional opioids. The most
frequent adverse drug reactions related to INFS were
nausea and dizziness, reported by 10 (22%) and 9
(20%) of the patients, respectively, during 12 weeks of
treatment. These are typical side effects reported in
other studies of rapid-acting fentanyl formulations
varying from 3% to 8% for dizziness and 5% to
13% for nausea.21,22,2428 Somnolence and constipation are often reported as AEs in studies of rapidacting fentanyl, varying from 4% to 9% and 2% to
6%, respectively. Somnolence was reported by only
one patient in the present study, although constipation
was not reported. Nausea and dizziness might seem to
be more frequently reported in the present study
compared with studies of other fast-acting fentanyl
formulations, but this might be explained by the long
follow-up compared with most of the other studies
referred. Nausea, constipation, and somnolence are
among the most frequent reported adverse effects of
traditional opioids as well.29 It is difcult to separate
which opioid caused the different side effects in the
treatment of BTP.
Adverse effects in the nasal cavity are of special
concern for drugs with this mode of administration.
Two studies have explored this for fentanyl pectin nasal
spray.20,22 In these studies, AEs were assessed by
examination of the nose cavity by the study clinician
combined with patient assessment noted in a diary. No
AEs were detected by the clinicians in these studies. In the
present study, the examination of the nasal cavity was
conducted by an otorhinolaryngologist and a predened
checklist was lled in. This might lead to a more proper
examination and could explain why more adverse effects
were reported in the present study. It should be noted
that patient-reported adverse effects of the nasal cavity
were few in the present study and similar to reports in
studies of fentanyl pectin nasal spray.
One patient experienced an SAE of respiratory
depression. This patient received a fentanyl patch as
BGP medication, tramadol, and double doses of INFS
400 g. The patient received nalorphine and recovered
rapidly without sequelae. The patient reporting this
SAE did not have any concomitant illnesses that could
explain this reaction. Respiratory depression has also
been associated with the use of other fentanyl

594

preparations and might be a challenge with high doses

of these rapid-acting drugs. Therefore, these fentanyl
preparations should be used with caution. Both
patients and caregivers should be well informed on
signs of serious adverse reactions, such as severe
drowsiness and somnolence. A nasal formulation of
naloxone for self administration given in 100 L (0.8
mg) by a disposable sprayer system is under development in our research group (personal communication
Professor Ola Dale). In the future, medication like this
should perhaps be offered to patients using rapidacting fentanyl, leading to increased safety for this
group of patients.

Limitations
The present study is relatively small and was designed
to explore the efcacy difference between INFS 400 g
and placebo 10 minutes post administration. Of 46
patients that were included, 16 completed the tolerability
phase. In a larger study, more information about the
tolerability might have been identied.
The examination of the nasal cavity was conducted
in line with common procedures by otorhinolaryngologists to explore local adverse effects due to the use of
INFS. In addition to INFS, these patients received
other drugs, including chemotherapy, which might
have inuenced the ndings in the nasal mucosa.
The patients used INFS as BTP medication. All of
them used opioids as BGP medication. It is difcult to
distinguish between the adverse effects of BGP medication and INFS.
The differences between rapid-acting fentanyl formulations and placebo are relatively small for several of the
efcacy measurements in clinical studies of patients with
BTP and placebo shows effect as well. This might be due,
in part, to the nature of BTP, explained by the rapid pain
ares and sometimes rapid pain relief, even if untreated.
An additional explanation might obviously be the
placebo effect itself.30
The efcacy of the drug is reported in an experimental setting. The numbers of patients were limited,
as were the numbers of BTP episodes treated for each
patient in the efcacy phase. From clinical studies and
clinical experience, we know that the intra- and
interpersonal experiences of pain and BTP vary. As
for other drugs in the investigational phase, this
should be taken into account. In order to prove the
clinical signicance of this drug, systematic observations of larger patient cohorts are needed.

Volume 37 Number 3
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

M. Throns et al.

CONCLUSIONS
This study found that INFS 400 g is more effective
than placebo and leads to clinically signicant pain
relief 10 minutes after administration. The 400-g
dose of INFS will therefore be an important supplement to other dose strengths of INFS in patients
requiring more powerful treatment for their BTP
episodes. However, AEs must be monitored during
treatment with such a potent drug. Long-term followup data show few adverse effects during 12 weeks of
use of INFS, including only minor local tolerability
adverse effects related to the nasal cavity.

ACKNOWLEDGMENTS
The authors would like to thank Tora Skeidsvoll
Solheim, MD for important contribution during the
writing process and Zoa Polya, MD for important
contributions in including patients in Hungary.
There are no grants or other support given to the
authors.
Morten Throns contributed in all parts of the
study. Lars popper and Martin Eeg contributed in the
study design and the writing process. Ellen Jaatun
contributed in the study design, data collection and
the writing process. Magnar Kvitberg contributed in
the data collection and the writing process. Stein
Kaasa contributed in the study design, data interpretation and writing process.

CONFLICTS OF INTEREST
This study was funded by Takeda Pharmaceutical
Company Limited. Lars Popper and Martin Eeg were
employees of Takeda at the time the study was
conducted and during the development of the manuscript. The authors have full control of all primary
data, and have agreed to allow the Journal to review
these data if requested. The authors have indicated
that they have no other conicts of interest regarding
the content of this article.

REFERENCES
1. van den Beuken-van Everdingen MH, de Rijke JM, Kessels
AG, et al. Prevalence of pain in patients with cancer: a
systematic review of the past 40 years. Ann Oncol.
2007;18:14371449.
2. Caraceni A, Hanks G, Kaasa S, et al. Use of opioid
analgesics in the treatment of cancer pain: evidence-based
recommendations from the EAPC. Lancet Oncol. 2012;13:
e58e68.

3. Portenoy RK, Hagen NA. Breakthrough pain: denition

and management. Oncology (Williston Park). 1989;3:2529.
4. Deandrea S, Corli O, Consonni D, et al. Prevalence of
Breakthrough Cancer Pain: A Systematic Review and a
Pooled Analysis of Published Literature. J Pain Symptom
Manage. 2014;47:5776.
5. Davies A, Buchanan A, Zeppetella G, et al. Breakthrough
Cancer Pain: An Observational Study of 1000 European
Oncology Patients. J Pain Symptom Manage. 2014;47:5776.
6. Collins SL, Faura CC, Moore RA, McQuay HJ. Peak
plasma concentrations after oral morphine: a systematic
review. J Pain Symptom Manage. 1998;16:388402.
7. Mercadante S. Managing breakthrough pain. Curr Pain
Headache Rep. 2011;15:244249.
8. Smith H. A comprehensive review of rapid-onset opioids
for breakthrough pain. CNS Drugs. 2012;26:509535.
9. Zeppetella G, Davies A, Eijgelshoven I, Jansen JP. A
Network Meta-Analysis of the Efcacy of Opioid Analgesics for the Management of Breakthrough Cancer Pain
Episodes. J Pain Symptom Manage. 47:772785.e5.
10. Dale O, Hjortkjaer R, Kharasch ED. Nasal administration
of opioids for pain management in adults. Acta Anaesthesiol Scand. 2002;46:759770.
11. Striebel HW, Kramer J, Luhmann I, et al. [Pharmacokinetics of intranasal Fentanyl.]. Schmerz. 1993;7:122125.
12. Kaasa S, Moksnes K, Nolte T, et al. Pharmacokinetics of
intranasal fentanyl spray in patients with cancer and
breakthrough pain. J Opioid Manag. 2010;6:1726.
13. Foster D, Upton R, Christrup L, Popper L. Pharmacokinetics and pharmacodynamics of intranasal versus intravenous fentanyl in patients with pain after oral surgery.
Ann Pharmacother. 2008;42:13801387.
14. Christrup LL, Foster D, Popper LD, et al. Pharmacokinetics efcacy, and tolerability of fentanyl following
intranasal versus intravenous administration in adults
undergoing third-molar extraction: a randomized, double-blind, double-dummy, two-way, crossover study. Clin
Ther. 2008;30:469481.
15. Kress HG, Oronska A, Kaczmarek Z, et al. Efcacy
and tolerability of intranasal fentanyl spray 50 to 200
microg for breakthrough pain in patients with cancer:
a phase III, multinational, randomized, double-blind,
placebo-controlled, crossover trial with a 10-month,
open-label extension treatment period. Clin Ther. 2009;
31:11771191.
16. Mercadante S, Radbruch L, Davies A, et al. A comparison
of intranasal fentanyl spray with oral transmucosal
fentanyl citrate for the treatment of breakthrough cancer
pain: an open-label, randomised, crossover trial. Curr Med
Res Opin. 2009;25:28052815.
17. Plock N, Facius A, Hartmann L, et al. An innovative phase
I population pharmacokinetic approach to investigate the
pharmacokinetics of an intranasal fentanyl spray in

March 2015

595
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

Clinical Therapeutics

18.

19.

20.

21.

22.

23.

24.

healthy subjects. Int J Clin Pharmacol Ther. 2013;51:495508.

Farrar JT, Portenoy RK, Berlin JA,
et al. Dening the clinically important difference in pain outcome
measures. Pain. 2000;88:287294.
Farrar JT. Advances in clinical research methodology for pain clinical trials. Nat Med. 2010;16:1284
1293.
Davies A, Sitte T, Elsner F, et al.
Consistency of efcacy, patient
acceptability, and nasal tolerability
of fentanyl pectin nasal spray
compared with immediate-release
morphine sulfate in breakthrough
cancer pain. J Pain Symptom Manage. 2011;41:358366.
Fallon M, Reale C, Davies A, et al.
Efcacy and safety of fentanyl
pectin nasal spray compared with
immediate-release morphine sulfate tablets in the treatment of
breakthrough cancer pain: a multicenter, randomized, controlled,
double-blind, double-dummy multiple-crossover study. J Support Oncol. 2011;9:224231.
Portenoy RK, Burton AW, Gabrail
N, Taylor D. Fentanyl Pectin Nasal
Spray 043 Study G. A multicenter,
placebo-controlled, double-blind,
multiple-crossover study of Fentanyl Pectin Nasal Spray (FPNS)
in the treatment of breakthrough
cancer pain. Pain. 2010;151:617
624.
Webster LR, Slevin KA, Narayana
A, et al. Fentanyl buccal tablet
compared with immediate-release
oxycodone for the management of
breakthrough pain in opioidtolerant patients with chronic cancer and noncancer pain: a
randomized, double-blind, crossover study followed by a 12-week
open-label phase to evaluate patient outcomes. Pain Med. 2013;14:
13321345.
Rauck R, North J, Gever LN, et al.
Fentanyl buccal soluble lm
(FBSF) for breakthrough pain in
patients with cancer: a randomized,

596

double-blind, placebo-controlled
study. Ann Oncol. 2010;21:1308
1314.
25. Rauck RL, Tark M, Reyes E, et al.
Efcacy and long-term tolerability
of sublingual fentanyl orally disintegrating tablet in the treatment
of breakthrough cancer pain. Curr
Med Res Opin. 2009;25:2877
2885.
26. Rauck R, Reynolds L, Geach J,
et al. Efcacy and safety of fentanyl sublingual spray for the
treatment of breakthrough cancer
pain: a randomized, double-blind,
placebo-controlled study. Curr
Med Res Opin. 2012;28:859870.
27. Taylor D, Radbruch L, Revnic J,
et al. A Report on the Long-term
Use of Fentanyl Pectin Nasal Spray

in Patients With Recurrent Breakthrough Pain. J Pain Symptom Manage. 2014;47:10011007.

28. Nalamachu S, Hassman D, Wallace
MS, Dumble S. Derrick R, Howell J.
Long-term effectiveness and tolerability of sublingual fentanyl orally
disintegrating tablet for the treatment of breakthrough cancer pain.
Curr Med Res Opin. 2011;27:519
530.
29. Cherny N, Ripamonti C, Pereira J,
et al. Strategies to manage the
adverse effects of oral morphine:
an evidence-based report. J Clin
Oncol. 2001;19:25422554.
30. Colloca L, Benedetti F. Placebos
and painkillers: is mind as real as
matter? Nat Rev Neurosci. 2005;
6:545552.

Address correspondence to: Morten Throns, MD, St Olavs Hospital,

Kunnskapssenteret 4th Floor, 7006 Trondheim, Norway. E-mail: morten.
thrones@ntnu.no

Volume 37 Number 3
Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 13, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.