Exhibit 1008
`
`Coalition For Affordable Drugs XI LLC
`Exhibit 1008
`Coalition For Affordable Drugs XI LLC v Insys Pharma, Inc.
`IPR2015-01800
`
`

`
` PAIN®
`
`.elsevier.com/locate/pain
`
`A multicenter, placebo—controlled, double—blind, multiple—crossover study of
`Fentanyl Pectin Nasal Spray (FPNS) in the treatment of breakthrough cancer pain
`
`Russell K. Portenoy 3*, Allen W. Burton b, Nashat Gabrail“, Donald Taylor E‘, on behalf of the Fentanyl Pectin
`Nasal Spray 043 Study Group
`aBeti1 Israel Medical Center, New York, NY, USA
`bThe University of Texas MD. Anderson Cancer Cenier, Houston, TX, USA
`‘Gabrail Cancer Center, Canton, OH, USA
`“ Georgia Center for Cancer Pain Management 8 Palliative Medicine, it-iaiietta, GA, USA
`
`ARTICLE INFO
`
`ABSTRACT
`
`Article history:
`Received 20 November 2009
`Received in revised form 26 July 2010
`Accepted 27 July 2010
`
`Keywords:
`Breakthrough pain
`Cancer pain
`Fentanyl Pectin Nasal Spray
`intranasal opioid
`Rapidonset opioid
`
`This randomized, double—blind, crossover study assessed the efficacy and tolerability of a new rapid onset
`nasal fentanyl formulation (Fentanyl Pectin Nasal Spray; FPNS) for breakthrough cancer pain {BTCP).
`Eighty—three of114 patients experiencing one to four BTCP episodes/day while taking 260 mg/day of oral
`morphine or equivalent successfully identified an effective dose of FPNS during a titration phase and
`entered a double—blincl phase in which 10 BTCP episodes were treated with this effective dose (7) or pla-
`cebo (3). Compared with placebo, FPNS significantly improved mean summed pain intensity difference
`CSPID) from 10 min (P<0.0S) until 60 min {P<0.0001).
`including the primary endpoint at 30 min
`(P<0.0001). FPNS significantly improved pain intensity (Pl) scores as early as 5 min (P< 0.05); pain
`intensity difference (PID) from 10 min (P < 0.01 J; and pain relief(PR) scores from 10 min (P < 0.001). More
`patients showed a clinically meaningful ( 22-point reduction in Pl) pain reduction from 10 min onward
`(P g 0.01) and 90.6% of the FPNS—treated versus 80.0% of placebo—treated BTCP episodes did not require
`rescue medication (P < 0.001). Approximately 70% of patients were satisfied or very satisfied with the
`convenience and ease of use of FPNS. Only 5.3% of patients withdrew from treatment due to adverse
`events, no significant nasal effects were reported, and 87% of patients elected to continue open—label
`treatment post—study. in this short—term study. FPNS was safe, well tolerated. and rapidly efficacious
`for BTCP.
`© 2010 International Association for the Study of Pain. Published by Elsevier BV. All rights reserved.
`
`1. Introduction
`
`Breakthrough pain, defined as a transitory exacerbation of pain
`that occurs on a background of otherwise controlled persistent
`pain [16]. has been reported to occur in 33-95% of populations
`with cancer pain [1 9,2026]. Typically, patients with chronic cancer
`pain experience three to four breakthrough cancer pain (BTCP) epi-
`sodes daily. and the typical BTCP peaks within minutes and per-
`sists for a short period [typically less than 45 min)
`{19——21}.
`Patients with BTCP are more likely to have severe pain, psycholog-
`ical distress, impaired function, and poorer quality of life {Z21}.
`BTCP also has been associated with higher health care costs {10].
`Historically, BTCP has been managed with doses of short—acting
`oral opioid drugs, offered “as needed" to supplement a fixed-
`schedule opioid regimen {L16}. Although such treatment with a
`
`* Corresponding author. Address: Department of Pain Medicine and Palliative
`Care, Beth lsrael Medical Center, First Avenue at 16th Street, New York NY 10003,
`USA. Tel.: +1 (212) 844 1505: fax: +1 (212) 844 1503.
`E-mail address: rportenoy@chpnet.org (RR. Portenoy).
`
`“rescue dose" is commonplace, its known pharmacodynamic rela-
`tionship does not closely match the typical time course of a BTCP
`episode. For example, the onset of effect of oral short—acting formu-
`lations of morphine or oxycodone usually begins at least 20 min
`after the close and the peak effect does not occur for nearly an hour
`{L4}. This recognized mismatch between oral drug pharmacody—
`namics and the time course of a typical BTCP episode has led to ef-
`forts to identify alternative drugs and delivery systems to improve
`pain control.
`New fentanyl delivery systems for BTCP have focused on the
`transmucosal route of administration. which is capable of yielding
`pharmacokinetic profile characterized by a high early fentanyl con-
`centration and enhanced early systemic fentanyl exposure [(337].
`Various transmucosal routes have been studied, including buccal,
`sublingual, and intranasal. Among these,
`the intranasal route
`may yield particularly rapid absorption owing to the high vascular-
`ity and permeability of nasal tissues £5.12]. Rapid absorption is
`supported by pharmacokinetic measurements demonstrating a
`short arterial Tmax and a significant arteriovenous difference in
`fentanyl concentration after intranasal administration [17]. The
`
`0304—3959/$36.00 © 2010 International Association for the Study of Pain. Published by Elsevier B.\/1 All rights reserved.
`doi:10.1016;j.pain.2010.07.028
`
`

`
`618
`
`RJC Portenoy et a1./PAIN" I5} (2020 61 7-524
`
`pharmacokinetic inconsistencies related to swallowing part of a
`dose, which could lead to dose—to—dose variability in effects during
`repeated oral transmucosal administration. also might be limited
`with intranasal drug delivery [12}. This potential advantage may
`be enhanced by new technologies that modulate drug release
`and reduce the risk of nasal drip or unintentional swallowing B0}.
`Fentanyl Pectin Nasal Spray (FPNS) uses a proprietary pectin-
`based transmucosal delivery system (PecSys“’1) to modulate drug
`release. FPNS is delivered as a low—volume fine mist of uniform
`droplets that form a gel on contact with the calcium ions present
`in the mucosal membrane secretions. Compared with oral trans-
`mucosal fentanyl citrate, the pharmacokinetics of FPNS are charac-
`terized by reduced time to peak plasma values and significantly
`increased bioavailability [30].
`The primary objective of this study was to demonstrate the effi—
`cacy of FPNS in the treatment of BTCP in patients who are receiving
`regular opioid therapy. Secondary objectives were to demonstrate
`FPNS onset of action, time to clinically meaningful pain relief,
`safety, tolerability, and acceptability.
`
`2. Methods
`
`2.1. Study design
`
`randomized, placebo—controlled, double-
`This multicenter,
`blind, multiple—crossover study was conducted at 36 centers in
`the United States, Costa Rica, and Argentina. The study protocol
`was executed in accordance with regulatory requirements and
`good practice guidelines. and was approved by institutional review
`boards at the participating institutions. All participating patients
`provided signed informed consent. The maximum study duration
`for individual patients was set at 8 weeks.
`
`2.2. Patients
`
`Adult men or women were eligible if they had a histologically
`confirmed diagnosis of cancer, were receiving a fixed—schedule opi-
`oid regimen at a total daily dose equivalent to or greater than 60 mg
`oral morphine per day for background pain, and had one to four epi-
`sodes of moderate to severe BTCP per day. If a patient had more than
`one type of BTCP or had breakthrough pain in more than one loca-
`tion, only one of the pains was identified as a “target" BTCP.
`Patients who had uncontrolled or rapidly escalating background
`pain and those who were medically unstable were not eligible for
`the study. Other exclusion criteria included breakthrough pain not
`primarily related to cancer, past inability to tolerate fentanyl or
`other opioids, history of alcohol or substance abuse, treatment
`with monoamine oxidase inhibitors, and treatment with radiother-
`apy or other investigational drug within the previous 30 days. The
`concomitant use of other medications or interventions that might
`have impacted the patients experience of pain between and during
`episodes (such as analgesic or antiepileptic medication, radiother-
`apy. or chemotherapy) was to be avoided during the double—blind
`period or. in case of medications of these types, the dose had to
`have been stable for between 2 and 3 weeks and was to remain sta-
`
`ble during the study. Treatment with specific medications with a
`known potential for hazardous interaction with fentanyl (such as
`monoamine oxidase inhibitors} was also excluded. Additionally,
`patients with any disorder or medication use likely to adversely af-
`fect the normal functioning of the nasal mucosa were not eligible.
`
`2.3. Procedures
`
`Consenting patients who met inclusion and exclusion criteria
`were allowed to enter an open—label titration phase, the objective
`
`of which was to treat a series of BTCP episodes with successively
`higher doses of FPNS until either an effective dose was found,
`drop—out occurred due to adverse events (AEs), or the drug was
`demonstrated to be ineffective at the highest dose tested. A dose
`was considered “effective" if two episodes of target BTCP were suc-
`cessfully treated (defined as acceptable pain relief [PR] without
`unacceptable adverse effects) with the same dose of FPNS. If PR
`was unacceptable 30 min after taking FPNS, the patient could use
`his or her usual rescue medication.
`During this open—label phase, FPNS doses were titrated from an
`initial dose of 100 pg. Doses were sequentially escalated to 200 pg,
`400 pg, and then 800 ug, if necessary, to identify the effective dose.
`If the 800 pg dose was ineffective, the patient was discontinued
`from the study.
`Only patients who identified an effective dose were eligible to
`continue into the double—blind phase. The objective of this phase
`was to treat a total of 10 BTCP episodes with either the effective
`dose of FPNS (seven episodes) or an identically appearing placebo
`(three episodes).
`During the double—blind phase, patients received 10 separate
`“blinded” bottles, each of which contained either FPNS at the effec-
`tive dose or placebo, identified only by a number, 1-10, by random
`assignment. Patients were instructed to use the bottles in the order
`designated, which was established by a computer—generated sche-
`dule of active drug and placebo in a 7:3 ratio. The patient and all
`personnel
`involved with the study (including investigators and
`investigation site personnel) were blinded to the medication codes.
`The randomization code for each study site was kept in a sealed
`envelope (one per drug pack), to be opened only in a medical
`emergency.
`Patients were instructed to treat no more than four BTCP epi-
`sodes per day and to have an interval of at least 4 h between doses.
`Each episode was treated with a single dose. Pain that continued to
`require treatment 30 min after the dose of study medication could
`be treated with the patient’s usual rescue medication. Patients also
`were instmcted that an interval of at least 4 h was to elapse be-
`tween the use of rescue medication and the next dose of FPNS.
`No protocol violations were identified by use of the e—diary. Any
`occurrence of acute pain other than the target BTCP could be trea-
`ted using the patient's usual rescue drug.
`
`2.4. Ejficacy outcome measures
`
`Electronic diaries (e—diariesj were used to collect patient data
`during the dose—titration and double—blind phases. Baseline pain
`intensity (Pl) prior to treating a BTCP episode was recorded using
`an 11-point numeric scale (0 = no pain to 10 = worst possible pain).
`After this baseline measurement, the study drug was taken. The e-
`diary then provided cues so that both Pi and PR scores were re-
`corded at S, 10, 15, 30, 45, and 60 min. PR was measured on a 5-
`point numeric scale (0 = none to 4 = complete). Use of rescue med-
`ications was recorded throughout the study.
`During the double—blind phase, patients also were asked to rate
`overall satisfaction with the nasal spray at 30 and 60 min after
`each treated BTCP episode. The rating was obtained using a 4-point
`scale (1 = not satisfied to 4 = very satisfied}. Similarly, at the end of
`the study (after the last treated BTCP episode), patients also rated
`the ease of use and convenience of the nasal spray on separate 4-
`point scales.
`
`2.5. Safety and tolerabiiity assessments
`
`Alis were recorded throughout the study. Objective visual nasal
`assessments were performed by the study physician at screening
`and at the end of treatment. Subjective nasal assessments were
`performed by the patient using a 10—item questionnaire (each item
`
`

`
`R.1<. Portenoy er a1.,/PAIN“ :51 (2010) 617-624
`
`619
`
`rated on a 4-point scale: 0 = absent to 3 = severe) before the first
`use of study drug, 1 h after each dose of study medication. and at
`the final study visit. The items rated were stuffyjblocked nose. run-
`ny nose,
`itchinglsneezing, crusting/dryness, burning/discomfort,
`bleeding of nose, cough. postnasal drip, sore throat, and taste
`disturbance.
`
`2.6. Statistical analysis
`
`The primary endpoint was the patient—averaged summed pain
`intensity difference 30 min after dosing (SPlD30), defined as the
`cumulative sum of the recorded difference between PI and baseline
`at each time point from 5 to 30 min post dose. This endpoint was
`chosen because of the likelihood that it would best reflect the effi-
`
`cacy of the dose; at 30 min. it would be expected that the full dose
`would be absorbed, though the underlying pain related to the
`breakthrough episode would still be present for most patients.
`Approximately 88 patients were required for the double—blind
`phase of the study to detect a mean 1* SD treatment difference of
`2.25 i 4.35 between FPNS and placebo in SPlD30, with a 90% power
`and a significance level of 0.05.
`Secondary endpoints included SPID at 10, 15, 45, and 60 rnin; PI
`scores at 5, 10, 15, 30, 45, and 60 min; and the pain intensity dif-
`ference (PID) between scores at specific time points (5, 10, 15,
`30, 45, and 60 min) and the baseline score.—
`
`_ Other secondary endpoints included the PR
`scores at 5, 10, 15, 30, 45, and 60 min; total pain relief (TOTPAR)
`calculated as the cumulative sum of the recorded PR scores at 10,
`15, 30, 45, and 60 min. respectively; the percentage of episodes
`of BTCP that
`required additional
`rescue medication within
`60 min, and the extent to which each treated episode was followed
`by clinically meaningful pain relief (defined as a 22-point reduc-
`tion in Pl {9} ) after FPNS versus placebo therapy. A 22-point reduc-
`tion in SPID also was evaluated in these analyses.
`The statistical analysis used a modified intent—to—treat (ITT) ap-
`proach, which included all patients in the randomized population
`who treated at least one pain episode with FPNS and one pain epi-
`sode with placebo, and, for these episodes, had at least a baseline
`and one post—baseline Pl measurement. The safety analysis set in-
`cluded all patients who received at least one dose of FPNS. Analy-
`ses were performed at {1} the patient—level (patient averages,
`percentage of patients} and (2) at the episode—level (percentage
`of episodes) as an indicator of the consistency of effect. The last-
`
`observation—carried—forward (LOCF} method was used to input
`missing data prior to calculating the average values for each pa-
`tient. The mean value of each variable for each patient was deter-
`mined (up to seven target BTCP episodes per patient treated with
`FPNS and up to three target BTCP episodes per patients treated
`with placebo), giving two numbers — the mean score for episodes
`treated with FPNS and the mean score for episodes treated with
`placebo — per variable, per patient.
`For the primary endpoint, analysis of covariance (AN COVA) was
`used to compare treatments, with the SPlD3O score as the depen-
`dent variable and both treatment (FPNS or placebo) and pooled
`study center included as covariates. Secondary endpoints compar-
`ing treatment differences in Pl. PID, SPID, PR. and TOTPAR at each
`time point were analyzed using a model similar to the primary
`endpoint. In addition, the number and percentage of (a) patients
`and (b) episodes in each treatment group achieving Pl scores 21
`and >2 and SPID scores >2 were summarized. Tests for associa-
`tion between endpoint and treatment were performed using the
`McNemar test for correlated 2 >< 2 binomial endpoints for the pa-
`tient—level analysis and a multilevel model for binary data with
`random effects for the episode—level analysis.
`For the ease—of—use and convenience assessments performed
`after the last treated episode, patient—averagecl scores by treatment
`were categorized as neither satisfied nor dissatisfied (score <3} and
`satisfied or very satisfied (score >3). Safety data during the titra-
`tion and double—blind phases were summarized by treatment.
`
`3. Results
`
`3.}. Patient disposition and baseline demographics
`
`A total of 139 patients were screened for the study and 114
`were enrolled in the titration phase (Fig. 1). Of these 114 patients.
`113 took study medication and were included in the safety popu-
`lation. The mean: SE age of this group was 53.8 i 1.1, 53.1% were
`male. and 68.1% were Caucasian (Table 1).
`Eighty—three patients (73.5%) identified an effective and tolera-
`ble FPNS dose during the titration phase and 31 discontinued the
`study,
`including seven who withdrew for lack of efficacy; six
`who withdrew because of AES, and five who withdrew consent.
`The remaining 13 were either lost to follow—up, did not continue
`to meet study requirements, violated protocol, or had another rea-
`son to discontinue.
`
`Screenect, N:-1:39
`
`Entered open, c1o$e~t§trat:o,_n phase, N='l .14
`
`Rand-m>.':ized‘to d_‘eubie-mind phase, N:-‘83
`
`
`
`Gosnpiieted sit;-233:, 53:78
`
`
`
`m:3T‘t .-= m::L:ti:'«s;:' meat to Seat
`
`Fig. 1. Patient disposition. FPNS = Fentanyl Pectin Nasal Spray.
`
`

`
`RJC Portenoy et a1./PAIN" 151 (2020 617-624
`
`3.2. Eflicacy
`
`The analysis ofthe primary endpoint — patient—avcraged SPlD30
`— revealed a significant difference between episodes treated with
`FPNS and episodes treated with placebo. The mean i SD was
`6.57 :r 4.99 for FPNS doses and 4.45 i 5.51 for placebo (mean 1‘ SD
`treatment difference 2.12 $3.91, P< 0.0001; 95% CI, 1.21-3.03).
`
`Mean baseline PI scores for patient—averaged FPNS—treated and
`placebo—treated episodes were comparable (6.89 versus 6.96,
`respectively).
`
` The percentage of patients who reported a 21-point reduction
`
`in P1 score at each time point, comparing FPNS—treated episodes
`and placebo episodes, were calculated to evaluate onset of effect.
`0
`
`o..2 r-A O E :1 U1 on M\ o“'1 “:1in:1m:1r-rm E6‘E:3an '11rm2 U’) m:1D.
`
`
`38.4% of patients following placebo reported this degree of relief
`(P<0.01), and at 15 min. 72.6% of patients receiving FPNS and
`52.1% of patients receiving placebo had this onset of effect
`(P < 0.01). Analysis by episodes revealed that. compared with pla-
`cebo, 33% of FPNS—treated episodes showed onset of effect (21-
`point reduction in PI) at 5 min (P< 0.05), 61% at 10 min, and 73%
`at 15 min (both P < 0.0001).
`
`0. DJI"?DJ
`"UND.rm:1‘T’5fl
`I"V':r('0U)(D
`m<‘J.r:WD. :1on
`0‘< V)EU
`(cumulative relief rather than relief at one point in time) demon-
`strated that a significantly higher percentage of patients reported
`a mean reduction in SPID score 22 following administration of
`FPNS compared with administration of placebo at each time point
` (Supplementary Fig. 2). Evaluating this
`outcome at the level of each BTCP episode revealed a significant
`difference in favor of FPNS—treated episodes in providing a reduc-
`
`620
`
`Table 1
`Baseline demographics.
`Parameter‘
`
`Age {_years) mean 1 SE (range)
`Race, re {%)
`Caucasian
`Black
`Southeast Asian
`Other
`% Male
`
`Weight (kg), mean: SE (range)
`Primary tumor type
`Breast
`Lung
`Reticuloenclothelial
`Bowel
`Prostate
`Musculoskeletal
`Primary not specified/known
`Upper gastrointestinal
`Pancreas
`Renai
`Throat
`CNS
`Ovary
`Uterus
`Primary hepatic
`Cervix
`Testicular
`Melanoma
`Neumendocrine
`Bladder
`
`Summary statistics (N = 113)
`53.8 1 1.1 [$11-86]
`
`77 (68.1)
`13 (11.5)
`2 (1.8)
`21 (18.6)
`53.1
`
`38.8 -_+ 1.‘? [45.0—147.7}
`(N = 139)“
`24 (1 ?.3)
`18 (12.9)
`1 7 (12.2)
`16 (11.5)
`9 (6.5)
`7 (5.0)
`7 (5.0)
`5 (3.6)
`4 (2.9)
`4 (2.9)
`4 (2.9)
`4 (2.9)
`4 (2.9)
`3 (2.6)
`3 (2.6)
`2 (1.4)
`2 (1.4)
`2 (1.4)
`2 (1.4)
`2 (1.4)
`
`Opioid use, :2 (%)“
`1 (0.9)
`Acetaminophimpropoxyphene
`23 (20.4)
`Methadone/methadone hydrochloride
`7 (6.2)
`1-lydromorphone
`45 (39.9)
`Morphine
`9 (3.0)
`0xycodone—acetaminophen
`26 (23.0)
`Oxycoclone
`7 (6.2)
`Hydrococlone bitartrate—acetaminophen
`5 (4.4)
`Hydrocodone
`1 (0.9)
`Tramadol
`27 (23.9)
`Fentanyl
`Eastern Cooperative Oncology Group (ECOG) scores, n (76)
`0
`10 (13.7)
`1
`42 (57.5)
`2
`21 (28.8)
`
`“ Data for the population screened.
`‘’ Some subjects used >1 opioid medication.
`
`The most common opioids used for background pain were mor-
`phine. fentanyl, oxycodone. and methadone (Table 1). Many pa-
`tients (26%) were taking multiple opioid medications. Of the 83
`patients who were successfully titrated, a subset of 28 patients
`was taking only morphine;
`in this group, the mean close was
`252.9 mg (range, 60-1200 mg).
`Of the 83 patients who identified an effective FPNS dose and
`proceeded into the double—blind phase, 76 (91.6%) completed the
`study (Fig. 1). Of the seven patients who discontinued during the
`double—blind phase, three withdrew consent and one each discon-
`tinued due to AEs, lack of efficacy, lost to follow—up and patient
`death, respectively.
`Seventy—three of the 76 patients who completed the study met
`criteria for inclusion in the modified ITT population. This included
`8 (11.0%) who found that the effective dose was 100 pg, 7 (9.6%)
`who required 200 pg, 24 (32.9%) who required 400 pig. and 34
`(46.6%) who required 800 pg. The patients in the modified 11'1" pop-
`ulation had a median number of BTCP episodes per day of 3 (range
`1-25): all reported that the BTCP episodes were characterized by
`moderate or severe pain. A total of 459 BTCP episodes were treated
`with FPNS and 200 episodes were treated with placebo.
`
`
`
`
`
`summedPainintensityDifference
`
`‘ZB
`
`~; g;
`
`*9‘ FPNS
`
`-I-' Placebo
`
`(meantSE)
`
`5
`
`“£0 15
`
`30
`.
`.
`Time (minutes)
`
`45
`
`60
`
`0
`*P<t3.(!1
`..;,\.mm_
`“"P<O.GG(3"l.
`
`Fig. 2. SPID scores at all time points. FPNS = Fentanyl Pectin Nasal Spray.
`
`

`
`R.1<. Portenoy er a1.,/PAIN“ :51 (2010) 617-624
`
`621
`
`
`
`PainlntensityScore3:»-(maan:t3ElC)-Ato03A0‘:
`I33‘-4
`
`The overall mean patient—averaged acceptability assessment
`score was significantly greater for the active nasal spray compared
`with placebo at 30 min post close (2.63 versus 2.01 ; P < 0.0001 ) and
`at 60 min post close (2.73 versus 2.02: P<0.0001). Acceptability
`assessments after the last treated episode demonstrated that 50
`(68.5%) patients reported an overall acceptability assessment score
`23 (satisfied to very satisfied) for the ease of use, and 51 (69.9%)
`patients reported an overall acceptability assessment score 23
`for convenience with the nasal spray (Supplementary Fig. 3). In to-
`tal, 87% of patients opted to continue FPNS into a long—term. open-
`label safety study.
`
`5
`0
`'P=~1).?)5_
`...;,~_a_m§
`We-0 com.
`
`‘$0
`
`$5
`
`30
`.
`.
`Time {minutes}
`
`$5
`
`30
`
`3.4. Safety and tolerabilitjx
`
`
`
`
`
`
`
`PainlntenrsityUifferencie3*‘
`
`4
`.0’ 01
`
`-G-FPNS
`
`..t
`
`(A)
`
`
`
`{meanififil.0.“’*NO0‘:—*~(.11TR)(.21
`
`Treatment—related AES were more frequent with FPNS than
`placebo and were mainly consistent with the pharmacologic ef-
`fects of fentanyl. They were mostly mild to moderate in severity
`and did not increase in frequency or severity with increasing dose.
`Table 2 shows the most common treatment—emergent AEs in all
`phases. Overall, 14 serious AES (eight following last treatment with
`1-‘PNS and six following last treatment with placebo) were reported
`by nine patients during this study. There was no apparent relation-
`ship between FPNS dose and the serious AES (100 pg. 11 = 5; 200 pg.
`ti = 1; 400 pg, 11 = 1; 800 pg, 11 = 0). Except for the event of noncar—
`diac chest pain. which followed last treatment with FPNS, all
`events were considered by the investigators working directly with
`the patients to be unrelated to study drug. Eight deaths occurred
`during the study (from screening to 1 month after completion).
`Four patients died during the screening period prior to taking
`any study drug. Of the remaining four patients, two died during
`the open dose—titration phase. one died during the clouble—blind
`phase and one died within a month of completing this study. None
`of the deaths were assessed by the investigators as related to study
`drug. Overall. nine patients (eight patients following last treatment
`with HJNS and one patient following last treatment with placebo}
`reported 16 treatment—emergent AES resulting in study drug dis-
`continuation. No patients were suspected of abuse or diversion
`by investigators at any of the centers involved in the trial.
`There were no changes on objective clinical assessment of the
`nose to suggest tolerability problems with FPNS over periods of
`up to 4 weeks. No patient in the safety population reported any na-
`sal problems of severe intensity either before the first use of study
`drug or at the final study visit. For each item, fewer than 10 pa-
`tients reported nasal tolerability events of mild or moderate inten-
`sity. Mean symptom scores were extremely low (<02). however.
`and no clinically significant differences were noted between FPNS—
`and placebo—treated episodes.
`
`0
`
`5
`
`10 ‘£5
`
`30
`
`45
`
`60
`
`*P<0 03.
`“‘F‘+=D.L‘t0U‘-E.
`
`Time {minutes}
`
`Fig. 3. Patient—averaged efficacy measures at all time points: (A) pain intensity (Pl)
`score; (B) pain intensity difference (PID) score. FPNS = Fentanyl Pectin Nasal Spray.
`
`tion in Pl score 22 at 10 min (P= 0.01) and at 15 min and time
`points thereafter (P < 0.0001) (Fig. 4).
`
`33. Patient acceptability
`
`Overall, 90.6% (41 5 of 459} of episodes treated with FPNS versus
`80.0% (160 of 200) of episodes treated with placebo did not require
`additional rescue medication within 60 min (P< 0.001). No rescue
`medications were required following episodes with either treat-
`ment later than 60 rnin.
`
`1:38
`
`2:1 FPN5
`
`4. Discussion
`
`at:
`us‘
`
`%’.£
`
`:1.
`as
`
`
`
`5
`
`$0
`
`15
`
`38
`
`-$5
`
`$6
`
`Time From Dosing
`_
`,
`{minutes}
`..F,":3'G'f:;f’3::
`Fig. 4. Percentages of episodes with clinically meaningful pain relief (>2—point
`reduction in pain intensity). FPNS = Fentanyl Pectin Nasal Spray.
`
`This is the first study to examine the efficacy, safety, and toler-
`ability of FPNS in the treatment of BTCP. The study met its primary
`endpoint: FPNS was efficacious for pain, as indicated by a statisti-
`cally significant improvement in SPID30 compared with placebo
`(P< 0.0001). Moreover. statistically significant differences in pain
`scores were reported with FPNS compared with placebo within
`5 min of dosing. and significant improvements in pain versus pla-
`cebo were maintained for 60 min after dosing.
`At present, the usual approach to the treatment of BTCP in-
`volves the supplemental administration of an oral immediate—re—
`lease opioid formulation, typically morphine or oxycodone. The
`time—action relationship of these drugs — which is characterized
`by an onset that may be delayed for 20 min or more. a peak that
`occurs at about 1 h, and a duration of effect that may extend for
`many hours [L4] — may be unable to provide optimal effectiveness
`
`

`
`622
`
`RJC Portenoy et a1./PAIN" 151 (2020 617-624
`
`Table 2
`Treatment—emergent adverse events (1‘EAEs)a by type (all phases).
`
`TEAEs, rt (%)“
`.
`,.
`..
`,.
`
`.
`
`,..
`
`,.
`
`Vomiting
`Nausea
`Dizziness
`Disease progression
`Epistaxis
`Headache
`Nasopharyngitis
`Somnolence
`Dysgeusia
`Overall {all A135)
`
`Fentanyl Pectin Nasal Spray
`.Numbe1,_.0fTEAESa.{%),.
`.
`,.
`100 pg (n= Q5)
`200 pg (n= 82}
`6 (6.3)
`1 (1.2)
`5 (5.3)
`3 (3.7)
`5 (5.3)
`3 (3.?)
`2 (2.1)
`O (0)
`1 (1.1)
`2 (2.21)
`3 (3.2)
`1 (1.2)
`2 (2.1)
`0 (O)
`2 (2.1)
`2 (2.4)
`2 (2.1)
`0 (O)
`31 (32.6)
`11 (13.4)
`
`.
`.
`.
`.
`,
`400 pg (fl = 78}
`4 (5.1)
`2 (2.6)
`1 (1.3)
`O (0)
`2 (2.6)
`0 (0)
`0 (0)
`1 (1.3)
`1 (13)
`16 (20.5)
`
`.
`
`.
`
`.
`.
`,
`800 pg (n = 53)
`1 (13)
`10(0)
`1 (153)
`3 (5.7)
`2 (3.8)
`0(0)
`2 (3.8)
`1 (1.9)
`0(0)
`16 (30.2)
`
`. ..
`.
`,
`All {n = 113)
`12 (10.6)
`10 (8.8)
`9 (8.0)
`5 (4.4)
`5 (4.4)
`4 (3.5)
`4 (3.5)
`4 (3.5)
`3 (2.7)
`58 (51.3)
`
`Placebo (n = 78)
`,..
`,.
`..
`,
`..
`
`,..
`
`0 (D)
`O (0)
`0 (0)
`O (0)
`O (0)
`O (0)
`O (G)
`0 (D)
`O {(3}
`4 (5.1)
`
`9 Treatment assignment was to the most recent active breakthrough cancer pain treatment within the previous 24 h or to placebo if no active breakthrough cancer pain
`treatment had been taken.
`
`when treating a BTCP that has a more rapid onset and briefer dura-
`tion. This mismatch between the pharmacodynamics of conven-
`tionally used oral rescue medications and the rapid onset and
`short duration of a typical BTCP episode has been the impetus for
`the development of new drug formulations designed to provide a
`more rapid onset of analgesia and duration of action that may be
`more consistent with the temporal profile of a typical BTCP. Most
`of these formulations have been based on the delivery of the lipo-
`philic opioid fentanyl through mucous membranes. Commercially
`available transmucosal fentanyl formulations for BTCP deliver the
`opioid in the mouth, through buccal, gingival, or sublingual muco-
`sa, and have been able to achieve onsets substantially more rapid
`than is possible with a standard orally administered opioid
`[3,'l4,22—24].
`A goal of drug development programs for BTCP has been to
`achieve progressively more rapid onsets of action on the assump-
`tion that the typical BTCP episode, which usually has an onset over
`just a few minutes, would be optimally treated, in at least some pa-
`tients, using a formulation that can produce meaningful effects in
`the same time frame. This is the first study of a transmucosal fen-
`tanyl formulation to observe significant relief from pain relative to
`placebo as early as 5 min. From the 10-min time point onward,
`FPNS was significantly better than placebo in several measures of
`PI and PR, and these differences were maintained for at least
`60 min. Using a commonly accepted metric of a 22-point reduc-
`tion in P1 as an indicator of a clinically meaningful response [9],
`33% of episodes had this level of relief within 10 min ofa FPNS dose
`and 51% experienced it by 15 min. These analyses confirm the effi—
`cacy of FPNS and provide a foundation for predicting the outcomes
`that are most likely to be clinically favorable. at least for patients
`whose episodes of BTCP are characterized by onset over a few
`minutes.
`Overall, more treatment—emergent AES were reported following
`FPNS treatment than following placebo, but no dose—dependent
`pattern could be identified. The most commonly reported AES asso-
`ciated with FPNS were consistent with opioid treatment and were
`mild to moderate in severity. It is difficult to determine whether
`these events were caused by the treatment of BTCP or by their
`fixed—schedule opioid. More specifically,
`treatment assignment
`was to the most recent active BTCP treatment within the previous
`24 h or to placebo if no active BTCP treatment had been taken. The
`four deaths recorded following receipt of study drug were classi-
`fied by the investigators as associated with the progression of dis-
`ease and not related to study drug.
`When questioned about the acceptability of different routes of
`administration of analgesia for BTCP, patients in one survey indi-
`cated that they feared that the nasal route would be difficult to
`administer, catch in the throat, or have a bad taste, and that they
`
`were unfamiliar with the idea [28]. The results of this study refuted
`these concerns, demonstrating that FPN S caused no significant na-
`sal—related symptoms, as assessed by both objective examination
`and subjective recording, and the majority of patients rated FPNS
`as easy to use and convenient. The assessment specifically included
`items on nasal drip and taste disturbances. The nasal route may be
`an alternative for patients with advanced cancer who find oral
`administration difficult and/or uncomfortable due to oral problems
`such as xerostomia, mucositis, or previous surgery [7,8,11].
`The design of this study was comparable to that employed in
`trials to evaluate the efficacy and safety of other fentanyl formula-
`tions for BTCP. Although the use of an open—label dose—titration
`phase to identify a tolerable but effective dose {an enrichment ap-
`proach) has been criticized (1525), its feasibility and robustness
`have allowed the development of a class of rapid onset drugs for
`BTCP, and the enrichment itself may increase the relevance to clin-
`ical practice, during which patients‘ doses are titrated to yield the
`best outcome possible. Almost three—quaiters (72.8%) of patients
`were able to find an effective dose during open—label titration (only
`6% failed to do so because of lack of efficacy),
`indicating that
`the enrichment process did not exclude a large nonresponding
`group.
`The study had several limitations, and the data must be inter-
`preted appropriately. As