A dose titration study of fentanyl buccal soluble film for breakthrough cancer pain in Taiwan

Abstract

Background

Fentanyl buccal soluble film (FBSF), a new formulation of fentanyl, is developed for the treatment of breakthrough pain (BTP) in opioid‐tolerant patients with cancer.

Aims

This study aimed to assess the feasible dose range of FBSF required for Taiwanese population.

Methods and results

This was an open‐label, multicenter, noncomparative study. Cancer patients who were aged 20 years or older and had a stable regimen equivalent to 60 to 1000 mg/day of oral morphine, 20 to 120 mg/day of intravenous morphine, or 25 to 300 μg/h of transdermal fentanyl for at least 1 week were enrolled. The primary endpoint was the feasible dose range of FBSF. Secondary endpoints included difference in pain intensity at 30 minutes (PID30), percentage of episodes requiring rescue medication, and overall satisfaction. Adverse events (AEs) and serious AEs (SAEs) were recorded for safety measurements. The final effective dose in the per‐protocol (PP) population (n = 30) ranged from 200 to 800 μg, of which 26 subjects (86.7%) achieved an effective dose range of 200 to 400 μg. Among the 283 BTP episodes recorded in the maintenance period, the mean PID30 was 4.0, and only 13 events (4.6%) required rescue medication. For 63.6% of the BTP episodes, patients rated their satisfaction as good to excellent. Only 5% of AEs were considered drug‐related.

Conclusions

Individualized dose titration is recommended for BTP management for patients' benefit. Overall, FBSF was effective and well tolerated and was positively correlated with patients' background opioid dose for persistent pain management.

1. INTRODUCTION

Breakthrough pain (BTP) is a type of transitory exacerbation of pain superimposed on a stable and adequately controlled pain experienced by cancer patients.1, 2 It could occur spontaneously or be induced by some known, such as movement or physical activities, or unknown triggers.2, 3 The prevalence of BTP has been reported to range from 40% to 80% 4 and is associated with patients' morbidity.5 In consideration of its impacts on cancer patients, adequate management of BTP is required but is still suboptimal.6

Administration of rapid‐onset opioids (ROOs) can provide rapid pain relief and is thus used to treat BTP episodes.7 As a new formulation of fentanyl and characterized by its quick‐acting feature, fentanyl buccal soluble film (FBSF) may serve as a preferable option for BTP management in cancer patients. FBSF (Taiwan: Painkyl; United States: Onsolis; Europe: Breakyl) is a potent form of opioid analgesics that uses the dual‐layer technology of BioErodible MucoAdhesive (BEMA) to achieve quick delivery of fentanyl citrate through buccal mucosa.8 It was first approved by the United States Food and Drug Administration in 2009 and was approved by Taiwan Food and Drug Administration (TFDA) in 2013. While FBSF was shown to alleviate BTP effectively, no results were obtained in the ethnic group of Asians.9 As FBSF is the only ROO available in Taiwan, it is of clinical importance to obtain information specifically for Taiwanese cancer patients.

Given the high prevalence of BTP and its impact on cancer patients, a well‐managed BTP could significantly improve patients' quality of life. In order to find a suitable and practical dose range, the primary objective of the present study was to prospectively assess the feasible dose range of FBSF for BTP treatment in Taiwanese cancer patients. The secondary efficacy endpoints were estimated by the changes in pain intensity, subjects' satisfaction, and the percentage of episodes requiring rescue medication at each FBSF administration. The adverse events (AEs) and serious AEs (SAEs) were recorded for safety evaluations.

2. METHODS

2.1. Ethics statement

This study was carried out in accordance with the World Medical Association's Declaration of Helsinki as amended in 1996 and the International Conference on Harmonization (ICH) guidelines for Good Clinical Practice. The study protocol was reviewed and approved by Institutional Review Boards (IRBs) of each study center (Keelung Chang Gung Memorial Hospital IRB No.102‐2922A3; Taipei Veterans General Hospital IRB No.2013‐09‐003A; Shuang Ho Hospital JIRB No.201310006; and Mackay Memorial Hospital IRB No.13CT040b). Written informed consent was obtained from all patients.

2.2. Subjects

2.2.1. Inclusion criteria

Subjects eligible for inclusion were 20 years of age or older and were in a stable current regimen of oral opioids equivalent to 60 to 1000 mg/day of oral morphine, 20 to 120 mg/day of intravenous morphine, or 25 to 300 μg/h of transdermal fentanyl for 1 week or longer. Subjects regularly experienced one to three BTP episodes per day that required additional opioids from pain control, and at least partial relief of BTP by use of opioid therapy was achieved. The ability to understand and willingness to sign a written informed consent document, the ability to self‐administer the study medication correctly or the availability of a responsible adult caregiver to administer the study medication correctly, and the willingness and ability to complete a patient diary with each pain episode were required.

2.2.2. Exclusion criteria

Subjects who had rapidly escalating pain (eg, regularly more than three BTP episodes per day) that were hard to be controlled by analgesics were excluded from the study. A history of hypersensitivity or intolerance to fentanyl or having a cardiopulmonary disease that, in the opinion of the investigator, would significantly increase the risk of respiratory depression was also exclusive. Subjects were not allowed to enter the study if they had (a) a psychiatric/cognitive or neurological impairment that would limit their ability to understand or complete the diary and (b) a moderate (grade 3) to severe (grade 4) mucositis (subjects with less than moderate mucositis were permitted and were to be instructed to not apply an FBSF film at a site of inflammation). Other exclusion criteria included (a) abnormal oral mucosa that would impede drug absorption, (b) being under other treatments that might alter the effect of pain control based on investigator's judgment, (c) recent history or current evidence of alcohol or other drug substance (licit or illicit) abuse, and (d) use of an investigational drug within 4 weeks preceding this study. Subjects who were pregnant, nursing, or with positive pregnancy test were also excluded from the study.

2.3. Study design

This was an open‐label, multicenter, single‐arm study including a screening period, a dose titration period (maximum of 2 weeks), and a 7‐day treatment maintenance period. After screening, eligible subjects were individually titrated to an adequate dose of FBSF (titration period) and continued on this dose as required to control their BTP throughout the maintenance period of the study.

During the dose titration period, subjects were administered with FBSF (Painkyl/Onsolis; TTY Biopharm Company Limited/BioDelivery Sciences International, Inc. [BDSI]) in a dose escalation manner until a treatment dose was identified (defined as an adequate relief of BTP observed for at least two consecutive episodes). All patient started with a dose of 200 μg and increased by 200 μg in each subsequent episode until an adequate pain relief with tolerable side effects was achieved. Doses above 1200 μg were not allowed. For patients who entered the maintenance period, treatment was administrated for 1 week unless there was a lack of effect, intolerable toxicity, or consent withdrawal. There was no dose modification during the maintenance period.

Usage of FBSF was limited to one to three episodes of BTP per day. If a subject experienced more than three episodes per day, he or she was withdrawn from the study. During the study period, all subjects continued their scheduled opioid regimens. Subjects were allowed to use their usual rescue medication if adequate pain relief was not achieved within 30 minutes after application of FBSF. Subjects must separate the FBSF doses by a period of at least 2 hours. The subsequent dose of FBSF should be for the emergence of a new target BTP episode and not for a previous episode that was treated and not resolved.

The study was conducted at four study centers in Taiwan, namely, Keelung Chang Gung Memorial Hospital, Taipei Veterans General Hospital, Shuang Ho Hospital, and Mackay Memorial Hospital.

2.4. Assessments

2.4.1. Efficacy

The primary endpoint was the feasible range of FBSF required for the Taiwanese population. The secondary endpoints were the difference in pain intensity at 30 minutes (PID30) after FBSF administration, subjects' satisfaction, and the percentage of episodes requiring rescue medications.

Pain intensity was determined using an 11‐point numeric scale from 0 = “no pain” to 10 = “worst pain.” Patients were assessed with baseline pain as well as pain intensity at 30 minutes after dosing. The PID30 was obtained by baseline pain score minus score rated 30 minutes after dosing.

Patient's satisfaction was assessed using a 5‐point (poor, fair, good, very good, and excellent) categorical scale at 30 minutes after taking FBSF with the following question: “What was your overall satisfaction with the medication?”

At each episode of BTP, subjects recorded whether a rescue medication was taken after administration of FBSF.

2.4.2. Safety

Physical examination was performed, and vital signs were monitored throughout the study period. The occurrence of AEs and SAEs upon treatment was documented. The relationship of AEs to study medication was assessed as not related, unlikely related, possibly related, and probably related by the investigators.

2.5. Statistical analysis

The safety evaluation was performed using the safety population defined as all subjects who were exposed to at least one dose of FBSF and were available for follow‐up safety information. The primary endpoint was analyzed using the per‐protocol (PP) population, the patients who entered treatment maintenance period. The secondary endpoints were analyzed using BTP episodes in both the full analysis set (FAS; defined as subjects who received at least one dose of study medication without major protocol violation) and the PP population. The demographic data, the correlation between around‐the‐clock (ATC) dose, and baseline pain score as well as between ATC dose and effective FBSF dose were analyzed using the FAS population. The efficacy endpoints and safety data were summarized using descriptive statistics. The number of observation, mean, median, standard deviation, and minimum and maximum were presented for continuous variables. For categorical variables, frequencies and proportions were provided. Spearman rank correlation was utilized for analyzing correlation coefficient between ATC dose and effective dose. A P value of less than .05 was considered statistically significant, and all tests were two‐tailed. All analyses were performed using SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA).

3. RESULTS

3.1. Patient disposition and characteristics

The present study was conducted in four clinical sites from October 2014 to June 2016. A total of 36 subjects were screened and enrolled in this study (Figure 1). Of the 36 subjects, 34 subjects received at least one dose of FBSF and were included in both the FAS and the safety populations. Thirty subjects who entered the maintenance period were included in the PP population. Among the FAS population, 19 were male and 15 were female, with an average age of 57.2 (SD 12.0) (Table 1). All subjects were inpatients. The cancer types of the subjects were also summarized in Table 1.

Figure 1.

Flow of patients through the study (FAS, N = 34). FAS, full analysis set; PP, per‐protocol

Table 1.

Patient characteristics (FAS, N = 34)

Gender, n (%)
Male	19	(55.9)
Female	15	(44.1)
Age, y
Mean (SD)	57.2	(12.0)
Median (range)	58.0	(28.6, 80.4)
Height, cm
Mean (SD)	161.9	(7.9)
Median (range)	163.0	(146.0, 188.0)
Weight, kg
Mean (SD)	53.8	(10.9)
Median (range)	52.2	(35.7, 77.0)
BMI, kg/m2
Mean (SD)	20.6	(4.6)
Median (range)	19.6	(13.8, 32.7)
Cancer type, n (%)
H&N/NPC/esophageal	9	(26.5)
GI/pancreatic/CRC/rectal/HCC	8	(23.5)
Lung cancer	7	(20.6)
Breast cancer	5	(14.7)
GU/GYN/others	5	(14.7)

Abbreviations: BMI, body mass index; CRC, colorectal; FAS, full analysis set; GI, gastrointestinal; GU, genitourinary; GYN, gynecologic; H&N, head and neck; HCC, hepatocellular carcinoma; NPC, nasopharyngeal carcinoma; SD, standard deviation.

3.2. Efficacy

The efficacy of FBSF was evaluated using the PP population, subjects who reached a treatment dose after titration and entered the maintenance period. Among the PP population, 13 (43.4%) subjects needed further dose titration, of which nine (30.0%), three (10.0%), and one (3.3%) subjects were identified with a final dose of 400, 600, and 800 μg, respectively (Table 2).

Table 2.

Effective dose of FBSF (PP, N = 30)

Final dosage, μg	n	%
200	17	56.7
400	9	30.0
600	3	10.0
800	1	3.3
1200	‐	‐
	n	% (95% CI)
Requiring dose titration	13	43.3 (25.5, 62.6)

Abbreviations: CI, confidence interval; FBSF, fentanyl buccal soluble film; PP, per‐protocol.

To explore whether patients with higher ATC dose would have higher baseline pain scores for BTP, the correlation between these two variables was analyzed. It was found that the majority of episodes (281 events; 71.1%) occurred in patients receiving an ATC dose of oral morphine equivalent to 60 to 120 mg/day, and no correlation was identified between baseline pain scores and corresponding ATC doses (Figure S2).

To further elucidate whether the effective doses identified were associated with the patients' ATC dose (ATC opioid regimens in oral morphine or equivalent doses of opioids 10), the distribution of each ATC dose and its corresponding effective doses were analyzed (Figure 2). The results suggest that, for patients who required a higher ATC dose for pain management, a higher effective FBSF dose was needed to treat their BTP (Spearman rank correlation, r_s = 0.56, P = .001).

Figure 2.

The correlation between around‐the‐clock (ATC) dose and effect dose (per protocol [PP], N = 30). A total of 30 patients who entered the treatment maintenance period was included in the PP population. For each patient, the ATC dose of oral morphine (mg/day) and its corresponding effective treatment dose of fentanyl buccal soluble film (FBSF) (μg) are illustrated. A positive correlation was observed between the ATC and FBSF doses (Spearman rank correlation, r_s = 0.56, P = .001).

A total of 283 BTP episodes were recorded during the maintenance period for the PP population (Table 3). The median number of BTP episodes was 9.5 per patient, with a range of one to 21 episodes. The PID30 after FBSF administration was 4.0 (SD 2.0) with a median of 3. Of the 283 BTP episodes recorded, 94.0% and 79.2% of the events had reductions in pain scores greater than or equal to 30% and 50%, respectively. After FBSF administration, a postmedication pain score less than or equal to 3 was achieved in 250 (88.3%) events, in which 105 (86.1%) and 145 (90.0%) episodes had premedication scores of 7 to 10 (severe) or 4 to 6 (moderate), respectively. Thus, it suggests that the relief of BTP was achieved in both severe and moderate episodes. Rescue medication was required in 13 (4.6%) out of 283 episodes. For the patients who needed rescue medications, morphine was used in all 13 episodes. With respect to the overall satisfaction, 63.6% of the BTP episodes were rated as good to excellent. Only eight (2.8%) episodes were rated as poor (Figure S1). Notably, although one‐third of the events were rated as fair, the majority of these episodes (79%; 76 out of 96 events) displayed a postmedication pain score less than or equal to 3, indicating the pain was effectively relieved in these subjects.

Table 3.

PID30 (PP, BTP = 283, maintenance period)

BTP event = 283	Mean (SD)	Median (range)
PID30	4.0 (2.0)	3 (0, 10)

	Event (%)
≥30% reduction in pain score	266 (94.0)
≥50% reduction in pain score	224 (79.2)

BTP event = 283	Event	% (95% CI)
Pain score ≤ 3	250	88.3(84.0, 91.8)
Pain relief from severe pain (7 to 10)	105/122	86.1%
Pain relief from moderate pain (4 to 6)	145/161	90.0%

BTP event = 283	Event	% (95% CI)
Requiring rescue medication	13	4.6(2.5, 7.7)

Abbreviations: BTP, breakthrough pain; CI, confidence interval; PID30, difference in pain intensity at 30 minutes; PP, per‐protocol; SD, standard deviation.

3.3. Safety

Sixty AEs were reported by 34 patients (the FAS population). Three (8.8%) patients reported three drug‐related AEs, including skin itching (2.9%), dizziness (2.9%), and constipation (2.9%), all of which were of grade 1 to 2 in severity. Most AEs (57 out of 60; 95.0%) were not related to the study drug. Seven SAEs were reported by seven subjects. None of them were considered to be study drug–related in the opinion of the investigator.

4. DISCUSSION AND CONCLUSIONS

As a new formulation of fentanyl, only a few studies are currently available for FBSF, and these are primarily focused on its pharmacokinetics.11, 12, 13, 14 FBSF was found to be more effective in treating BTP as compared with placebo treatment in a phase III study,9 which was the only published clinical study using dose titration to evaluate the efficacy of FBSF in cancer patients. Nevertheless, the study only included one Asian in the safety population, and the subject was not included in the intent‐to‐treat (ITT) population. Of the 81 subjects who entered the double‐blind portion of the study, four (4.9%), 15 (18.5%), 23 (28.4%), 19 (23.5%), and 20 (24.7%) subjects were dosed at 200, 400, 600, 800, and 1200 μg, respectively. Most patients had an effective dose of greater than or equal to 400 μg. On the contrary, the present study found that over one‐half of the patients had a final effective dose of 200 μg FBSF after titration, and another 30% of the subjects reached an effective dose of 400 μg FBSF. Although the difference in effective dose distribution between the two studies might be caused by sample size, study design, or other reasons, our results provide necessary data for Asian patients suffering from BTP.

In a recent study conducted by Yen et al,15 the authors assessed the efficacy of FBSF at a dose proportional to the ATC opioid regimen in Taiwanese patients with BTP. By applying proportional dosing, the number of patients identified with effective dose of 200, 400, 600, 800, and 1200 μg was 15 (60%), three (12%), two (8%), four (16%), and one (4%), respectively. As the majority of the patients had an FBSF effective dose of 200 μg, the dose distribution was similar to the results of the present study. In our study, over 50% of the patients had an ATC dose equal to 60 mg/day of oral morphine. Around 53% of the BTP episodes were experienced by the same group of patients. Because patients with higher ATC doses were more likely to be excluded because of physical condition or status of disease progression, their BTP episodes and associated pain intensity were relatively unstable. Thus, only a few patients with high ATC dose were enrolled in the study. On the other hand, because the ATC dose was not allowed to change after entry into the study, subjects recruited in this study tended to have a relatively low ATC dose (equivalent to 60 to 120 mg/day of oral morphine), which may contribute to the high percentage of subjects identified with a feasible dose of 200 to 400 μg FBSF.

Both the study performed by Yen and colleagues and the present study enrolled a limited number of patients with similar patient characteristics.15 Unfortunately, the number of patients in the present study was partially limited by the study design. In the present study, one of the inclusion criteria asked that subjects must be in a stable current regimen for 1 week or longer. Additionally, because ATC dose was not allowed to change during the study period, subjects might have to maintain the same ATC dose for a maximum of 4 weeks. These reasons might discourage patients from entering the study and thereby limited the number of subjects enrolled. However, prohibition of ATC dose change allowed us to identify whether there was a relationship between FBSF dose and background ATC dose and to control possible influences of ATC dose on effective FBSF dose.

In the present study, the effective dose of FBSF was found to be positively correlated with the ATC dose for baseline pain control. To the best of our knowledge, there are only limited studies reporting the correlation between the dose of oral transmucosal fentanyl formulations and ATC doses, especially in the population of Asians. In contrast to our study, several publications showed that a lack of relationship was recognized between these two variables.16, 17, 18 In favor of our result, a similar relationship has been found between the effective dose of transmucosal fentanyl citrate lozenges (OTFC) and ATC opioid dose with statistical significance.19 Nevertheless, with large interindividual variability in patients' dose requirements, the study suggested that OTFC should be routinely titrated and separated from the ATC medication. With respect to FBSF, the correlations between effective FBSF dose and ATC dose has only been previously reported by Yen and colleagues.15 While our study aimed to assess the feasible dose range of FBSF, the study by Yen and colleagues aimed to evaluate the efficacy of proportional doses of FBSF by identifying the percentage of patients requiring dose titration. Since the fundamental study design was different between the study conducted by Yen and colleagues and the present study, the results, especially the correlation between FBSF and ATC doses, cannot be compared directly. With a starting dose proportional to the basal opioid regimen instead of completing a full titration process, it is logically expected that the FBSF and ATC doses would be highly correlated (r_s = 0.8 as reported by Yen et al15). By using a standard dose titration process, we found that there was a positive but relatively weaker correlation between the same two variables (r_s = 0.56). In fact, dose titration is commonly recommended nowadays for patients initiating ROOs treatment.20 Our results further provide necessary information for developing future strategies in BTP management for Asian patients using the new formulation of fentanyl, ie, FBSF.

In general, FBSF was effective in BTP relief. The mean PID30 was 4 in the PP population. Among 283 BTP episodes recorded in the maintenance phase, 250 episodes achieved a postmedication score equal to or below 3, which reflects the efficacy of FBSF. Notably, the effective dose was identified after an adequate relief of BTP observed for at least two consecutive episodes within a relatively short period. As mentioned earlier, the study protocol required that the effective dose not to be adjusted after entry into the maintenance period; a patient who needed a higher FBSF dose was prohibited from dose escalation during that period. Therefore, the ATC dose and effective FBSF dose identified in the titration period may not meet patients' need following disease progression, and it may adversely affect the overall satisfaction obtained. Nevertheless, of the 283 episodes recorded, 63.6% of the events were rated as good to excellent, which is consistent with the results published by Rauck et al, with 67.1% of episodes rated as good to excellent.9 Even in events rated as fair (33.9%), 76 out of 96 episodes (79%) achieved a postmedication score less than or equal to 3. Therefore, although FBSF dose could not be adjusted during the maintenance period, the effectiveness of FBSF was observed through the results of patients' satisfaction.

The incidence of mucositis varies with tumor types and treatments and may impede the administration of therapeutic medications.21 A previous study evaluated the impact of grade 1 oral mucositis on the absorption of FBSF in opioid‐naïve cancer patients.22 The authors suggested that application of FBSF to an area of grade 1 mucositis does not increase fentanyl exposure as compared with the matching controls. In the present study, patients with moderate (grade 3) to severe (grade 4) mucositis were excluded. Only one patient with mild mucositis was enrolled. For this specific patient, the final effective dose of FBSF was 200 μg, with four BTP episodes recorded and a mean PID30 of 6, indicating that FBSF was efficacious for this patient. Nevertheless, the number of patients with mild mucositis was too limited to draw any conclusion.

For BTP treatment, ROOs are recommended to quickly resolve an episode. FBSF is currently the only ROO available in Taiwan. Globally, a number of fentanyl‐based ROOs are available on the market, including OTFC, fentanyl pectin nasal spray (FPNS), fentanyl buccal tablet (FBT), and FBSF.23 These ROOs have different pharmacokinetics and lead to their earlier onsets of action as compared with short‐acting opioids.23 FBSF was shown to have a more favorable pharmacokinetic profile than OTFC.13 Nevertheless, no other comparison has been assessed between FBSF and other ROOs. Moreover, for the issue of optimizing BTP management, clinical practices regarding BTP treatment are greatly affected by the medical care providers as well as health insurance policy. Factors other than the choice of the medication itself may make the situation more complex. An article has discussed and reviewed the related issues.17

Collectively, the present study assessed the feasible range of FBSF effective dose through dose titration in cancer patients suffering from BTP. The final FBSF dose obtained was effective with acceptable side effects. Although the results showed that a positive correlation existed between the effective dose of FBSF and baseline opioid dose for persistent chronic pain, individualized dose titration is recommended for BTP management for patients' benefit. In the future, studies focusing on specific types of cancer with a larger sample size may be conducted to further explore the effectiveness and safety profile of FBSF.

AUTHORS' CONTRIBUTIONS

All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, CH.W., TJ.C., TC.C., TY.C., JS.H., and YF.C.; Methodology, CH.W., TJ.C., TY.C., and YF.C.; Investigation, CH.W., TJ.C., TC.C., TY.C., JS.H., and YF.C.; Formal Analysis, CH.W., TJ.C., TY.C., and YF.C.; Resources, CH.W., TJ.C., TC.C., TY.C., JS.H., and YF.C.; Writing ‐ Original Draft, CH.W., TJ.C., and TC.C; Writing ‐ Review & Editing, CH.W., TJ.C., TC.C., TYC., JS.H., and YF.C.; Visualization, CH.W., TC.C, and TJ.C.; Supervision, CH.W.

FUNDING SOURCES

None declared.

CONFLICT OF INTEREST

None declared.

Supporting information

Fig. S1 Overall satisfaction of FBSF (PP, BTP=283) Patients’ satisfaction was assessed at each BTP episode. In the PP population, 283 events were recorded. The majority of the episodes (63.6%) was rated as good to excellent. Only 2.8% of the events were rated as poor.

Fig. S2 The correlation between ATC dose and baseline pain scores (FAS, BTP=395) A total of 395 episodes was recorded in the FAS population. For each episode, the ATC dose of oral morphine (mg/day) and the corresponding baseline pain score are illustrated. No correlation was observed between the two variables.

Chiou T‐J, Chao T‐C, Chao T‐Y, Huang J‐S, Chang Y‐F, Wang C‐H. A dose titration study of fentanyl buccal soluble film for breakthrough cancer pain in Taiwan. Cancer Reports. 2019;2:e1179. 10.1002/cnr2.1179

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.