PDF
Citation
Print
Introduction
Breakthrough cancer pain (BTcP) is a transient exacerbation of pain occurring on a background of stable, persistent pain in cancer patients [1]. Usually, BTcP has a rapid onset and does not last long, but its intensity is severe [2]. Because of the lack of a unified definition, the exact prevalence of BTcP is hard to estimate [3-5]. However, published data suggest that BTcP can be experienced by up to 89% of cancer patients [1,6-10].
Given its detrimental effect on the quality of life, BTcP constitutes an important therapeutic target [1,11]. The primary treatment strategy nowadays is the combination of long-acting analgesics administered around-the-clock (ATC) to adequately control the background pain with immediate-release opioids as a rescue medication in the case of a BTcP episode [12]. While immediate-release opioids are effective and their dose can be adjusted by the patient, many of those drugs are hydrophilic, which delays the onset of their effect [6,13,14].
The abovementioned limitation was overcome by developing highly lipophilic opioids, absorbed through the oral mucosa and producing rapid-onset analgesia without undergoing the first-pass metabolism. Fentanyl is a highly lipophilic opioid suitable for oral transmucosal administration [6,11]. The agent is available in various formulations, among them as oral transmucosal fentanyl citrate (OTFC), fentanyl buccal tablet, and fentanyl buccal soluble film [15-17]. Another form of transmucosal fentanyl is a sublingual fentanyl tablet (SFT) marketed as Abstral (A.Menarini Korea, Ltd.). SFT reaches its peak blood concentrations within 10 minutes of administration, providing rapid-onset analgesia [18]. Long-term effectiveness and tolerability of SFT in the control of BTcP has already been confirmed in two phase III trials involving opioid-tolerant cancer patients, and the safety profile of the drug was shown to be similar as for other rapid-onset opioids [19,20].
This prospective observational study was designed to verify the abovementioned findings in a real-world setting, using less stringent criteria than in randomized clinical trials. The aim of the study was to analyze the effectiveness, safety, and tolerability of SFT in the control of BTcP in a cohort of Korean cancer patients.
Materials and Methods
1. Study design
This was an observational, prospective, open, single-cohort study conducted in 13 referral hospitals in South Korea. Patients who were prescribed SFT, marketed as Abstral (A.Menarini Korea, Ltd.), for BTcP were followed up for 12 weeks to collect data mainly regarding SFT dose titration, pain intensity, quality of life, and adverse events in a real-world clinical setting. The study was registered at ClinicalTrials.gov (identifier: NCT03895762).
2. Participants
An adult male or female cancer patient aged 19 years or older for whom SFT was indicated for uncontrolled BTcP was invited to the study. After obtaining written informed consent, the following eligibility criteria were confirmed: (1) history of previous unsuccessful treatment for BTcP with other fentanyl products based on the investigator’s judgment or the patient’s dissatisfaction; (2) BTcP ≥ 1 episode/day during the week preceding the enrollment; (3) receiving ATC opioids for persistent cancer pain; and (4) opioid-tolerant, i.e. receiving, for more than a week, oral morphine ≥ 60 mg/day, fentanyl patch ≥ 25 μg/hr, oxycodone ≥ 30 mg/day, oral hydromorphone ≥ 8 mg/day, or an equianalgesic dose of another opioid. Patients with the following conditions were excluded from the study: previous use of SFT within one month before enrollment; contraindicated conditions for SFT; history of participation in an opioid-related clinical trial within 30 days before enrollment; or neuropathic pain.
3. Outcomes
The primary effectiveness endpoint was the SFT titration success rate. A titration was considered successful if all the following conditions were met: (1) no additional dose was administered during each maintenance phase (within two hours after taking SFT), (2) pain intensity reduction ≥ 2 points on the numeric rating scale (NRS) at 30 minutes post-dosing, and (3) adverse drug reactions, if any, were tolerable. As for secondary effectiveness endpoints, the following outcomes were examined: proportion of patients achieving personalized pain goal assessed via NRS at the first week, SFT maintenance dose, maximum pain intensity among all incidences of BTcP, pain intensity difference at 30 and 60 minutes from the pre-dose score, Brief Pain Inventory (BPI) scores, and sleep quality measured as the number of awakenings during sleep. The relationship between SFT dose and ATC opioid dose was investigated as a post-hoc exploratory endpoint. For safety endpoints, treatment-emergent adverse events (TEAEs), blood pressure, and pulse rate were monitored. The severity of TEAEs were classified as mild, moderate, and severe based on their impact on daily activities. Mild TEAEs caused no interference with daily activities, while severe TEAEs resulted in a complete inability to perform daily activities. Seriousness and causality of TEAEs were assessed following the guidelines of International Conference on Harmonisation (ICH), ICH E2A and E2E, respectively.
4. Statistical analysis
The sample size of 262 was calculated to secure the precision of the confidence interval (CI) for the primary endpoint (assumed as 64.4%) at the width of 0.12. The success rate of the primary endpoint was referenced in the previous study [21]. A total of 309 patients were targeted for enrollment, assuming a 15.0% drop-out rate.
The safety set comprised all enrolled patients who took at least a single dose of SFT and had safety follow-up data. Among patients included in the safety set, those with at least one effectiveness datum were included in the effectiveness set. Study data were summarized primarily with descriptive statistics. Mean, standard deviation (SD), median, and ranges (minimum, maximum, and interquartile range) were presented for continuous variables. Multiple categorical or dichotomous variables were summarized with the number of patients and percentage. The titration success rate and proportion of patients achieving pain goals were also presented with 95% CIs. The relationship between SFT dose and ATC opioid dose was tested using Pearson’s correlation coefficient test. Differences in pain intensity between post-dosing (30 and 60 minutes after dosing) and pre-dosing were tested with Wilcoxon signed rank test. BPI scores at follow-up visits were compared with the baseline or week 1 score using repeated measures analysis of variance. Additional details of study design, procedures, and statistical analysis were published previously [22].
Results
1. Participants
During the period between 4 July 2017 and 28 June 2019, 143 patients who received SFT for BTcP were screened for the study, 136 were enrolled, and 36 completed the 12-week follow-up study schedule (Fig. 1). Although just about half of the planned number of patients were enrolled, the enrollment was not extended as to finish the study within the period approved by the institutional review boards. The most common reason for early withdrawal was death (31.0%) followed by consent withdrawal (24.0%), lost to follow-up (18.0%), and physician’s discretion (18.0%). After excluding patients without available data, 133 and 113 patients, respectively, were included in the safety and effectiveness sets. Mean (±SD) age was 60.37±11.54 years, with 36.8% being at least 65 years old or older (Table 1). More male patients were enrolled (63.2% vs. 36.8%), and the mean (±SD) body mass index was 21.60±3.11 kg/m2. Primary cancers diagnosed in more than 10% of patients within 5 years prior to the study enrollment were, in the order of frequency, head and neck cancer (15.0%), pancreatic cancer (12.0%), and colorectal cancer (11.3%). As treatment for cancer, 91.7% of the patients received chemotherapies, 42.9% surgical operations, and 35.3% radiotherapies. Patients mostly marked the front torso (62.8%), back torso (46.0%), or front side of head/neck/shoulder (26.5%) as painful areas, and these responses were consistent with the most painful areas (Table 2). Baseline ATC opioid dose (mg/day in oral morphine equivalent) was distributed most frequently between 100 mg (inclusive) and 400 mg (exclusive) (53.9%), and 19.5% of patients were receiving ATC opioid ≥ 400 mg among 128 patients with available data (mean±SD, 277.80±248.68 mg). The most frequently used concomitant opioid was fentanyl patch (78.9%), followed by oxycodone (44.5%), tramadol (14.1%), and morphine (13.3%).
2. Effectiveness
Among 113 patients in the effectiveness set, 103 had a successful SFT titration. Hence, the titration success rate was 91.2% (95% CI, 85.9 to 96.4). The majority of the patients (n=100) obtained successful titration of SFT by week 1, and three more patients reached it by week 4 (Fig. 2A). Initial titration dose and subsequent maintenance dose of SFT ranged from 100 μg to 400 μg, and the most frequently used dose was 100 μg (Fig. 2B). The proportions of patients using 100 μg of SFT were 65.0%, 72.1%, and 81.8% at week 1, 4, and 12, respectively. The proportions of patients who used 300 μg or 400 μg of SFT were less than 10% across all timepoints (4%-9%). Meanwhile, the mean ATC opioid dose (mg/day in oral morphine equivalent) gradually increased over time, with mean doses of 292 mg, 385 mg, 526 mg, and 529 mg at baseline, week 1, week 4, and week 12, respectively. The increased mean dose of ATC opioids at later time points was contributed mainly by the increased dose of long-acting opioids, with mean doses of 289 mg, 377 mg, 533 mg, and 526 mg at baseline, week 1, week 4, and week 12, respectively, whereas mean doses of short-acting opioids were 25 mg, 34 mg, 28 mg, and 27 mg, respectively, at the same time points. Statistically significant correlations were observed between the mean ATC opioid dose and maintenance dose of SFT (Pearson’s correlation coefficient, 0.21; p=0.030) as well as between mean ATC opioid dose and final dose of SFT (Pearson’s correlation coefficient, 0.25; p=0.010); however, the levels of correlation were too weak to bear a clinical significance.
The target pain goal set by each patient at baseline ranged from score 0 to 5 on an 11-point NRS, where 0 represented no pain. Specifically, 19.5% of patients set a NRS goal of 0, 15.9% of 1, 16.8% of 2, 36.3% of 3, 6.2% of 4, and 5.3% of 5. At week 1, the proportion of patients achieving the personalized pain goal assessed with NRS was 75.2% (95% CI, 67.3 to 83.2). The total number of incidences of BTcP per patient during the first week of the study ranged from 1 to 41, with a median of 9. Among all incidences of BTcP recorded on the pain diary during the first week of the study, maximum pain intensity on the NRS ranged from 7 to 10 (interquartile range, 9 to 10), and its mean (±SD) was 9.37±0.84. Mean (±SD) of BTcP intensity right before taking SFT was 6.14±1.68, and SFT resulted in significant reduction in the pain (Fig. 3). Mean (±SD) change in pain intensity at 30 and 60 minutes after taking SFT was –2.57±1.53 and –3.62±1.89, respectively (p < 0.001 for both).
The number of awakenings during sleep was recorded on the pain diary during the first week of the study, and its mean (±SD) frequency was 1.70±1.20 times/day.
Responses on the BPI questionnaire showed that the worst pain within the previous 24 hours (NRS mean) was reduced from baseline (7.21) at week 1 (6.62, p=0.035) and week 4 (6.50, p=0.029) but no statistical difference was found at week 12 (6.83, p=0.378). NRS mean of least pain within the previous 24 hours at baseline was 2.29, and thereafter, the mean values ranged from 2.16 to 2.63. NRS mean values of average and current pain ranged from 3.58 to 4.21 and 2.92 to 3.40, respectively. The median of pain relief (%) from pain medication used within 24 hours at baseline was 50% (interquartile range, 30% to 70%), and median values and interquartile ranges at later visits implied the level of pain relief modestly increased with time (e.g., at week 4, median 60%; interquartile range, 50% to 80%); however, no statistical difference was detected. In terms of pain interference with daily activities, general trend of improvement in quality of life was observed till week 12 and the BPI scores were lowest at week 4 (Fig. 4). Compared with baseline BPI score, mean BPI scores of general activity at week 4 (7.09 vs. 6.17, p=0.016), mood at week 4 (7.07 vs. 5.91, p=0.005), normal work at week 4 and 12 (7.14 vs. 6.02 [Week 4], p=0.005; vs. 6.10 [Week 12], p=0.048), relations with other people at week 4 (6.38 vs. 5.31, p=0.015), sleep at week 1, 4, and 12 (6.04 vs. 5.20 [Week 1], p=0.040; vs. 4.11 [Week 4], p < 0.001; vs. 4.70 [Week 12], p=0.033), and enjoyment of life at week 4 (6.77 vs. 5.53, p=0.004) were lower with statistical significance. Furthermore, mean BPI scores of normal work, sleep, and enjoyment of life in week 4 were lower than in week 1 with statistical significance (p=0.046, p=0.025, and p=0.038, respectively).
3. Safety
Among 133 patients, 91 (68.4%) experienced 268 TEAEs, of which 171, 61, and 36 events were respectively mild, moderate, and severe in severity. The most common TEAEs were gastrointestinal symptoms (Table 3). A total of 24 TEAEs were reported to have fatal outcomes, most of which were due to exacerbation of the underlying diseases. Two TEAEs, which were confusional state and vomiting, reported in two patients led to permanent discontinuation of SFT. According to the investigator’s assessment of the causality of TEAEs, 19 events reported in 12 patients (9.0%) were classified as adverse drug reactions (probable/likely, 1 event; possible, 9 events; conditional/unclassified, 2 events; and unassessable/unclassifiable, 7 events). The most common adverse drug reactions were vomiting (4 events) and nausea (3 events). Six serious adverse drug reactions (hepatocellular carcinoma, nasal sinus cancer, dyspepsia, multiple organ dysfunction syndrome, oral candidiasis, and headache) were reported in five patients. No clinically significant findings or statistically significant changes were detected in blood pressure or pulse rate.
Discussion
The aim of the present study was to analyze the effectiveness, safety, and tolerability of SFT in Korean cancer patients with BTcP. Most of the study patients successfully completed the titration phase, with effective doses of SFT of 100-400 μg. SFT provided rapid analgesia, which, aside from a significant reduction in BTcP severity, also contributed to an improvement in the quality of life. SFT was generally tolerated well, with a safety profile similar to other transmucosal fentanyl formulations.
The titration success rate in the present study (91.2%) differed considerably from the results of two published randomized trials evaluating the efficacy and tolerability of SFT for the treatment of BTcP in opioid-tolerant cancer patients (59.5% and 69.1%) [19,20]. Titration success rates in clinical trials of other transmucosal fentanyl formulations, OTFC, fentanyl buccal tablet, and fentanyl buccal soluble film, were also different from the present study, with the effective doses obtained in 71%, 65%, and 54% of patients, respectively [15-17]. The high titration success rate in our present study seems to be primarily related to its real-world character. In the two trials of SFT mentioned above [19,20], primary reasons behind the titration phase failure were administrative factors, such as protocol violation, the sponsor’s decision or the subject’s withdrawal of consent. After excluding those factors, which do not play a role in real-world clinical practice, the proportions of patients who completed the titration phase in those two trials would be similar as in the present study, at 80.1% and 84.2% [19,20].
In the present study, the effective dose of SFT, defined as the single dosage strength providing effective relief (pain reduction ≥ 2 on NRS) of all BTcP episodes with an acceptable and tolerable adverse event (AE) profile, was 100-400 μg, with the most frequently used dose being 100 μg. A similar definition of effective dose was also used in the previously mentioned randomized, placebo-controlled, multicenter phase III trial of SFT [19]. In that study, effective analgesia was achieved at a broader SFT dose strength spectrum, 100-800 μg, with the effective dose ≥ 300 μg for over 85% of patients [19]. Similar dose distributions were also reported in the studies analyzing other transmucosal fentanyl formulations, OTFC, fentanyl buccal tablet, and fentanyl buccal soluble film [15-17]. However, those studies used less stringent effective dose definitions, with no relief of all BTcP required, which precludes a head-to-head comparison of their results with our findings [21]. Thus, it is unclear if the difference between the effective dose spectra of SFT in the present real-world study and the previous clinical trial is associated with different patient characteristics or whether it is clinically relevant [19]. Nevertheless, having a choice between dosing strengths, clinical practitioners should select the one that provides an optimal balance between analgesic effects and acceptable adverse effects in a given patient.
At 30 and 60 minutes after administration of SFT, our patients reported significant pain reduction, expressed as a pain intensity difference compared to the pre-dose pain. Hence, the study confirmed that SFT can provide a rapid analgesic effect during a BTcP episode. This observation is consistent with the data from the multicenter phase III trial of SFT [19]. In that study, the mean sum of pain intensity difference at 30 and 60 minutes after the administration of SFT was significantly greater than with placebo. Additionally, the administration of SFT was associated with significantly greater improvements in pain intensity difference and significantly greater pain relief, starting 10 minutes post-dose and lasting throughout the 60-minute assessment period [19].
Our analysis showed the absence of a strong correlation between the mean ATC opioid dose and SFT dose. The results of previous studies in this matter are inconclusive. While some authors found a moderate to strong correlation between ATC opioid dose and the effective dose of OTFC [23,24], many others observed weak correlations between those variables or no correlations at all [25-28]. One important aspect that needs to be considered when using SFT for BTcP is starting titration with the lowest dose of the drug to avoid unnecessary side effects, as recommended in the current label of the product. Most of our patients could manage BTcP with the lowest dose (100 μg), which might explain a weak correlation between the ATC opioid dose and the final or maintenance dose of SFT documented in the present study.
Up to 75.2% of patients participating in the present study reached their target goal of pain level. The analysis of the BPI scores in the sensory dimension adds to this observation, showing that the worst pain within the previous 24 hours (NRS mean) was significantly reduced from baseline at weeks 1 and 4. The lack of significant difference in NRS mean at week 12 was likely associated with 73.5% of patients being withdrawn prematurely, mostly due to high mortality, which made the sample underpowered. Nevertheless, the results of the non-randomized, open-label, multicenter phase III study, with a maintenance phase lasting up to 12 months [20], suggest that SFT can also be effective in a longer perspective. In that trial, mean pain levels generally remained stable throughout the 12-month follow-up period, and the current pain level at six months was even significantly lower than at the baseline [20].
Published evidence suggests that merely a modest correlation exists between pain intensity and everyday functioning [29]. In this context, our findings for the reactive dimensions of BPI are worth emphasizing. In the present study, the scores for the reactive dimensions of BPI generally decreased with time, in some cases significantly, with the lowest values observed at week 4. Especially, the interference of sleep due to pain was significantly improved at all timepoints. Similar to the sensory dimension, the lack of significant improvement in most reactive dimensions of the BPI at week 12 might be associated with the low statistical power of the sample at this time point. In the adequately powered multicenter phase III study [20], the composite score for interference of pain with daily activities was significantly improved at six months, and so were five out of seven individual aspects of this dimension. The same study showed that analgesia achieved with SFT exerted a favorable effect on other dimensions of the quality of life, measured with the depression, anxiety, and positive outlook scale (DAPOS) [20]. Taken altogether, both our findings and the published evidence show that the analgesic effect of SFT is also associated with a beneficial effect on the quality of life.
In this study, 268 TEAEs were recorded in 91 out of 133 patients (68.4%), with the vast majority (232 out of 268 events) being mild or moderate in severity. A total of 19 events reported in 12 patients (9.0%) were classified as adverse drug reactions, with the most common being vomiting and nausea. This safety profile was similar to those reported previously in the phase III trials of SFT [19,20]. In those studies, AEs were documented in 43.3% of the patients and were predominantly mild or moderate in severity. The most common AEs were nausea (12.2%), vomiting (5.3%), and somnolence (4.3%) [21]. The pattern of AEs documented in the present study was also consistent with those reported in the trials of other transmucosal fentanyl preparations and rapid-onset opioids [16,21].
The primary limitation of this study is associated with a high withdrawal rate, which is unfortunately unavoidable in the case of research involving terminally ill patients. Also, the short duration of the study did not allow us to analyze the long-term effectiveness, safety, and tolerability of SFT. To address these limitations in future research, larger sample sizes and longer study durations could be considered. Leveraging the analysis of the large-scale data set, such as the electronic medical records or health claims data, could provide valuable insights into real-world effectiveness. Despite some shortcomings mentioned above, the results of the present study are valuable from the practical point of view because of its real-world character. Because of less stringent eligibility criteria, participants of this study were probably more representative of the population of cancer patients suffering from BTcP than subjects included in previous phase III trials of SFT. Moreover, all treatment-related decisions were left solely at the investigator’s discretion, allowing us to analyze the treatment outcomes in various clinical scenarios unlikely to be tested in a traditional clinical trial.
To summarize, the results of this real-world study stay in agreement with the evidence from previous phase III trials, showing that SFT provides rapid and effective analgesia in BTcP without compromising the patient’s safety.
XML Download