Cebranopadol is a novel strong analgesic that coactivates the nociceptin/orphanin FQ receptor and classical opioid receptors. There are indications that activation of the nociceptin/orphanin FQ receptor is related to ceiling in respiratory depression. In this phase 1 clinical trial, we performed a pharmacokinetic-pharmacodynamic study to quantify cebranopadol’s respiratory effects.
Twelve healthy male volunteers received 600 μg oral cebranopadol as a single dose. The following main endpoints were obtained at regular time intervals for 10 to 11 h after drug intake: ventilation at an elevated clamped end-tidal pressure of carbon dioxide, pain threshold and tolerance to a transcutaneous electrical stimulus train, and plasma cebranopadol concentrations. The data were analyzed using sigmoid Emax (respiration) and power (antinociception) models.
Cebranopadol displayed typical opioid-like effects including miosis, analgesia, and respiratory depression. The blood-effect-site equilibration half-life for respiratory depression and analgesia was 1.2 ± 0.4 h (median ± standard error of the estimate) and 8.1 ± 2.5 h, respectively. The effect-site concentration causing 50% respiratory depression was 62 ± 4 pg/ml; the effect-site concentration causing 25% increase in currents to obtain pain threshold and tolerance was 97 ± 29 pg/ml. The model estimate for minimum ventilation was greater than zero at 4.9 ± 0.7 l/min (95% CI, 3.5 to 6.6 l/min).
At the dose tested, cebranopadol produced respiratory depression with an estimate for minimum ventilation greater than 0 l/min. This is a major advantage over full μ-opioid receptor agonists that will produce apnea at high concentrations. Further clinical studies are needed to assess whether such behavior persists at higher doses.
Nociceptin/orphanin FQ peptide (NOP) receptor activation reduces μ-opioid peptide (MOP) receptor–induced respiratory depression, probably due to a functional interaction of the two receptor systems
Cebranopadol coactivates NOP and MOP receptors
In a phase 1 clinical trial of oral cebranopadol, its potency for respiratory depression was three times that for analgesia measured by an experimental electrical pain model
The blood-effect-site equilibration half-life for respiratory depression was estimated to be 1.2 h, while that for analgesia was 8.1 h
The model estimate for minimum ventilation was more than zero
MODERATE to severe pain is commonly treated with opioid analgesics. Opioid therapy, however, can be associated with serious side effects. Most common adverse events are central nervous system related (e.g., dizziness, sedation, respiratory depression) or gastrointestinal tract related (e.g., constipation, nausea or vomiting). Respiratory depression, which occurs due to activation of μ-opioid peptide (MOP) receptors expressed in the respiratory centers in the brainstem, is potentially life-threatening and leads to significant morbidity and mortality.1–5 There are currently various developments aimed at preventing or reducing the occurrence of hazardous opioid-induced respiratory depression.1,6,7 One possibility is the use of opioid analgesics that coactivate receptor systems that counteract MOP-related opioid-induced respiratory depression. For example, we previously postulated that opioids that act at MOP receptor and coactivate the nociceptin/orphanin FQ peptide (NOP) receptor, an atypical member of the opioid receptor family, will display limited respiratory depression.8 Various animal studies provide evidence for this.9–11 NOP receptor activation reduces MOP receptor-induced respiratory depression,11 most probably due to a functional interaction of the two receptor systems.12,13
The novel analgesic agent cebranopadol targets the NOP receptor in addition to classical opioid receptors and showed potent analgesia in experimental studies11,14,15 ; various clinical trials are currently underway (see ClinicalTrials.gov and EU Clinical Trial Register).16–18 Animal studies indicate that cebranopadol is especially effective in experimental models of chronic neuropathic pain (e.g., streptozotocin-induced diabetic polyneuropathy and spinal nerve ligation models) compared to acute nociceptive pain (bone cancer pain and tail-flick test).15 In the rat, there are strong indications that cebranopadol shows limited depression of breathing.11,15 In a first step to improve our understanding of the respiratory behavior of cebranopadol in humans, we performed a population pharmacokinetic–pharmacodynamic (PK–PD) modeling study. Twelve healthy male volunteers ingested a single oral dose of 600 μg cebranopadol and subsequently ventilation was measured for 11 h both without and with clamping of the end-tidal pressure of carbon dioxide (Petco2). Additionally, we measured antinociceptive and pupil diameter responses. We here focus on the PK–PD and physiologic analysis of the respiratory data and PK–PD analysis of the experimental pain data.
Materials and Methods
This study was performed from March to June 2010 and was registered in the EU Clinical Trials Register (identification number 2009-010893-39). All procedures were performed according to the Declaration of Helsinki and were approved by the local medical ethics committee (Commissie Medische Ethiek, Leiden, The Netherlands) and the Central Committee on Research Involving Human Subjects (Centrale Commissie Mensgebonden Onderzoek, The Hague, The Netherlands). The study protocol consisted of three experimental sessions. On two separate days, the effect of the MOP receptor agonist fentanyl was investigated on respiration and pain responses; on the third day, the effect of cebranopadol was tested. The order of study days was randomized. The fentanyl data were published previously.19 Here we present the cebranopadol data.
Twelve healthy male volunteers completed the study. Inclusion criteria were male sex; age 18 to 45 yr; body mass index, 20 to 28 kg/m2; body weight, not less than 50 kg; absence of any medical disease; and ability and willingness to give written informed consent to participate in the trial. Exclusion criteria included the inability to communicate meaningfully with the investigators; smoking of more than 10 cigarettes/day (or equivalent); a history of asthma or other respiratory diseases; conditions known to interfere with the absorption, distribution, metabolism, or excretion of drugs; a history of drug allergy; participation in another clinical trial within 90 days of the current study; and a history of drug abuse or psychiatric illness.
Subjects were asked to refrain from food containing caffeine (e.g., coffee, tea, cola-containing drinks, and chocolate) for 48 h before dosing, drinks containing quinine (e.g., bitter lemon and tonic water) for 2 weeks before dosing, grapefruit juice and alcohol for 48 h before dosing, and poppy seed-containing food for 72 h before each dosing. All subjects fasted from the evening before dosing until 7 h after dosing.
All subjects came to the clinical research unit (Centre for Human Drug Research, Leiden, The Netherlands) on the day before the study and were transported to the Anesthesia and Pain Research Unit at Leiden University Medical Center (Leiden, The Netherlands) at 7:30 am on the study day. In the laboratory, one intravenous access catheter for administration of fluids (100 ml/h NaCl, 0.9%) and one venous sample catheter were placed, and pretreatment baseline respiratory, pupil diameter, and pain responses were obtained. At 9:00 am, subjects received an oral dose of 1 ml cebranopadol (cebranopadol hemicitrate oral solution in Macrogol 400, containing 600 μg/ml cebranopadol; Grünenthal GmbH, Germany). Cebranopadol was injected into the mouth of the subject from a prefilled syringe and washed down with 300 ml noncarbonated water. Next, respiration, pupil diameter, and pain responses were measured at regular intervals.
Breath-to-breath inspired minute ventilation (VE) was measured with the use of the dynamic end-tidal forcing technique.6,8,19,20 During the respiratory studies, subjects breathed through a facemask (fitted over nose and mouth). The airway gas flow was measured with a pneumotachograph (model 4813; Hans Rudolph, USA) connected to a pressure transducer, which yields a volume signal. The pneumotachograph was heated throughout the study period. This signal was calibrated with a 1-l calibration syringe (Hans Rudolph). The pneumotachograph was connected to a T-piece; one arm of the T-piece received a gas mixture with a flow of 45 l/min from a gas mixing system, consisting of three mass-flow controllers (Bronkhorst High-Tech, The Netherlands) via which the flow of oxygen, nitrogen, and carbon dioxide could be set individually at any desired level. A computer provided control signals to the mass-flow controllers allowing adjustment of the inspired gas mixture to force the end-tidal gas concentrations of oxygen and carbon dioxide to follow a specific pattern in time. Gas concentrations were measured with a gas analyzer (Datex Multicap, Finland); arterial hemoglobin oxygen saturation was measured via a finger probe with a Masimo pulse oximeter (Masimo Corporation, USA).
In the current study, the end-tidal pressure of oxygen was maintained at 110 mmHg. The end-tidal Pco2 (Petco2)value to be maintained (clamped) in the study was determined before drug administration. After a 1- to 2-min period of breathing with no additional inspired carbon dioxide, the Petco2 was stepwise and slowly increased such that the VE reached a steady-state value of 22 ± 2 l/min for 3 min. This elevated Petco2 level was the target Petco2 to be used in the study. VE measurements were obtained at 1-h intervals for 11 h postdosing. Each measurement was performed by stepwise and slow increases of Petco2 to the target value. Thereafter, steady-state VE was measured for another 5 to 7 min at the target Petco2 level. The investigator determined whether a steady state in ventilation was reached by inspection of the real-time breath-to-breath data plotted on the computer screen. One-minute average VE values obtained at clamped Petco2 were used in the PK–PD analysis. The complete data set (1-min averages of resting and carbon dioxide-clamped data) was used in the mechanistic data analysis.
Experimental Acute Pain.
Acute pain was induced by an electrical current through two surface electrodes (Red Dot; 3M, Canada) placed on the skin overlaying the left tibial bone.21 The electrodes were attached to a computer-interfaced current stimulation (Leiden University Medical Center). The intensity of the noxious stimulation was increased from 0 mA in steps of 0.5 mA/s. The stimulus train consisted of a square-wave pulse of 0.2-ms duration applied at 10 Hz and had a cutoff of 128 mA. The subjects were instructed to press a control button when they felt pain (pain threshold) and when no further increase in stimulus intensity was acceptable (pain tolerance; this ended the stimulus train). Pain responses were obtained at baseline and at t = 90, 150, 210, 270, 330, 390, 450, 510, 570, and 630 min after dosing.
Blood Sampling and Plasma Drug Concentration Measurements.
Blood samples (4 ml) were obtained for determination of the plasma cebranopadol concentrations. To that end, venous samples were obtained from the left or right cubital vein (on the arm opposite of the intravenous catheter placed on the hand of fluid administration). Plasma was separated within 30 min of blood collection and stored at −20°C until analysis. Blood samples were obtained at t = 15 min, 1, 2, 3, 4, 5, 6, 8, 10, 12, 18, 24, and 32 h after dosing.
The plasma concentrations of cebranopadol were analyzed by liquid chromatography with tandem mass spectrometric detection after liquid–liquid extraction with t-butyl methyl ether. The internal standard used was D5-cebranopadol. The matrix-based calibration curves were linear in the measured range from 0.002 to 2 ng/ml using a sample volume of 500 μl. The values for the overall accuracy (expressed as percentage of nominal value) and the overall precision (expressed as coefficient of variation) for quality controls assayed during analysis of study samples were 94.0 to 100.8% and 3.1 to 5.9%, respectively. The samples were analyzed in three batches with interbatch accuracy of 86.4 to 104.3% and coefficient of variation of 1.1 to 7.3% over the calibration range.
The pupil diameter in one eye (the same eye during all measurements) was measured using a CIP Pupillometer (AMTech Pupilknowlogy GmbH, Germany). Three measurements at 15-s intervals were obtained per time point. Before each measurement, the lights in the laboratory were dimmed and the eyes of the subjects were allowed to adapt for 3 min. Pupillometry was performed at baseline and at t = 75, 135, 195, 255, 315, 375, 435, 495, 555, and 615 min after dosing.
Since there are no previous data on the effect of cebranopadol on respiration in humans, we had no indication of possible effect sizes in this phase 1 trial. A sample size of 12 was considered adequate to meet the trial objectives (to obtain quantitative PK–PD data on cebranopadol for the three designated endpoints) and was based on previous trials performed in our laboratory. Furthermore, the number of subjects complies with safety requirements by minimizing the number of exposed subjects in line with conventional sample sizes used for phase 1 trials.
The PK–PD data were analyzed with the statistical package NONMEM VII (ICON Development Solutions, USA). The PK–PD analysis was performed in two stages. From the first stage (pharmacokinetic [PK] analysis), empirical Bayesian estimates of the PK parameters were obtained. In the second stage (pharmacodynamic [PD] analysis), the PK parameters were fixed to those obtained in the first stage.
One-, two-, three-, and four-compartmental absorption models with identical or distinct rate constants were tested. The distribution model component always consisted of two compartments as we deemed one compartment very likely to be insufficient and three compartments unlikely to be supported by the data. Model selection was based on the goodness-of-fit criterion (i.e., the magnitude of the decrease in the minimum objective function value; chi-square test with P < 0.01 was considered significant). Model parameters were assumed to have log-normal distributions. The residual error had an additive and a relative error term, σ1 and σ2, respectively. Weight and height were considered as covariates. Only the best model will be described here. Since we were unaware of the bioavailability of the oral cebranopadol solution, PK parameters V (compartmental volume) and CL (clearance) are scaled against F (bioavailability).
To eliminate a possible hysteresis between plasma concentration and effect, an effect compartment was postulated that equilibrates with the plasma compartment with half-life t½ke0 (i.e., the blood-effect-site equilibration half-life). Distinct values for t½ke0 were estimated for analgesia and respiratory depression: t½ke0A for analgesia and t½ke0R for respiratory depression.
Ventilation was modeled as19
where Effect is the effect at time t (minute ventilation); Emax, maximum or predrug effect (baseline ventilation); Emin, minimum effect (Emin = 0 indicates that apnea is reached); effect-site concentration at time t (CE[t]); C50, effect-site or steady-state concentration causing 50% depression of ventilation; and γv, a shape parameter.
where Pain response (t) is the stimulus intensity at which a pain threshold or pain tolerance response occurs at time t; Baseline response, predrug stimulus intensity at which a pain threshold or pain tolerance response occurs; C25, effect-site or steady-state concentration causing 25% stimulus intensity for a response (threshold or tolerance); and γp, a shape parameter. Pain threshold and tolerance were simultaneously analyzed.
Model parameters were assumed to be log-normally distributed. For all effect parameters, the residual term was additive. P values less than 0.01 were considered significant.
Visual Predictive Check.
Visual predictive checks (VPCs; using 1,000 simulations) were performed to assess the adequacy of the description of both fixed and random effects by simulating data using the models and calculating their 2.5th, 50th, and 97.5th percentile at all sampling times.
We constructed utility functions (UFs) as previously described.19 In brief, the utility of drug effect was defined as the probability of obtaining the desired effect minus the probability of obtaining a side effect. Here, we present the UF of cebranopadol as a function of its effect-site concentration:
where P(A) is the probability for analgesia; P(R), probability for respiratory depression; and α and β, threshold values. So the UF for at least 50% respiratory depression and an increase in current of at least 50% above baseline (i.e., an increase in analgesia by 50%) equals P(A > 50%) − P(R > 50%). The probabilities P were calculated by simulating the PD models with their estimated typical values and interindividual variabilities (using 100,000 simulations) and counting how often the thresholds α and β were crossed. These calculations were performed in R (The R Foundation for Statistical Computing; http://www.r-project.org, accessed January 18, 2017). The value of UF ranges from +1 to −1. UF < 0 indicates that the probability for respiratory depression exceeds the probability for analgesia; UF > 0 indicates that the probability for analgesia exceeds the probability for respiratory depression. UF > 0.4 or UF < −0.4 indicate large effects, while values −0.2 < UF < 0.2 indicate small effects, i.e., absence of selectivity of effect with similar probabilities for analgesia and respiratory depression.19
Physiologic Analysis of Respiratory Depression.
where VE is the inspired minute ventilation; VR, resting ventilation (or ventilation obtained without any addition of inspired carbon dioxide); G, ventilatory carbon dioxide sensitivity; Ptco2, carbon dioxide concentration at the site of chemoreception (and instantaneously related to VE); and Pbco2, baseline value of Ptco2. To cover the delay of Ptco2 relative to Petco2, it is assumed that
where τ is the time constant of the dynamic ventilatory response to carbon dioxide (i.e., the combined central and peripheral chemoreflex loops).
A population analysis was performed in NONMEM VII. The data were expressed as median ± standard error of the estimate.
The occurrence of the following side effects was scored (yes or no) in the study: nausea, vomiting, somnolence, sedation, and headache. Oxygen saturation was monitored continuously, and the lowest value obtained in-between respiratory measurements were noted.
All 12 subjects completed the study without serious adverse events. All subjects were white men with a mean age of 22.1 yr (range, 19 to 27 yr), height 186 cm (range, 176 to 200 cm), weight 80.5 kg (range, 64 to 111 kg), and body mass index of 23.2 kg/m2 (range, 21 to 28 kg/m2).
The mean plasma concentration of cebranopadol over time is given in figure 1. The final models used to analyze the PK data are given in figure 2 and consisted of two absorption compartments (VA1 and VA2), one central compartment (V1), and one peripheral compartment (V2). A delay in the absorption was modeled with a tank-in-series approach, i.e., two absorption compartments in series, both with rate constants KA (for absorption from the first into the second absorption compartment and from the second absorption compartment into the central [systemic] compartment). The PK parameter estimates are given in table 1. No effect of the covariates weight, height, and age was observed on any of the model parameters. Two error terms were incorporated into the model, one relative (σ1) and one additive (σ2).
Goodness-of-fit plots (measured vs. individual predicted concentrations and individual weighted residuals vs. time) are given in figure 2, B and C. Examples of data fits are given in figures 3 and 4. The goodness-of-fit plots and inspection of the individual data fits indicate that the PK models adequately describe the PK data. The VPC is given in figure 2D.
PD Data Analysis
Mean ventilatory, pupil size, and pain responses are given in figure 1, B–D. Cebranopadol had a typical opioid effect on all three endpoints with increases in pain threshold and tolerance, miosis, and respiratory depression. Best, median, and worst PD data fits (and corresponding PK fits) for ventilation and pain responses are given in figures 3 and 4. The goodness-of-fit plots are given in figure 5, A–D. Goodness-of-fit plots and inspection of the individual data fits indicate that the PD models adequately describe the PD data. Model parameter values are given in table 2. The most important observation in the respiratory data is that minimum ventilation, Emin, has a significant value greater than zero, 4.9 ± 0.7 l/min with 95% CI 3.5 to 6.6 l/min. Estimates of the shape parameter γ were not significantly different from 1 for either respiratory depression (γv = 1.1 ± 0.1) or analgesia (γp = 1.4 ± 0.5). Consequently, both parameters were fixed to 1. No effect of the covariates weight, height, and age was observed on any of the model parameters. The results of the VPCs for ventilation, pain threshold, and pain tolerance are plotted in figure 5, E–G.
The UF for P(A > 50%) and P(R) at a range of thresholds (from greater than 50% to greater than 80%) are given in figure 6. The higher the threshold for respiratory depression, the more positive the UF is, which is putatively due to the ceiling in respiratory depression (table 2). At 50% threshold values (i.e., U = P(A > 50%) − P(R > 50%), red line in fig. 6A), the UF becomes negative at cebranopadol concentrations greater than 130 pg/ml with a nadir of −0.3 at 220 pg/ml after which it slowly returns to 0. The negative part of the function signifies a greater probability of at least 50% respiratory depression than for an increase in analgesia by at least 50% over the concentration range greater than 130 pg/ml.
Physiologic Analysis of Respiratory Depression
The results of the analysis are given in figure 7. Resting ventilation (i.e., ventilation without any addition of inspired carbon dioxide) showed an average decrease over time of 30% that lasted for 9 h from drug intake. The decrease in ventilation was due to a decrease in tidal volume without any change in breathing frequency. The changes in resting Petco2 were small with a nonsignificant increase at the end of the study (t = 10 h) of 3.8 mmHg (0.4 kPa) from 39.0 to 42.8 mmHg (from 5.2 to 5.7 kPa).
Cebranopadol caused a large and significant reduction of the slope of the VE–Pco2 response curve (G in eqn. 2) with an average reduction of 70% throughout the study period. The intercept of the VE–Pco2 response curve with the x-axis was shifted to the left by cebranopadol from 32.3 ± 3.0 mmHg at baseline to 26.3 ± 4.8 mmHg (mean ± SD) at t = 3 h. An example of the effect of cebranopadol on the response is given in figure 8 for subject 208 (baseline and t = 3 h). The estimated value of time constant τ was 1.7 ± 0.2 min.
No unexpected or serious side effects occurred. Lowest oxygen saturation was 96.9% occurring 5 h after cebranopadol administration. Sedation occurred in eight subjects, somnolence in three subjects, and nausea, dizziness, and headache in two subjects each. One subject vomited.
The results of the PK analysis are expressed as a function of cebranopadol bioavailability (F in table 1), and parameter estimates are realistic considering a two-compartment model for disposition and a sample time of 32 h. For example, if we assume a bioavailability of 40% (data on file; Grünenthal GmbH), V1 would equal 492 l and CL1 56 l/h. The estimated variability of F was 26% (table 1), which we consider well acceptable. The PD parameter estimates show an almost threefold greater potency for respiratory depression than for analgesia as measured by the experimental pain model (table 2). The t½ke0 of 1.2 h (respiratory depression) and 8.1 h (analgesia) indicate that the temporal profile of these two endpoints are pulled apart with a more rapid onset and offset of respiratory effect against a much slower but longer acting analgesic effect. In order to enhance the initial moment of analgesia in nociceptive pain, a greater dose than used by us is required.
For both respiratory and analgesia models, we fixed the values of the Hill parameter γ to 1 as both parameters were not significantly different from 1. This indicates that both parameters were estimable but with great uncertainty. The complexity of any model is governed by the information in the data, and hence to prevent overinterpretation of the data, we fixed both parameters. Fixing parameters to a certain value (i.e., 1 or 0) is common practice in modeling studies and adheres to the principle of parsimony. One other limitation of the study is the use of venous rather than arterial plasma concentrations in the PK analysis. Large concentration differences may cause misspecifications of the PK and PK–PD models, when linking the venous concentration to effect.24 Further studies are needed to assess whether a significant cebranopadol arteriovenous concentration difference is present and whether this affects parameter estimates derived from venous sampling.
Cebranopadol is a synthetic drug derived from spiroamine compounds,14 which targets NOP and classical opioid receptors.11,15 It has affinity for all four opioid receptor subtypes with one third the affinity for the κ-opioid receptor and one twentieth the affinity for the δ-opioid receptor compared to the MOP receptor (MOP Ki = 0.9 nM, NOP Ki = 0.7 nM, κ-opioid receptor Ki = 2.6 nM, δ-opioid receptor Ki = 18 nM; data from human receptors).15 Animal studies indicate a major advantage of simultaneous NOP and MOP receptor activation, causing synergistic enhancement of antinociception without side effects.10 The NOP receptor is an atypical opioid receptor (e.g., it is not antagonized by naloxone) first described in 199525,26 and has been implicated in a variety of biologic functions including nociception, reward, cardiovascular control, and immunity.27 Although the systemic administration of NOP receptor agonists produces analgesia, local administration at supraspinal sites causes hyperalgesia (probably via activation of nociceptive off-cells in the brainstem), while local administration at the spinal level is associated with potent analgesia through inhibitory actions at dorsal horn nociceptive neurons.27 The pronociceptive effect of the NOP receptor system, which is also confirmed from data obtained in mice without NOP receptor,9 led us to speculate previously that the ceiling effect observed in respiratory depression from agonists at the NOP and MOP receptors is related to reduced respiratory depression due to NOP receptor activation.8,28 Our current data and the related animal work are in further agreement with this hypothesis.11 In rats, Linz et al.11 show that cebranopadol produces analgesia (tail-flick test) and respiratory depression (as measured from arterial Po2 and Pco2). For analgesia, the ED50 (dose causing a half-maximum effect) was reached at 7.4 μg/kg. For respiratory depression at doses causing full analgesic responses (up to 17.1 μg/kg), the increase in arterial Pco2 remained less than 5 mmHg with arterial Po2 values greater than 70 mmHg. We argue that this is an indication of limited respiratory depression or ceiling of cebranopadol. Furthermore, blocking the NOP receptor with the selective NOP receptor antagonist J-113397 further increased cebranopadol-induced respiratory depression, an effect that was fully antagonized by naloxone.11 If one interprets our current human data in light of these animal experiments, the restricted respiratory effect that we observed in our modeling study seems likely due to the interactive effect of cebranopadol at the MOP and NOP receptors, with NOP receptor activation reducing the respiratory effect of MOP receptor activation.12,13 If such behavior persists at higher cebranopadol doses, we may conclude that this analgesic drug produces ceiling in respiratory depression in humans like that observed in animals. We remain uninformed on the presence of a plateau in analgesia. However, there are no indications from animal data that ceiling in analgesia exists.11,15 Also, in our analgesia data, there were no indications for a plateau as assessed by our analysis of the pain responses with an inverted sigmoid function (note that our analgesia PD model contains a sigmoid Emax model as electric currents are modeled as specified in Sarton et al.22 ). Further investigations at higher cebranopadol doses are needed to come to definite conclusions.
In several PK–PD studies, we previously characterized the respiratory behavior of the full MOP receptor agonists alfentanil,6 fentanyl,17,26 morphine,29,30 and morphine-6-glucuronide29 and observed—as expected—full respiratory depressant effect with a value of Emin not significantly different from zero. We relate this to the activation of MOP receptors expressed on the respiratory neurons in the brainstem, for example in the pre-Bötzinger complex, an area that is involved in respiratory rhythmogenesis.31 In contrast, cebranopadol has an Emin greater than 0 at the tested oral dose of 600 μg. While such behavior was considered previously a characteristic of partial agonism at the MOP receptor, our current observations suggest that activity at the NOP receptor may be held responsible. NOP activation prevents the full respiratory depressant effect of the MOP receptor activation. As respiratory depression is potentially a life-threatening complication of high-dose opioid treatment,1 this is theoretically a major advantage beyond opioid analgesics with full respiratory depressant behavior.
Opioid-induced respiratory depression is best considered in context of its beneficial effect.19,32 In order to quantify the risk–benefit of cebranopadol, we constructed utility or safety functions, which were defined by the probability of analgesia minus the probability of respiratory depression (eqn. 3). The UF at threshold values of 50% (α and β in eqn. 3) was negative at cebranopadol effect-site concentrations greater than 130 pg/ml (fig. 6A, dark red line). At higher threshold values for respiratory depression (β), the UF becomes positive, with values approaching 1 at a threshold for respiratory depression greater than 80%. This is due to the ceiling effect that precluded respiratory depression greater than 70% (table 2). As was demonstrated in rodents, cebranopadol is superior in counteracting neuropathic rather than nociceptive or inflammatory pain.15 A 10-fold potency difference was observed for reduction of neuropathic pain compared to nociceptive pain. Extrapolating these findings to humans would suggest a greater potency of cebranopadol in treating neuropathic pain. To get an indication of the effect of a greater cebranopadol analgesic potency, we simulated UFs with a twofold greater cebranopadol potency for analgesia (fig. 6B) and now observed predominantly positive UFs. Considering the fact that cebranopadol seems best suited for treatment of neuropathic pain, our use of a nociceptive pain model and testing just one dose are evident limitations. Still, this phase 1 study is an initial approach in understanding the respiratory behavior of a single dose of cebranopadol in relation to its analgesic effect, and these first results encourage further studies.
Apart from the PK–PD study, we performed a physiologic analysis of the respiratory data (fig. 7). We observed a 30% reduction of baseline ventilation, a 70% depression of the slope (G in eqn. 4) without an increase in Petco2. This latter finding is best explained by a cebranopadol-induced downward movement of the metabolic hyperbola, suggestive of a reduction in metabolism or energy expenditure. The NOP receptor plays an important role in thermoregulation and energy expenditure through actions at the hypothalamic level.33 Our observations may additionally be explained by a reduction in physiologic dead space or an increase in end-tidal to arterial gradient. We have no indication that any of these two possibilities occurred in our set of healthy volunteers. Nonetheless, these data indicate that respiratory depression may be significant even when Petco2 remains in the normal range and hence that observational studies (i.e., studies without inhaled carbon dioxide) are often of limited value in the assessment of the respiratory effects of drugs, especially when these drugs also negatively affect metabolic rate. An interesting observation is the finding that cebranopadol did not affect respiratory rate (fig. 7D). In this respect, cebranopadol differs from pure MOP receptor agonists.34
In conclusion, we observed that cebranopadol, a MOP and NOP receptor agonist, produces respiratory depression with an estimated Emin value of greater than 0 l/min, suggestive of ceiling. While this is an encouraging beneficial effect, studies at higher cebranopadol doses are needed to confirm such behavior at higher brain concentrations. Finally, it is important to note that the laboratory assessment of ventilatory depression in healthy volunteers may not always correlate with ventilatory depression in the clinical setting.
This study was supported in part by Grünenthal GmbH, Aachen, Germany.
Dr. Neukirchen and Dr. Bothmer are employees of Grünenthal GmbH, Aachen, Germany. Dr. Dahan received speaker and consultancy fees from Grünenthal GmbH. The other authors declare no competing interests.