Brain PET imaging using 11 C-flumazenil and 11 C-buprenorphine does not support the hypothesis of a mutual interaction between buprenorphine and benzodiazepines at the neuroreceptor level

Abstract

Among opioids, buprenorphine presents a favorable safety profile with a limited risk of respiratory depression. However, fatalities have been reported when buprenorphine is combined to a benzodiazepine. Potentiation of buprenorphine interaction with opioid receptors (ORs) with benzodiazepines, and/or vice versa, is hypothesized to explain this drug-drug interaction (DDI). The mutual DDI between buprenorphine and benzodiazepines was investigated at the neuroreceptor level in nonhuman primates (n = 4 individuals) using brain PET imaging and kinetic modelling. The binding potential (BP_ND) of benzodiazepine receptor (BzR) was assessed using ¹¹C-flumazenil PET imaging before and after administration of buprenorphine (0.2 mg, i.v.). Moreover, the brain kinetics and receptor binding of buprenorphine were investigated in the same individuals using ¹¹C-buprenorphine PET imaging before and after administration of diazepam (10 mg, i.v.). Outcome parameters were compared using a two-way ANOVA. Buprenorphine did not impact the plasma nor brain kinetics of ¹¹C-flumazenil. ¹¹C-flumazenil BP_ND was unchanged following buprenorphine exposure, in any brain region (p > 0.05). Similarly, diazepam did not impact the plasma or brain kinetics of ¹¹C-buprenorphine. ¹¹C-buprenorphine volume of distribution (V_T) was unchanged following diazepam exposure, in any brain region (p > 0.05). To conclude, our PET imaging findings do not support a neuropharmacokinetic or neuroreceptor-related mechanism of the buprenorphine/benzodiazepine interaction.

Keywords

Allosteric neuroimaging neuroreceptor positron emission tomography respiratory depression

Introduction

The opioid overdose crisis started in 1999 with the increase of fatalities attributed to the irrational and over-prescription of opioids.¹ In this context, buprenorphine has become the leading opioid maintenance treatment prescribed in the US, ranking before methadone.² First-line prescription of buprenorphine is related to its respiratory safety profile. Compared to methadone, which is a full mu-opioid receptor (MOR) agonist, buprenorphine is described as a partial MOR and a kappa-opioid receptor (KOR) antagonist, resulting in ceiling effects regarding respiratory depression but not analgesia, even at high doses.^3,4 The slow association/dissociation of buprenorphine at its target receptors may also contribute to its particular safety profile.⁵

However, as soon as marketed in Europe, where it was first approved in the mid-1990s, deaths involving buprenorphine were reported.⁶ Forensic studies have clearly shown that most of deaths occurred when buprenorphine was co-administered with another psychotropic drug, primarily a benzodiazepine.^6
–8 Consistently, animal studies have shown that buprenorphine or benzodiazepines alone do not induce respiratory depression while the combination of both drugs does,⁹ strongly suggesting a drug-drug interaction (DDI), which mechanisms still remains unclear. Several preclinical studies, mainly performed in rodents, do not support the hypothesis of peripheral pharmacokinetic DDI. Buprenorphine is metabolized by cytochrome P450 (CYP) 3 A to norbuprenorphine, which is a ∼10-fold more potent MOR agonist compared to buprenorphine in vitro.¹⁰ In the rat, CYP3A induction increased the proportion of circulating norbuprenorphine but did not worsen buprenorphine-induced respiratory depression.¹¹ This experimental finding could be explained by the low brain penetration of norbuprenorphine, which transport across the blood-brain barrier (BBB) is negligible compared to the transport of buprenorphine itself.^12,13 Therefore, it can be hypothesized that the DDI between buprenorphine and benzodiazepines may occur at the CNS level. This may involve interaction, cross-regulation or allosteric modulation of the neuroreceptors mediating the effects of each respective drug.

In vivo quantification of benzodiazepine receptors (BzR) is possible with Positron Emission Tomography (PET) imaging using ¹¹C-flumazenil, a reversible BzR antagonist.¹⁴ ¹¹C-flumazenil has been used to quantify the engagement of BzR by different pharmacological doses of benzodiazepines.^15,16 Moreover, ¹¹C-flumazenil PET imaging has been widely used to investigate regulation of BzRs in various pathological conditions such as epilepsy,¹⁷ Alzheimer disease,¹⁸ multiple sclerosis,¹⁹ or ischemic stroke.²⁰ ¹¹C-flumazenil benefits from extensive research to validate pharmacokinetic models and quantitatively describe the availability of BzR in vivo.²¹ In this setting, the partial saturation method has been validated in baboons ²² and humans,^23,24 aiming to estimate both the density and affinity to BzRs from a single-scan session.

Several PET radioligands have been developed for quantification of opioid receptors (ORs) in the living brain.²⁵ The MOR ¹¹C-carfentanil has been useful to investigate the dose and time-dependent occupancy of MORs by buprenorphine in humans.²⁵ We recently validated ¹¹C-buprenorphine in rodents and non-human primates as a mean to specifically address the neuropharmacokinetics and binding of buprenorphine to its CNS targets in vivo.^26,27 Blocking experiments performed in rats suggested that the specific binding of ¹¹C-buprenorphine in the brain predominantly corresponds to its interaction with MORs rather than with other OR subtypes.²⁶

In the present study, the hypothesis of a mutual DDI between buprenorphine and diazepam, a widely used benzodiazepine, was investigated at the neuroreceptor level using brain PET imaging, according to a cross-over design in non-human primates. The impact of a pharmacological dose of buprenorphine on BzR was assessed using ¹¹C-flumazenil PET imaging. Then, the impact of a pharmacological dose of diazepam on the brain kinetics and receptor binding of buprenorphine was investigated in the same individuals using ¹¹C-buprenorphine PET imaging.

Material and methods

Chemicals and radiochemicals

Buprenorphine hydrochloride for intravenous (i.v.) injection (0.3 mg/mL) was obtained from Axience (Pantin, France). Naloxone hydrochloride for i.v. injection (0.4 mg/mL) was obtained from Aguettant (Lyon, France). Diazepam for intravenous (i.v.) injection (5 mg/mL) was obtained from Roche (Boulogne-Billancourt, France). Ketamine was obtained from Virbac (Carros, France). Propofol was purchased from Fresenius laboratory (Sèvres, France). Isoflurane was obtained from Abbvie (Rungis, France). Flumazenil (0.1 mg/mL) was obtained from Mylan (Paris, France).

¹¹C-Buprenorphine was synthesized in-house according to the method described by Lever et al.²⁸ with minor modifications previously described.²⁶ Ready-to-inject ¹¹C-buprenorphine (1.8–3.5 GBq, decay-corrected radiochemical yield: 15%) was obtained with a radiochemical purity above 98% and a molar activity of 28 GBq/μmol at end of synthesis (EOS).

¹¹C-Flumazenil was synthesized based on a previously published method.²⁹ Ready-to-inject ¹¹C-flumazenil (1.1–2.8 GBq, decay-corrected radiochemical yield: 23%) was obtained with a radiochemical purity above 98% and a molar activity of 21 GBq/μmol EOS.

Animals

All animal of the procedures were used in accordance with the recommendations of the European Community for the care and use of laboratory animals (2010/63/UE) and the French National Committees (French Decret 2013–118). Experimental protocols were validated by a local ethics committee for animal use (CETEA/A15–002 and A18–065) and approved by the French government. Four adult male Rhesus macaques (Macaca Mulatta; weight = 7.2 ± 1.2 kg, age = 5.6 ± 0.5 years during the study), which were obtained from Silabe (Simian Laboratory Europe, France). Animals were housed and acclimatized for at least 7 days before experiments. Macaques were housed under standard experimental conditions: room temperature (22 ± 1°C); light/dark cycle (14 h light/10 h dark); free access to water and food (fresh fruit preparations, pellets). Animals were kept in social groups with enrichment (toys and structures for resting, hiding and climbing). Animal data are reported in compliance with the ARRIVE guidelines 2.0.

Study design

The present study is composed of two sequential sets of experiments with ¹¹C-buprenorphine and ¹¹C-flumazenil, respectively designated as study 1 and study 2. All four macaques participated to study 1 and study 2, Allowing for paired comparison of outcome data. The number of individuals was set based on previous studies using a similar design.^26,30 A minimum interval of two weeks was respected between two PET imaging experiments in the same individual.

Study 1 consisted of two ¹¹C-buprenorphine PET scans, with or without pretreatment with diazepam (10 mg, i.v. bolus), administered 15 min before microdose ¹¹C-buprenorphine injection (n = 4).

Study 2 consisted of two ¹¹C-flumazenil PET scans with or without pre-treatment with unlabeled buprenorphine (0.2 mg, i.v. bolus), administered 15 min before ¹¹C-flumazenil injection (n = 4). ¹¹C-Flumazenil was systematically co-injected with unlabeled flumazenil (10 µg/kg) according to the partial saturation method.²⁴

Baseline scans were performed before pretreatement scans, assuming a possibility for long-term regulation of investigated neuroreceptor by the pretreatment.

Arterial input function and metabolism

During PET acquisition, in both studies 1 and 2, arterial blood samples (500 µL) were obtained at selected times after the radiotracer injection. Samples were centrifuged (5 min; 2,054 g; 4 °C) and the supernatant (200 µL) was gamma-counted for total plasma radioactivity. Additional plasma samples were obtained at 0, 5, 10, 15, 30, 60, and 90 min to measure the percentage of parent (unmetabolized) radiotracer using radio-HPLC and state of-the-art methodology.³¹ The fraction of parent radiotracer in each sample was used to generate the metabolite-corrected arterial input function.

PET acquisition procedure

First, each monkey underwent an anatomical T1-weighted brain MR scan using an Achieva 1.5 T scanner (Philips Healthcare, Suresnes, France) on ketamine anesthesia (intramuscular injection, i.m.). PET acquisitions were performed using a HR+ Tomograph (Siemens Healthcare, Knoxville, TN, USA) in anesthetized macaques, as previously described.³² Briefly, the macaque received ketamine (10 mg/kg, i.m.) to induce anesthesia. After intubation in supine position, venous catheters were inserted for radiotracer injection (sural vein), propofol infusion (sural vein), and drug injection for pre-treatment conditions (brachial vein). Another catheter was inserted in the femoral artery for arterial blood sampling. Macaques were positioned under the camera before the administration of a 2 mL i.v. bolus of propofol followed by a 1 mL/kg/h i.v. infusion under oxygen ventilation. The heart rate, oxygen saturation, respiratory rate, and endtidal CO₂ were monitored during the PET scan procedure.

In study 1, macaques were i.v. injected with a microdose of ¹¹C-buprenorphine (280 ± 60 MBq, molar activity at the time of injection (MA_inj) = 19.3 ± 8.6 GBq/µmol).

In study 2, macaques were i.v. injected with ¹¹C-flumazenil (277 ± 44 MBq, MA_inj = 10.8 ± 4.9 GBq/µmol), co-injected in the same syringe with 10 µg/kg of unlabeled flumazenil to enable quantification of receptologic parameters using the partial saturation method, as previously described in humans.²⁴

Imaging data reconstruction and segmentation

The person analysing the data was not aware of the baseline/pretreatment status of each animal. PET data were analyzed using PMOD software (PMOD Technologies, Zürich, Switzerland). PET images were co-registered to corresponding T1-weigted MR images for each macaque. A macaque T1-weighted MR template³³ was generated from individual MR images. Transformation matrices were thereafter applied to the segmentation obtained from the template to generate time-activity curves in 13 brain structures selected as volumes of interest (VOI): frontal, orbital, cingulate, temporal, parietal and occipital cortices, and caudate nucleus, putamen, amygdala, thalamus, hypothalamus, pons and cerebellum.

PET quantification and data modeling

For ¹¹C-buprenorphine and ¹¹C-flumazenil PET experiments, kinetics of radioactivity in the selected brain regions and in arterial plasma samples were corrected for radioactive decay, injected dose of radioactivity and animal weight, and expressed in standardized uptake values (SUV). Brain exposure was estimated by the area under the time-activity curve (AUC) of the regional brain kinetics of PET radioligands.

In study 1, the brain distribution of ¹¹C-buprenorphine (V_T; mL/cm³) was estimated using the Logan plot analysis taking the metabolite-corrected input function into account.³⁴

In study 2, the regional brain distribution of ¹¹C-flumazenil was estimated by V_T. The binding potential of ¹¹C-flumazenil was also estimated using the simplified reference tissue model (SRTM) using the pons as the reference region (BP_ND). Moreover, a graphical analysis based on the partial saturation method was tested for estimation of i) B_Avail, which represents the concentration of receptors in the VOI, estimated graphically with the Scatchard analysis as the X-axis intercept and ii) _AppKd represents the in vivo affinity, estimated graphically with the Scatchard analysis as the Y axis-intercept, as previously described.²⁴

Statistical analysis

Comparisons between the different experimental conditions were performed using R³⁵ and GraphPad Prism (v9.5, USA) softwares. Plasma exposures were compared using a Wilcoxon matched pairs test. Outcome parameters of brain PET data were compared using a Two-Way Repeated Measure ANOVA, assuming that both factors “Brain regions” and “Treatment” were matched for each individual. The distribution of residues was normal (Levene’s test) and homogeneous (Bartlett’s test). A Tukey’s post-hoc was performed for multiple comparisons test. A result was deemed significant when a two-tailed p value was <0.05.

Results

Impact of diazepam pretreatment on the kinetics of ¹¹C-buprenorphine

In the study 1, diazepam did not impact the metabolism of ¹¹C-buprenorphine. The proportion of parent ¹¹C-buprenorphine was not different in the absence and presence of diazepam pre-treatment, at any time-point (Figure 1(a)). Plasma concentrations of parent ¹¹C-buprenorphine peaked rapidly, followed by a fast washout of radioactivity to reach equilibrium (∼0.12–0.20; SUV unit) at 15 min after ¹¹C-buprenorphine injection (Figure 1(b)). Diazepam did not impact the plasma exposure of ¹¹C-buprenorphine with AUC_plasma = 26.9 ± 1.9 and 27.3 ± 1.7 SUV.min in the absence and presence of diazepam, respectively (p = 0.89, Figure 1(c)).

Figure 1.

Metabolism and plasma pharmacokinetics of ¹¹C-buprenorphine (¹¹C-BUP) and ¹¹C-flumazenil (¹¹C-FMZ), with or without diazepam (DZP) or buprenorphine (BUP) pre-treatment, respectively. (a) Percentage of parent ¹¹C-buprenorphine vs time in plasma. Arterial input function of unmetabolized ¹¹C-buprenorphine (b) and corresponding areas under the curve (c). Percentage of parent ¹¹C-flumazenil vs time (d). Arterial input function of unmetabolized ¹¹C-flumazenil (e) and corresponding areas under the curve (f). ¹¹C-Flumazenil vs time is shown in a and c. Data are shown as mean ± SD (n = 4). Statistical significance was set at p < 0.05 with ns: non-significant.

The brain kinetics and distribution of ¹¹C-buprenorphine in macaques was consistent with previous findings²⁶ (Figure 2), with a significant difference between brain regions (p < 0.001) and maximal distribution in the OR-rich regions such as the putamen (V_T = 11.2 ± 1.6 mL.cm⁻³) and caudate nucleus and (V_T = 10.5 ± 1.4 mL.cm⁻³), and minimal distribution in the regions with poor expression of ORs such as the cerebellum (V_T = 5.0 ± 0.7 mL.cm⁻³) and the occipital cortex and (V_T = 5.0 ± 0.9 mL.cm⁻³, Figure 2). Pretreatment with diazepam slightly increased the V_T of ¹¹C-buprenorphine in some brain regions, although statistical significance was not reached (p > 0.05, Figure 2). Maximum change in the V_T was observed in the cerebellum (+8.8 ± 14.8%), the orbital cortex (+7.4 ± 11.2%), the temporal cortex (+7.4 ± 13.5%) and the pons (+7.0 ± 8.1%).

Figure 2.

PET data of ¹¹C-buprenorphine obtained with or without diazepam (DZP) pre-treatment in macaques. (a) ¹¹C-Buprenorphine brain kinetics in the putamen and occipital cortex. (b) Regional V_T estimated using the Logan plot analysis in each investigated brain region, with or without diazepam pre-treatment (b). Data are shown as mean ± SD (n = 4) with ns = non-significant.

Impact of buprenorphine pretreatment on the kinetics of ¹¹C-flumazenil

Buprenorphine did not impact the plasma metabolism of ¹¹C-flumazenil according to the results of the study 2. At each of the sampling times and up to 90 min, the proportion of the parent ¹¹C-flumazenil fraction is not different with or without pre-treatment with buprenorphine (Figure 1(d)). Buprenorphine did not impact the plasma exposure of ¹¹C-flumazenil (Figure 1(e)), with AUC_plasma of 57.4 ± 5.5 and 55.0 ± 15.9 SUV.min with or without buprenorphine pre-treatment, respectively (p = 0.88, Figure 1(f)).

Regional V_Ts of ¹¹C-flumazenil were significantly different across region (p < 0.001), with the lowest uptake in the pons. The V_T of ¹¹C-flumazenil in the pons was not different with or without buprenorphine pre-treatment (p = 0.99). Regional BP_ND using the pons as the reference region were significantly different among brain regions (p < 0.001) but were not different with or without buprenorphine pretreatment (p = 0.56) (Figure 3). In most experiments, linearity was not obtained for the Scatchard plot so that reliable estimation of B_Avail and _AppKd was not possible.

Figure 3.

PET data of ¹¹C-flumazenil obtained with or without buprenorphine (BUP) pre-treatment in macaques. (a) ¹¹C-flumazenil brain kinetics for pons and frontal cortex regions. (b) Regional V_T estimated using the Logan plot analysis in each investigated brain region, with or without diazepam pre-treatment and (c) Regional BP_ND measured using the SRTM with the pons as the reference region for each investigated brain region. Data are shown as mean ± SD (n = 4) with ns = non-significant.

Discussion

Previous studies have shown that combination of buprenorphine and benzodiazepines may lead to severe and often fatal respiratory depression, suggesting a DDI which mechanism remains poorly understood.⁹ The functional interplay between the GABAergic and opioidergic systems, with the presence of MORs and BzRs on the same neurons, suggest an interactive role in controlling brain function.³⁶ Direct agonist effect of diazepam on MORs, and/or effect of buprenorphine on BzRs, is very unlikely, although this remained to be addressed in vivo.^9,37 Moreover, the hypothesis of a positive allosteric modulation of BzRs and/or MORs by each respective drug remained to be investigated. The possibility for an allosteric modulation of MORs, which increased the affinity of orthosteric ligands has been reported.^38,39 Translational studies have shown that some drugs like tiagabine or yohimbine may act as positive allosteric modulators of BzRs, which was confirmed in vivo using ¹¹C-flumazenil PET imaging.^40
–42 A cross-over PET study using ¹¹C-buprenorphine and ¹¹C-flumazenil was therefore conducted in non-human primates to address the hypothesis of a mutual DDI between diazepam and buprenorphine at the neuroreceptor level.

The impact of diazepam on the brain kinetics of buprenorphine was first investigated using ¹¹C-buprenorphine PET imaging. Most imaging studies focusing on OR use dedicated PET radioligands such as ¹¹C-carfentanil, which offer an insight into regulation of MORs in physiological conditions and disease settings.²⁵ In previously conducted rats and macaques studies, we showed that ¹¹C-buprenorphine is suitable to investigate the brain exposure and receptor binding of buprenorphine, since predominantly driven by its affinity for MORs.²⁶ Therefore, this pharmacokinetic imaging approach was performed to more specifically investigate the impact of diazepam on both the brain exposure and binding to ORs in vivo. Only a microdose (∼10 µg) of ¹¹C-buprenorphine could be administered to avoid respiratory depression in the presence of diazepam. The selected dose of diazepam used to precipitate the DDI (i.e. 10 mg i.v. bolus, 15 min before PET) corresponded to half the dose used in rats in this situation,⁹ and five to ten-fold the human dose when i.v. injected. A previous PET study using ¹¹C-flumazenil in baboons showed that a similar dose of diazepam (1.4 mg/kg), administered 20 min before PET, was associated with ∼70% occupancy of BzR in baboons.³⁰ This suggests that the buprenorphine-diazepam interaction, which is observed in the therapeutic range,⁷ occurs in the situation of partial occupancy of BzRs in humans. Our results and kinetic modeling data suggested no or limited impact of diazepam on the brain exposure and binding of ¹¹C-buprenorphine to brain regions, regardless of the local density of MORs, thus not supporting the hypothesis of an enhanced brain delivery of buprenorphine or regulation of the availability of MORs in the presence of diazepam. Our finding obtained in macaques is consistent with a previous study in rats showing that diazepam (20 mg/kg, s.c.) does not significantly change the brain kinetics of ¹¹C-buprenorphine.⁹ Of note, diazepam did not increase the binding of ¹¹C-buprenorphine in the pons which contains important MOR-dependent loci for respiratory control such as the parabrachial nuclei or the Kölliker fuse.⁴³

The impact of a pharmacological dose of buprenorphine on the availability of BzR was thereafter investigated using ¹¹C-flumazenil PET imaging. The selected dose of buprenorphine used to precipitate the DDI (0.2 mg i.v. bolus, which corresponds to 0.03 mg/kg in our animals) was administered in anesthetized macaques and shown to achieve ∼80% receptor occupancy.²⁶ ¹¹C-flumazenil is widely used in animals and humans and benefits from extensive research in pharmacokinetic modelling with the aim to better estimate the receptologic profile of BzR in vivo.²² Moreover, flumazenil is a BzR antagonist and a pharmacological dose of unlabeled flumazenil could be safely administered after an unlabeled buprenorphine to explore DDIs using ¹¹C-flumazenil PET, described as the partial saturation approach.²⁴ The selected dose of unlabeled flumazenil (10 µg/kg) used for partial saturation was previously shown to occupy ∼60% of available BzRs in baboons.³⁰ This corresponds to the recommended level for using the partial saturation method.²⁴ Consistently, regional BP_ND are lower than those reported in macaques using microdose ¹¹C-flumazenil, in the absence of partial saturation.⁴⁴ In the partial saturation condition, level of available receptor is still substantial (∼40%).⁴⁵ However, it may be hypothesized that partial saturation, which reduces the dynamic range of available receptors, may reduce the sensitivity of ¹¹C-flumazenil PET to detect changes in BzR availability. Nevertheless, buprenorphine did not change the regional V_T of ¹¹C-flumazenil. Neither was BP_ND, which was estimated using the pons as a reference region with low expression of BzRs, increased by buprenorphine pretreatment. Although microdose ¹¹C-flumazenil PET has not been performed to confirm this result, this suggests a limited impact of buprenorphine pre-treatement on the brain distribution and binding of ¹¹C-flumazenil to BzRs.

The partial saturation approach proposed by Delforge et al. has been described in baboons²² and humans²⁴ aiming to estimate the density and the affinity of ¹¹C-flumazenil from a single scanning section.⁴⁶ However, linearity of the Scatchard plot was not obtained in many macaques, in both baseline and buprenorphine conditions, so that B_Avail and _AppKd could not be reliably estimated (data not shown). The selected dose of unlabeled flumazenil used for partial saturation may be to low to achieve equilibrium during the scan in macaques. Further work is probably needed to optimize the dose of unlabeled flumazenil for partial saturation in the different animal species.

This study suggests a negligible impact of diazepam on the brain kinetics of buprenorphine, and the absence of regulation of MORs. Buprenorphine seems also not to impact the availability of BzRs. It is highly however likely that the interaction between buprenorphine and diazepam occurs at the level of their intraneuronal signalling pathways.⁸ Vodovar et al. have investigated the diazepam/buprenorphine DDI in rats using plethysmography and suggested a possible pharmacodynamic mechanism.⁹

This study has limitations, which may question the accuracy of our model to mimic the DDI between buprenorphine and diazepam observed in Human. First, only microdoses of the MOR agonist ¹¹C-buprenorphine could be administered for safety reasons. However, non-linearity was not observed regarding the non-specific brain uptake of ¹¹C-buprenorphine in the presence of co-injected doses of buprenorphine up to 0.03 mg/kg, whereas dose-dependent occupancy of ORs by buprenorphine was observed.²⁶ This suggests that microdose ¹¹C-buprenorphine PET imaging reliably describes the neuropharmacokinetics of buprenorphine. Moreover, a single dose of buprenorphine and diazepam was tested. It may be hypothesized that lower or higher doses of each drug may have differently impacted the outcome of the study. Nevertheless, the selected doses of buprenorphine and diazepam were higher than those observed in reported cases of death associated with the combination of buprenorphine and diazepam.⁷ The resolution of the PET scanner used in this study (∼7 mm) is too low to specifically investigate the binding of ¹¹C-buprenorphine to the small intrapontine loci involved in respiratory control.⁴³ However, it can be assumed that subregion-specific regulation of MOR, that would not occur in any other MOR-containing brain regions, is very unlikely. Flumazenil is a BzR antagonist, which could be safely co-injected with ¹¹C-flumazenil to investigate the DDI at pharmacological doses. However, this method only provides an insight on the regulation of BzRs and does not inform on the neuropharmacokinetic brain uptake of benzodiazepines such as diazepam used by patients. Finally, general anesthesia using ketamine followed by propofol was required to induce and maintain anesthesia during PET acquisition in non-human primates. PET acquisition in awake macaques is possible, although very limited by technical and ethical issues.^44,47 The impact of propofol anesthesia on ¹¹C-buprenorphine PET data is not known. Propofol anesthesia, but not ketamine, was shown to increase the binding of ¹¹C-flumazenil.^44,48 The same standardized protocol was used for all experiments, which enables comparison of PET data obtained in control and pretreated animals. However, the presence of anesthesia should be considered for clinical translation of the present PET data. The diazepam/buprenorphine interaction could be further investigated using ¹¹C-buprenorphine and ¹¹C-flumazenil PET imaging in humans, without anesthesia, to confirm the negative results obtained in anaesthetized macaques.

Conclusion

Using a cross-over study PET imaging study in nonhuman primates, our results suggest a limited impact of the regulation of BzR and MOR and do not support a neuropharmacokinetic or receptologic mechanism to explain the DDI between buprenorphine and benzodiazepines. This negative result supports that respiratory depression induced by diazepam/buprenorphine combination may result from a pharmacodynamic interaction induced by the combined effect of each respective drug on ventilation.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Acknowledgements

We gratefully thank Jérôme Cayla, Maud Goislard and Vincent Brulon for technical assistance.

Data availability statement

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

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Authors’ contributions

SA, DV, SC, BM and NT designed the research; SA, DV, SG, AS and FC performed the research; SA, DV, SC, LC, MB, BM and NT analyzed and interpreted the the data; SA, DV and NT wrote the initial draft of the manuscript. All authors revised the manuscript critically for important intellectual content and approved the submitted version.

ORCID iDs

Salvatore Cisternino

Nicolas Tournier

References

Centers for Disease Control and Prevention (CDC). Vital signs: overdoses of prescription opioid pain relievers – United States, 1999–2008. MMWR Morb Mortal Wkly Rep 2011; 60: 1487–1492.

Olfson

Zhang

Schoenbaum

, et al. Trends in buprenorphine treatment in the United States, 2009–2018. JAMA 2020; 323: 276–277.

Dahan

Yassen

Romberg

, et al. Buprenorphine induces ceiling in respiratory depression but not in analgesia. Br J Anaesth 2006; 96: 627–632.

Raffa

Haidery

Huang

H-M

, et al. The clinical analgesic efficacy of buprenorphine. J Clin Pharm Ther 2014; 39: 577–583.

Yassen

Olofsen

Romberg

, et al. Mechanism-based pharmacokinetic-pharmacodynamic modeling of the antinociceptive effect of buprenorphine in healthy volunteers. Anesthesiology 2006; 104: 1232–1242.

Pirnay

Borron

Giudicelli

, et al. A critical review of the causes of death among post-mortem toxicological investigations: analysis of 34 buprenorphine-associated and 35 methadone-associated deaths. Addiction 2004; 99: 978–988.

Häkkinen

Launiainen

Vuori

, et al. Benzodiazepines and alcohol are associated with cases of fatal buprenorphine poisoning. Eur J Clin Pharmacol 2012; 68: 301–309.

Dahan

van Lemmen

Jansen

, et al. Buprenorphine: a treatment and cause of opioid-induced respiratory depression. Br J Anaesth 2022; 128: 402–404.

Vodovar

Chevillard

Caillé

, et al. Mechanisms of respiratory depression induced by the combination of buprenorphine and diazepam in rats. Br J Anaesth 2022; 128: 584–595.

10.

Yassen

Kan

Olofsen

, et al. Pharmacokinetic-pharmacodynamic modeling of the respiratory depressant effect of norbuprenorphine in rats. J Pharmacol Exp Ther 2007; 321: 598–607.

11.

Hreiche

Mégarbane

Pirnay

, et al. Dexamethasone hepatic induction in rats subsequently treated with high dose buprenorphine does not lead to respiratory depression. Toxicol Appl Pharmacol 2006; 217: 352–362.

12.

Alhaddad

Cisternino

Declèves

, et al. Respiratory toxicity of buprenorphine results from the blockage of P-glycoprotein-mediated efflux of norbuprenorphine at the blood-brain barrier in mice. Crit Care Med 2012; 40: 3215–3223.

13.

Auvity

Breuil

Goislard

, et al. An original radio-biomimetic approach to synthesize radiometabolites for PET imaging. Nucl Med Biol 2020; 90–91: 10–14.

14.

Price

Mayberg

Dannals

, et al. Measurement of benzodiazepine receptor number and affinity in humans using tracer kinetic modeling, positron emission tomography, and [11C]flumazenil. J Cereb Blood Flow Metab 1993; 13: 656–667.

15.

de Laat

Morris

ED.

A local-neighborhood lassen plot filter for creating occupancy and non-displaceable binding images. J Cereb Blood Flow Metab 2021; 41: 1379–1389.

16.

Jucaite

Cselényi

Lappalainen

, et al. GABAA receptor occupancy by subtype selective GABAAα2,3 modulators: PET studies in humans. Psychopharmacology (Berl) 2017; 234: 707–716.

17.

Froklage

Postnov

Yaqub

, et al. Altered GABAA receptor density and unaltered blood-brain barrier [11C]flumazenil transport in drug-resistant epilepsy patients with mesial temporal sclerosis. J Cereb Blood Flow Metab 2017; 37: 97–105.

18.

Meyer

Koeppe

Frey

, et al. Positron emission tomography measures of benzodiazepine binding in Alzheimer’s disease. Arch Neurol 1995; 52: 314–317.

19.

Freeman

Garcia-Lorenzo

Bottin

, et al. The neuronal component of gray matter damage in multiple sclerosis: a [(11) C]flumazenil positron emission tomography study. Ann Neurol 2015; 78: 554–567.

20.

Heiss

W-D

Sobesky

Smekal

U. v

, et al. Probability of cortical infarction predicted by flumazenil binding and diffusion-weighted imaging signal intensity: a comparative positron emission tomography/magnetic resonance imaging study in early ischemic stroke. Stroke 2004; 35: 1892–1898.

21.

Salmi

Aalto

Hirvonen

, et al. Measurement of GABAA receptor binding in vivo with [11C]flumazenil: a test-retest study in healthy subjects. NeuroImage 2008; 41: 260–269.

22.

Delforge

Syrota

Bottlaender

, et al. Modeling analysis of [11C]flumazenil kinetics studied by PET: Application to a critical study of the equilibrium approaches. J Cereb Blood Flow Metab 1993; 13: 454–468.

23.

Delforge

Spelle

Bendriem

, et al. Quantitation of benzodiazepine receptors in human brain using the partial saturation method. J Nucl Med 1996; 37: 5–11.

24.

Delforge

Pappata

Millet

, et al. Quantification of benzodiazepine receptors in human brain using PET, [11C]flumazenil, and a single-experiment protocol. J Cereb Blood Flow Metab 1995; 15: 284–300.

25.

Cumming

Marton

Lilius

, et al. A survey of molecular imaging of opioid receptors. Mol Basel Switz 2019; 24: 4190.

26.

Auvity

Goutal

Caillé

, et al. Pharmacokinetic neuroimaging to study the dose-related brain kinetics and target engagement of buprenorphine in vivo. Neuropsychopharmacology 2021; 46: 1220–1228.

27.

Soyer

Leterrier

Breuil

, et al. Validation of a pharmacological imaging challenge using 11C-buprenorphine and 18F-2-fluoro-2-deoxy-D-glucose positron emission tomography to study the effects of buprenorphine to the rat brain. Front Neurosci 2023; 17: 1181786.

28.

Lever

Mazza

Dannals

, et al. Facile synthesis of [11C]buprenorphine for positron emission tomographic studies of opioid receptors. Int J Rad Appl Instrum A 1990; 41: 745–752.

29.

Canales-Candela

Riss

Aigbirhio

FI.

Synthesis of [¹¹ C]flumazenil([¹¹ C]FMZ). In: Scott

PJH

Hockley

(eds) Radiochemical syntheses. Hoboken, NJ: John Wiley & Sons, Inc., pp.221–231.

30.

Brouillet

Chavoix

Bottlaender

, et al. In vivo bidirectional modulatory effect of benzodiazepine receptor ligands on GABAergic transmission evaluated by positron emission tomography in non-human primates. Brain Res 1991; 557: 167–176.

31.

Bentourkia

Determination of the input function at the entry of the tissue of interest and its impact on PET kinetic modeling parameters. Mol Imaging Biol 2015; 17: 748–756.

32.

Tournier

Goutal

Mairinger

, et al. Complete inhibition of ABCB1 and ABCG2 at the blood-brain barrier by co-infusion of erlotinib and tariquidar to improve brain delivery of the model ABCB1/ABCG2 substrate [11C]erlotinib. J Cereb Blood Flow Metab 2021; 41: 1634–1646.

33.

Rohlfing

Kroenke

Sullivan

, et al. The INIA19 template and NeuroMaps atlas for primate brain image parcellation and spatial normalization. Front Neuroinform 2012; 6: 27.

34.

Logan

Fowler

Volkow

, et al. Graphical analysis of reversible radioligand binding from time-activity measurements applied to [N-11C-methyl]-(-)-cocaine PET studies in human subjects. J Cereb Blood Flow Metab 1990; 10: 740–747.

35.

R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, 2020. https://www.R-project.org/ (2020, accessed 11 December 2023).

36.

Hosseinzadeh Sahafi

Sardari

Alijanpour

, et al. Shared mechanisms of GABAergic and opioidergic transmission regulate corticolimbic reward systems and cognitive aspects of motivational behaviors. Brain Sci 2023; 13: 815.

37.

Cox

Collins

MA.

The effects of benzodiazepines on human opioid receptor binding and function. Anesth Analg 2001; 93: 354–358, 3rd contents page.

38.

Burford

Clark

Wehrman

, et al. Discovery of positive allosteric modulators and silent allosteric modulators of the μ-opioid receptor. Proc Natl Acad Sci U S A 2013; 110: 10830–10835.

39.

Kandasamy

Hillhouse

Livingston

, et al. Positive allosteric modulation of the mu-opioid receptor produces analgesia with reduced side effects. Proc Natl Acad Sci 2021; 118: e2000017118.

40.

Frankle

Cho

Narendran

, et al. Tiagabine increases [11C]flumazenil binding in cortical brain regions in healthy control subjects. Neuropsychopharmacology 2009; 34: 624–633.

41.

Matsunaga

Tsukada

Nishiyama

, et al. Yohimbine increases the binding potential for [11C]flumazenil in the monkey brain. J Neural Transm (Vienna) 2001; 108: 1375–1382.

42.

Frankle

Cho

Mason

, et al. [11C]flumazenil binding is increased in a dose-dependent manner with Tiagabine-Induced elevations in GABA levels. PLoS ONE 2012; 7: e32443.

43.

Ramirez

J-M

Burgraff

Wei

, et al. Neuronal mechanisms underlying opioid-induced respiratory depression: our current understanding. J Neurophysiol 2021; 125: 1899–1919.

44.

Sandiego

Jin

Mulnix

, et al. Awake nonhuman primate brain PET imaging with minimal head restraint: Evaluation of GABAA benzodiazepine binding with [11C]flumazenil in awake and anesthetized animals. J Nucl Med 2013; 54: 1962–1968.

45.

Bouvard

Costes

Bonnefoi

, et al. Seizure-related short-term plasticity of benzodiazepine receptors in partial epilepsy: a [11C]flumazenil-PET study. Brain J Brain 2005; 128: 1330–1343.

46.

Delforge

Spelle

Bendriem

, et al. Parametric images of benzodiazepine receptor concentration using a partial-saturation injection. J Cereb Blood Flow Metab 1997; 17: 343–355.

47.

Milham

Petkov

Margulies

, et al. Accelerating the evolution of nonhuman primate neuroimaging. Neuron 2020; 105: 600–603.

48.

Salmi

Kaisti

Metsähonkala

, et al. Sevoflurane and propofol increase 11C-flumazenil binding to gamma-aminobutyric acidA receptors in humans. Anesth Analg 2004; 99: 1420–1426.

Brain PET imaging using 11 C-flumazenil and 11 C-buprenorphine does not support the hypothesis of a mutual interaction between buprenorphine and benzodiazepines at the neuroreceptor level

Abstract

Keywords

Introduction

Material and methods

Chemicals and radiochemicals

Animals

Study design

Arterial input function and metabolism

PET acquisition procedure

Imaging data reconstruction and segmentation

PET quantification and data modeling

Statistical analysis

Results

Impact of diazepam pretreatment on the kinetics of 11C-buprenorphine

Impact of buprenorphine pretreatment on the kinetics of 11C-flumazenil

Discussion

Conclusion

Footnotes

Funding

Acknowledgements

Data availability statement

Declaration of conflicting interests

Authors’ contributions

ORCID iDs

References

Impact of diazepam pretreatment on the kinetics of ¹¹C-buprenorphine

Impact of buprenorphine pretreatment on the kinetics of ¹¹C-flumazenil