Usefulness of Olanzapine as an Adjunct to Opioid Treatment and for the Treatment of Neuropathic Pain

• Chronic pain is often associated with sleep disturbances

• Severe side effects of opioids given for pain treatment include nausea and vomiting

• In ferrets, olanzapine, an atypical thienobenzodiazepine antipsychotic drug, suppressed morphine-induced emesis and improved pain-related sleep disturbances

THE World Health Organization1has stated that morphine is the “gold standard” for the treatment of moderate to severe pain caused by cancer. However, the use of morphine for this purpose is often associated with distressing side effects, such as drowsiness, constipation, emesis, and delirium.2,3Many clinicians consider that dopamine receptor antagonists, including prochlorperazine, are the preferred drugs for combating opioid-induced nausea and vomiting.2,3However, these drugs often produce adverse effects, including extrapyramidal symptoms.4Therefore, new approaches are needed to prevent opioid-induced emesis, as is a better understanding of the mechanism of drug action.

Nausea and vomiting are controlled by the “vomiting center” in the medulla oblongata,5which receives signals from the chemoreceptor trigger zone (CTZ) in the area postrema, the gastrointestinal tract, the vestibular apparatus in the temporal lobe, and the cerebral cortex.6Opioids have emetogenic effects by stimulating the CTZ and the vestibular apparatus and by inhibiting gut motility.7Although stimulation of the CTZ by opioids involves opioid μ and δ receptors,8signals from the CTZ to the vomiting center mainly involve dopamine D²and serotonin (5-HT³) receptors in the former. However, opioid-induced stimulation of the vestibular apparatus and subsequent sensory input to the vomiting center have both been suggested to involve histamine H¹and muscarinic acetylcholine pathways.9 

Atypical antipsychotic medications treat the positive symptoms of schizophrenia, such as hallucinations and delusions, more effectively than the negative symptoms, such as lack of motivation and social withdrawal. Olanzapine is a newer atypical antipsychotic that blocks dopaminergic, serotonergic, adrenergic, histaminergic, and muscarinic receptors for multiple neurotransmitters. Because it affects neurotransmitters that are associated with nausea, it may have potential as an antiemetic medication.10 

In addition, patients with chronic pain commonly experience sleep disturbance11,–,13and may benefit from its treatment.13Sleep problems and daytime sleepiness seem to be related to depression and the severity of pain.14It has been suggested that olanzapine may improve sleep disturbance.15 

The primary endpoint of the study was to investigate whether olanzapine at doses lower than those that would induce catalepsy could suppress morphine-induced emesis with few side effects. We also examined if olanzapine could improve sleep dysregulation under a neuropathic pain-like state.

The present study was conducted in accordance with the Guiding Principles for the Care and Use of Laboratory Animals at Hoshi University, as adopted by the Committee on Animal Research of Hoshi University (Tokyo, Japan). Every effort was made to minimize the numbers and suffering of animals used in the following experiments.

The observer was not blinded in all of the experiments.

In the present study, male Institute of Cancer Research mice (20–25 g) (Tokyo Laboratory Animals Science, Tokyo, Japan), male C57BL/6J mice (25–30 g) (CLEA Japan, Tokyo, Japan), and Sprague-Dawley rats (200–300 g) (Tokyo Laboratory Animals Science) were used. Animals were housed in a room maintained at 22°± 1°C with a 12-h light–dark cycle. Food and water were available ad libitum . Each animal was used only once. Male ferrets weighing 1–1.5 kg were obtained from Marshall Research Labs (North Rose, NY) and housed individually in a room kept at 23°± 1°C under a 12-h light–dark cycle (lights on 8:00 AM-8:00 PM). They were given a standard cat diet (70–80 g/animal, Oriental Yeast Co. Ltd., Chiba, Japan) and allowed free access to water.

Mouse whole brain was treated as described previously,16and the resulting pellet was resuspended and used as the membrane fraction. The binding assay was performed in triplicate with [³H]clozapine (specific activity, 70–87 Ci/mmol; American Radiolabeled Chemicals, St. Louis, MO) at 0.2 nM, [³H]ketanserin hydrochloride (specific activity, 67 Ci/mmol; PerkinElmer, Waltham, MA) at 0.5 nM, [³H] BRL-43694 (granisetron) (specific activity, 85.3 Ci/mmol; PerkinElmer) at 0.5 nM, [³H]GR113808 (specific activity, 78.3 Ci/mmol; PerkinElmer) at 0.5 nM, [³H]pyrilamine (specific activity, 30 Ci/mmol; PerkinElmer) at 0.5 nM, and [³H]pirenzepine (specific activity, 72.8 Ci/mmol; PerkinElmer) at 0.5 nM, in a final volume of 500 ml that contained 50 mM Tris-HCl buffer, pH 7.4, and 200 ml homogenized membrane fraction. Ninety to 140 mg membrane proteins were used in each assay. Specific binding was defined as the difference in binding observed in the absence and presence of 1 mM unlabeled clozapine, ketanserin, granisetron, or GR113808, 10 mM unlabeled pyrilamine, or 100 mM unlabeled pirenzepine, respectively. All samples were incubated as described previously,16and radioactivity in the samples was determined with a liquid scintillation analyzer. All receptor binding curves were fitted using Prism software (version 5.0a; GraphPad Software, La Jolla, CA).

Emesis in ferrets after the administration of morphine (0.6 mg/kg, subcutaneous injection) was evaluated by counting the number of retching or vomiting behaviors as described elsewhere,17where retching was defined as a rhythmic abdominal contraction without expulsion and vomiting was the oral expulsion of solid or liquid from the gastrointestinal tract. Emesis was assessed for 30 min after the injection of morphine.18To determine the effect of olanzapine on morphine-induced emesis, groups of ferrets were pretreated with olanzapine 30 min before the injection of morphine.

An interval of at least 7 days was allowed between testing for each animal to allow for drug washout and to minimize the development of tolerance.

Catalepsy19,20was evaluated using the horizontal bar test as described previously.21Briefly, animals were placed so that both forepaws were over a horizontal bar 5 cm above the floor, and the amount of time (s) the animal maintained this position was recorded for as long as 60 s. Catalepsy was considered to have finished when a forepaw touched the floor or when the mouse climbed on the bar. Scores were assigned based on the duration of the cataleptic posture (score 1: 15 to 29 s, score 2: 30 to 59 s, score 3: 60 s or more).

After 3 days of habituation to the main animal colony, all of the rats were anesthetized with sodium pentobarbital (50 mg/kg, intraperitoneal administration) for surgery as described previously.21Briefly, the anesthetized animals were placed in a stereotaxic apparatus, the skull was exposed, and a small hole was made using a dental drill. A guide cannula (AG-8; Eicom, Kyoto, Japan) was implanted into the nucleus accumbens (from the bregma: anterior, +4.0 mm; lateral, −0.8 mm; ventral, −6.8 mm; angle 16 degrees) according to the atlas of Paxinos and Watson22and fixed to the skull with cranioplastic cement. Three to 5 days after surgery, microdialysis probes (A-I-8–02; 2 mm membrane length; Eicom) were slowly inserted into the nucleus accumbens through guide cannulas during anesthesia with diethyl ether, and the rats were placed in experimental cages (30 cm wide × 30 cm deep × 30 cm high). The probes were perfused continuously (2 ml/min) with artificial cerebrospinal fluid: 0.9 mM MgCl², 147.0 mM NaCl, 4.0 mM KCl, and 1.2 mM CaCl². Outflow fractions were collected every 20 min. After three baseline fractions were collected from the rat nucleus accumbens, rats were given olanzapine (0.3 mg/kg, intraperitoneal administration), vehicle (5% dimethyl sulfoxide [DMSO]); 1 ml/kg, intraperitoneal administration) or saline (1 ml/kg, intraperitoneal administration) 30 min before treatment with morphine (10 mg/kg, intraperitoneal administration). Dialysis samples were collected for 180 min after treatment and analyzed by high-performance liquid chromatography (Eicom) with electrochemical detection (Eicom). Dopamine and its metabolites, 3,4-dihydroxyphenylacetic acid and 3-methoxy-4-hydroxyphenyl acetic acid, were separated by column chromatography and identified and quantified by the use of standards, as described previously.21 

In the study of gastrointestinal transit,23Institute of Cancer Research mice were fasted for 12 h before the experiments. Groups of mice were pretreated with olanzapine (0.03–1 mg/kg, subcutaneous injection) or vehicle (5% DMSO) 30 min before the administration of morphine (0.7 mg/kg, subcutaneous injection) or saline, and ink (0.3 ml/mouse) was orally administered 20 min after the injection of morphine or saline. Twenty minutes after the administration of ink, the animal was killed by cervical dislocation, and the small intestine was removed. The percentage inhibition of gastrointestinal transit was calculated as follows: (distance traveled by the ink/length from the pylorus to the cecum) × 100.

C57BL/6J mice were administered olanzapine (0.03–1 mg/kg, subcutaneous injection) or vehicle (5% DMSO) once a day for 1 week. At 60 min after the final injection, the tail was cut and blood was collected. Blood glucose was measured using a self-monitoring blood glucose meter (Medisafe-Mini; Terumo, Tokyo, Japan). The Medisafe-Mini system is based on the optoelectric colorimetry method.

C57BL/6J mice were anesthetized with 3% isoflurane. A partial sciatic nerve ligation model was made by tying a tight ligature with 8–0 silk suture around approximately one third to one half the diameter of the sciatic nerve on the right side (ipsilateral side) under a light microscope (SD30; Olympus, Tokyo, Japan). In sham-operated mice, the nerve was exposed without ligation.

The sensitivity to thermal stimulation was measured as described previously.24Briefly, the right plantar surface of mice was exposed to a well-focused radiant heat light source (model 33 Analgesia Meter; IITC/Life Science Instruments, Woodland Hills, CA) that had been adjusted so that the average baseline latency of paw withdrawal in naive mice was approximately 8–10 s. Only quick movements of the hind paw away from the stimulus were considered to be a withdrawal response: paw movements associated with locomotion or weight shifting were not counted as a response. The paws were measured alternating between left and right with an interval of more than 3 min between measurements. Before testing, mice were placed in a clear acrylic cylinder (15 cm high and 8 cm in diameter) for at least 30 min. The data represent the average latency of paw withdrawal for the right hind paw.

Electroencephalogram and electromyogram recordings were obtained as described previously.24Briefly, electroencephalogram signals were monitored with two stainless-steel electroencephalogram recording screws 1 mm anterior to the bregma or λ, both 1.5 mm lateral to the midline, and electromyogram activity was monitored by stainless steel, nonstick-coated wires placed bilaterally into both trapezius muscles. Electroencephalogram and electromyogram signals were amplified, filtered (0.5–30 Hz and 20–200 Hz, respectively), digitized at a sampling rate of 128 Hz, and recorded using SleepSign software (Kissei Comtec, Nagano, Japan), which was also used to automatically classify vigilance over 4-s epochs as wakefulness, rapid eye movement (REM) sleep, or non-REM sleep using standard criteria. Finally, defined sleep–wake stages were examined visually and corrected, if necessary. For each epoch, the electroencephalogram power density in the δ (0.75–4.0 Hz) and θ bands (6.25–9.0 Hz) and the integrated electromyogram value were displayed on a computer monitor. Electroencephalogram and electromyogram activities were monitored for 24 h at 7 days after sciatic nerve ligation. Recordings were started at 8:00 PM. Vehicle (5% DMSO) or olanzapine (0.06 mg/kg, intraperitoneal administration) was injected every day at 8:00 AM.

The drugs used in the current study were morphine hydrochloride (Daiichi-Sankyo, Tokyo, Japan), prochlorperazine maleate (Sigma–Aldrich, St. Louis, MO), clozapine (Wako Pure Chemical Industries, Osaka, Japan), olanzapine (Toronto Research Chemicals, Toronto, Ontario, Canada), telenzepine dihydrochloride hydrate (Sigma–Aldrich), ritanserin (Tocris Biotechnology, Ellisville, CA), pyrilamine maleate salt (Sigma–Aldrich), ketanserin tartrate (Wako Pure Chemical Industries), granisetron (Sigma–Aldrich), GR113808 (Sigma–Aldrich), haloperidol (Sigma–Aldrich), L745870 (Research Biochemicals International, Natick, MA), raclopride (Santa Cruz Biotechnology, Santa Cruz, CA), pirenzepine (Toronto Research Chemicals), and DL-trihexyphenidyl hydrochloride (Sigma–Aldrich).

Data are expressed as the mean with SEM. The statistical significant of differences between the groups was assessed with one-way and two-way ANOVA followed by the Bonferroni multiple comparisons test or Student t  test (unpaired, two-tailed). The concentration of the test compound that caused 50% inhibition of specific binding (IC⁵⁰value) was determined from each concentration-response curve. All statistical analyses and IC⁵⁰values were calculated by Prism software (version 5.0a, GraphPad Software). A P  value of <0.05 was considered to reflect significance.

In mouse brain membranes, we determined the competitive displacement binding of [³H]clozapine with graded concentrations (10⁻¹¹–10⁻⁴M) of unlabeled clozapine, olanzapine, telenzepine, ritanserin, pyrilamine, ketanserin, GR113808, granisetron, haloperidol, L745870, and raclopride. The binding of [³H]clozapine was displaced by olanzapine in a concentration-dependent manner (fig. 1A). In addition, the binding of [³H]clozapine was partially displaced by telenzepine (M¹), ritanserin (5-HT^2A), pyrilamine (H¹), ketanserin (5-HT^2C), GR113808 (5-HT⁴), granisetron (5-HT³), haloperidol (D²), L745870 (D⁴), and raclopride (D²) (fig. 1B).

In mouse brain membranes, we determined the competitive displacement binding of [³H]ketanserin, [³H]BRL-43694 (granisetron), [³H]pyrilamine, [³H]GR113808, and [³H]pirenzepine with graded concentrations (10⁻¹¹–10⁻⁴M) of unlabeled ketanserin, granisetron, pyrilamine, GR113808, telenzepine, pirenzepine and olanzapine. The binding of [³H]ketanserin and [³H]pirenzepine was displaced by olanzapine in a concentration-dependent manner (fig. 2, A and B). The binding of [³H]pyrilamine, [³H]BRL-43694, and [³H]GR113808 was partially displaced by olanzapine (fig. 2, C, D, and E).

Pretreatment with either olanzapine (0.03 mg/kg, subcutaneous injection) 30 min before the injection of morphine (0.6 mg/kg, subcutaneous injection) or prochlorperazine (0.3 mg/kg and 1.0 mg/kg, subcutaneous injection) 60 min before the injection of morphine reduced the number of retching and vomiting behaviors induced by morphine (fig. 3).

The results from the horizontal bar test as a measure of catalepsy were obtained at 15, 30, 45, and 60 min after the subcutaneous injection of vehicle, prochlorperazine (0.3–1 mg/kg), haloperidol (0.03–0.3 mg/kg), risperidone (0.01–0.1 mg/kg), or olanzapine (0.03–0.3 mg/kg). Although the catalepsy scores were not affected by a single subcutaneous injection of olanzapine (0.03–0.3 mg/kg), catalepsy was observed with the other antipsychotics within this dose range (fig. 4).

In the microdialysis study, the concentrations of dopamine and its metabolites 3,4-dihydroxyphenylacetic acid and 3-methoxy-4-hydroxyphenyl acetic acid in dialysate from the rat nucleus accumbens were markedly increased by the intraperitoneal administration of morphine at 10 mg/kg compared with those under the administration of saline. The increased concentrations of dopamine, 3,4-dihydroxyphenylacetic acid, and 3-methoxy-4-hydroxyphenyl acetic acid in the nucleus accumbens after the administration of morphine were not affected by olanzapine at 0.3 mg/kg (olanzapine-morphine vs.  vehicle-morphine: F^(1,77)= 0.1516, P = 0.7086 fig. 5A; F^(1,77)= 0.06326, P = 0.8086 fig. 5B; F^(1,77)= 1.851, P = 0.2158 fig. 5C).

Morphine slowed gastrointestinal transit, and this effect was not significantly altered by the coadministration (subcutaneous injection) of olanzapine at 0.03–1 mg/kg (fig. 6A). Olanzapine itself did not slow gastrointestinal transit at doses of 0.03 and 0.1 mg/kg but significantly inhibited gastrointestinal transit at 0.3 and 1 mg/kg (fig. 6B).

Blood glucose was measured after long-term treatment with olanzapine, saline, or vehicle (5% DMSO) in mice. Hyperglycemia was not observed under treatment with olanzapine at 0.03, 0.1, or 0.3 mg/kg (subcutaneous injection) (fig. 7).

Sciatic nerve ligation markedly decreased the latency of paw withdrawal in response to a thermal stimulus on the ipsilateral side. This state of persistent pain caused by partial ligation of the sciatic nerve was suppressed by treatment with olanzapine at 0.06 mg/kg (fig. 8).

We next investigated the changes in sleep patterns in sciatic nerve-ligated mice. Vigilance was classified automatically offline as wakefulness, REM sleep, or non-REM sleep. Mice with sciatic nerve ligation showed a statistically significant increase in wakefulness (P = 0.0009 vs.  sham operated mice with vehicle, fig. 9) and a decrease in non-REM sleep (P = 0.0067 vs.  sham-operated mice with vehicle, fig. 9C) during the light phase. REM sleep during the light phase was not affected by sciatic nerve ligation (P = 0.2896 vs.  sham-operated mice with vehicle, fig. 9B). On the other hand, there was no statistically significant difference in the sleep conditions during the dark phase between the two groups (wakefulness: P = 0.6170 vs.  sham operated mice with vehicle, fig. 9D; REM: P = 0.3936 vs.  sham operated mice with vehicle, fig. 9E; non-REM: P = 0.5479 vs.  sham operated mice with vehicle, fig. 9F).

To ascertain the hypnotic effect of olanzapine in a neuropathic pain-like state, we performed electroencephalogram and electromyogram recordings. The increased wakefulness and decreased non-REM during the light phase in nerve-ligated mice were significantly attenuated compared with those in sham-operated mice by the intraperitoneal administration of olanzapine (wakefulness: P = 0.0006 vs.  nerve-ligated mice with vehicle, fig. 9A; non-REM: P = 0.001 vs.  nerve-ligated mice with vehicle, fig. 9C).

The use of opioids for pain management is often associated with nausea and vomiting. Opioids induce emesis by stimulating the CTZ in the brainstem and by enhancing vestibular sensitivity.25,26Although several compounds are known to act on receptors in the CTZ, opioid-induced nausea and vomiting are attributable primarily to the transmission of dopamine. Many clinicians consider that typical antipsychotics such as prochlorperazine and haloperidol, which mainly act as dopamine D²receptor antagonists, are the drugs of choice for preventing the nausea and vomiting induced by morphine.27,–,29However, such compounds often produce extrapyramidal symptoms.4 

Olanzapine is an atypical thienobenzodiazepine antipsychotic that is clinically indicated for schizophrenia and mania.30It blocks multiple neurotransmitters, including dopaminergic, serotonergic, adrenergic, histaminergic, and muscarinic receptors.31In the current binding study, olanzapine showed the highest affinity for muscarinic M¹receptors. To understand its affinity in greater detail, we investigated the affinity of olanzapine toward serotonin 5-HT^2A/2C, 5-HT³, histamine H¹, dopamine D², dopamine D⁴, and 5-HT⁴receptors. Olanzapine also showed affinity for each of these receptors. Because of its effect on several neurotransmitters that are associated with nausea and vomiting, we expected that olanzapine would have potential as an antiemetic medication. In a behavioral study, we found that morphine-induced nausea and vomiting were decreased in a dose-dependent manner by pretreatment with olanzapine at a dose that was almost 1/200 of that at which an antipsychotic effect is observed,32whereas olanzapine at a dose that had antiemetic effects failed to induce catalepsy. However, although the dopamine D²receptor antagonist prochlorperazine suppressed morphine-induced nausea and vomiting, it did so at a dose that caused a dose-dependent increase in the expression of catalepsy. Furthermore, olanzapine had no effect on the morphine-induced release of dopamine in the nucleus accumbens. Muscarinic M¹receptors have been suggested to be responsible for the opioid-induced stimulation of the vestibular apparatus.6In addition, sensory input from the vestibular apparatus to the vomiting center follows muscarinic M¹receptor pathways. Taken together with the fact that olanzapine showed the highest affinity toward muscarinic M¹receptors, these findings suggest that, although the exact mechanism by which olanzapine suppresses morphine-induced emesis remains unclear, muscarinic M¹receptors may be a critical target for morphine-induced emesis. To prove our hypothesis, we next investigated whether the selective muscarinic M¹receptor antagonist trihexyphenidyl could affect morphine-induced nausea and vomiting. Trihexyphenidyl significantly suppressed morphine-induced retching and vomiting in ferrets (data not shown), which indicates that M¹receptors play an important role in the opioid-sensitive emetic pathway. However, trihexyphenidyl significantly enhanced the morphine-induced increase in the release of dopamine in the nucleus accumbens (data not shown). If we consider the risk of the overexcitation of dopamine transmission with the use of drug combinations, a specific M¹receptor antagonist might not be a better choice as an adjunct agent in combination with opioids. Because olanzapine acts not only on muscarinic M¹receptors, but also partly on histamine H¹and dopamine D²receptors as an antagonist,10it is likely that olanzapine at a dose lower than that at which it has antipsychotic effects could be useful for strongly preventing opioid-induced emesis without severe side effects.

Constipation is another adverse effect of treatment with morphine. In the current study, olanzapine at doses that had antiemetic effects had no effect on the morphine-induced inhibition of gastrointestinal transit. This may be attributable to the high central transitivity of olanzapine.

Long-term treatment with olanzapine is most commonly associated with increased weight gain, obesity, and diabetes mellitus.33Therefore, we evaluated the effect of chronic treatment with olanzapine on blood glucose. As a result, hyperglycemia was not observed during treatment with olanzapine at a dose at which it had an antiemetic effect (0.03 mg/kg). However, long-term treatment with olanzapine at a dose of 1 mg/kg was associated with a slight but statistically significant increase in blood glucose concentrations. These findings support the idea that olanzapine may have a wide margin of safety when used as an antiemetic.

With regard to sleep disturbance in severe pain, we confirmed that mice with sciatic nerve ligation showed a statistically significant increase in wakefulness and a decrease in non-REM sleep during the light phase. Under the current conditions, treatment with olanzapine at doses at which the pain threshold was restored improved the sleep disturbance after sciatic nerve ligation. Histamine and serotonin are the key neurotransmitters that regulate wakefulness, and their receptors are the ultimate targets of many wakefulness- and sleep-promoting drugs. In particular, histamine H¹receptor antagonist and serotonin 5-HT^2A/2Creceptor antagonist are known to shift one's arousal state from hyperactivity to sleep.34Therefore, the improvement of sleep disturbance during treatment with olanzapine may result from the agent's antagonistic effects toward histamine H¹and serotonin 5-HT^2A/2C.

Overall, the current results suggest that olanzapine may be useful for the treatment of morphine-induced emesis, reducing neuropathic pain, and improving pain-related sleep disturbance. Against a background of increasing concern about “polypharmacy,” olanzapine can be used as a single adjunct agent and can be given in a state-dependent dose, which should improve the quality of life for patients while greatly reducing the side effects of opioids.

In conclusion, we propose that treatment with olanzapine may lead to a new strategy for controlling emesis when patients are given opioid medications.

In addition, the current study provides evidence that olanzapine may be a useful agent for improving the sleep disturbance caused by neuropathic pain that is observed in some patients with cancer.

The authors thank Daisuke Takei, M.Sc., Kotaro Takeda, B.S., Daiki Masukawa, B.S., Mr. Hiroshi Yoneyama, and Mr. Taizo Ishiguro (Graduate Students, Department of Toxicology, Hoshi University School of Pharmacy and Pharmaceutical Sciences, Shinagawa-ku, Tokyo, Japan) for their expert technical assistance.