Attenuation of the Positive-Reinforcing Effects of Ultra-Potent Fentanyl Analogs, Along with Those of Fentanyl and Heroin, During Daily Treatment with Methocinnamox in Rhesus Monkeys

Abstract

Without substantial intervention, the opioid crisis is projected to continue, underscoring the need to develop new treatments for opioid use disorder (OUD). One drug under development is the µ opioid receptor antagonist methocinnamox (MCAM), which appears to offer advantages over currently available medications; however, some questions remain about its potential utility, including its ability to block the effects of ultra-potent fentanyl analogs. The goal of this study was to examine its effectiveness in attenuating the abuse-related effects of the fentanyl analogs carfentanil and 3-methylfentanyl in monkeys responding for food or intravenous infusions under a choice procedure. These drugs were compared with fentanyl, heroin, methamphetamine, and cocaine. Food was preferred over saline, and there was a dose-dependent increase in responding for drug over food with no marked decrease in response rates or number of choice trials completed for any of the six drugs studied. Naltrexone (0.032 mg/kg) antagonized choice of µ opioid receptor agonists, producing rightward shifts in dose-effect curves ranging from 27-fold (carfentanil) to 71-fold (heroin). In contrast, naltrexone was less effective in attenuating choice of methamphetamine or cocaine with curves obtained in the presence of naltrexone shifted <3-fold. Daily treatment with 0.032 mg/kg MCAM also antagonized the effects of opioids, shifting curves 20-fold (fentanyl) to 72-fold (heroin) rightward; MCAM did not significantly change dose-effect curves for methamphetamine or cocaine. Thus, antagonism by MCAM is similar across a variety of µ opioid receptor agonists, including ultra-potent fentanyl analogs, further supporting its potential utility as a treatment for OUD.

SIGNIFICANCE STATEMENT

Treatments for opioid use disorder (OUD) should attenuate the effects of a variety of opioids, including emerging threats like the ultra-potent fentanyl analogs. The novel µ opioid receptor antagonist MCAM is being developed to treat OUD because it provides long-lasting blockade of the reinforcing effects of heroin and fentanyl. The current study shows that MCAM attenuates the abuse-related effects of the fentanyl analogs carfentanil and 3-methylfentanyl, further supporting the utility of MCAM as a treatment for OUD.

Introduction

Opioid use disorder (OUD) is escalating in the US, leading to a steep rise in overdose deaths. Between October 2019 and October 2020, provisional data showed a 16% increase in overdose fatalities compared with the previous 12 months (Ahmad et al., 2021). While the COVID-19 pandemic clearly contributed to these soaring numbers (American Medical Association 2020; Haley and Saitz 2020; Ochalek et al., 2020; O’Donnell et al., 2020; Wainwright et al., 2020), recent projections suggest that overdose deaths and the prevalence of OUD will continue to increase unless treatment is improved and made more widely available (Ballreich et al., 2020). Moreover, in response to this growing problem, the Stanford-Lancet Commission on the North American Opioid Crisis was formed to identify strategies to reduce the impact of the opioid epidemic; one of their recommendations was to develop novel medications to treat OUD and overdose (Humphreys et al., 2022).

Among the three medications currently approved for treating OUD are the opioid receptor antagonist naltrexone and opioid receptor agonists buprenorphine and methadone. While these pharmacotherapies can decrease opioid use in some patients, they are not universally effective, and other options are needed. One candidate is methocinnamox (MCAM), a novel drug that retains some advantages of naltrexone and buprenorphine. Like naltrexone, MCAM is a µ opioid receptor antagonist, and like buprenorphine, it has a long duration of action. A single dose of MCAM can attenuate the abuse-related and ventilatory-depressant effects of heroin and fentanyl for at least a week (Gerak et al., 2019; Maguire et al., 2019, 2020; Maguire and France 2022). Binding of MCAM to µ opioid receptors has been shown to be functionally irreversible (i.e., pseudoirreversible; Broadbear et al., 2000; Zamora et al., 2021), which likely accounts for its long duration of action and could offer important advantages over naltrexone. For example, an extended-release formulation of naltrexone (Vivitrol) increases its duration of antagonism, although it does not mitigate the surmountability of its antagonist effects when larger doses of an opioid receptor agonist are administered (Darke et al., 2019). At an appropriate dose, pseudoirreversible binding of MCAM to µ opioid receptors should result in insurmountable antagonism, thereby protecting patients from the risk of overdose, including those who continue taking opioids.

MCAM has been shown to attenuate the abuse-related effects of heroin and fentanyl (Maguire et al., 2019, 2020; Maguire and France 2022); however, additional information is needed to determine its full potential for treating OUD, including its ability to block the effects of other opioids. One emerging threat involves fentanyl analogs, which have been implicated in overdose deaths and cases of impaired driving and are often detected in combination with fentanyl and other opioids, thereby increasing the likelihood of toxic effects (Delcher et al., 2020; Brunetti et al., 2021; Chhabra et al., 2021; Kiely and Juhascik 2021). The situation is further complicated by reports that overdose caused by ultra-potent fentanyl analogs is difficult to reverse with naloxone. Having a medication that reliably attenuates the abuse-related and toxic effects of ultra-potent fentanyl analogs would dramatically improve clinical outcomes and save lives.

Given the increased availability and emerging threat of ultra-potent fentanyl analogs, the clinical utility of MCAM would be increased if it antagonizes the abuse-related effects of these drugs. Consequently, the goal of the current study was to examine the ability of MCAM to attenuate the reinforcing effects of two ultra-potent fentanyl analogs, carfentanil and 3-methylfentanyl. Several other abused drugs were studied for comparison, including fentanyl, heroin, methamphetamine, and cocaine. A choice procedure was used with monkeys responding on one lever to receive food or on a second lever to receive an intravenous infusion (Gerak and France 2021). This procedure was selected because reinforcing effectiveness is measured by dose-dependent increases in choice of drug infusions over food pellets, resulting in monotonic dose-effect curves that are shifted rightward by antagonists. In contrast, single-response self-administration procedures examine changes in response rate or number of reinforcers delivered, resulting in biphasic dose-effect curves that can be shifted rightward or downward by antagonists. For this choice procedure, a single unit dose was available during each session, and dose-effect curves were determined by changing the unit dose across sessions. MCAM produces long-lasting, but not permanent, antagonism (e.g., Gerak et al., 2019); to ensure that MCAM antagonism changed as little as possible across determinations of dose-effect curves, MCAM was administered daily at a dose that has been shown to produce a ∼20-fold shift to the right in the fentanyl dose-effect curve in monkeys responding under a single-response self-administration procedure (Maguire and France 2022).

Material and Methods

Subjects

Four male (HU, KI, MU, and LA) and one female (JA) adult rhesus monkeys previously participated in a variety of behavioral pharmacology studies and had been responding under the experimental conditions described here for at least one year before this study began (e.g., Weed et al., 2017; Gerak and France 2021). In the current study, monkeys weighed between 8.5 and 13.0 kg. Their weights were maintained with primate chow (High Protein Monkey Diet; Harlan Teklad, Madison, WI) and fresh fruit provided daily in the home cage, along with peanuts and food pellets during sessions. They were housed individually in rooms maintained on a 14-hour light/10-hour dark cycle, and water was available continuously in home cages. Monkeys were maintained in accordance with the Institutional Animal Care and Use Committee, The University of Texas Health Science Center at San Antonio, and the 2011 Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources on Life Sciences, National Research Council, National Academy of Sciences).

Surgery

Chronic indwelling venous catheters were implanted for drug self-administration. After sedation with 10 mg/kg ketamine (subcutaneously; Henry Schein Animal Health, Dublin, OH), anesthesia was maintained with isoflurane (Butler Animal Health Supply, Grand Prairie, TX) in oxygen and delivered at a rate of 2 l/min. One end of a polyurethane catheter (SIMS Deltec, Inc., St. Paul, MN) was secured in a vein (e.g., femoral or jugular), and the other end was tunneled to the midscapular region where it was connected to a subcutaneous access port (Access Technologies, Skokie, IL). Penicillin B&G (40,000 IU/kg) and meloxicam were given postoperatively.

Apparatus

Experimental sessions were conducted in custom-built, ventilated, sound-attenuating chambers. Monkeys were seated in chairs (Primate Products, Miami, FL), which were placed inside the chambers facing response panels equipped with stimulus lights, response levers, and pellet troughs. Completion of the response requirement on one lever resulted in the delivery of food pellets (5TUT, Test Diet, Richmond, IN) to the troughs from pellet dispensers (Med Associates, Inc., St. Albans, VT) located outside of chambers. Completion of the response requirement on the other lever resulted in the delivery of an intravenous infusion to the catheter from syringes mounted in syringe drivers (model PHM-100; Med Associates, Inc., St. Albans, VT) outside of chambers. Infusions (vehicle or drug) were delivered from syringes to ports through 183-cm catheter extension sets (Baxter Healthcare, Deerfield, IL) and 20-g Huber-point needles (Access Technologies, Skokie, IL) at a rate of 3.4 ml/min. Experimental events were controlled and data were collected using a computer running MedPC IV software (Med Associates, Inc., St. Albans, VT). During sessions, monkeys were continuously monitored by cameras, and white noise was provided in each chamber.

Procedure

Monkeys received subcutaneous injections before daily experimental sessions during which intravenous infusions and food pellets were available concurrently under fixed-ratio 30 schedules. At the beginning of each session, syringe pumps were activated to fill ports and catheters with the solution available for self-administration after which green stimulus lights were illuminated to signal the beginning of trials. Two forced trials were followed by up to 28 choice trials and monkeys could satisfy the response requirement by completing 30 consecutive responses on a lever. During forced trials, one light was illuminated, and monkeys could respond on the lever below that light for the reinforcer associated with that lever; responding on the other lever reset the response requirement. A forced trial occurred for each lever (i.e., food or infusion) with the order of presentation varying nonsystematically across sessions. During choice trials, both lights were illuminated, and monkeys could respond on either lever to receive a reinforcer. Completion of the response requirement extinguished green lights, illuminated the red light above the lever on which the response requirement was completed for 5 seconds, delivered the reinforcer associated with responding on that lever, and initiated a 180-second timeout. The chamber was dark for the last 175 seconds of each trial, and responding during this period was recorded but had no programmed consequence. Sessions ended 90 minutes after the start of the first forced trial.

For all sessions, responding on one lever delivered a food pellet and responding on the other lever delivered an infusion. The lever assignments remained the same for individual monkeys throughout these studies, although they were counterbalanced across monkeys. Before testing began, the solution available for self-administration varied among saline, fentanyl, or methamphetamine to confirm that monkeys chose food over saline and chose drug over food when unit doses of 0.32 µg/kg/infusion fentanyl or 10 µg/kg/infusion methamphetamine were available; monkeys self-administered these unit doses under other conditions (Maguire et al., 2020; Gerak et al., 2016). Test sessions were not different from baseline sessions except that the subcutaneous injection given before sessions could change along with the unit dose and/or solution available for self-administration during sessions. Dose-effect curves were obtained for fentanyl and methamphetamine followed by heroin, cocaine, and two ultra-potent fentanyl analogs (carfentanil and 3-methylfentanyl) with saline administered subcutaneously 15 minutes before sessions. Initial unit doses were selected based on previous data (e.g., heroin and cocaine; Gerak and France 2021) and estimates of the relative potency of carfentanil and 3-methylfentanyl to fentanyl from unpublished studies in monkeys in this laboratory. Unit doses were increased and/or decreased as necessary to obtain the entire dose-effect curve. Once a complete dose-effect curve was obtained following subcutaneous administration of saline, it was redetermined after subcutaneous administration of 0.032 mg/kg naltrexone, which replaced the saline injection given 15 minutes before the start of sessions. Under other conditions, this dose of naltrexone shifts dose-effect curves for µ opioid receptor agonists 10-fold rightward (e.g., Gerak et al., 2019). During this first set of studies, test sessions were conducted in a nonsystematic order across monkeys and drugs, with tests occurring during two or three nonconsecutive sessions each week and baseline sessions conducted on the other 4 or 5 days. Test sessions were scheduled as long as monkeys chose food over saline or drug over food when unit doses of 0.32 µg/kg/infusion fentanyl or 10 µg/kg/infusion methamphetamine were available during the baseline session that immediately preceded the test session. The testing criterion had to be satisfied during at least 80% of choice trials to schedule the test session; otherwise, another baseline session was conducted. Each unit dose of agonist was tested once in the presence of naltrexone in each monkey.

The ability of MCAM to attenuate the reinforcing effects of opioids, especially ultra-potent fentanyl analogs, was investigated in four monkeys (HU, MU, LA, JA) by administering 0.032 mg/kg/d MCAM 60 minutes before sessions; the fifth monkey did not participate in this portion of the study for reasons not related to the experiment. Daily dosing was used because of the long duration of action of MCAM. To be sure that the effects of a single dose of MCAM had dissipated completely, tests with MCAM would need to be separated by at least 3 weeks (e.g., 0.1 and 0.32 mg/kg MCAM; Gerak et al., 2019), which was impractical in a study in which agonist dose-effect curves were obtained across multiple sessions. While a smaller MCAM dose (i.e., 0.032 mg/kg) has a shorter duration of action, a single administration of the smaller dose does not reliably attenuate the reinforcing effects of fentanyl; in contrast, daily administration of the smaller MCAM dose was found to produce consistent antagonism throughout the treatment period (Maguire and France 2022). In the current study, dose-effect curves were determined periodically during MCAM treatment with the unit dose increasing across consecutive sessions. For the first 3 days of MCAM treatment, monkeys could choose between food and a 0.32 µg/kg/infusion of fentanyl, and on the 4^th day, a 10 µg/kg/infusion of methamphetamine was available instead of fentanyl; that pattern was repeated once after which dose-effect curves were redetermined. Fentanyl dose-effect curves were obtained beginning on the 10^th and 17^th days of treatment starting with a unit dose of 0.32 µg/kg/infusion of fentanyl; during MCAM treatment, monkeys chose food over that unit dose of fentanyl. Dose-effect curves for the other five drugs were generated twice (see Table 1 for order of testing). For baseline sessions between dose-effect curves, the solution available for self-administration rotated nonsystematically between saline and a 10 µg/kg/infusion of methamphetamine. Two weeks after the beginning of the last fentanyl dose-effect curve, vehicle injections replaced daily MCAM. For 3 days after MCAM treatment ended, monkeys could choose between food and a 0.32 µg/kg/infusion of fentanyl, with 10 µg/kg/infusion of methamphetamine available on the 4^th day instead of fentanyl; this pattern was repeated three times, and a fentanyl dose-effect curve was determined beginning 15 days after the last dose of MCAM. Dose-effect curves for the remaining five drugs were redetermined after discontinuation of MCAM treatment.

TABLE 1.

Order of testing during daily MCAM administration. Dose-effect curves were obtained by increasing the unit dose across consecutive sessions up to the unit dose that resulted in at least 80% responding for infusions; the first day of the dose-effect curve determination, during which the smallest unit dose was available for self-administration, is indicated in the left column.

Days after start of MCAM treatment	Dose-effect curve
10	fentanyl
17	fentanyl
24	methamphetamine
31	3-methylfentanyl
38	heroin
45	fentanyl
52	cocaine
73	fentanyl
87	carfentanil
101	methamphetamine
115	3-methylfentanyl
129	heroin
143	cocaine
157	carfentanil
171	fentanyl

Drugs

Fentanyl hydrochloride, carfentanil hydrochloride, 3-methylfentanyl hydrochloride, heroin hydrochloride, naltrexone hydrochloride, and cocaine hydrochloride were provided by the National Institute on Drug Abuse Drug Supply Program, Bethesda, MD, USA. Methamphetamine hydrochloride was purchased from Sigma-Aldrich Corporation (St. Louis, MO, USA), and MCAM hydrochloride was purchased from Syncom (Groningen, NL). All drugs were dissolved in saline except MCAM, which was dissolved in a vehicle of 10% w/v β-cyclodextrin in saline. Unit doses were obtained by preparing drug concentrations that were 10-fold larger than unit doses and adjusting infusion durations to account for differences in body weights. Drugs administered subcutaneously were given in volumes of 0.2–0.8 ml.

Data Analyses

Three dependent variables were monitored for each session; these variables were the percentage of choice trials during which monkeys responded for infusions, overall response rate during choice trials, and the number of choice trials completed. Response rates were obtained by counting the total number of responses that occurred while green lights were illuminated during choice trials and dividing that number by the total time that the green lights were illuminated. Each dependent variable is plotted as a function of dose; each data point represents results obtained during a single session and averaged across monkeys (±1 S.E.M.).

Potency estimates for each drug were determined by calculating ED₅₀ values for each dose-effect curve generated in these studies. The unit dose expected to produce responding for infusions during 50% of choice trials was obtained for individual monkeys by fitting a line to the linear portion of the curve and estimating the dose that would produce a 50% effect. For each drug, ED₅₀ values obtained following subcutaneous administration of naltrexone were compared with values obtained following subcutaneous administration of saline in five monkeys using paired t tests. ED₅₀ values obtained during and after daily MCAM treatment were compared with values obtained before MCAM treatment in four monkeys using one-way repeated measures ANOVAs with Geisser-Greenhouse correction; separate analyses were performed for each agonist and significant differences were further analyzed using Dunnett’s multiple comparisons tests. Data analyses were performed using GraphPad Prism (version 9.2.0 GraphPad, La Jolla, California, USA), and results were considered statistically significant when P < 0.05.

Results

When saline was administered before sessions and saline was available for self-administration, monkeys responded predominantly for food, receiving an infusion of saline during 5.1 ± 3.1% of choice trials; response rates were, on average, 2.04 ± 0.52 responses/s, and 24.8 ± 1.5 trials were completed in the session (stars above “S”, Fig. 1). Each drug dose-dependently increased choice of infusions with at least one unit dose of each drug producing more than 80% choice of drug. The rank order potency was carfentanil > 3-methylfentanyl > fentanyl > heroin > methamphetamine > cocaine (top panel, Fig. 1). For opioids, response rates increased as monkeys switched from responding for food to responding for drug with the number of trials completed was not markedly changed (middle and bottom panels, Fig. 1). In contrast, response rates were not changed or decreased as the unit dose of methamphetamine and cocaine, respectively, increased and monkeys chose drug over food; the number of trials completed was not markedly changed (middle and bottom panels, Fig. 1).

Fig. 1.

Effects of saline, carfentanil, 3-methylfentanyl, fentanyl, heroin, methamphetamine, and cocaine in five monkeys choosing between food and intravenous infusions. Each data point represents results from a single session averaged across monkeys (±1 S.E.M.). Ordinates indicate the percentage of trials in which monkeys received an infusion (top panels), response rates in responses/s (middle panels), and number of trials completed (bottom panels). Abscissae indicate the unit dose of drug (µg/kg/inf) available for self-administration; points above S indicate effects obtained when monkeys could choose between food and saline.

Naltrexone attenuated choice of carfentanil, 3-methylfentanyl, fentanyl, and heroin, which was evident by rightward shifts in dose-effect curves. Antagonism by 0.032 mg/kg naltrexone was similar for fentanyl and its analogs with the magnitude of shifts varying from 27- to 35-fold (top panels, Fig. 2). Naltrexone also antagonized heroin, although the 71-fold shift in the heroin dose-effect curve was greater than the shift of the curves for the other opioids (top, left panel, Fig. 3). In contrast, naltrexone produced a small (<3-fold), but significant, shift in choice of methamphetamine and no significant shift in choice of cocaine. Despite shifts in choice of drug over food and the availability of larger unit doses for self-administration, effects of opioids, methamphetamine, and cocaine on response rates and number of trials completed were not markedly changed by naltrexone (middle and bottom panels, Figs. 2 and 3).

Fig. 2.

Effects of 0.032 mg/kg naltrexone on responding for carfentanil, 3-methylfentanyl, and fentanyl in five monkeys choosing between food and intravenous drug infusions. Each data point represents results from a single session averaged across monkeys (±1 S.E.M.). Ordinates indicate the percentage of trials in which monkeys received an infusion (top panels), response rates in responses/s (middle panels), and number of trials completed (bottom panels). Abscissae indicate the unit dose of drug (µg/kg/inf).

Fig. 3.

Effects of 0.032 mg/kg naltrexone on responding for heroin, methamphetamine, and cocaine in five monkeys choosing between food and intravenous drug infusions. Each data point represents results from a single session averaged across monkeys (±1 S.E.M.). Ordinates indicate the percentage of trials in which monkeys received an infusion (top panels), response rates in responses/s (middle panels), and number of trials completed (bottom panels). Abscissae indicate the unit dose of drug (µg/kg/inf).

Daily treatment with MCAM also attenuated choice for µ opioid receptor agonists and not those of methamphetamine and cocaine. When vehicle was given 60 minutes before sessions in which monkeys could respond for food and 0.32 µg/kg/infusion of fentanyl, all four monkeys chose fentanyl during 100% of choice trials with a mean response rate of 3.75 ± 0.36 responses/s and 27 ± 0 trials completed. On the next day (i.e., day 1 of MCAM treatment), 0.032 mg/kg MCAM was given 60 minutes before sessions, and choice of 0.32 µg/kg/infusion fentanyl was decreased, on average, to less than 20% of control and remained low throughout MCAM treatment. In contrast, average choice of methamphetamine was more than 80% on days 4 and 8 of treatment. MCAM decreased the potencies of µ opioid receptor agonists and did not change the potencies of methamphetamine and cocaine. Across multiple determinations, carfentanil, 3-methylfentanyl, and fentanyl dose-effect curves were shifted between 20- and 48-fold rightward (top panels, Fig. 4), whereas heroin dose-effect curves were shifted further (71- to 72-fold; top left panel, Fig. 5). In contrast, the potencies of methamphetamine and cocaine were not changed during daily MCAM administration (top middle and top right panels, Fig. 5). Despite shifts in choice of opioids over food and the availability of larger unit doses of those drugs for self-administration, effects of opioids, methamphetamine, and cocaine on response rates and number of trials completed were not markedly changed by daily MCAM administration (middle and bottom panels, Figs. 4 and 5). Treatment ended after 184 days, and daily MCAM was replaced with vehicle injections before sessions; choice of 0.32 µg/kg/infusion of fentanyl over food increased across days with monkeys choosing drug during 14.5 ± 10.9% of choice trials on the day after the last dose of MCAM and all monkeys choosing drug exclusively within 7 days of the last dose of MCAM. Monkeys continued to respond for 10 µg/kg/infusion methamphetamine over food after MCAM treatment. Dose-effect curves obtained after MCAM treatment were similar to those obtained before treatment (inverted triangles, top panels, Figs. 4 and 5).

Fig. 4.

Effects of treatment with 0.032 mg/kg/d MCAM on responding for carfentanil, 3-methylfentanyl, and fentanyl in four monkeys choosing between food and intravenous drug infusions. Dose-effect curves were generated once before treatment began, at least twice during daily MCAM treatment, and once after discontinuation of treatment. The first and last dose-effect curves obtained during treatment are shown. Ordinates indicate the percentage of trials in which monkeys received an infusion (top panels), response rates in responses/s (middle panels), and number of trials completed (bottom panels). Abscissae indicate the unit dose of drug (µg/kg/inf).

Fig. 5.

Effects of treatment with 0.032 mg/kg/d MCAM on responding for heroin, methamphetamine, and cocaine in four monkeys choosing between food and intravenous drug infusions. Dose-effect curves were generated once before treatment began, twice during daily MCAM treatment, and once after discontinuation of treatment. Ordinates indicate the percentage of trials in which monkeys received an infusion (top panels), response rates in responses/s (middle panels), and number of trials completed (bottom panels). Abscissae indicate the unit dose of drug (µg/kg/inf).

Naltrexone and MCAM significantly increased the unit dose of opioids estimated to produce 50% responding for drug. A dose of 0.032 mg/kg naltrexone significantly increased ED₅₀ values of carfentanil (t₄ = 15.28, P = 0.0001), 3-methylfentanyl (t₄ = 8.22, P = 0.0012), fentanyl (t₄ = 11.18, P = 0.0004), and heroin (t₄ = 15.58, P < 0.0001); in contrast, naltrexone produced a small, but significant, increase in ED₅₀ value of methamphetamine (t₄ = 4.37, P = 0.012) and no significant change in the ED₅₀ value of cocaine (t₄ = 1.43, P = 0.23; data to the left of the solid vertical line, all panels, Fig. 6). Daily MCAM administration also increased ED₅₀ values for carfentanil (F_{1.431, 4.293} = 28.32, P = 0.004), 3-methylfentanyl (F_{1.428, 4.283} = 20.38, P = 0.0076), fentanyl (F_{2.159, 6.478} = 39.89, P = 0.0002), and heroin (F_{1.557, 4.670} = 61.74, P = 0.0005); ED₅₀ values for methamphetamine (F_{2.248, 6.745} = 0.72, P = 0.53) and cocaine (F_{1.175, 3.526} = 0.55, P = 0.53) were not significantly changed (data to the right of the solid vertical line, all panels, Fig. 6). After discontinuation of daily MCAM treatment, ED₅₀ values for choice of µ opioid receptor agonists were not significantly different from to those obtained before treatment began for any drug.

Fig. 6.

ED₅₀ values obtained from dose-effect curves for each drug alone, in combination with naltrexone, and during and after daily MCAM administration in monkeys choosing between food and intravenous infusions. Full curves are shown for some determinations in the top panels of Figs. 2–5. Each symbols represents the value for an individual monkey with horizontal lines indicating group means. Naltrexone antagonism was analyzed using paired t-tests whereas MCAM antagonism was analyzed using one-way repeated measures ANOVA followed by Dunnett’s multiple comparisons tests; * denotes significant differences from values obtained in the absence of an antagonist. Ordinates indicate log ED₅₀ values. Abscissae indicate the injection given before sessions (left of solid vertical line) or the day of MCAM treatment (right of solid vertical line; bef=before, aft=after).

Discussion

Improving treatment of OUD is needed to reduce the ongoing opioid crisis in the US. MCAM is being developed as a medication for OUD because of its novel pharmacological profile of activity. Specifically, MCAM provides long-lasting, selective antagonism at µ opioid receptors (Broadbear et al., 2000; Gerak et al., 2019; Maguire et al., 2019, 2020). As MCAM continues to advance as a potential treatment for OUD, additional data are required to fully establish its usefulness as a medication. The current study used a choice procedure in which monkeys responded for food or intravenous infusions to examine the ability of MCAM to attenuate the abuse-related effects of the ultra-potent fentanyl analogs carfentanil and 3-methylfentanyl.

In previous studies, MCAM decreased the positive reinforcing effects of heroin and fentanyl using a single-response self-administration procedure (Maguire et al., 2019; 2020; Maguire and France 2022). Results of the current study are consistent with those findings. In this choice procedure where monkeys could respond for food on one lever and intravenous infusions on the other lever, monkeys chose food over saline and switched to responding for drug as unit doses increased, generating a monotonic curve for reinforcing effects. MCAM significantly attenuated choice of µ opioid receptor agonists, including heroin and fentanyl, shifting their dose-effect curves rightward up to 72- and 38-fold, respectively. Similar results were obtained in monkeys responding under the single-response self-administration procedure, which used number of infusions as a measure of reinforcing effects (Maguire and France 2022), a dependent variable that produces a biphasic dose-effect curve. When MCAM was administered under treatment conditions identical to those used in the current study, the fentanyl dose-effect curve was shifted 20-fold rightward. MCAM did not alter the reinforcing effects of cocaine in either study. Despite the use of different types of self-administration procedures, similar effects were obtained, thereby replicating the earlier results and providing further support for the development of MCAM for treating OUD.

Another feature of the current study involves multiple determinations of dose-effect curves during daily treatment to assess the consistency of antagonism by MCAM over time. In the current study, a small dose of MCAM was given daily, and there was no evidence that antagonism by MCAM changed over time. Dose-effect curves for each µ opioid receptor agonist were generated at least twice with no evidence that antagonism varied markedly across determinations or among the various agonists, particularly for fentanyl and its derivatives. Thus, daily MCAM treatment produced rapid and sustained blockade of the positive-reinforcing effects of opioids, suggesting that repeated administration of a small dose or an injectable extended-release formulation could be effective for treating OUD. Two studies have used the same MCAM treatment regimen; however, the duration of treatment differed between the studies. In the earlier study (Maguire and France 2022), treatment ended after 52 days, whereas in the current study, treatment lasted 184 days. While monkeys in both groups recovered completely after treatment was discontinued, recovery of responding for 0.32 µg/kg/infusion of fentanyl occurred within 2 days in the earlier study and within 7 days in the current study. Although effects obtained during daily MCAM administration appear to be comparable to those expected during repeated or sustained-released treatment with naltrexone, long-term treatment with MCAM might provide protection from opioid abuse after treatment ends, with the duration of this protection dependent on MCAM dose as well as frequency and duration of treatment; in contrast, the effects of naltrexone will not linger once the naltrexone has been eliminated.

As indicated above, the current and previous studies have shown that repeated administration of a small dose of MCAM attenuated the positive-reinforcing effects of opioids and did not antagonize choice of methamphetamine and cocaine. This selectivity indicates that patients using cocaine or methamphetamine along with opioids would need additional treatment to address polydrug abuse. In fact, the current study was designed to exploit this selectivity by having methamphetamine available periodically for self-administration. Sessions in which monkeys chose between food and methamphetamine were conducted several times each week, thereby ensuring that monkeys responded for infusions during MCAM treatment. Otherwise, monkeys would have responded for food in most sessions, and a lever bias could have developed. Because monkeys alternated between responding on one lever for food and on the other lever for methamphetamine, persistent responding on one lever was avoided.

Importantly, this study demonstrated the ability of MCAM to antagonize the positive-reinforcing (abuse-related) effects of ultra-potent fentanyl analogs. A number of different fentanyl analogs have been detected under a variety of conditions, such as overdose deaths, patients with positive urine screens for fentanyl and other opioids, and cases of impaired driving (Brunetti et al., 2021; Chhabra et al., 2021; Kiely and Juhascik 2021). Fentanyl analogs, such as carfentanil and 3-methylfentanyl, are much more potent than fentanyl, an observation that was supported by results from the current study (Fig. 1). Moreover, overdose involving these drugs is reportedly difficult to reverse, often requiring multiple administrations of naloxone. In rats, naltrexone is more potent in antagonizing the discriminative-stimulus effects of fentanyl and heroin compared with antagonism of the effects of carfentanil (Flynn and France 2022). Consequently, demonstrating that MCAM can block the abuse-related effects of ultra-potent fentanyl analogs would dramatically improve the clinical utility of MCAM. In the current study, both naltrexone and MCAM antagonized choice of opioids with similar rightward shifts in dose-effect curves for fentanyl, carfentanil, and 3-methylfentanyl, suggesting that both antagonists similarly attenuate the positive-reinforcing effects of fentanyl and its ultra-potent analogs.

To investigate the ability of MCAM to block the reinforcing effects of ultra-potent fentanyl analogs, a treatment regimen of MCAM was selected based on previous studies in this laboratory. Doses of MCAM larger than 0.032 mg/kg produce long-lasting antagonism of the effects of µ opioid receptor agonists such that tests with MCAM would have to be separated by at least 3 weeks to ensure complete recovery between acute MCAM administrations (Gerak et al., 2019). Generating complete dose-effect curves would have been impractical in a study in which each dose-effect curve was obtained by increasing the unit dose across sessions. The smaller dose of MCAM (i.e., 0.032 mg/kg) has a shorter duration of action; however, acutely that dose does not reliably attenuate the positive-reinforcing effects of fentanyl (Maguire and France 2022). In that earlier study, daily administration of the smaller MCAM dose was found to produce consistent antagonism throughout the treatment period (Maguire and France 2022), and that treatment regimen was used in the current study so that dose-effect curves could be generated reliably and efficiently. Antagonism produced by daily treatment with this small dose of MCAM was surmountable in the earlier study and was expected to be surmountable in the current study. Nevertheless, this treatment regimen was used because it satisfied several criteria that were important for the current study. First, under these treatment conditions, MCAM antagonized the reinforcing effects of fentanyl, producing a significant (20-fold) shift to the right in the fentanyl dose-effect curve. This magnitude of shift would allow for detection of differences in the ability of MCAM to attenuate the reinforcing effects of various opioids. In the current study, MCAM antagonized choice of fentanyl and two analogs similarly, suggesting that MCAM would be equally effective in blocking abuse of these drugs. Second, antagonism by MCAM was evident by the second day of daily treatment. This rapid emergence of antagonism was desirable in the current study, which was predicted to require months of treatment to determine dose-effect curves for six different drugs on multiple occasions. Moreover, by demonstrating that ED₅₀ values do not change over 6 months of daily MCAM administration (Fig. 6), the current study adds another dimension to the understanding of long-term treatment with a small dose of MCAM that was not examined in the earlier study, indicating that antagonism remains consistent over long periods and would be expected to produce persistent and steady attenuation of abuse-related effects of opioids in OUD patients. Another reason for selecting this dosing regimen for MCAM is related to the rapid recovery after discontinuation of treatment. A single administration of a larger dose of MCAM can antagonize effects of opioids for 2 weeks. While daily administration of a larger dose than the one given in this study might have demonstrated insurmountable antagonism, recovery after treatment ended would be expected to take much longer. Thus, 0.032 mg/kg/d MCAM adequately addressed the specific goals of this current study and allowed for a rapid recovery after discontinuation, making these treatment conditions ideal for this study.

In summary, daily administration of a small dose of MCAM attenuates the abuse-related effects of µ opioid receptor agonists and not those of methamphetamine or cocaine. In addition to replicating, extending, and supporting previous findings obtained under different experimental conditions, these studies underscore the potential usefulness of MCAM by showing that repeated administration does not diminish its effectiveness and that MCAM blocks the effects of ultra-potent fentanyl analogs. Collectively with other studies on MCAM, data from this study support the continued development of MCAM as a medication for OUD.

Abbreviations

MCAM

methocinnamox

OUD

opioid use disorder

Authorship Contributions

Participated in research design: Gerak, France.

Conducted experiments: Gerak.

Performed data analysis: Gerak.

Wrote or contributed to the writing of the manuscript: Gerak, France.