Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data

Abstract

Background:

Antipsychotic-induced weight gain (AIWG) is a major concern in psychiatry, where there is a mortality gap between those with mental illness, particularly schizophrenia, and the general population. One development proposed is using centrally-acting opioid receptor antagonists (CORAs) such as naltrexone and samidorphan.

Objective:

The systematic review and meta-analysis evaluated the available human clinical trial data on the effect of CORA on AIWG.

Methodology:

Four online databases (MEDLINE, EMBASE, PsycINFO, and Cochrane) were searched for randomized-controlled trials (RCTs) on the topic. The primary outcome was change in bodyweight. Secondary anthropometric outcomes included percentage bodyweight change, BMI change, and absolute risk of weight gain. Meta-analysis was conducted on primary outcome.

Results:

Nine RCT articles (samidorphan = 6, naltrexone = 3) and two extension studies from RCTs (both samidorphan) were identified. Meta-analysis of four RCTs (n = 1416) found olanzapine/samidorphan was associated with less weight gain than olanzapine alone (mean difference in bodyweight change: −1.18 kg; 95% CI: −1.67 to −0.68). Olanzapine/samidorphan was also superior to olanzapine for changes in BMI (−0.65 kg/m²; 95% CI: −1.1 to −0.28), waist circumference (−1.5 cm; 95% CI: −2.67 to −0.32), and risk reduction for gaining 7% body weight (−12.4%; 95% CI: −18.27 to −6.54) or 10% body weight (−10.8%; 95% CI: −16.21 to −5.45). Naltrexone did not separate from placebo for change in weight or BMI.

Conclusion:

CORA, specifically samidorphan, was effective at reducing weight gain in individuals prescribed olanzapine. The small effect sizes and discrepancy between samidorphan and naltrexone suggest effects may be timing dependent, not a class effect, or dependent on the antipsychotic combination.

Keywords

Antipsychotic-induced weight gain opioid receptor antagonists samidorphan oplanzapine overweight

Introduction

Antipsychotic-induced weight gain (AIWG) and obesity in severe mental illness (SMI) are major concerns in psychiatry (Correll et al., 2011; NHS England, 2024; Woodall et al., 2024) and are often associated with second-generation antipsychotics (SGAs), such as olanzapine, clozapine, quetiapine, and risperidone (Huhn et al., 2019; Pillinger et al., 2020). SGAs are increasingly used in psychiatry, both in preference over first-generation antipsychotics and an increasing range of uses including off-label use (Højlund et al., 2019; Pillarella et al., 2012; Rhee et al., 2018, 2020). This is a concern as obesity is more prevalent in SMI, including schizophrenia (25.5%; Afzal et al., 2021), and it can lead to life-limiting diseases, including diabetes, heart disease, stroke, and cancer (Centers for Disease Control and Prevention, 2022). This may contribute to the mortality gap seen in SMI (reported estimates from 5 to 15 years) (Laursen et al., 2019; Lawrence et al., 2013; Plana-Ripoll et al., 2019), where reduction in life-years-lost from suicide has been offset by more life-years-lost from physical illnesses (Plana-Ripoll et al., 2020). In schizophrenia—where SGAs are most often used—this includes higher rates of obesity-related causes, including diabetes, heart disease, and stroke (Chen et al., 2022; Correll et al., 2022; Peritogiannis et al., 2022; Saha et al., 2007).

There have been calls for psychiatrists to play a more active role in addressing this health inequality by being more involved in supporting and/or managing their patients’ physical health care (Lawrence and Kisely, 2010; Lee et al., 2024; NHS England, 2024). SMI patients may form a vulnerable group, who may experience deprivation, marginalization, and have limited access to services. They may also lack insight and/or capacity or be otherwise hard to engage with due to the SMI. Hence, as doctors with regular contact with them, psychiatrists have a special duty of care to address their needs and are uniquely positioned to do so. This includes managing the adverse effects from SGAs that they prescribe such as weight gain and obesity (Lee et al., 2024; Saha et al., 2007).

Weight management typically involves making appropriate lifestyle adjustments (e.g., diet, exercise). Although there is meta-analytic evidence that behavioral interventions for AIWG are efficacious, the actual effect sizes are small and unlikely to translate into meaningful improvements in overall health (Álvarez-Jiménez et al., 2008). There has therefore been growing interest in the possible role of adjunctive pharmacotherapy to help combat antipsychotic-induced weight gain. Currently, antidiabetic medications are typically utilized for this purpose (Cooper et al., 2016; De Silva et al., 2016; Lee et al., 2022), though some of these medications (e.g., metformin) remain off-label. This can be limiting as not all patients may be suitable for these treatments. Additionally, psychiatrists do not typically prescribe or oversee such medications and may be thus less familiar with its management.

More recently, there have been developments in the use of centrally-acting opioid receptor antagonists (CORAs). This includes the approval of the combination medication naltrexone/bupropion for general weight management as well as the recent marketing of a olanzapine/samidorphan combination treatment, intended to mitigate the metabolic effects of olanzapine.

Background

Primarily indicated for the treatment of schizophrenia and bipolar affective disorder, antipsychotics are associated with weight gain (Huhn et al., 2019). A variety of mechanisms have been postulated to explain this propensity to increased appetite and subsequent weight gain. The traditionally understood mechanism of antipsychotic action—namely D2 dopamine receptor antagonism—is itself postulated to increase food consumption by diminishing the hedonic response to satiety mediated by dopamine transmission in the nucleus accumbens of the ventral striatum (Reynolds and Kirk, 2010).

The endogenous opioids, endorphin and enkephalin, have been implicated both in the homeostatic regulation of eating and the hedonic response to food consumption. Modulation of their effects at their primary receptors—mu-, kappa-, and delta-opioid receptors (MOR, KOR, and DOR)—has become an area of interest as a potential target for weight management pharmacotherapy (Mazereel et al., 2020; Nielsen et al., 2021; Reynolds and Kirk, 2010). Opioid receptor expression and stimulation in a variety of brain regions have been associated with increased food consumption and preference for high-calorie food in animal models (Murray et al., 2014; Nogueiras et al., 2012). Particularly, increased expression and binding of MOR in the nucleus accumbens is associated with binge eating and increased food consumption, as is direct administration of MOR agonists to this area, suggesting causal effects of MOR activation on behavioral drive to consume food (Murray et al., 2014). Furthermore, the administration of opioid antagonists, such as naloxone, nalmefene, and naltrexone, reduces both the amount of high-calorie food consumed in a single sitting and the rated pleasantness of high-calorie food in human trials, but without affecting reported levels of hunger (Yeomans and Gray, 2002). We can therefore speculate that MOR antagonists may redress the deficits in satiety signaling caused by D2 receptor antagonism, reducing overall food consumption.

The homeostatic regulatory effects of endogenous opioids on food consumption and energy expenditure are mediated by their actions on proopiomelanocortin (POMC) positive neurons found in the arcuate nucleus of the hypothalamus (ARC). POMC is a precursor peptide, the cleavage of which produces a range of neuropeptides and hormones, including adrenocorticotropic hormone, melanocyte-stimulating hormones, and the endogenous opiates beta-enkephalin and metenkephalin. Expressed widely during development, in adulthood, POMC is expressed in only in two specific locations; the POMC-positive neuronal population in ARC and the nucleus tractus solitarius of the brainstem (Zhan, 2018). These neuronal populations are sensitive to a range of neurotransmitters associated with appetite and feeding behaviors, including serotonin, endocannabinoids, and acetylcholine, as well as circulating hormone and nutrient molecules including insulin, leptin, ghrelin, glucose, and amino acids (Toda et al., 2017), making them a central node of convergent appetite regulation pathways. Activation of POMC neurons in ARC is associated with reduced food consumption and increased energy expenditure. This is evidenced by mice who have undergone selective ablation of POMC neurons in ARC, and in humans who have specific dysfunction of this neuronal population; both of whom exhibit hyperphagia and obesity (Toda et al., 2017; Zhan, 2018).

The Dopamine, Serotonin and Noradrenaline Receptor antagonists (DSNRAs), clozapine, and olanzapine are consistently found to have the greatest effect on bodyweight despite having low affinities for the D2 receptor (Dayabandara et al., 2017; Huhn et al., 2019). Instead, their obesogenic effects are thought to result from the high degree of antagonism they exert at the H1 histamine receptors and 5-HT2C serotonin receptors, effecting increased appetite and reducing satiety, respectively (Balt et al., 2011). H1 receptors are then further implicated, along with M1 muscarinic acetylcholine receptors by causing sedation, which causes reduced physical activity and therefore caloric expenditure (Green et al., 2000).

POMC neurons express serotonin 5-HTC2 receptors (Romanova et al., 2018), agonism of which results in depolarization of the neurons with anorectic consequences having demonstrated in both mice (Doslikova et al., 2013) and humans (Thomas et al., 2018). The obesogenic effects of medications used to treat psychosis with potent 5HT2C receptor antagonism, such as olanzapine and clozapine (Balt et al., 2011), are therefore mediated by directly reducing activity in this neuronal population and therefore its anorectic effects.

POMC neurons in ARC also express MOR, a G-coupled inhibitory receptor, on post-synaptic dendritic membranes. Agonism at this receptor reduces the activity of POMC, the behavioral consequence of which being increased food consumption (Nogueiras et al., 2012). The administration of MOR antagonists like naltrexone and samidorphan would therefore counter this action, reducing the inhibitory effects of endogenous opioids on POMC neurons, increasing their activity and so reducing caloric intake while increasing expenditure. By co-administering MOR antagonists with 5HT2C-antagonizing antipsychotic medications, imbalances in neurochemical regulation of POMC activity may be corrected, with a net normalization of homeostatic drive toward food consumption.

Taken together, this suggests a role for endogenous opioids in the control of feeding behaviors where release in ARC increases food consumption, and release in the nucleus accumbens reinforces feeding behaviors based on experienced pleasantness, possibly reinforcing a preference for high-calorie foods. Demonstration that opiate antagonists can attenuate these effects therefore suggests that this is a potentially helpful mechanism by which obesogenic effects of dopamine and serotonin receptor antagonists can be mitigated. As the existing evidence demonstrates that opioid antagonists reduce the drive to consume highly palatable foods, administration of these agents at the beginning of antipsychotic treatment may limit the degree of weight gain this treatment causes. It is important to recognize that the mechanism by which opioid antagonists may attenuate weight gain is distinct from those by which antipsychotics are thought to cause it. As the obesogenic effects of DSNRAs may not be dissociable from their therapeutic effects, a distinct mechanism acting at a different neurotransmitter system may be essential to avoid the loss of antipsychotic efficacy. As the endogenous opioid system appears to be central pathways on which other systems converge to regulate metabolism, this may represent an ideal candidate.

Objective

This systematic review and meta-analysis aimed to identify and analyze the available human clinical trial data for the effectiveness of prescribing CORA to manage AIWG.

Methodology

This systematic review and meta-analysis was conducted in accordance with guidelines set by the Preferred Reporting Items for Systematic Reviews and Meta-analysis (Page et al., 2021). The protocol was registered with PROSPERO (ID: CRD42023455663) (Lee et al., 2023).

Eligibility criteria

Only randomized-controlled studies (RCTs), and studies continuing from those trials, investigating the effect of CORA on weight in individuals receiving antipsychotic medication were included for this review. The primary outcome was mean bodyweight change, while secondary outcomes included other anthropometric parameters (e.g., body mass index (BMI), percentage achieving target weight change, etc), cardiometabolic parameters (e.g., lipid profile, glycated hemoglobin (HbA1c) levels, etc), and reported adverse outcomes. Further requirements for inclusion were that all participants were consenting adults (aged 16 and above) and received antipsychotics either shortly prior to or throughout the duration of the trial.

More specifically, the primary outcome examined was the difference in change in bodyweight in kilograms between antipsychotic plus CORA versus antipsychotic alone or antipsychotic plus placebo at the end of study follow-up. Secondary anthropometric outcomes examined were difference in changes in percentage bodyweight, BMI, and waist circumference, and the absolute risk difference (ARD) in gaining more than 7% bodyweight and gaining more than 10% bodyweight. The choice of ARD rather than odds ratio (the most common form of reporting in retrieved manuscripts) was made to improve interpretability and avoid overstating effect sizes (Chu et al., 2021). Also examined were secondary metabolic outcomes for the difference in change in total cholesterol (mg/dL), triglycerides (mg/dL), glucose (mg/dL), and HbA1c (%).

Other studies, such as observational studies and quasi-experimental studies, were not included in this review, to better facilitate meta-analysis of the data. Other exclusions were articles that were only available in abstract form and studies that lacked clear ethical approval. Trials involving children were not part of the review’s inclusion criteria, though no trials were excluded solely on this basis. Likewise, no language restrictions were placed to be included in the review.

Search strategy and selection process

Literature search was performed using MEDLINE, EMBASE, PsycINFO, and the Cochrane Library online databases. Search terms used included a combination of CORA medication names and terms relating to antipsychotics and weight or AIWG. A detailed list has been included in Appendix A.

An initial search was first performed on September 01, 2023, covering a period from the databases’ inception to the search date. A second, updated search was subsequently performed on August 07, 2024, covering from January 01, 2023 to the second search date to identify any articles published since the start of the review.

Initial screening was based on the articles’ titles and abstracts, with the search and screening completed independently by two authors. Differences in eligibility findings were discussed, and decision (screened in or out) was made through consensus.

Full-text evaluation for suitability based on the established inclusion and exclusion criteria was undertaken for all articles initially screened in, with all authors involved in this process. Each article was evaluated independently by at least two authors, and final decision for inclusion in the review was made through consensus agreement of all authors.

Data were extracted using a self-developed data extraction tool, which has been included in Appendix B.

Risk of bias assessment

Risk of bias assessment was undertaken for all studies included. RCTs were assessed using the Cochrane risk-of-bias tool for randomized trials, version 2 (RoB-2) (Sterne et al., 2019) while extension studies were assessed using the Risk Of Bias In Non-randomized Studies—of Interventions (Sterne et al., 2016); both developed by the Cochrane Bias Methods Group. Each study was assessed by at least two authors, with classifications determined through consensus.

Data analysis

Random effects meta-analyses for each outcome were performed using the metan command in Stata 13.0 (Fisher et al., 2024), applying the Restricted Maximum Likelihood ( Corbeil and Searle, 1976) method to estimate between-study variance. Continuous effect sizes were expressed as mean differences for changes in bodyweight (primary outcome), BMI, waist circumference, HbA1c, blood glucose, total cholesterol, and serum triglyceride. Dichotomous effect sizes were expressed as ARD for changes in probability of ⩾7% bodyweight gain and probability of ⩾10% bodyweight gain. All used confidence intervals (CIs) of 95%. Weighting was by the inverse variance method, ensuring that larger studies contributed more to the pooled estimate.

Due to incompatibility in controls (antipsychotic with placebo, placebo), samidorphan and naltrexone RCTs could not be combined for meta-analysis and have been analyzed separately. Insufficient studies were retrieved to perform meta-regression. The included studies were tested for heterogeneity using the I² statistic and assessed subjectively.

Although too few studies permit sensitive assessment of publication bias, a funnel plot and trim-fill analysis were consulted using JASP (Version 0.18.3) (JASP Team, 2024) for the primary outcome and did not show evidence of publication bias. Egger’s test was not carried out due to the small number of trials (n < 10), limiting its reliability.

Results

Initial search of the online databases on September 01, 2023 yielded 450 records (Figure 1), with 321 records remaining for manual screening after duplicates were removed. Of these, 297 were screened out, and full-text assessment was undertaken for the remaining 24 records. Thirteen were excluded (abstract only = 9, unrelated outcome measures = 4), with 11 articles accepted for inclusion in the narrative synthesis. This consisted of nine RCT articles (Correll et al., 2020, 2023; Kahn et al., 2023; Lyu et al., 2018; Martin et al., 2019; Meyer et al., 2023; Silverman et al., 2018; Taveira et al., 2014; Tek et al., 2014) and two open-label study articles (Kahn et al., 2021; Yagoda et al., 2021) that were extensions of RCTs. Four of the articles (RCT analyses = 3 (Correll et al., 2023; Meyer et al., 2023); open-label extension = 1 (Kahn et al., 2021)) were from a single trial (ENLIGHTEN-2).

Figure 1.

Article selection process.

One notable exclusion was Potkin et al. (2020), which was the original clinical trial (ENLIGHTEN-1) for its open-label extension study Yagoda et al. (2021). The aim of ENLIGHTEN-1 was to assess efficacy and safety with respect to psychiatric outcomes, demonstrating the non-inferiority of olanzapine/samidorphan to olanzapine alone for schizophrenia. A second search of the same online databases on August 07, 2024 yielded 56 records (Figure 1) but all were screened out either for lack of relevance or for already having been identified from the original search.

The descriptions and demographics of the accepted articles are described in Tables 1 and 2, and their results are in Table 3. In total, 9 separate trials were identified across 11 articles; totaling 1416, participants (intervention = 779; control = 637; extension studies only = 431). Of the included trial analyses, five investigated samidorphan across eight articles (Correll et al., 2020, 2023; Kahn et al., 2021, 2023; Martin et al., 2019; Meyer et al., 2023; Silverman et al., 2018; Yagoda et al., 2021) and three investigated naltrexone (naltrexone monotherapy = 2 (Taveira et al., 2014; Tek et al., 2014); naltrexone/bupropion = 1 (Lyu et al., 2018)). No other CORA (e.g., naloxone, nalmefene) was investigated, and there were no trials investigating multiple CORAs. No studies had a comparator arm that used a weight-managing agent/intervention (e.g., metformin, GLP-1 receptor agonist), and all blinded participants to treatment allocation. Naltrexone studies compared to placebo and combination olanzapine/samidorphan compared to olanzapine alone. Study periods ranged from 21 days to 52 weeks (RCTs only = 21 days to 24 weeks). All but two studies (Lyu et al., 2018; Silverman et al., 2018) were conducted in outpatient settings. One samidorphan study (Martin et al., 2019) conducted subgroup analysis based on dose (samidorphan 5–20 mg), though it should be noted that samidorphan 10 mg is the only approved dose (combined with variable doses of olanzapine) and that it is not approved for doses above 10 mg.

Table 1.

Article characteristics.

Authors, year	Article type	Study and study site(s) (country)	Setting	Objective	Study duration	Methodology
Correll et al. (2020)	RCT	ENLIGHTEN-2, USA	Outpatient	Assess efficacy (weight gain reduction)	24 weeks	Phase 3, multicenter, randomized, double-blind trial
Correll et al. (2023)	Post hoc analysis of RCT	ENLIGHTEN-2, USA	Outpatient	Compare OLZ/SAM with olanzapine alone for clinically relevant cardiometabolic risk factors, such as obesity, hypertension, and metabolic syndrome	24 weeks	Double-blind study. Adults with schizophrenia randomized into olanzapine-samidorphan or olanzapine groups. Post hoc analyses assessed changes from baseline to week 24 in cardiometabolic risk parameters, including BMI, risk of developing obesity (BMI ⩾ 30 kg/m²) or metabolic syndrome, waist circumference, along with mean and potentially clinically significant changes in blood pressure, glucose, and lipids
Kahn et al. (2021)	Extension study	ENLIGHTEN-2, USA	Outpatient	Assess the long-term safety and tolerability of OLZ/SAM in patients with schizophrenia	52 weeks	52-week open-label treatment with study visits every 2 weeks to assess for adverse event, weight, waist circumference, metabolic parameters, PANNS, and CGI-S
Kahn et al. (2023)	RCT	ENLIGHTEN-EARLY, multi-national (USA, UK, Ireland, Germany, Poland, Austria, Italy, Israel, Russia, South Korea, Ukraine)	Outpatient	Evaluate weight effects of combination olanzapine and samidorphan (OLZ/SAM) versus olanzapine in early-phase illness	12 weeks	Double-blind study. Young adults with early-phase (<4 years symptom onset) schizophrenia, schizophreniform disorder or bipolar I disorder; BMI < 30 kg/m²; and <24 weeks’ antipsychotic exposure were randomized to olanzapine-samidorphan or olanzapine groups. Primary endpoint was percent change from baseline body weight at week 12. Secondary endpoints were proportions with ⩾10% or ⩾7% weight gain; waist circumference change; and CGI-S change
Lyu et al. (2018)	RCT	China	Inpatient	Examine whether naltrexone and bupropion combination treatment can help weight loss and smoking cessation in patients with schizophrenia	24 week	Double-blind study. Obese patients with schizophrenia, receiving antipsychotic treatment, and who were current smokers were randomized into adjunctive naltrexone and bupropion combination or placebo groups. Body weight, BMI, fasting lipids, smoking urge, expired carbon monoxide (CO) level and cigarettes smoked per week were measured at baseline and week 24
Martin et al. (2019)	RCT	Multi-national (USA, Czechia, Bulgaria)	Outpatient	Assess if antipsychotic efficacy of olanzapine plus samidorphan is comparable to olanzapine plus placebo. Assess effect of combining olanzapine with samidorphan on olanzapine-induced weight gain. Assess overall safety and tolerability of olanzapine plus samidorphan	12 weeks following initial 1-week lead-in of olanzapine only	Double-blind, 4-arm study (olanzapine plus placebo, olanzapine plus samidorphan 5 mg, olanzapine plus samidorphan 10 mg, and olanzapine plus 20 mg). Initial 1-week open-label lead-in period of olanzapine alone followed by blinded doses of samidorphan, followed by a 12-week extension period where all patients received samidorphan, and 4-week safety follow-up period
Meyer et al. (2023)	Post-hoc analysis of RCT	USA	Outpatient	Subgroup analysis to explore the consistency of weight mitigation effect of olanzapine-samidorphan versus olanzapine across demographic subgroups	24 weeks	Double-blind study. Adults with schizophrenia randomized into olanzapine-samidorphan or olanzapine groups. Exploratory subgroup analyses were conducted by sex, age, self-reported race, and baseline BMI
Silverman et al. (2018)	RCT	USA	Other	Proof of concept	21 days	Multi-arm placebo-controlled trial in healthy volunteers. Randomized to OLZ alone, OLZ + SAM, SAM alone, or placebo in a 2:2:1:1 ratio
Taveira et al. (2014)	RCT	USA	Outpatient	Assess the effects of opioid blockade on olanzapine-treated patients’ metabolic status	12 weeks	Double-blind study. Adult patients established on olanzapine were randomized using simple coin toss into naltrexone group or placebo group. Primary outcome was change in BMI. Secondary measures were body fat and fat-free mass, homeostasis model assessment-estimated insulin rsistance (HOMA-IR), plasma lipids, and liver function test
Tek et al. (2014)	RCT	USA	Outpatient	Assess change in body weight from baseline in patients randomized to placebo versus naltrexone	8 weeks	Randomized, double-blind, placebo-controlled, fixed dose, parallel study
Yagoda et al. (2021)	Extension study	Multi-national (USA, Bulgaria, Ukraine, Serbia)	Outpatient	Assess the long-term safety and tolerability of OLZ/SAM in patients with schizophrenia	52 weeks	Open label extension of RCT. Variable OLZ dosing with fixed SAM dose. Those on OLZ at the end of ENLIGHTEN-1 switched to OLZ + SAM

Table 2.

Participant demographics.

Authors, year	Intervention	Participants (n)	Gender (% female)	Ethnicity (% White)	Nationality (% USA)	Mean/median age	Comparator	Participants (n)	Gender (% female)	Ethnicity (% White)	Nationality (% USA)	Mean/median age	Total (N)
Correll et al. (2020)	1 week OLZ 10 mg + SAM 10 mg, followed by OLZ 20 mg + SAM 10 mg	274	29.6	23	100	40.3	Olanzapine 10 mg for 1 week, then 20 mg	276	25	23.6	100	40.1	550
Correll et al. (2023)	OLZ/SAM (10/10 OR 20/10)	266	29.3	22.9		40.3	OLZ	272	25.4	23.5		40.1	538
Kahn et al. (2021)	Olanzapine max dose 20 mg; Samidorphan max dose 10 mg	265	27.5	24.2	100	40.7	No comparator	NA	NA	NA	NA	NA	265
Kahn et al. (2023)	OLZ/SAM	211	32.7	65.9	48.3	26	OLZ	215	34.9	67	48.4	25.7	426
Lyu et al. (2018)	Buproprion/naltrexone 300 mg/25 mg	11	0	0	0	54.3	Placebo	10	0	0	0	56.2	21
Martin et al. (2019)	Olanzapine 5–20 mg with samidorphan 5–20 mg, daily. Proportion samidorphan 5/10/20 (n): 75, 83, 67	225	25.2	35.9	84.6	38.4	Olanzapine 5–20 mg with placebo, daily	74	29.3	37.3	80	40.3	299
Meyer et al. (2023)	Olanzapine 10–20 mg with samidorphan 10 mg, once daily	266	29.3	22.9	100	40.3	Olanzapine 10–20 mg once daily	272	25.4	23.5	100	40.1	538
Silverman et al. (2018)	OLZ 10 mg + SAM 5 mg	34	0	67.6	100	27.3	OLZ 10 mg	35	0	60	100	25.6	69
Taveira et al. (2014)	Olanzapine 5–30 mg and naltrexone 50 mg, daily	14	35.7	Not specified	100	43.6	Olanzapine 5–30 mg and placebo, daily	16	31.3	Not specified	100	45.6	30
Tek et al. (2014)	Naltrexone 25 mg/d	10	100	36.4	100	50	Placebo	11	100	58.3	Not known	41.4	21
Yagoda et al. (2021)	Daily doses of 10/10 mg (10/10), 15/10 mg (15/10), or 20/10 mg. 83.7% receiving 20/10 mg/day	266	41.9	78.7	27.1	41.4	No comparator	NA	NA	NA	NA	NA	266

Table 3.

Article results and conclusions.

Authors, year	Primary outcomes	Secondary outcomes	Adverse events (deaths and cardiac events)	Authors’ conclusions
Correll et al. (2020)	Change in weight (kg) OLZ 5.1 kg vs OLZ + SAM 3.2 kg. −1.9 kg (p < 0.05)	Number gaining >10% OLZ + SAM 17.8% vs OLZ 29.8% (p < 0.05); waist circ OLZ + SAM 2.4 cm vs OLZ 4.5 cm (p < 0.05); cholesterol mg/dL OLZ + SAM 0.9 vs OLZ 2.1; Triglycerides mg/dL OLZ + SAM 23.9 vs OLZ 24.5; Glucose mg/dL OLZ + SAM 4.5 vs OLZ 2.3; Insulin mIU/mL OLZ + SAM 3.2 vs OLZ 3.4; HbA1c % OLZ + SAM 0.06 vs OLZ 0.07	None reported	Not only were patients less likely to gain any weight with combined olanzapine/samidorphan, but also the risk of clinically significant weight gain (of 7% and of 10%) was reduced by 50% relative to olanzapine; a metabolic benefit for combined olanzapine/samidorphan relative to olanzapine was not observed in this study
Correll et al. (2023)	Risk difference of weight gain ⩾10%: −13.7% (95% CI: −22.8 to −4.6)	BMI: BMI change from baseline; LS mean difference 95% CI: −0.65 kg/m² (−1.01 to −0.28). Risk of obesity (BMI ⩾ 30 kg/m²); odds ratio 95% CI: 0.52 (0.32–0.82); NNT 10. Risk difference of BMI ⩾ 30 kg/m²: −10.6% (95% CI: −18.0 to −3.2). Waist. Risk difference of waist circumference increase ⩾5 cm: −17.1% (95% CI: −26.3 to −7.8). BP: Change in systolic BP; LS mean difference (95% CI): −2.63 mmHg (−4.78 to −0.47). Change in diastolic BP; LS mean difference (95% CI): −0.75 mmHg (−22.31 to 0.80). Risk of shifting from normal to hypertensive; OR (95% CI): 0.48 (0.24–0.96); NNT 7. Risk of shifting from normal/elevated to stage 1 or 2 hypertension; OR (95% CI): 00.66 (0.38–1.17); NNT 12. Risk of developing metabolic syndrome; OR (95% CI): 0.55 (0.31–0.99); NNT 20	Tachycardia: 1 event in olanzapine group	OLZ/SAM mitigated olanzapine-associated increases in BMI and supine blood pressure and reduced the risk of developing stage ½ hypertension, obesity, and metabolic syndrome—all well-established risk factors for cardiovascular morbidity and mortality
Kahn et al. (2021)	Mean (SD) weight change from baseline to week 52 was −0.03 (6.22) kg	Mean (SD) change in BMI from baseline to week 52 was −0.03 (2.04) kg/m²; Mean (SD) change in waist circumference was −0.35 (6.12) cm; Table 3 of article	Hypertension: 6 patients (2.3%)	OLZ/SAM was well tolerated over 52 weeks of treatment. Weight, waist circumference, and metabolic parameters were stable over 52 weeks. Long-term durability of OLZ/SAM was evidenced by sustained symptom improvement.
Kahn et al. (2023)	Percent weight change from baseline; LS mean difference (SE): −1.87% (0.75); p = 0.012). Proportion weight gain ⩾10%; OR: 0.64 (95% CI: 0.39–1.05); not statistically significant. Proportion weight gain ⩾7%; OR: 0.61 (95% CI: 0.39–0.94)	Waist circumference change; LS mean difference (SE): −0.92 cm (0.58) (95% CI: −2.06 to 0.22). Change in metabolic parameters from baseline (olanzapine-samidorphan vs olanzapine)- Total chol; mg/dL (SE): 7.1 (2.73) vs 13.2 (2.69)- HDL; mg/dL (SE): −5.0 (0.95) vs −2.5 (0.94)- LDL; mg/dL (SE): 8.9 (2.40) vs 11.3 (2.35)- Trig; mg/dL (SE): 26.3 (5.86) vs 29.4 (5.78)- Glucose; mg/dL (SE): 3.6 (1.21) vs 2.8 (1.20)- HbA1c; % (SE): −0.03 (0.02) vs −0.00 (0.02)- Insulin; µIU/mL: 5.57 (1.74) vs 5.86 (1.74)	1 recorded death in olanzapine-samidorphan group. Suicidal ideation: 1 event recorded in olanzapine-samidorphan group	In patients who are early in the course of schizophrenia, schizophreniform disorder, or BD-I illness, and thus at high risk for weight gain, OLZ/SAM resulted in less weight gain at week 12 compared with olanzapine while providing similar antipsychotic efficacy. Small metabolic changes observed in both the OLZ/SAM and olanzapine groups were consistent with previously reported early effects of olanzapine on lipid and glucose metabolism. Taken together with previously published findings, these data suggest that OLZ/SAM mitigates olanzapine-associated weight gain across both early and later stages of illness
Lyu et al. (2018)	Change in body weight from baseline (treatment group vs placebo) (kg)- Mean (SD): 0.2 (5.3) vs −0.9 (3.1); p = 0.779	Change in metabolic parameters from baseline (treatment group vs placebo); Mean (SD)- BMI (kg/m²): 0.0 (1.9) vs −0.3 (1.1); p = 0.314- Trig: −1.0 (0.9) vs −0.4 (0.6); p = 0.591- HDL: 0.0 (1.1) vs 0.0 (0.1); p = 0.657- LDL: −0.4 (0.7) vs −0.2 (0.7); p = 0.900- Glucose: 0.5 (1.5) vs 1.0 (1.6); p = 0.264- HbA11c: 0.05 (0.6) vs 0.4 (1.1); p = 0.385	No recorded MI. Tachycardia: 1 event in placebo group	Study failed to demonstrate the benefit of naltrexone and bupropion combination treatment for weight loss or smoking cessation in patients with schizophrenia
Martin et al. (2019)	Body weight change (kg); LS mean difference: −1.9 kg (95% CI: −2.9 to −0.9, p < 0.001)	Subgroup calculations:- 5 mg: −1.3 (95% CI: −2.6 to −0.0, p = 0.049)- 10 mg: −1.9 (95% CI: −3.2 to −0.7, p = 0.003)- 20 mg: −1.2 (95% CI: −2.5 to 0.1, p = 0.072)Body weight change (%); LS mean difference: −2.7% (95% CI: −4.0 to −1.4, p < 0.001)Risk of patients gaining ⩾10% baseline weight; OR: 4.1 (95% CI: 1.4–12.3, p = 0.008)Risk of patients gaining ⩾7% baseline weight; OR: 2.21 (95% CI: 0.9–5.3, p = 0.07)	Ischemic stroke: 1 event in “OLZ + placebo” group. No recorded deaths. Suicidal ideation: 1 event recorded in “OLZ + SAM 5 mg” group	Olanzapine-samidorphan was associated with clinically meaningful and statistically significant mitigation of weight gain compared with olanzapine plus placebo
Meyer et al. (2023)	Percent change in weight from baseline; LS mean difference: −2.38% (95% CI: −3.88 to −0.88)	Subgroups; LS mean % (95% CI):- Male: −2.73 (−4.45 to −1.01)- Female: −1.53 (−4.43–1.38)- <30 years: −3.43 (−7.00 to 0.13)- ⩾30 years: −2.14 (−3.78 to −0.51)- Black: −2.37 (−4.10 to −0.63)- Non-black: −2.41 (−5.28 to 0.46)- BMI <27: −2.17 (−4.10 to −0.24)- BMI ⩾27: −2.70 (−5.01 to −0.39)Proportion ⩾10% weight gain; OR: 0.50 (95% CI: 0.31–0.80)Subgroups; OR (95% CI):- Male: 0.41 (0.23–0.73)- Female: 0.68 (0.30–1.55)- <30 years: 0.65 (0.25–1.70)- ⩾30 years: 0.46 (0.27–0.78)- Black: 0.47 (0.27–0.82)- Non-black: 0.55 (0.24–1.30)- BMI < 27: 0.57 (0.32–1.00)- BMI ⩾ 27: 0.38 (0.16–0.87)	None reported	Weight mitigating effects of olanzapine-samidorphan versus olanzapine were observed consistently across patient subgroups and were in line with results from the overall study population
Silverman et al. (2018)	Change in weight (kg) OLZ 3.1 kg versus OLZ + SAM 2.2 kg. −0.9 kg (p = 0.02)	Fasting glucose mmol/L (+0.1; n.s); fasting insulin pmol/L (+6.7; n.s); glucose to insulin ratio (−0.4; n.s); triglyceride mmol/L (−0.2; n.s); cholesterol mmol/L (−0.25; n.s)	None reported	Subjects taking OLZ + SAM gained significantly less weight over the 21 day treatment period than subjects taking OLZ alone. Although the magnitude of this difference over 21 days was relatively modest in this proof of concept study
Taveira et al. (2014)	Body weight (kg) (mean (SE))- OLZ + NTX: baseline: 92.4 (2.2); 12-week: 91.3 (2.6)- OLZ + placebo: baseline: 94.7 (2.5); 12-week: 93.8 (2.6)- p = 0.95Body fat mass (kg) (mean (SE))- OLZ + NTX: baseline: 34.7 (2.3); 12-week: 32.0 (2.9)- OLZ + placebo: baseline 33.7 (2.8); 12-week: 33.7 (2.8)- p = 0.001Fat-free mass (kg) (mean (SE))- OLZ + NTX: baseline: 57.7 (3.0); 12-week: 59.3 (3.5)- OLZ + placebo: baseline: 61.0 (3.4); 12-week: 60.0 (3.2)- p = 0.03	BMI (kg/m²) (mean (SE))- OLZ + NTX: baseline: 31.7 (1.0); 12-week: 31.3 (1.3)- OLZ + placebo: 33.0 (1.5); 12-week: 32.7 (1.5)- p = 0.83Chol (mg/dL) (mean (SE))- OLZ + NTX: baseline: 190 (14.5); 12-week: 170.5 (13.1)- OLZ + placebo: 193.1 (13.8); 12-week: 178.4 (15.5)- p = 0.60HDL (mg/dL) (mean (SE))- OLZ + NTX: baseline: 43.8 (3.1); 12-week: 39.7 (2.0)- OLZ + placebo: baseline: 51.7 (5.6); 12-week: 44.1 (4.0)- p = 0.09LDL (mg/dL) (mean (SE))- OLZ + NTX: baseline: 109 (16.1); 12-week: 170.5 (13.1)- OLZ + placebo: baseline: 100.8 (14.7); 12-week: 100.4 (17.8)- p = 0.56Trig (mg/dL) (mean (SE))- OLZ + NTX: baseline: 185.0 (45.3); 12-week: 157.5 (51.0)- OLZ + placebo: baseline: 202.5 (39.1); 12-week: 171.4 (31.5)- p = 0.85	None reported	Addition of naltrexone to olanzapine resulted in no change in BMI but showed significant attenuation of body fat mass and an increase in fat-free mass
Tek et al. (2014)	Weight change from baseline NTX (−3.40 kg); weight change placebo group (+1.37 kg)	BMI NTX was −1.37 versus BMI PLA 0.57; Waist circumference NTX −3 cm versus waist circumference PLA +1 cm; No significant difference in metabolic parameters from baseline and between groups. See Table 2	None reported	Study demonstrated that NTX produced weight loss in women with schizophrenia/schizoaffective who were acutely gaining weight on antipsychotics
Yagoda et al. (2021)	Weight change over 52 weeks 0.03 kg overall; −0.42 in those transitioning OLZ to OLZ + SAM at the start of the study versus +0.32 kg in those on continuous OLZ + SAM	Waist circumference −1.26 cm (p < 0.05) in those transitioning OLZ to OLZ + SAM at the start of the study versus +0.48 kg (n.s) in those on continuous OLZ + SAM; fasting glucose mg/Dl 1.3 (16.0); cholesterol mg/dL −2/4 (26.5); Triglycerides mg/dL −10.7 (65.6); insulin mIU/ml 2.5 (29.0); HbA1c % 0.03% (0.3)	None reported	The tolerability of OLZ/SAM is supported by the overall study completion rate, the low rate of discontinuation attributable to AEs and the low incidence of SAEs. Long-term treatment with OLZ/SAM was associated with sustained antipsychotic efficacy and minimal effects on weight, waist circumference, and metabolic parameters over 52 weeks of treatment

Risk-of-bias assessments for each study are described in Table 4. Gradings were varied among RCT articles (low risk = 3; some concerns = 5; high risk = 1) owing to heterogeneity in methodology. By contrast, the open-label trials fared more poorly; with both at high risk of bias, owing to risk of confounding (domain 1).

Table 4.

Risk-of-bias assessments.

Authors			Domain 1	Domain 2	Domain 3	Domain 4	Domain 5	Overall
RoB2
Correll et al. (2020)			1	1	1	1	1	1
Correll et al. (2023)			1	2	2	1	1	2
Kahn et al. (2023)			1	2	2	1	1	2
Lyu et al. (2018)			1	2	1	1	2	2
Martin et al. (2019)			1	1	3	1	1	3
Meyer et al. (2023)			1	1	1	1	1	1
Silverman et al. (2018)			1	1	1	1	1	1
Taveira et al. (2014)			1	1	2	1	2	2
Tek et al. (2014)			2	1	1	1	1	2
Authors	Domain 1	Domain 2	Domain 3	Domain 4	Domain 5	Domain 6	Domain 7	Overall
RoB-I
Kahn et al. (2021)	3	1	1	1	1	1	1	3
Yagoda et al. (2021)	3	1	3	1	1	2	1	3

Legend: Ratings (1 = low risk, 2 = some concerns/moderate risk, 3 = high risk/serious risk); RoB2 domains (1 = randomization process, 2 = deviation from the intended interventions, 3 = missing outcome data, 4 = measurement of outcome, 5 = selection of the reported result); RoB-I domains (1 = confounding, 2 = selection of participants, 3 = classification of interventions, 4 = deviation from the intended interventions, 5 = missing data, 6 = measurement of outcomes, 7 = selection of the reported results).

For primary outcome, mean difference in bodyweight change was statistically significant for samidorphan when trialed as olanzapine/samidorphan against olanzapine (MD: −1.18 kg; 95% CI: −1.67 to −0.68; p < 0.001) while naltrexone was not when evaluated against placebo. Due to the small number of studies using naltrexone-containing interventions (n = 3), it was thought more meaningful for naltrexone/bupropion to be analyzed with naltrexone monotherapy studies in the naltrexone subgroup. However, neither naltrexone monotherapy nor naltrexone/bupropion was significant when analyzed separately. Primary outcome funnel plot and sensitivity analysis have been included in Appendix C. No publication bias was found for the primary outcome (Figure 2).

Figure 2.

Forest plot for mean difference in bodyweight change (kg).

Secondary outcomes meta-analyzed included other weight and obesity-related outcomes (BMI, risk of weight gain) and cardiometabolic outcomes (cholesterol, triglyceride, glucose, glycated hemoglobin (HbA1c)), and tabulated in Tables 5 and 6, with their respective forest plots included in Appendix D. BMI change was found to be statistically significant, favoring intervention, for olanzapine/samidorphan but not significant in the naltrexone trials; though it is noted that the evidence for olanzapine/samidorphan around BMI consists of a single RCT (Correll et al., 2023) and has yet to be replicated. Three trials (Correll et al., 2023; Kahn et al., 2023; Martin et al., 2019) compared the risk of weight gain with olanzapine/samidorphan against standalone olanzapine and this was statistically significant for both, weight gain ⩾7% (risk difference: −12.40%; 95% CI: −18.27 to −6.54; p < 0.001) and ⩾10% (risk difference: −10.83%; 95% CI: −16.21 to −5.45; p < 0.001) on meta-analysis. However, neither were statistically significant in Martin et al. (2019) and only Correll et al. (2023) was statistically significant for ⩾10% weight gain when these results are expressed in risk difference.

Table 5.

Summary of meta-analyses of anthropometric outcome measures.

Outcome measures	Number of RCTs	N participants (intervention)	N participants (control)	Effect size	p	95% CI
Outcome measures	Number of RCTs	N participants (intervention)	N participants (control)	Effect size	p	Lower	Upper
Samidorphan
Mean difference in weight (kg) (Correll et al., 2020; Kahn et al., 2023; Martin et al., 2019; Silverman et al., 2018)	4	744	600	−1.18	<0.001	−1.671	−0.682
Mean difference in BMI change (kg/m²) (Correll et al., 2023)	1	274	276	−0.65	<0.001	−1.01	−0.28
Mean difference in waist circumference (cm) (Correll et al., 2020, 2023)	2	485	491	−1.50	0.013	−2.67	−0.32
Absolute risk difference for gaining ⩾7% body weight (%) (Correll et al., 2020; Kahn et al., 2023; Martin et al., 2019)	3	710	565	−12.40	<0.001	−18.27	−6.54
Absolute risk difference for gaining ⩾10% body weight (%) (Correll et al., 2020; Kahn et al., 2023; Martin et al., 2019)	3	710	565	−10.83	<0.001	−16.21	−5.45
Naltrexone
Mean difference in weight (kg) (Lyu et al., 2018; Taveira et al., 2014; Tek et al., 2014)	3	35	37	−1.72	0.458	−6.25	2.81
Mean difference in BMI change (kg/m²) (Lyu et al., 2018; Taveira et al., 2014; Tek et al., 2014)	3	35	37	−0.68	0.137	−2.45	1.08

Effect size: Magnitude of the observed effect; Effect size direction: negative values favor intervention over placebo; p: p-value indicating statistical significance (significance threshold: p < 0.05); Absolute risk difference has been presented in percentage and not in decimal form.

Table 6.

Summary of meta-analyses of laboratory cardiometabolic outcome measures.

Outcome measures	Number of RCTs	N participants (intervention)	N participants (control)	Effect size	p	95% CI
Outcome measures	Number of RCTs	N participants (intervention)	N participants (control)	Effect size	p	Lower	Upper
Samidorphan
Mean difference in HbA1c change (%) (Correll et al., 2020, 2023; Kahn et al., 2023)	2	485	491	−0.02	0.355	−0.06	0.02
Mean difference in blood glucose change (mg/dL) (Correll et al., 2020; Kahn et al., 2023; Martin et al., 2019; Silverman et al., 2018)	4	744	600	0.61	0.522	−1.25	2.46
Mean difference in total cholesterol change (mg/dL) (Correll et al., 2020; Kahn et al., 2023; Martin et al., 2019; Silverman et al., 2018)	4	744	600	−4.36	0.016	−7.901	−0.820
Mean difference in serum triglyceride change (mg/dL) (Correll et al., 2020; Kahn et al., 2023; Martin et al., 2019; Silverman et al., 2018)	4	744	600	−3.17	0.518	−12.783	6.442
Naltrexone
Mean difference in HbA1c change (%) (Lyu et al., 2018; Tek et al., 2014)	2	25	26	0.18	0.676	−0.67	1.04
Mean difference in blood glucose change (mg/dL) (Lyu et al., 2018; Tek et al., 2014)	2	25	26	−0.04	0.997	−23.19	23.11
Mean difference in total cholesterol change (mg/dL) (Lyu et al., 2018; Taveira et al., 2014; Tek et al., 2014)	3	35	37	−3.85	0.576	−17.33	9.64
Mean difference in serum triglyceride change (mg/dL) (Lyu et al., 2018; Taveira et al., 2014; Tek et al., 2014)	3	35	37	−9.032	0.666	−50.091	32.027

Analysis by baseline demographical subgroup was conducted by two olanzapine/samidorphan trials (Martin et al., 2019; Meyer et al., 2023) around treatment dose, sex, age, ethnicity, and baseline BMI. Here, it was found that weight change was not statistically significant in the following subgroups—samidorphan 20 mg (mean: −1.3 kg, 95% CI: −2.5 to 0.1) (Martin et al., 2019), female (mean: −1.53%, 95% CI: −4.43 to 1.38) (Meyer et al., 2023), under 30-year-old (mean: −3.43%, 95% CI: −7.00 to 0.13) (Meyer et al., 2023), and non-black subgroups (mean: −2.41%, 95% CI: −5.28 to 0.46) (Meyer et al., 2023).

In the open-label extension studies, participants taking established olanzapine treatment were initiated on samidorphan alongside the treatment group; with all participants then followed-up for 52 weeks. The authors found that olanzapine/samidorphan was well tolerated, with low rate of discontinuation attributable to adverse effects, and antipsychotic efficacy was maintained long-term (Kahn et al., 2021; Yagoda et al., 2021). After 52 weeks, mean weight change was −0.03 kg (SD: 6.22) (Kahn et al., 2021); with little meaningful difference in terms of weight change between participants newly initiated on samidorphan at the beginning of the trial extension, and those previously established on samidorphan (−0.42 vs 0.32 kg) (Yagoda et al., 2021).

Discussion

Interpretation

Our findings indicate that olanzapine/samidorphan partially mitigates AIWG when compared against olanzapine. We did not find any effect of naltrexone alone or in combination with bupropion on AIWG.

The effect on reducing weight gain appears smaller than comparable treatment options, such as GLP-1 receptor agonists (Khaity et al., 2023) and metformin, at face value (De Silva et al., 2016). However, this difference may be related to study design factors and direct head-to-head trials would be beneficial.

Overall, our findings potentially point to CORA only having a narrow use case in AIWG, in the presence of other available options. While treatment with GLP-1 receptor agonists and metformin may be better established, the former is administered by injection, and both are medication groups not typically prescribed by psychiatrists. Instead, they typically fall into the domain of endocrinologists and general practitioners, who prescribe them for Type 2 diabetes mellitus. This may limit access to those medications without the involvement of the appropriate specialists. Furthermore, there is developing concern regarding development of suicidal ideation among patients prescribed GLP-1 receptor agonists (Guirguis et al., 2024; McIntyre et al., 2024; Tobaiqy and Elkout, 2024) though such links remain unproven with surveillance remaining ongoing (European Medicines Agency, 2024; Food and Drug Association, 2024; Wang et al., 2024). Aripiprazole—whether as adjunct or in place of the patient’s regular antipsychotic—is another option described in guidelines by the British Association of Psychopharmacology, Maudsley, and American Psychiatric Association (2020; Cooper et al., 2016; Taylor et al., 2020). However, this may involve either reducing/discontinuing the patient’s regular antipsychotic or exposing the patient to high combined doses of antipsychotic, and so may not always be suitable. In that respect, it may be that CORA is an option when psychiatrists are unable to access metformin and/or GLP-1 receptor agonists; there is concern around risk to self (e.g., suicidal ideation, self-harm); and/or introduction of aripiprazole is unsuitable.

Limitations

The main limitation is that trials evaluating samidorphan and naltrexone used different comparators; preventing their results from being combined for meta-analysis. There was also a lack of head-to-head comparison against other advocated treatments (e.g., metformin, GLP-1 receptor agonists, aripiprazole, nonpharmacological interventions) to investigate superiority.

Furthermore, there was heterogeneity in the included studies’ methodology. This included study duration (8–52 weeks); medication and dose (naltrexone 25–50 mg, naltrexone/bupropion 25 mg/300 mg, samidorphan 5–20 mg); sample size (21–550 participants); and target population (individuals with diagnosed SMI, healthy individuals). This limited the interpretation and what conclusions could be drawn from the results, due to the variables’ potential to influence effect size and significance.

For instance, Lyu et al. (2018) was conducted in an inpatient setting and sought to investigate both, the weight-reducing and smoking-cessating properties of naltrexone/bupropion simultaneously. These are potential confounders as smoking cessation is associated with weight gain (Filozof et al., 2004), whereas the inpatient setting may limit participants’ access and opportunity to engage in regular physical activity.

Additionally, RCTs investigating naltrexone (including naltrexone/bupropion) were small in scale (n ⩽ 30) and may have been underpowered to detect significant results. This contrasts with the studies using samidorphan, which were generally larger, with all but Silverman et al. (2018)—a phase 1 trial—involving over 200 participants each. In the trials of olanzapine/samidorphan, the CORA was initiated at the same time as the antipsychotic. The trials of naltrexone versus placebo involved initiating the CORA in patients who had already experienced weight gain.

Samidorphan studies solely included participants receiving olanzapine, which is recognized as one of the most obesogenic antipsychotics (Huhn et al., 2019), as opposed to other antipsychotics. The observed effect may not generalize to other antipsychotics. It is plausible that samidorphan may have a greater or lesser effect on AIWG from other, less obesogenic antipsychotics. This limitation in the literature may be a consequence of samidorphan being developed as, and currently only available in, the olanzapine/samidorphan combined preparation. We are however unaware of any mechanistic reason to anticipate samidorphan working differently if paired with other antipsychotics. Combined, these differences in methodology between samidorphan and naltrexone trials render it impossible to say whether there is a class effect of CORAs.

RCT length was limited to 24 weeks or less. Longer-term data—which was limited only to samidorphan—was instead based on open-label, extension studies (Kahn et al., 2021; Yagoda et al., 2021) where all participants received the intervention (specifically, samidorphan), with there being no long-term RCTs comparing CORA recipients and non-CORA recipients. It is possible that the relatively modest weight reduction obtained for CORA may be due to the duration of the trials and that RCTs of longer duration could produce more pronounced results.

Geographically, study sites were heavily concentrated in the United States and Europe, with only two studies including participants from outside this region—specifically, Israel (Kahn et al., 2023), South Korea (Kahn et al., 2023), and China (Lyu et al., 2018). Populations from Africa, Australasia, South America, and South Asia were completely absent from the available data, and we were thus unable to include them to inform our review and meta-analysis.

Future directions

Our findings identify early but conflicting results around the efficacy of CORA at managing AIWG. We observed that different medications within the CORA class potentially produce significantly different clinical effects. One possible explanation may be differences in their pharmacological properties; with Tan et al. (2022) describing samidorphan as having a higher affinity to MOR and DOR receptors, and more potent antagonism at MOR, when compared to naltrexone, and Cunningham et al. (2019) suggesting that it may be contributed by samidorphan having peripheral—in addition to central—effects. Further research in these dimensions could shed light on the pathophysiology of AIWG and development of more effective medications for it. The olanzapine/samidorphan trials reported their primary outcomes with an emphasis on the number of participants gaining 7% or 10% bodyweight. The results from our analysis demonstrate rather modest mean reductions in weight gain, waist circumference, and BMI which are considerably less flattering to the medications overall efficacy.

Currently, samidorphan shows the most promise amongst existing CORA. However, there is a need for further study and replication. It is notable that all samidorphan trials identified share the same commercial funder. Independent study would be desirable to see; as would further trials involving non-Western populations and different antipsychotics; using more standardized methodology and consistent outcome measures that include its clinical usefulness.

Other avenues of research would be more detailed exploration of its safety profile, including any potential effects on mental health symptoms and/or interactions with antipsychotics; economic studies and/or cost-effectiveness; as well as long-term outcomes. Ideally, there would also be head-to-head trials comparing its efficacy against other treatment options to more directly investigate superiority, as this may inform clinical practice and where CORA sits—if at all—within treatment guidelines for AIWG.

Conclusion

Overall, our systematic review and meta-analysis found that CORA was effective at reducing AIWG. However, the evidence for this is moderate at best, with a lack of studies that were low risk of bias, and limited to samidorphan only—which produced only a small effect size on subgroup analysis. Studies using naltrexone had mixed results, whilst no data were available for naloxone and nalmefene. With samidorphan, evidence was limited to individuals prescribed olanzapine, with no data for any other antipsychotic. All of this highlights a need for further research around CORA in AIWG.

Supplemental Material

sj-docx-1-jop-10.1177_02698811251337374 – Supplemental material for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data

Supplemental material, sj-docx-1-jop-10.1177_02698811251337374 for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data by Kenn Lee, Matthew Twohig, Nguemo Pauline Idoko and Benjamin David Williams in Journal of Psychopharmacology

Supplemental Material

sj-docx-2-jop-10.1177_02698811251337374 – Supplemental material for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data

Supplemental material, sj-docx-2-jop-10.1177_02698811251337374 for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data by Kenn Lee, Matthew Twohig, Nguemo Pauline Idoko and Benjamin David Williams in Journal of Psychopharmacology

Supplemental Material

sj-docx-3-jop-10.1177_02698811251337374 – Supplemental material for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data

Supplemental material, sj-docx-3-jop-10.1177_02698811251337374 for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data by Kenn Lee, Matthew Twohig, Nguemo Pauline Idoko and Benjamin David Williams in Journal of Psychopharmacology

Supplemental Material

sj-docx-4-jop-10.1177_02698811251337374 – Supplemental material for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data

Supplemental material, sj-docx-4-jop-10.1177_02698811251337374 for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data by Kenn Lee, Matthew Twohig, Nguemo Pauline Idoko and Benjamin David Williams in Journal of Psychopharmacology

Supplemental Material

sj-docx-5-jop-10.1177_02698811251337374 – Supplemental material for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data

Supplemental material, sj-docx-5-jop-10.1177_02698811251337374 for Centrally-acting opioid receptor antagonists as a treatment for antipsychotic-induced weight gain: A systematic review and meta-analysis of clinical trial data by Kenn Lee, Matthew Twohig, Nguemo Pauline Idoko and Benjamin David Williams in Journal of Psychopharmacology

Footnotes

Acknowledgements

None.

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.

Funding

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

ORCID iDs

Kenn Lee

Benjamin David Williams

Supplemental material

Supplemental material for this article is available online.

References

Afzal

Siddiqi

Ahmad

, et al. (2021) Prevalence of overweight and obesity in people with severe mental illness: Systematic review and meta-analysis. Front Endocrinol 12: 769309.

Álvarez-Jiménez

Hetrick

González-Blanch

, et al. (2008) Non-pharmacological management of antipsychotic-induced weight gain: Systematic review and meta-analysis of randomised controlled trials. Br J Psychiatry 193: 101–107.

American Psychiatric Association (2020) The American Psychiatric Association Practice Guideline for the Treatment of Patients With Schizophrenia, 3rd edn. Washington, DC: American Psychiatric Association Publishing. Available at: https://psychiatryonline.org/doi/book/10.1176/appi.books.9780890424841 (accessed 11 August 2024).

Balt

Galloway

Baggott

, et al. (2011) Mechanisms and genetics of antipsychotic-associated weight gain. Clin Pharmacol Ther 90: 179–183.

Centers for Disease Control and Prevention (2022) Consequences of obesity. Available at: https://www.cdc.gov/obesity/basics/consequences.html (accessed 9 August 2024).

Chen

Perera

Shetty

, et al. (2022) Body mass index and mortality in patients with schizophrenia spectrum disorders: A cohort study in a South London catchment area. Gen Psychiatr 35: e100819.

Chu

Liu

Leas

, et al. (2021) Effect size reporting among prominent health journals: A case study of odds ratios. BMJ Evid Based Med 26: 184–184.

Cooper

Reynolds

With expert co-authors (in alphabetical order):, et al. (2016) BAP guidelines on the management of weight gain, metabolic disturbances and cardiovascular risk associated with psychosis and antipsychotic drug treatment. J Psychopharmacol 30: 717–748.

Corbeil

Searle

(1976) Restricted maximum likelihood (REML) estimation of variance components in the mixed model. Technometrics 18: 31.

10.

Correll

Lencz

Malhotra

(2011) Antipsychotic drugs and obesity. Trends Mol Med 17: 97–107.

11.

Correll

Newcomer

Silverman

, et al. (2020) Effects of olanzapine combined with samidorphan on weight gain in schizophrenia: A 24-week phase 3 study. Am J Psychiatry 177: 1168–1178.

12.

Correll

Solmi

Croatto

, et al. (2022) Mortality in people with schizophrenia: A systematic review and meta-analysis of relative risk and aggravating or attenuating factors. World Psychiatry 21: 248–271.

13.

Correll

Stein

Graham

, et al. (2023) Reduction in multiple cardiometabolic risk factors with combined olanzapine/samidorphan compared with olanzapine: Post hoc analyses from a 24-week phase 3 study. Schizophr Bull 49: 454–463.

14.

Cunningham

Eyerman

Todtenkopf

, et al. (2019) Samidorphan mitigates olanzapine-induced weight gain and metabolic dysfunction in rats and non-human primates. J Psychopharmacol 33: 1303–1316.

15.

Dayabandara

Hanwella

Ratnatunga

, et al. (2017) Antipsychotic-associated weight gain: Management strategies and impact on treatment adherence. Neuropsychiatr Dis Treat 13: 2231–2241.

16.

De Silva

Suraweera

Ratnatunga

, et al. (2016) Metformin in prevention and treatment of antipsychotic induced weight gain: A systematic review and meta-analysis. BMC Psychiatry 16: 341.

17.

Doslikova

Garfield

Shaw

, et al. (2013) 5-HT2C receptor agonist anorectic efficacy potentiated by 5-HT1B receptor agonist coapplication: An effect mediated via increased proportion of pro-opiomelanocortin neurons activated. J Neurosci 33: 9800–9804.

18.

European Medicines Agency (2024) Meeting highlights from the Pharmacovigilance Risk Assessment Committee (PRAC) 8–11 April 2024 | European Medicines Agency (EMA). Available at: https://www.ema.europa.eu/en/news/meeting-highlights-pharmacovigilance-risk-assessment-committee-prac-8-11-april-2024 (accessed 11 August 2024).

19.

Filozof

Fernandez Pinilla

Fernandez-Cruz

(2004) Smoking cessation and weight gain. Obes Rev 5: 95–103.

20.

Fisher

Harris

Bradburn

, et al. (2024) METAN: Stata Module for Fixed and Random Effects Meta-Analysis. Statistical Software Components. Boston, MA: Boston College Department of Economics.

21.

Food and Drug Association (2024) Update on FDA’s ongoing evaluation of reports of suicidal thoughts or actions in patients taking a certain type of medicines approved for type 2 diabetes and obesity. Food Drug Association. Available at: https://www.fda.gov/drugs/drug-safety-and-availability/update-fdas-ongoing-evaluation-reports-suicidal-thoughts-or-actions-patients-taking-certain-type (accessed 10 August 2024).

22.

Green

Patel

Goisman

, et al. (2000) Weight gain from novel antipsychotic drugs: Need for action. Gen Hosp Psychiatry 22: 224–235.

23.

Guirguis

Chiappini

Papanti

, et al. (2024) Exploring the association between suicidal thoughts, self-injury, and GLP-1 receptor agonists in weight loss treatments: Insights from pharmacovigilance measures and unmasking analysis. Eur Neuropsychopharmacol 82: 82–91.

24.

Højlund

Pottegård

Johnsen

, et al. (2019) Trends in utilization and dosing of antipsychotic drugs in Scandinavia: Comparison of 2006 and 2016. Br J Clin Pharmacol 85: 1598–1606.

25.

Huhn

Nikolakopoulou

Schneider-Thoma

, et al. (2019) Comparative efficacy and tolerability of 32 oral antipsychotics for the acute treatment of adults with multi-episode schizophrenia: A systematic review and network meta-analysis. Lancet 394: 939–951.

26.

JASP Team (2024) JASP (Version 0.18.3) [Computer Software]. Available at: https://jasp-stats.org

27.

Kahn

Silverman

DiPetrillo

, et al. (2021) A phase 3, multicenter study to assess the 1-year safety and tolerability of a combination of olanzapine and samidorphan in patients with schizophrenia: Results from the ENLIGHTEN-2 long-term extension. Schizophr Res 232: 45–53.

28.

Kahn

Kane

Correll

, et al. (2023) Olanzapine/samidorphan in young adults with schizophrenia, schizophreniform disorder, or bipolar I disorder who are early in their illness: Results of the randomized, controlled ENLIGHTEN-early study. J Clin Psychiatry 84: 22m14674.

29.

Khaity

Mostafa Al-Dardery

Albakri

, et al. (2023) Glucagon-like peptide-1 receptor-agonists treatment for cardio-metabolic parameters in schizophrenia patients: A systematic review and meta-analysis. Front Psychiatry 14: 1153648.

30.

Laursen

Plana-Ripoll

Andersen

, et al. (2019) Cause-specific life years lost among persons diagnosed with schizophrenia: Is it getting better or worse? Schizophr Res 206: 284–290.

31.

Lawrence

Kisely

(2010) Inequalities in healthcare provision for people with severe mental illness. J Psychopharmacol 24: 61–68.

32.

Lawrence

Hancock

Kisely

(2013) The gap in life expectancy from preventable physical illness in psychiatric patients in Western Australia: Retrospective analysis of population based registers. BMJ 346: f2539.

33.

Lee

Abraham

Cleaver

(2022) A systematic review of licensed weight-loss medications in treating antipsychotic-induced weight gain and obesity in schizophrenia and psychosis. Gen Hosp Psychiatry 78: 58–67.

34.

Lee

Idoko

Twohig

, et al. (2023) Systematic review of the efficacy of opioid receptor antagonists at treating antipsychotic-induced weight gain. Available at: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=455663 (accessed 10 August 2024).

35.

Lee

Akinola

Abraham

(2024) Antipsychotic-induced weight gain: exploring the role of psychiatrists in managing patients’ physical health—challenges, current options and direction for future care. BJPsych Bull 48: 24–29.

36.

Lyu

Zhan

, et al. (2018) Naltrexone and bupropion combination treatment for smoking cessation and weight loss in patients with schizophrenia. Front Pharmacol 9: 181.

37.

Martin

Correll

Weiden

, et al. (2019) Mitigation of olanzapine-induced weight gain with samidorphan, an opioid antagonist: A randomized double-blind phase 2 study in patients with schizophrenia. Am J Psychiatry 176: 457–467.

38.

Mazereel

Detraux

Vancampfort

, et al. (2020) Impact of psychotropic medication effects on obesity and the metabolic syndrome in people with serious mental illness. Front Endocrinol 11: 573479.

39.

McIntyre

Mansur

Rosenblat

, et al. (2024) The association between glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and suicidality: Reports to the Food and Drug Administration Adverse Event Reporting System (FAERS). Expert Opin Drug Saf 23: 47–55.

40.

Meyer

Simmons

Jiang

, et al. (2023) Olanzapine/samidorphan combination consistently mitigates weight gain across various subgroups of patients. CNS Spectr 28: 478–481.

41.

Murray

Tulloch

Gold

, et al. (2014) Hormonal and neural mechanisms of food reward, eating behaviour and obesity. Nat Rev Endocrinol 10: 540–552.

42.

NHS England (2024) Improving the physical health of people living with severe mental illness. Available at: https://www.england.nhs.uk/long-read/improving-the-physical-health-of-people-living-with-severe-mental-illness/ (accessed 9 August 2024).

43.

Nielsen

Banner

Jensen

(2021) Cardiovascular disease in patients with severe mental illness. Nat Rev Cardiol 18: 136–145.

44.

Nogueiras

Romero-Picó

Vazquez

, et al. (2012) The opioid system and food intake: Homeostatic and hedonic mechanisms. Obes Facts 5: 196–207.

45.

Page

Moher

Bossuyt

, et al. (2021) PRISMA 2020 explanation and elaboration: Updated guidance and exemplars for reporting systematic reviews. BMJ 372: n160.

46.

Peritogiannis

Ninou

Samakouri

(2022) Mortality in schizophrenia-spectrum disorders: Recent advances in understanding and management. Healthcare 10: 2366.

47.

Pillarella

Higashi

Alexander

, et al. (2012) Trends in use of second-generation antipsychotics for treatment of bipolar disorder in the United States, 1998–2009. Psychiatr Serv 63: 83–86.

48.

Pillinger

McCutcheon

Vano

, et al. (2020) Comparative effects of 18 antipsychotics on metabolic function in patients with schizophrenia, predictors of metabolic dysregulation, and association with psychopathology: A systematic review and network meta-analysis. Lancet Psychiatry 7: 64–77.

49.

Plana-Ripoll

Pedersen

Agerbo

, et al. (2019) A comprehensive analysis of mortality-related health metrics associated with mental disorders: A nationwide, register-based cohort study. Lancet 394: 1827–1835.

50.

Plana-Ripoll

Weye

Momen

, et al. (2020) Changes over time in the differential mortality gap in individuals with mental disorders. JAMA Psychiatry 77: 648.

51.

Potkin

Kunovac

Silverman

, et al. (2020) Efficacy and safety of a combination of olanzapine and samidorphan in adult patients with an acute exacerbation of schizophrenia: Outcomes from the randomized, phase 3 ENLIGHTEN-1 study. J Clin Psychiatry 81: 19m12769.

52.

Reynolds

Kirk

(2010) Metabolic side effects of antipsychotic drug treatment—pharmacological mechanisms. Pharmacol Ther 125: 169–179.

53.

Rhee

Mohamed

Rosenheck

(2018) Antipsychotic prescriptions among adults with major depressive disorder in office-based outpatient settings: National trends from 2006 to 2015. J Clin Psychiatry 79: 17m11970.

54.

Rhee

Olfson

Nierenberg

, et al. (2020) 20-year trends in the pharmacologic treatment of bipolar disorder by psychiatrists in outpatient care settings. Am J Psychiatry 177: 706–715.

55.

Romanova

Derkach

Mikhrina

, et al. (2018) The leptin, dopamine and serotonin receptors in hypothalamic POMC-neurons of normal and obese rodents. Neurochem Res 43: 821–837.

56.

Saha

Chant

McGrath

(2007) A systematic review of mortality in schizophrenia: Is the differential mortality gap worsening over time? Arch Gen Psychiatry 64: 1123.

57.

Silverman

Martin

Memisoglu

, et al. (2018) A randomized, double-blind, placebo-controlled proof of concept study to evaluate samidorphan in the prevention of olanzapine-induced weight gain in healthy volunteers. Schizophr Res 195: 245–251.

58.

Sterne

Hernán

Reeves

, et al. (2016) ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions. BMJ 355: i4919.

59.

Sterne

JAC

Savović

Page

, et al. (2019) RoB 2: A revised tool for assessing risk of bias in randomised trials. BMJ 366: l4898.

60.

Tan

Gajipara

Sun

, et al. (2022) In vivo characterization of the opioid receptor–binding profiles of samidorphan and naltrexone in rats: Comparisons at clinically relevant concentrations. Neuropsychiatr Dis Treat 18: 2497–2506.

61.

Taveira

W-C

Tschibelu

, et al. (2014) The effect of naltrexone on body fat mass in olanzapine-treated schizophrenic or schizoaffective patients: A randomized double-blind placebo-controlled pilot study. J Psychopharmacol 28: 395–400.

62.

Taylor

Gaughran

Pillinger

(eds) (2020) Treatment of antipsychotic-induced weight gain. In: The Maudsley Practice Guidelines for Physical Health Conditions in Psychiatry. Hoboken, NJ: John Wiley & Sons, Inc, pp. 125–130.

63.

Tek

Ratliff

Reutenauer

, et al. (2014) A randomized, double-blind, placebo-controlled pilot study of naltrexone to counteract antipsychotic-associated weight gain: Proof of concept. J Clin Psychopharmacol 34: 608–612.

64.

Thomas

Dourish

Tomlinson

, et al. (2018) The 5-HT2C receptor agonist meta-chlorophenylpiperazine (mCPP) reduces palatable food consumption and BOLD fMRI responses to food images in healthy female volunteers. Psychopharmacology 235: 257–267.

65.

Tobaiqy

Elkout

(2024) Psychiatric adverse events associated with semaglutide, liraglutide and tirzepatide: A pharmacovigilance analysis of individual case safety reports submitted to the EudraVigilance database. Int J Clin Pharm 46: 488–495.

66.

Toda

Santoro

Kim

, et al. (2017) POMC neurons: From birth to death. Ann Rev Physiol 79: 209–236.

67.

Wang

Volkow

Berger

, et al. (2024) Association of semaglutide with risk of suicidal ideation in a real-world cohort. Nat Med 30: 168–176.

68.

Woodall

Abuzour

Wilson

, et al. (2024) Management of antipsychotics in primary care: Insights from healthcare professionals and policy makers in the United Kingdom. PLoS One 19: e0294974.

69.

Yagoda

Graham

Simmons

, et al. (2021) Long-term safety and durability of effect with a combination of olanzapine and samidorphan in patients with schizophrenia: Results from a 1-year open-label extension study. CNS Spectr 26: 383–392.

70.

Yeomans

Gray

(2002) Opioid peptides and the control of human ingestive behaviour. Neurosci Biobehav Rev 26: 713–728.

71.

Zhan

(2018) POMC neurons: Feeding, energy metabolism, and beyond. In: Wu

Zheng

(eds) Neural Regulation of Metabolism. Singapore: Springer, pp. 17–29.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.03 MB

0.42 MB

0.17 MB

0.02 MB