ARTICLES
`
`Effects of Olanzapine Combined With Samidorphan on
`Weight Gain in Schizophrenia: A 24-Week Phase 3 Study
`
`Christoph U. Correll, M.D., John W. Newcomer, M.D., Bernard Silverman, M.D., Lauren DiPetrillo, Ph.D.,
`Christine Graham, Ph.D., Ying Jiang, Ph.D., Yangchun Du, Ph.D., Adam Simmons, M.P.H., Craig Hopkinson, M.D.,
`David McDonnell, M.D., René S. Kahn, M.D., Ph.D.
`
`Objective: A combination of olanzapine and the opioid re-
`ceptor antagonist samidorphan is under development for the
`treatment of schizophrenia and bipolar I disorder. The single-
`tablet combination treatment is intended to provide the efficacy
`of olanzapine while mitigating olanzapine-associated weight
`gain. In this phase 3 double-blind trial, the authors evaluated the
`weight profile of combined olanzapine/samidorphan com-
`pared with olanzapine in patients with schizophrenia.
`
`Methods: Adults (ages 18 55 years) with schizophrenia were
`randomly assigned to receive either combination treatment
`with olanzapine and samidorphan or olanzapine treatment
`for 24 weeks. Primary endpoints were percent change from
`baseline in body weight and proportion of patients with $10%
`weight gain at week 24. The key secondary endpoint was the
`proportion of patients with $7% weight gain. Waist cir-
`cumference and fasting metabolic laboratory parameters
`were also measured.
`
`Results: Of 561 patients who underwent randomization
`(olanzapine/samidorphan combination, N=280; olanzapine,
`N=281), 538 had at least one postbaseline weight assess-
`ment. At week 24, the least squares mean percent weight
`change from baseline was 4.21% (SE=0.68) in the olanzapine/
`samidorphan group and 6.59% (SE=0.67) in the olanzapine
`
`2.38% [SE=0.76] was significant).
`group (the difference of
`Significantly fewer patients in the olanzapine/samidorphan
`combination group compared with the olanzapine group had
`weight gain $10% (17.8% and 29.8%, respectively; number
`needed to treat [NNT]=7.29; odds ratio=0.50) and weight
`gain $7% (27.5% and 42.7%, respectively; NNT=6.29; odds
`ratio=0.50). Increases in waist circumference were smaller
`in the olanzapine/samidorphan combination group com-
`pared with the olanzapine group. Schizophrenia symptom
`improvement was similar between treatment groups. Ad-
`verse events (in $10% of the groups) in the olanzapine/
`samidorphan and olanzapine groups included weight gain
`(24.8% and 36.2%), somnolence (21.2% and 18.1%), dry
`mouth (12.8% and 8.0%), and increased appetite (10.9% and
`12.3%). Metabolic changes were small and similar between
`treatments.
`
`Conclusions: Olanzapine/samidorphan combination treat-
`ment was associated with significantly less weight gain and
`smaller increases in waist circumference than olanzapine
`and was well tolerated. The antipsychotic efficacy of the
`combination treatment was similar to that of olanzapine
`monotherapy.
`
`Am J Psychiatry 2020; 177:1168 1178; doi: 10.1176/appi.ajp.2020.19121279
`
`Mental disorders are a leading cause of ill health and disability
`worldwide (1). Schizophrenia affects 2.4 million people in the
`United States (2) and is associated with a 3.7-fold increased
`mortality risk (3). Antipsychotics, including olanzapine, are the
`cornerstone of treatment of schizophrenia. In long-term ef-
`fectiveness studies, olanzapine treatment was associated with
`lower rates of hospitalization for disease exacerbation (4, 5),
`higher rates of remission (6), and consistently longer time to
`all-cause discontinuation (4, 5, 7). Despite olanzapine’s robust
`efficacy, the associated risk of significant weight gain and
`metabolic sequelae (4, 8) has limited its overall clinical utility (9).
`Antipsychotic-associated weight gain reduces adherence
`and leads to treatment switches (10, 11), placing patients
`
`at significant risk of relapse, hospitalization, and disease
`progression (12). In patient populations already predisposed
`to shortened lifespan and cardiometabolic risks, weight gain
`exacerbates this risk. Weight gain also profoundly affects
`quality of life, psychosocial adaptation, body image, and self-
`esteem (13, 14), an impact superimposed on the challenges
`accompanying a psychiatric diagnosis, including stigma and
`social isolation (15).
`Evidence suggests that opioid receptor antagonists may
`mitigate medication-associated weight gain and/or metabolic
`dysregulation (16–19). Samidorphan, a new molecular entity,
`binds in vitro with high affinity to human m-, k-, and d-opioid
`receptors and acts as an antagonist at m-opioid receptors and a
`
`See related feature: Editorial by Dr. Buchanan (p. 1113)
`
`1168 ajp.psychiatryonline.org
`
`Am J Psychiatry 177:12, December 2020
`
`1
`
`Exhibit 2064
`Slayback v. Sumitomo
`IPR2020-01053
`
`

`

`partial agonist at k- and d-opioid receptors (20, 21). In vivo,
`samidorphan functions as an opioid receptor antagonist (22).
`A combination of olanzapine and samidorphan adminis-
`tered as a single tablet is under development for treatment of
`schizophrenia and bipolar I disorder, and it is hypothesized
`that the combination treatment would be associated with
`significantly less weight gain than olanzapine monotherapy.
`Antipsychotic efficacy of the combination treatment has been
`established in a phase 3 study in patients with an acute exac-
`erbation of schizophrenia (23). In preclinical studies, co-
`administration of olanzapine and samidorphan was found to
`attenuate olanzapine-associated weight gain and mitigate several
`olanzapine-associated metabolic abnormalities, independently
`of effects on weight (24). In a 12-week phase 2 dose-ranging
`study in patients with schizophrenia, combined olanzapine/
`samidorphan treatmentresulted ina 37%reduction inweightgain
`compared with olanzapine (25). Thus, combined olanzapine/
`samidorphantreatmentmayhaveanimprovedbenefit-riskprofile
`compared with olanzapine, providing an important long-term
`treatment option with antipsychotic efficacy and the benefit of
`significantly reduced weight gain (23, 26). This 24-week phase
`3 study was specifically designed to evaluate the weight profile
`of combined olanzapine/samidorphan compared with olanza-
`pine at clinically relevant dosages in adults with schizophrenia.
`
`METHODS
`
`Ethics
`The study (ClinicalTrials.gov identifier: NCT02694328) was
`conducted in accordance with the Declaration of Helsinki,
`1964, and Good Clinical Practice principles (International
`Conference on Harmonization, 1997). The study protocol and
`all amendments were approved by an institutional review
`board at each study site.
`
`Patients
`Patients 18–55 years of age meeting DSM-5 criteria for a primary
`diagnosis of schizophrenia were enrolled. Patients were re-
`quired to be outpatients, have a baseline body mass index (BMI)
`between 18 and 30, and have stable body weight (self-reported
`change #5%) for at least 3 months before study initiation.
`Key exclusion criteria included a history of treatment-
`resistant schizophrenia, ,1 year elapsed since initial onset
`of symptoms, naive to antipsychotic medication, active al-
`cohol or substance use disorders (excluding marijuana/
`tetrahydrocannabinol), or any clinically significant or un-
`stable medical illness (e.g., diabetes, hypo- or hypertension,
`thyroid dysfunction, and history of seizure disorder or brain
`tumor) that might compromise patient safety or study end-
`point assessments or interfere with the ability to fulfill study
`requirements. Opioid agonist use within 14 days of screening,
`opioid antagonist use within 60 days of screening, or anti-
`cipated need for opioid treatment during the study were
`exclusionary, as was the use of olanzapine in the 60 days
`before screening. All patients provided written informed
`consent after receiving a complete description of the study.
`
`CORRELL ET AL
`
`Study Design
`This was a phase 3 multicenter, randomized, double-blind
`study conducted in the United States. Candidates were
`screened within 30 days of randomization; eligible patients
`were randomly assigned in a 1:1 ratio to receive treatment
`with either combined olanzapine/samidorphan or olanza-
`pine for 24 weeks (see Figure S1A in the online supplement).
`Study completers were eligible to enroll in a long-term open-
`label safety study evaluating treatment with combined
`olanzapine/samidorphan over 52 weeks (ClinicalTrials.gov
`identifier: NCT02873208); those who elected not to enroll
`(or who prematurely discontinued the double-blind study)
`entered a 4-week safety follow-up period.
`
`Study Treatment
`The daily doses of combined olanzapine/samidorphan used
`in this study (10 mg olanzapine/10 mg samidorphan [10/10]
`and 20 mg olanzapine/10 mg samidorphan [20/10]) represent
`the lowest and highest approved maintenance dosages of
`olanzapine for schizophrenia treatment and the intended
`therapeutic fixed dosage of samidorphan, representing the
`optimal weight and safety profile when combined with
`olanzapine (25, 27).
`In general, the use of psychotropic medications other than
`study drug was prohibited, except for beta-blockers, anti-
`histamines, benzodiazepines, and anticholinergics for aka-
`thisia or extrapyramidal symptoms. Patients who were taking
`other antipsychotic medications at study entry were cross-
`tapered off these medications and titrated onto either com-
`bined olanzapine/samidorphan treatment or olanzapine
`monotherapy over the course of 2 weeks. In the first week,
`patients received daily doses of combined olanzapine/
`samidorphan 10/10 or 10 mg olanzapine. The olanzapine
`dosage was increased to 20 mg/day beginning at week 2. At
`the end of week 2, 3, or 4, the olanzapine dosage could be
`lowered to 10 mg/day for tolerability reasons. No dosage
`adjustments were permitted beyond week 4.
`
`Assessments
`Patient visits occurred weekly through week 6, then biweekly
`for the remaining 18 weeks. Assessments included body
`weight and waist circumference (both measured in tripli-
`cate), vital signs, ECG, adverse events, extrapyramidal
`symptoms (Abnormal Involuntary Movement Scale [AIMS]
`[28], Simpson-Angus Scale [29], Barnes Akathisia Rating
`Scale [30]), the Columbia-Suicide Severity Rating Scale (31),
`the Positive and Negative Syndrome Scale (PANSS) (32), and
`the Clinical Global Impressions severity scale (CGI-S) (33).
`Blood samples for fasting ($8 hours by self-report) metabolic
`laboratory parameters (triglycerides, cholesterol, glucose,
`and insulin) and nonfasting hemoglobin A1c were collected.
`
`Primary and Secondary Endpoints
`The co–primary endpoints were percent change from
`baseline at week 24 in body weight and the proportion of
`patients with $10% weight gain from baseline at week 24.
`
`Am J Psychiatry 177:12, December 2020
`
`ajp.psychiatryonline.org 1169
`
`2
`
`

`

`EFFECTS OF OLANZAPINE COMBINED WITH SAMIDORPHAN ON WEIGHT GAIN IN SCHIZOPHRENIA
`
`The key secondary endpoint was proportion of patients
`with $7% weight gain at week 24. The cutoffs of 10% and 7%
`were selected on the basis of commonly accepted thresholds
`of clinically significant changes in weight for weight man-
`agement and psychiatric treatments, respectively.
`
`Statistical Analysis
`The initial target sample size was 200 patients per treatment
`group. This sample size was estimated to provide $90%
`power to detect significant differences in mean percent
`change in body weight of 4% (SD 9%) and in the proportion
`of patients with $10% weight gain of 13% at week 24, as-
`suming a cumulative dropout rate of 40%. A prespecified
`unblinded interim analysis for sample size reestimation was
`conducted by an independent statistical center when 50%
`of patients completed the double-blind treatment period
`or discontinued. Because the conditional power of the co–
`primary endpoints was less than 90% based on the interim
`results, the sample size was subsequently increased to
`540 patients.
`Safety was assessed in patients who received at least one
`dose of study drug. Weight and antipsychotic efficacy were
`assessed in all patients who had at least one postbaseline
`weight assessment.
`To control for multiplicity, both co–primary endpoints
`were tested at an alpha of 0.05 based on the method described
`by Cui et al. to adjust for the unblinded interim analysis (34).
`The key secondary endpoint would be tested only if both
`co–primary endpoints were met.
`Missing weight assessments were imputed by multiple
`imputation sequentially for each visit, using a regression
`method. The imputation regression model included treat-
`ment, race, and baseline age group as factors and body weight
`at all previous visits (including baseline) as covariates. Five
`hundred imputations were carried out. The co–primary
`endpoint of percent change from baseline in body weight at
`week 24 was analyzed by analysis of covariance (ANCOVA)
`based on the imputed data sets. The ANCOVA model in-
`cluded treatment, race, and age group as factors and base-
`line weight as a covariate. Results were combined using
`Rubin’s method. Additional details are provided in the online
`supplement.
`Analysis of the other co–primary endpoint and the key
`secondary endpoint (proportion of patients with $10%
`and $7% weight gain at week 24, respectively) was carried
`out using a logistic regression model based on the same
`multiply imputed weight data as the percent change from
`baseline in body weight at week 24.
`Analyses of change from baseline in body weight and waist
`circumference at each visit were similar to those of the
`co–primary endpoint of percent change in body weight at
`week 24. Change from baseline in metabolic laboratory pa-
`rameters, PANSS score, and CGI-S score were analyzed using
`a mixed model with repeated measures; the model included
`treatment, visit, treatment-by-visit interaction, race, and age
`group as categorical fixed effects and baseline weight as a
`
`covariate, with an unstructured covariance structure and
`Kenward-Roger approximation to adjust the denominator
`degree of freedom.
`
`RESULTS
`
`Patient Disposition and Baseline Characteristics
`Of 561 patients who underwent randomization, 550 (com-
`bined olanzapine/samidorphan group, N 274; olanzapine
`group, N 276) entered double-blind treatment (see Figure
`S1B in the online supplement). In all, 352 (64%) patients
`completed treatment, with similar completion rates in the
`two treatment groups. The most common reasons for dis-
`continuation with combined olanzapine/samidorphan and
`with olanzapine were adverse events (12.0% and 9.8%, re-
`spectively), withdrawal by participant (8.4% and 9.8%), and
`lost to follow-up (8.0% and 9.4%) (see Figure S1B). Ran-
`domization was balanced between groups at baseline for
`demographic characteristics, including race and BMI (Table
`1). Only minimal differences were noted between groups
`on prior antipsychotic medications used before randomiza-
`tion (see Table S1 in the online supplement).
`
`Drug Exposure
`Mean olanzapine dosage, calculated as time-weighted aver-
`age dosage of olanzapine during the entire study, was similar
`between groups (combined olanzapine/samidorphan group:
`16.8 mg/day, SD 3.94; olanzapine group, 16.9 mg/day,
`SD 3.57), with most patients (79.6%) taking 20 mg/day as the
`final dosage.
`
`Concomitant Medications
`Overall, patients in the combined olanzapine/samidorphan
`group and in the olanzapine group had similar concomitant
`medication use during the treatment period (74.1% [203/274]
`and 76.4% [211/276], respectively). The most frequently used
`concomitant medications ($10% of patients in either treat-
`ment group) in the combined olanzapine/samidorphan and
`olanzapine groups were risperidone (20.1% [55/274] and
`20.3% [56/276], respectively), ibuprofen (12.4% [34/274] and
`10.9% [30/276]), quetiapine fumarate (10.6% [29/274] and
`12.0% [33/276]), and aripiprazole (5.8% [16/274] and 10.5%
`[29/276]). The concomitant use of non-olanzapine antipsy-
`chotics reflects patients tapering off their prior antipsychotic
`medications during the first 2 weeks.
`
`Weight and Metabolic Effects
`Weight effects. For the co–primary endpoint, percent change
`from baseline in body weight at week 24, the least squares
`mean percent change was 4.21% (SE 0.681%) for the com-
`bined olanzapine/samidorphan group and 6.59% (SE 0.668%)
`for the olanzapine group. The least squares mean difference
`between the combined olanzapine/samidorphan group and
`the olanzapine group was
`2.38% (SE 0.765%; p 0.003)
`(Figure 1A; see also Table S2 in the online supplement). At each
`visit from week 6 through week 22, the 95% confidence
`
`1170 ajp.psychiatryonline.org
`
`Am J Psychiatry 177:12, December 2020
`
`3
`
`

`

`intervals for
`
`the between-
`
`TABLE 1. Baseline characteristics of participants in a study of weight gain with combined olanzapine/
`samidorphan in schizophrenia“
`
`CORRELL ET AL
`
`Characteristic
`
`Age (years)
`
`Sex
`Effile
`
`Raggck or African American
`White
`Asian
`American Indian or Alaska Native
`Native Hawaiian or other Pacific
`S an er
`I
`l
`d
`Other
`Multiple”
`
`group difference in percent
`change in weight did not in-
`clude zero, consistent with
`the primary endpoint at week
`24 (Figure 1A). Weight gain in
`the
`combined olanzapine/
`samidorphan group stabilized
`from week 6 onward, whereas
`weight continued to increase
`in the olanzapine group over
`the 24-week treatment pe-
`riod. At week 24, the least
`squares mean change in body
`weight was 3.18 kg in the
`.
`.
`olanzapine/samidorphan group
`and 5.08 kg 1n the olanzapine
`gI‘OUP-
`Fewer patients receiv-
`ing combined olanzapine/
`samidorphan experienced
`weight gain across a wide
`range of thresholds for per-
`cent change in body weight
`compared with olanzapine
`(Figure lB,C). The proportion of patients with 210% weight
`gain at week 24 (co—primary endpoint) was significantly
`lower in the combined olanzapine/samidorphan group
`(N 47 [17.8%]) than in the olanzapine group (N 81 [29.8%];
`number needed to treat
`[NNT] 7.29);
`the odds ratio
`for 210% weight gain at week 24 in the combined olanza-
`pine/samidorphan group compared with the olanzapine
`group was 0.50 (95% CI 0.31, 0.80; p 0.003) (see Table S2
`in the online supplement).
`For the key secondary endpoint, 27.5% of patients in the
`combined olanzapine/samidorphan group and 42.7% in the
`olanzapine group experienced 27% weight gain (N NT 6.29)
`(Figure 1B,C; see also Table 82 in the online supplement). The
`odds ratio for 27% weight gain at week 24 with combined
`olanzapine/samidorphan compared with olanzapine was 0.50
`(95% CI 0.33, 0.76; p 0.001) (see Table 82).
`
`_
`B°dy we'ghtlkg)
`BOdy mass Index
`5 Baseline was defined as the last nonmissing observation before the first dose of study drug.
`b Patients who reported more than one race were counted once under the multiple races category.
`
`77.17
`25.38
`
`13.69
`3.13
`
`77.57
`25.52
`
`Combined Olanzapine/
`Samidorphan (N=274)
`Mean
`SD
`
`Olanzapine (N=276)
`Mean
`SD
`
`40.3
`
`N
`
`193
`81
`
`199
`63
`4
`2
`1
`
`2
`3
`
`Mean
`
`9.79
`
`%
`
`70.4
`29.6
`
`72.6
`23.0
`1.5
`0.7
`0.4
`
`0.7
`1.1
`
`SD
`
`40.1
`
`N
`
`207
`69
`
`193
`65
`4
`2
`0
`
`4
`8
`
`Mean
`
`10.01
`
`“A
`
`75.0
`25.0
`
`69.9
`23.6
`1.4
`0.7
`0
`
`1.4
`2.9
`
`SD
`
`13.47
`3.19
`
`a finding associated with increased mortality risk (35); the risk
`difference was
`17.1% (95% CI
`26.3,
`7.8; odds ratio 0.47;
`NNT 5.86).
`
`Metabolic laboratory parameters. Changes from baseline to
`week 24 in glycemic and lipid laboratory parameters were
`generally small and similar for the two treatment groups and
`tended to occur early (Table 2). The largest changes in any
`of these parameters were in triglyceride levels, where least
`squares mean increases of 26.77 mg/dL (SE 5.78) and
`29.36 mg/dL (SE 5.69) were observed in the combined
`olanzapine/samidorphan and olanzapine groups, respectively.
`
`Safety
`Adverse events were reported in 74.1% and 82.2% of the
`combined olanzapine/samidorphan and olanzapine groups,
`respectively. The most commonly reported adverse events
`(210%) in the two groups were weight increase (24.8% and
`36.2%, respectively), somnolence (21.2% and 18.1%), dry
`mouth (12.8% and 8.0%), and increased appetite (10.9% and
`12.3%) (Table 3). Most adverse events were mild to moderate
`in severity. Twelve percent of patients in the combined
`olanzapine/samidorphan group discontinued treatment be-
`cause of an adverse event, compared with 9.8% in the
`olanzapine group (Table 3; see also Figure SlB in the online
`supplement).
`No deaths occurred during the study, and serious ad-
`verse events were reported in 3.6% and 2.5% of patients in
`the combined olanzapine/samidorphan and olanzapine
`groups, respectively (Table 3). The only serious adverse
`event occurring in more than one patient was worsening/
`
`Waist circumference. At week 24, the least squares mean
`change from baseline in waist circumference was 2.36 cm
`(SE 0.561) in the combined olanzapine/samidorphan group and
`4.47 cm (SE 0.546) in the olanzapine group (least squares
`mean difference:
`2.12 cm [SE 0.628]; 95% CI
`3.35,
`0.89).
`Separation in waist circumference occurred early, with smaller
`increases in waist circumference in the combined olanzapine/
`samidorphan group compared with the olanz apine group at all
`visits. Initially at week 1, and then at each visit from week 4
`through week 24, the 95% confidence intervals for the
`between-group difference in waist circumference did
`not include zero (Figure 2). At week 24, 26.8% of the
`combined olanzapine/samidorphan group and 43.2% of the
`olanzapine group had a waist circumference increase of 25 cm,
`
`Am J Psychiatry 177:12, December 2020
`
`ajp.psychiatryonline.org
`
`1171
`
`

`

`EFFECTS OF OLANZAPINE COMBINED WITH SAMIDORPHAN ON WEIGHT GAIN IN SCHIZOPHRENIA
`
`FIGURE 2. Least squares mean change from baseline in waist
`circumference by visit in a study of weight gain with combined
`olanzapine/samidorphan in schizophrenia“
`
`FIGURE 1. Change from baseline in body weight in a study of
`weight gain with combined olanzapine/samidorphan in
`schizophrenia‘3
`A. Least Squares Mean of Percent Change From Baseline in Body Weight
`by Visit
`
`
`
`
`
`PercentChangeFromBaseline
`
`
`
`
`
`(leastsquaresmeanandSE)
`
`0._.Nwau:or\1
`
`O 1 2
`
`4
`
`6
`
`+ Olanzapine
`-V- Combined olanzapine/samidorphan
`8
`12
`16
`20
`24
`
`StudyWeek
`
`
`
`
`
`ChangeFromBaselineinWaistCircumfa'aice
`
`
`
`
`
`
`
`(cm)(leastsquarsmeanandSE)
`
`0 1 2
`
`4
`
`6
`
`8
`
`12
`
`16
`
`20
`
`24
`
`Study Week
`
`+ Olanzapine
`+ Combined olanzapine/samidorphan
`
`aWaist circumference analysis was based on an analysis of covariance
`(ANCOVA) model with a multiple imputation approach for missing
`postbaseli neassessments. TheANCOVA modelincluded treatment, race
`(black/African American, nonblack/non African American). and age
`group (<30 years. 230 years) as factors and baseline body weight as the
`covariate. Baseline was defined as the last nonmissing value before the
`first dose of study drug.
`
`Akathisia Rating Scale), or Columbia-Suicide Severity Rating
`Scale scores for patients in the two treatment groups.
`The overall safety profile of combined olanzapine/
`samidorphan in this 24-week study was consistent with that
`of olanzapine, except for fewer adverse events of weight
`increase in the combined olanzapine/samidorphan group.
`
`Antipsychotic Efficacy
`PANSS and CGI-S scores. The mean PANSS total score at
`
`baseline was 68.2 (SD 9.51) in the combined olanzapine/
`samidorphan group and 70.2 (SD 9.47) in the olanzapine
`group. The PANSS total score improved similarly in both
`groups; the least squares mean change from baseline to week
`24 was
`8.2 (SE 0.73)
`in the combined olanzapine/
`samidorphan group and
`9.4 (SE 0.72) in the olanzapine
`group. The least squares mean difference between treatments
`was 1.2 (95% CI
`0.9, 3.2). Reductions in CGI-S score from
`baseline to week 24 were similar between the two treatment
`
`groups, consistent with the changes in PAN SS total score (see
`Figure S4 in the online supplement).
`
`DISCUSSION
`
`B. Cumulative Frequency Distribution of Percent Change From Baseline
`in Body Weight at Week 24
`100
`90
`80
`7O
`60
`50
`40
`
`......-...-.-...... IV\IR IVpoR
`
`30 — Olanzapine
`20 — Combined
`olanzapine]
`samidorphan
`
`—10
`—5
`—15
`O
`5
`10
`15
`20
`25
`
`1°
`0
`
`Percent Change From Baseline in Body Weight at Week 24
`
`C. Proportion of Patients With Weight Changes at Week 24
`100
`
`I Olanzapine
`I Combined olanzapine/samidorphan
`
`80
`
`Q
`i"
`,5
`E 60
`"6
`s
`'fl
`
`40
`
`310%
`37%
`20%
`Percent Change From Baseline in Body Weight at Week 24
`
`20
`
`O
`
`88
`
`-
`t
`
`3Missing postbaseline assessments were imputed based on multiple
`imputation. Data in panelA were analyzed using an analysis ofcovariance
`model with treatment, race (black/African American, nonblack/non
`African American), and age group (<30 years, 230 years) as factors and
`baseline body weight as the covariate. Baseline was defined as the last
`nonmissing value before the first dose of study drug.
`
`
`
`CumulativePercentageofPatient
`
`exacerbation of schizophrenia symptoms (one patient [0.4%]
`in the combined olanzapine/samidorphan group and three
`patients [1.1%] in the olanzapine group).
`There were no clinically meaningful changes or differ-
`ences observed in vital signs, ECG results, movement dis-
`order scale scores (AIMS, Simpson-Angus Scale, or Barnes
`
`In this 24-week study in patients with schizophrenia,
`treatment with combined olanzapine/samidorphan resulted
`in significantly less weight gain compared with olanzapine
`monotherapy. Differences in weight gain were apparent at
`week 6 and remained lower in the combined olanzapine/
`samidorphan group at each subsequent visit through week
`24. The weight distribution profile in the combined
`
`1172 ajp.psychiatryonline.org
`
`Am J Psychiatry 177:12, December 2020
`
`

`

`CORRELL ET AL
`
`TABLE 2. Analysis of change from baseline in fasting glycemic and lipid parameters in a study of weight gain with combined olanzapine/
`samidorphan in schizophreniaa
`
`Baseline
`
`Change From
`Baseline to Week 2
`
`Change From
`Baseline to Week 4
`
`Change From
`Baseline to Week 8
`
`Change From
`Baseline to Week 12
`
`Change From
`Baseline to Week 24
`
`Parameter
`
`OLZ/SAM Olanzapine OLZ/SAM Olanzapine OLZ/SAM Olanzapine OLZ/SAM Olanzapine OLZ/SAM Olanzapine OLZ/SAM Olanzapine
`
`Total
`cholesterol
`(mg/dL)
`N
`Mean
`SD
`HDL
`cholesterol
`(mg/dL)
`N
`Mean
`SD
`LDL
`cholesterol
`(mg/dL)
`N
`Mean
`SD
`Triglycerides
`(mg/dL)
`N
`Mean
`SD
`Glucose
`(mg/dL)
`N
`Mean
`SD
`Insulin
`(mIU/mL)
`N
`Mean
`SD
`b (%)
`HbA1c
`N
`Mean
`SD
`
`265
`183.4
`34.74
`
`270
`185.2
`37.27
`
`230
`9.3
`26.25
`
`256
`6.8
`24.60
`
`223
`7.3
`23.87
`
`234
`7.5
`25.60
`
`206
`5.8
`27.62
`
`222
`5.1
`26.85
`
`198
`3.4
`26.42
`
`207
`4.5
`24.68
`
`162
`0.9
`28.18
`
`166
`2.1
`28.88
`
`265
`62.4
`22.42
`
`270
`62.1
`21.02
`
`230
`0.4
`10.39
`
`256
`0.2
`10.11
`
`223
`1.7
`12.00
`
`234
`0.8
`12.22
`
`206
`3.7
`11.48
`
`222
`2.5
`11.59
`
`198
`3.6
`12.97
`
`207
`3.3
`12.90
`
`162
`5.1
`15.20
`
`166
`4.5
`11.35
`
`264
`109.6
`32.26
`
`265
`114.4
`93.96
`
`265
`90.3
`11.60
`
`265
`12.65
`20.87
`
`266
`5.40
`0.38
`
`270
`112.7
`33.98
`
`270
`107.1
`62.14
`
`270
`91.4
`12.03
`
`269
`12.12
`15.85
`
`272
`5.40
`0.42
`
`229
`8.3
`22.48
`
`230
`12.7
`71.16
`
`231
`6.2
`17.43
`
`227
`4.23
`19.78
`
`243
`0.03
`0.20
`
`256
`3.6
`20.80
`
`256
`24.0
`70.17
`
`256
`1.9
`14.28
`
`249
`5.49
`28.61
`
`264
`0.04
`0.21
`
`222
`6.7
`22.11
`
`223
`15.7
`59.52
`
`225
`3.4
`13.42
`
`220
`4.43
`26.32
`
`235
`0.04
`0.26
`
`234
`5.6
`22.08
`
`206
`6.1
`25.80
`
`234
`17.2
`64.83
`
`237
`2.3
`15.54
`
`230
`2.57
`19.41
`
`247
`0.02
`0.21
`
`206
`21.3
`78.17
`
`206
`3.6
`14.69
`
`205
`3.41
`20.16
`
`218
`0.01
`0.30
`
`222
`3.7
`24.30
`
`222
`24.1
`77.55
`
`220
`1.7
`15.12
`
`217
`3.23
`17.61
`
`234
`0.03
`0.26
`
`197
`4.4
`24.24
`
`198
`16.6
`75.91
`
`197
`4.2
`15.49
`
`196
`2.32
`26.43
`
`211
`0.00
`0.30
`
`207
`4.1
`21.62
`
`207
`19.8
`65.43
`
`206
`1.9
`13.75
`
`203
`1.96
`20.72
`
`216
`0.04
`0.27
`
`161
`0.6
`26.37
`
`162
`23.9
`78.29
`
`160
`4.5
`15.05
`
`162
`3.22
`28.72
`
`173
`0.06
`0.27
`
`166
`0.9
`26.51
`
`166
`24.5
`71.49
`
`166
`2.3
`15.70
`
`161
`3.40
`15.60
`
`173
`0.07
`0.27
`
`a Analyzed using a mixed model with repeated measures, which includes the treatment, visit, treatment by visit interaction, race (black or African American, non
`black or non African American), and age group (,30 years, $30 years) as factors and the corresponding baseline value as the covariate. OLZ/SAM=combination
`olanzapine/samidorphan.
`b Nonfasting.
`
`olanzapine/samidorphan group was shifted compared with
`the olanzapine group, such that fewer patients gained weight
`across a range of cutoffs. 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.
`The results of this study extend those of a previous study in
`which combined olanzapine/samidorphan treatment was
`associated with significantly less weight gain compared with
`olanzapine over 12 weeks (25). In both studies, combined
`olanzapine/samidorphan treatment resulted in weight gain
`over the first 4 to 6 weeks before stabilizing. The mean in-
`crease in weight in this study was 3.18 kg in the combined
`olanzapine/samidorphan group and 5.08 kg in the olanzapine
`group at 24 weeks. Some degree of weight gain is commonly
`reported in clinical trials with other atypical antipsychotics
`during periods beyond 6 months. For example, weight in-
`creases with brexpiprazole (1.3 kg) (36), cariprazine (1.7 kg)
`(37), quetiapine (2.4–3.7 kg) (38, 39), and risperidone (4.3 kg)
`(40) have been reported. Comparisons to other antipsy-
`chotics must be interpreted cautiously owing to different
`study objectives, designs, populations, and analyses. None-
`theless, the data suggest that while combined olanzapine/
`samidorphan may be associated with more weight gain than
`
`some atypical antipsychotics, it may be within the range of
`weight gain seen in commonly used treatments, such as
`risperidone and quetiapine.
`In addition to less weight gain, combined olanzapine/
`samidorphan was also associated with significantly smaller
`increases in waist circumference compared with olanzapine.
`Waist circumference is a proxy for central fat accumulation
`(41), and increases in waist circumference have been as-
`sociated with a greater risk of cardiovascular disease and
`diabetes,evenindependentlyofweight (41,42). The separation
`in waist circumference for combined olanzapine/samidorphan
`versus olanzapine occurred earlier than the separation in
`weight, suggesting that while weight gain was similar during the
`first 6 weeks, it may have been distributed differently. A shift
`away from increasing abdominal adiposity may potentially lead
`to areductionofcardiovascularanddiabetes riskwithcombined
`olanzapine/samidorphan in comparison to olanzapine, espe-
`cially long-term. For patients treated with combined olanza-
`pine/samidorphan, the risk of having a waist circumference
`increase of $5 cm was reduced by approximately 50% com-
`pared with olanzapine. A 5-cm increase in waist circumference
`is clinically relevant, as it not only represents a change in pant
`size for the average adult but also is associated with an increased
`mortality risk (9% in women and 7% in men) (35).
`
`Am J Psychiatry 177:12, December 2020
`
`ajp.psychiatryonline.org 1173
`
`6
`
`

`

`EFFECTS OF OLANZAPINE COMBINED WITH SAMIDORPHAN ON WEIGHT GAIN IN SCHIZOPHRENIA
`
`TABLE 3. Overview of adverse events during the double-blind treatment period in a study of weight
`gain with combined olanzapine/samidorphan in schizophreniaa
`Combined Olanzapine/
`Samidorphan (N=274)
`
`Olanzapine (N=276)
`
`N
`
`227
`
`125
`95
`7
`
`27
`
`7
`
`100
`50
`22
`34
`22
`12
`
`17
`
`%
`
`82.2
`
`45.3
`34.4
`2.5
`
`9.8
`
`2.5
`
`36.2
`18.1
`8.0
`12.3
`8.0
`4.3
`
`6.2
`
`the timing of
`Notably,
`changes in metabolic labo-
`ratory parameters was an
`important factor in this study.
`For both treatments, changes
`in metabolic laboratory pa-
`rameters generally occurred
`early, within the first few
`weeks, with little additional
`change over the remainder
`of the treatment period. The
`continued weight increases
`observed with olanzapine did
`not
`translate to any clear
`metabolic worsening beyond
`these initial effects. As a re-
`sult, mitigation of weight gain
`by
`combined olanzapine/
`samidorphan did not lead to a
`discernible metabolic benefit
`in this study. This finding
`points to a potential pitfall in
`trying to relate olanzapine-
`associated metabolic changes
`solely to changes in body
`weight. Several studies have
`documented the effects of
`olanzapine on lipid and
`glucose metabolism within days of initiating treatment
`(51–53). These early metabolic effects appear to be more
`directly related to the drug itself rather than to secondary
`consequences of weight gain. The time frame over which
`these initial changes persist or give way to the longer-term
`secondary changes associated with accumulating weight is
`not well understood. The 6-month duration of this study may
`have been insufficient to detect these secondary weight-
`related changes in metabolic risk factors; thus, any poten-
`tial metabolic effects derived from mitigating weight gain
`with combined olanzapine/samidorphan may have been
`missed.
`While a metabolic benefit for combined olanzapine/
`samidorphan relative to olanzapine was not observed in this
`study, extensive evidence supports the expectation that
`mitigation of olanzapine-associated weight gain should ul-
`timately lead to metabolic benefit. Olanzapine-associated
`weight gain poses such a relevant safety risk because it
`continues over long-term treatment, with a majority of pa-
`tients gaining substantial amounts of weight over periods well
`exceeding the 6-month time frame of this study (27, 54, 55). In
`293 patients treated with olan