Binge-eating disorder (BED), a prevalent psychiatric disorder associated strongly with obesity and psychosocial/functional impairments, is characterized by high persistence/chronicity (Udo & Grilo, Reference Udo and Grilo2018). While there exists an evidence base for treatments with acute efficacy for BED (Grilo, Ivezaj, & Gueorguieva, Reference Grilo, Ivezaj and Gueorguieva2024), controlled research examining maintenance treatments for responders to initial interventions is limited. This research gap is especially important for pharmacological approaches, which – in contrast to the longer-term durability of specific psychological treatments such as cognitive-behavioral therapy (Grilo, Masheb, Wilson, Gueorguieva, & White, Reference Grilo, Masheb, Wilson, Gueorguieva and White2011; Wilson, Wilfley, Agras, & Bryson, Reference Wilson, Wilfley, Agras and Bryson2010) – are associated with high relapse rates following acute treatment (Grilo, Crosby, Wilson, & Masheb, Reference Grilo, Crosby, Wilson and Masheb2012a).
Only two placebo-controlled trials have tested pharmacotherapy maintenance treatments for BED (Grilo, Lydecker, & Gueorguieva, Reference Grilo, Lydecker and Gueorguieva2023; Hudson, McElroy, Ferreira-Cornwell, Radewonuk, & Gasior, Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017). Hudson et al. (Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017) found that continued lisdexamfetamine reduced relapse risk relative to placebo (3.7% v. 32.1%) in patients with BED who responded to acute treatment with lisdexamfetamine (Hudson et al., Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017). Grilo et al. (Reference Grilo, Lydecker and Gueorguieva2023) found that naltrexone + bupropion following response to acute treatment with naltrexone + bupropion was associated with good maintenance of low binge-eating frequency and remission, and with significant additional weight loss. In contrast, maintenance medication (naltrexone + bupropion v. placebo) was not associated with any significant changes in binge-eating or weight in patients who responded to initial treatments with behavioral therapy (i.e. acute behavioral therapy – which showed good maintenance – significantly moderated the outcomes of maintenance pharmacotherapy).
This study was a prospective, randomized, double-blind, placebo-controlled test of the efficacy of lisdexamfetamine maintenance treatment following successful acute treatments with either lisdexamfetamine or lisdexamfetamine combined with cognitive-behavioral therapy (LDX v. CBT + LDX) for BED comorbid with obesity. This study logically examined maintenance pharmacotherapy following the ‘best-established’ (only FDA-approved) pharmacological (McElroy et al., Reference McElroy, Hudson, Mitchell, Wilfley, Ferreira-Cornwell, Gao and Jonas2015, Reference McElroy, Hudson, Ferreira-Cornwell, Radewonuk, Whitaker and Gasior2016) and psychological (Grilo et al., Reference Grilo, Ivezaj and Gueorguieva2024) treatments for BED. The present prospective controlled maintenance treatment study follows the acute treatment trial examining LDX v. CBT + LDX for BED comorbid with obesity for BED comorbid with obesity (Grilo et al., Reference Grilo, Ivezaj, Tek, Yurkow, Wiedemann and Gueorguievain press). Patients categorized as ‘responders’ following acute treatments were randomized, in double-blind fashion, with initial acute treatment (LDX alone or CBT + LDX) as a stratifying variable, to either maintenance LDX or placebo for 12 weeks without any additional psychotherapeutic interventions.
We hypothesized that participants receiving LDX would maintain reductions in binge-eating frequency and weight-loss attained with acute treatment while participants receiving placebo would show increases in binge-eating frequency and weight. We tested whether initial acute treatment with CBT predict or moderate outcomes of the maintenance treatment.
Method
Procedure
This single-site RCT was approved by the Yale institutional review board and included a data safety and monitoring plan with a physician safety monitor. Participants provided written informed consent.
Participants
Participants for this controlled maintenance trial were eligible if they were categorized as ‘responders’ following initial acute 12-week treatments with LDX-alone or with CBT + LDX in a RCT for BED with obesity (Grilo et al., Reference Grilo, Ivezaj, Tek, Yurkow, Wiedemann and Gueorguievain press). When participants enrolled in the treatment study, they were informed and they consented to two treatment stages – i.e. the acute treatment stage (Grilo et al., Reference Grilo, Ivezaj, Tek, Yurkow, Wiedemann and Gueorguievain press) and a second stage that would test pharmacotherapy maintenance if they responded to the initial treatments comprising pharmacotherapy. Accordingly, participation in this maintenance treatment study was anticipated when participants consented to the 2-stage (initial acute ‘Stage 1’ plus maintenance ‘Stage 2’) treatments without an opt-out (i.e. if they responded to Stage 1, they participated in Stage 2, unless medically contraindicated).
Eligibility for the acute trial required meeting DSM-5 (APA, 2013) BED criteria, ages 18–64 years old, and a body mass index (BMI; kg/m2) ≥30.0 and ≤50.0 (or ≥27.0 with obesity-related comorbidity). Exclusion criteria included clinical issues that required alternative treatments or represented contraindications to lisdexamfetamine, including: current evidence-based treatment for eating/weight disorders, taking contraindicated medications (e.g. opiates, MAOIs, NDRIs, and stimulants; SSRIs and strong inhibitors of CYP2D6 were considered on a case-by-case basis for safety/risk and depending on dosing), participation in another clinical trial/study, uncontrolled medical conditions, contraindications to LDX (e.g. alcohol and substance use disorders, psychosis/bipolar disorder, seizure history, cardiovascular disease, cerebrovascular disease, history renal, hepatic, or chronic pulmonary disease, systolic blood pressure > 160 mmHg, diastolic blood pressure > 100 mmHg, or heart rate > 100 beats/min), and pregnancy/breastfeeding.
Responder to acute treatment definition
Successful ‘response’ to the initial acute 12-week treatments was defined as ≥65% reduction (relative to baseline) in past-month frequency of binge-eating at posttreatment. This definition for response to treatment was used previously in the maintenance trial testing naltrexone + bupropion (Grilo et al., Reference Grilo, Lydecker and Gueorguieva2023) and in an adaptive stepped-care treatment trial to determine whether to continue treatment (for ‘responders’) or to switch to an alternative treatment in the case of ‘non-responders’ (Grilo et al., Reference Grilo, White, Masheb, Ivezaj, Morgan and Gueorguieva2020). This definition was adopted based on several studies reporting that this cut-point, originally defined empirically using signal detection methods, consistently predicted treatment outcomes across both pharmacological and psychological treatments for BED (Grilo, Masheb, & Wilson, Reference Grilo, Masheb and Wilson2006; Grilo, White, Masheb, & Gueorguieva, Reference Grilo, White, Masheb and Gueorguieva2015; Grilo, White, Wilson, Gueorguieva, & Masheb, Reference Grilo, White, Wilson, Gueorguieva and Masheb2012b).
Participants with ≥65% or greater reduction in binge-eating frequency were categorized as ‘responders’ and included in the present maintenance treatment trial. Binge-eating frequency (during previous month – i.e. 28 calendar days) was assessed using the Eating Disorder Examination Interview (EDE; Fairburn, Cooper, & O'Connor, Reference Fairburn, Cooper, O'Connor and Fairburn2008) at the posttreatment evaluation (Grilo et al., Reference Grilo, Ivezaj, Tek, Yurkow, Wiedemann and Gueorguievain press). The EDE was administered by doctoral-level assessors blinded to the treatment conditions. Posttreatment evaluation was performed immediately following completion of the initial acute treatments and eligible participants were randomized and began this maintenance treatment study within a week.
The 61 participants had mean age of 44.33 (s.d. = 10.86) years and mean BMI of 36.06 (s.d. = 4.49) kg/m2; 83.6% (N = 51) were female, 88.5% (N = 54) attended some/finished college, and 75.4% (N = 46) were White. Table 1 summarizes the participants' sociodemographic characteristics as well the specific Stage 1 treatments (LDX or CBT + LDX) received for the Stage 2 RCT participants, overall (N = 61) and separately for the LDX (N = 32) and placebo (N = 29) treatments.
Demographic characteristics overall and across treatment conditions
Note: LDX, lisdexamfetamine; M, mean; s.d., standard deviation; N, number; BED, binge-eating disorder; CBT, cognitive behavioral therapy.
Assessments
Assessment procedures were performed by trained/monitored doctoral-level research-clinicians who were independent from and blinded to treatments (both acute Stage 1 and maintenance Stage 2). The Eating Disorder Examination Interview (EDE; 16th-edition; Fairburn et al., Reference Fairburn, Cooper, O'Connor and Fairburn2008) was administered to assess binge-eating frequency and eating-disorder psychopathology (Global score) at baseline and posttreatment. The EDE has demonstrated good inter-rater and test–retest reliability in studies with BED (Grilo et al., Reference Grilo, Lydecker, Fineberg, Moreno, Ivezaj and Gueorguieva2022; Grilo, Masheb, Lozano-Blanco, & Barry, Reference Grilo, Masheb, Lozano-Blanco and Barry2004). Weight and height were measured at baseline and weight was measured monthly and at post-treatment. When being weighed using digital scales, participants wore lightweight clothing without shoes.
To complement the primary interview measures at baseline and posttreatment, two self-report measures were administered monthly to capture data throughout the course of treatments. The Eating Disorder Examination-Questionnaire (Fairburn & Beglin, Reference Fairburn and Beglin1994), which has good test–retest reliability (Reas, Grilo, & Masheb, Reference Reas, Grilo and Masheb2006), obtained binge-eating frequency and eating-disorder psychopathology (Global score) data during the past 28 days. The Beck Depression Inventory-II (Beck, Steer, & Brown, Reference Beck, Steer and Brown1996) (alpha = 0.92) is well-established measure of depression symptoms/levels (Beck, Steet, & Garbin, Reference Beck, Steer and Garbin1998).
Randomization and blinding procedures
The randomization schedule, developed by a biostatistician, assigned participants who received LDX treatment (i.e. either LDX or CBT + LDX) and were categorized as responders to the acute Stage 1 treatment to either LDX or placebo (double-blind) for 12-weeks. Responder status to Stage 1 acute treatments was defined as ≥65% reduction in binge-eating frequency. This cut-point for response was selected based on consistent findings regarding the positive prognostic significance of such reductions across both pharmacological and psychological treatments (Grilo et al., Reference Grilo, Masheb and Wilson2006, Reference Grilo, Crosby, Wilson and Masheb2012b). This cut-point has been utilized in recent sequential designs (Grilo et al., Reference Grilo, White, Masheb, Ivezaj, Morgan and Gueorguieva2020, Reference Grilo, Lydecker and Gueorguieva2023) and is supported by time course data for LDX (McElroy et al., Reference McElroy, Hudson, Gasior, Herman, Radewonuk, Wilfley and Busner2017). ‘Responders’ were randomized in equal proportions to maintenance LDX or placebo, using stratified block randomization with Stage 1 acute treatment (LDX or CBT + LDX) as a stratifying variable. Assessors of outcomes were blinded to whether participants had received CBT during their prior Stage 1 treatment in addition to the (double-blind) medication in current trial.
Treatments
Maintenance treatment involved solely the double-blind medication (LDX or placebo). There were no additional behavioral or psychotherapeutic interventions during this maintenance trial. At the end of Stage 1 acute treatment, down-titration of LDX occurred over a period of four days (50 mg for two days and 30 mg for two days). Stage 2 maintenance treatment began within a week of having completed Stage 1 acute treatment and having completed the down-titration. LDX treatment during this maintenance trial was administered following the dose-optimization (targeting 50–70 mg/day) protocol found superior to placebo (McElroy et al., Reference McElroy, Hudson, Ferreira-Cornwell, Radewonuk, Whitaker and Gasior2016). During week 1, LDX was 30 mg/day for initial titration. During week 2, LDX was titrated to 50 mg/day. If the dose of 70 mg/day was not tolerated during the Stage 1 acute treatment, those participants were dosed at 50 mg/day for the remainder of Stage 2. If the dose of 70 mg/day was tolerated in Stage 1, during weeks 3–4, LDX was increased to 70 mg based on acceptable tolerability and clinical need. If patients developed intolerable side-effects on 70 mg in Stage 2, downward titration to 50 mg/day could occur, and if patients experienced adverse events and/or could not tolerate the medication, they were withdrawn from the medication. For the remaining treatment (weeks 3–12), the optimized LDX dose (50 or 70 mg/day) was maintained. Similar to Stage 1 acute treatment, down-titration consisted of four days (50 mg for two days and 30 mg for two days). Placebo was taken in capsules matched in appearance and frequency.
At the first (initial) maintenance trial study visit, study physicians delivered the pharmacotherapy, which focused on medication management (compliance, safety, and side-effects). Additional psychotherapeutic or behavioral interventions were proscribed. Monthly medication refills were accompanied by re-reviewing medication compliance and dosing schedules. Side-effect and safety checklists were performed monthly by research clinicians.
Statistical analysis
Sample size for this maintenance trial was calculated for the comparison between LDX and placebo for maintenance (Hudson et al., Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017), based on estimated rates of response to acute treatments with LDX and CBT (Grilo et al., Reference Grilo, Masheb, Wilson, Gueorguieva and White2011; McElroy et al., Reference McElroy, Hudson, Ferreira-Cornwell, Radewonuk, Whitaker and Gasior2016), and considering clinically meaningful effect-sizes. We estimated enrolling 80 participants would yield 64 with complete data at the end of Stage 2 which, in turn, would yield 84% power at two-sided alpha level of 0.05 to detect differences in relapse rates reported by Hudson et al. (Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017) for LDX v. placebo (4% v. 32%). Our intent-to-treat sample (N = 61 total, with posttreatment data for N = 58) was only slightly lower than the target and thus we were well-powered (power of 80%) for a priori specified clinically meaningful effects.
Analyses comparing treatments were intention-to-treat, performed for all randomized patients who attended the first treatment session. Statistical testing was performed at 0.05 significance level and is reported with unadjusted p-levels.
The primary outcome variable was ‘relapse’ (to DSM-5-level of once-weekly binge-eating). Relapse was examined using two complementary ways (binge-eating frequency on EDE interview at posttreatment and EDE-Q monthly throughout).
The secondary outcome variables were binge eating, weight-loss, eating-disorder psychopathology, and depression. Binge eating was analyzed in two complementary ways – i.e. as a continuous variable (episode frequency monthly) and a categorical variable (binge-eating remission, defined as zero episodes in previous month (EDE-interview)) with missing data imputed as failure (i.e. non-remission). Remission was intended to complement the less stringent primary outcome relapse variable.
Weight (measured) was analyzed as a continuous outcome (percent weight-loss from start of maintenance trial). We further examined weight-loss as a categorical outcome (attaining ≥5% weight-loss during maintenance trial, with missing data imputed as failure) as ‘complementary’ analysis to percent weight-loss. Beyond intended ‘convergence,’ this categorical variable, a longstanding standard outcome in obesity trials (Greenway et al., Reference Greenway, Fujioka, Plodkowski, Mudaliar, Guttadauria, Erickson and Dunayevich2010) and increasingly in BED trials (Grilo et al., Reference Grilo, Lydecker, Fineberg, Moreno, Ivezaj and Gueorguieva2022), is clinically useful as it is associated with cardiometabolic benefits in BED/obesity (Yurkow, Ivezaj, & Grilo, Reference Yurkow, Ivezaj and Grilo2023).
For analyses of continuous variables, intention-to-treat analyses used all available data in mixed models without imputation. Variables not conforming to normality were log-transformed prior to analysis. Mixed models were constructed with fixed factors including maintenance medication treatment during stage 2 (LDX v. placebo), treatment during stage 1 (LDX v. CBT + LDX), time (all relevant time points), and all possible interactions. When interactions with stage 1 treatment were not significant, they were dropped from the models for parsimony. In each model we considered different error structures and selected the best-fitting structure based on the Schwarz' Bayesian Criterion. Tests of effect slices and focused comparisons of least square means were used to explain significant effects in the models.
For categorical variable outcomes, intention-to-treat analyses involved generalized estimating equations (GEE) models. Logit link function was used with binomial response distribution. Predictor variables were the same as in the mixed models above. If models encountered convergence problems with all possible interactions, interactions were dropped, and the model refitted until a model with no convergence issues was identified.
Results
Randomization and participant flow through treatment study
Figure 1 (CONSORT) summarizes participant flow throughout the study. Of the 94 participants randomized to LDX or CBT + LDX in Stage 1 treatment, 80 were categorized as treatment responders; of those, 16 were excluded due to adverse events during acute Stage 1 treatments and 3 were not interested in Stage 2 treatment. Thus, 61 were randomized and attended baseline treatment session for this RCT. Of the 61, 32 were randomized to LDX and 29 were randomized to Placebo. Overall, 5 (8.2%) dropped out, 3 (4.9%) were medically withdrawn, and 58 (95.1%) completed posttreatment assessments.
Participant flow throughout the study. Participant flow through this randomized double-blind controlled trial (RCT) testing lisdexamfetamine (LDX) v. placebo for maintenance treatment of patients with binge-eating disorder who responded successfully to acute treatments (Stage 1). Stage 1 treatment was a RCT testing LDX and cognitive-behavioral therapy (CBT) combined with LDX (i.e., CBT + LDX). Table 1 shows the specific Stage 1 treatments (a stratifying variable in the randomization) received by the participants in this Stage 2 maintenance RCT. Of the 94 participants randomized to LDX or CBT + LDX in Stage 1 treatment, 80 were categorized as treatment responders; of those, 16 were excluded due to adverse events during acute Stage 1 treatments and 3 were not interested in Stage 2 treatment. Thus, 61 were randomized and attended baseline treatment session for this RCT. Of the 61 participants in this Stage 2 maintenance RCT, 32 were randomized to LDX and 29 were randomized to Placebo. Overall, 5 (8.2%) dropped out and 3 (4.9%) were medically withdrawn (2 cases in LDX, one due to headaches and one due to dry eyes; 1 case in placebo due to increased blood pressure (See online Supplemental Table S1). Of the 61 participants, 58 (95.1%) completed posttreatment assessments.
Primary outcomes
Relapse rates (meeting diagnosis-level binge-eating frequency determined with the EDE) following maintenance treatments (10.0% (N = 3/30) for LDX and 17.9% (N = 5/28) for placebo) did not differ significantly (χ2 (1) = 0.66, p = 0.42) nor were they associated with Stage 1 acute treatments (χ2 (1) = 0.85, p = 0.36). Two sensitivity analyses converged with this primary finding. First, because at baseline of this maintenance trial, 9.4% (N = 3/32) of those assigned to LDX and 0% (N = 0/29) of those assigned to placebo met diagnosis-level binge-eating frequency despite being ‘responders’ to acute treatment, sensitivity analysis testing relapse restricted to those who attained remission from binge eating was performed and it converged in indicating LDX and placebo did not differ (χ2 (1) = 1.26, p = 0.26). Second, a series of parallel analyses comparing relapse rates at each month (determined with the EDE-Q) throughout the course of maintenance treatments also revealed no significant differences between LDX and placebo.
Secondary outcomes
The secondary outcome variables were binge eating, weight-loss, eating-disorder psychopathology, and depression. Table 2 shows descriptive statistics for secondary continuous outcomes and Figs 2 and 3 summarize statistical findings. There were no statistically significant effects for depression (BDI-II) scores; statistical findings for other outcomes are below.
Clinical measures across treatment conditions
Note: EDE, Eating Disorder Examination Interview; EDE-Q, Eating Disorder Examination Examination–Questionnaire; BDI-II, Beck Depression Inventory-II; LDX, lisdexamfetamine; N, number; M, mean; s.d., standard deviation.
Binge-eating across the treatment medication conditions.
2a (top panel). Binge-eating remission rates during maintenance treatment (Stage 2) at baseline and at post-treatment. Remission rates are defined as zero episodes of binge eating during the last 28 days assessed using the Eating Disorder Examination Interview. The rates are based on the intention-to-treat sample (N = 61) with any missing data imputed as failure to remit. St1 = Stage 1 (acute treatment); St2 = Stage 2 (maintenance treatment). The four lines show the rates of binge-eating remission separately for lisdexamfetamine (LDX) and placebo conditions during maintenance (Stage 2) treatment separately by LDX or cognitive-behavioral therapy plus LDX (i.e. CBT + LDX) during initial acute (Stage 1) treatments. There were no significant interaction effects between the acute (Stage 1) treatment condition (LDX and CBT + LDX) and maintenance (Stage 2) treatments (LDX and placebo). Maintenance LDX and placebo did not differ significantly.
2b (bottom panel). Least Square Means (LSM) for frequency of binge-eating episodes during the past 28 days for the maintenance treatment (Stage 2) at baseline and at post-treatment (assessed using the Eating Disorder Examination interview). St1 = Stage 1 (acute treatment); St2 = Stage 2 (maintenance treatment). The four lines show the LSMs (error bars indicate standard errors) of binge-eating episode frequency separately for LDX and placebo conditions during maintenance (Stage 2) treatment separately by LDX or CBT + LDX during initial acute (Stage 1) treatments. There were no significant interaction effects between the acute (Stage 1) treatment condition (LDX and CBT + LDX) and maintenance (Stage 2) treatments (LDX and placebo). Maintenance LDX and placebo did not differ significantly.
Percent weight loss across the treatment medication conditions. Least Square Means (LSM) of percent weight loss (from baseline start of maintenance treatment) calculated using measured values at baseline, measured monthly during maintenance treatment, and measured at post-treatment. St1 = Stage 1 (acute treatment); St2 = Stage 2 (maintenance treatment); Mth = Month. The four lines show the LSMs (error bars indicate standard errors) for lisdexamfetamine (LDX) and placebo conditions during maintenance (Stage 2) treatment separately by LDX or cognitive-behavioral therapy plus LDX (i.e. CBT + LDX) during initial acute (Stage 1) treatments. Analyses revealed significant interaction between maintenance treatment and time (p = 0.0002) and a significant main effect of maintenance treatment (p < 0.0001); acute (stage 1) treatment effects were not significant (p = 0.43). LDX maintenance had significant and progressive weight loss on average whereas placebo maintenance had significant and progressive weight gain on average. LDX and placebo were significantly different from each other at every monthly time-point through posttreatment (p < 0.0001).
Binge-eating outcomes
Intention-to-treat binge-eating remission rates following maintenance treatments were 59.4% (N = 19/32) for LDX and 65.5% (N = 19/29) for placebo. Figure 2a illustrates remission rates did not differ significantly (χ2 (1) = 0.24, p = 0.62) nor were they associated with Stage 1 acute treatments (χ2 (1) = 0.24, p = 0.62). Sensitivity analysis restricted to completers revealed convergent non-significant findings.
Mixed models of binge-eating frequency during the past month at posttreatment (based on the EDE) revealed no significant effects, including no significant treatment-by-time effects (F(1,56.5) = 2.08, p = 0.15). Figure 2b illustrates binge-eating frequency findings showing maintenance LDX and placebo conditions did not differ significantly (F(1,57.9) = 2.13, p = 0.15) nor were they associated with Stage 1 acute treatments (F(1,58.2) = 0.91, p = 0.34).
Mixed models of binge-eating frequency assessed monthly throughout the course of maintenance treatment revealed a statistically significant interaction between maintenance treatment and time (F(3157) = 8.32, p < 0.0001) and a statistically significant main effect of acute (Stage 1) treatment (F(1,57.7) = 5.72, p = 0.02). Stage 1 LDX-only, relative to CBT + LDX, had higher overall means of binge-eating frequency throughout Stage 2 maintenance. During Stage 2 maintenance trial, binge-eating frequency increased significantly with placebo maintenance (F(3155) = 9.39, p < 0.0001) and decreased non-significantly with LDX maintenance (F(3158) = 1.30, p = 0.28). The placebo maintenance group, which had significantly lower binge-eating frequency at baseline (F(1135) = 6.69, p = 0.01), had significantly higher frequency than the LDX maintenance group at posttreatment (F(1147) = 4.57, p = 0.03).
Percent weight loss
Table 2 shows weight values at baseline and post-treatment and changes. Mixed models of percent weight-loss at posttreatment during the maintenance trial revealed a significant main effect of maintenance treatment (F(1,55) = 40.19, p < 0.0001); acute stage 1 treatment was not significant (F(1,55) = 1.22, p = 0.27). The LDX group had significant weight-loss (−2.33%) whereas the placebo group had significant weight-gain (+2.24%) on average (both p < 0.0001).
Figure 3 summarizes percent weight-losses for the LDX and placebo treatments throughout the course of the maintenance trial shown separately for those who received LDX or CBT + LDX during the acute (Stage 1) treatments. Mixed models of percent weight-loss monthly during the maintenance trial revealed a significant interaction between maintenance treatment and time (F(2,53.8) = 9.90, p = 0.0002) and a significant main effect of maintenance treatment (F(1,53.5) = 53.08, p < 0.0001); acute (stage 1) treatment was not significant (F(1,53) = 0.63, p = 0.43). The LDX group had significant and progressive weight loss on average whereas the placebo group had significant and progressive weight gain on average. LDX and placebo were significantly different from each other at every monthly time-point through posttreatment (p < 0.0001).
Attaining 5% weight loss. The group receiving LDX maintenance treatment was significantly more likely than the placebo group to attain ≥5% weight loss specifically during the maintenance period (18.8% (N = 6/32) v. 0% (N = 0/29); Fisher's exact test value = 0.025).
Eating-disorder psychopathology
Table 2 shows the eating-disorder psychopathology global score values at baseline and at posttreatment and changes separately for the EDE interview and EDE-Q are shown in Figure 4. Mixed models of EDE global score revealed a statistically significant interaction between maintenance treatment and time (F(1,54.9) = 7.88, p = 0.01) and a statistically significant main effect of time (F(1,54.9) = 5.95, p = 0.02). EDE global score increased significantly with placebo maintenance (F(1,54.9) = 13.51, p = 0.0005) whereas it did not change significantly with LDX maintenance (F(1,55) = 0.07, p = 0.79).
Eating-disorder psychopathology across treatment medication conditions.
4a (top panel). Least Square Means (LSMs) for global score on the Eating Disorder Examination Interview during maintenance treatment (Stage 2) at baseline and at post-treatment. St1 = Stage 1 (acute treatment); St2 = Stage 2 (maintenance treatment). The four lines show the eating-disorder psychopathology global scores (LSMs) separately for lisdexamfetamine (LDX) and placebo conditions during maintenance (Stage 2) treatment separately by LDX or cognitive-behavioral therapy plus LDX (i.e. CBT + LDX) during initial acute (Stage 1) treatments. Analyses revealed a statistically significant interaction between maintenance treatment and time (p = .01) and a statistically significant main time effect (p = 0.02). EDE global score increased significantly with placebo maintenance (p = 0.0005) whereas it did not change significantly with LDX maintenance (p = 0.79).
4b (bottom panel). Least Square Means (LSMs) for global score on the Eating Disorder Examination-Questionnaire at monthly assessments throughout the maintenance treatment (Stage 2) at baseline and at post-treatment. The four lines show the eating-disorder psychopathology global scores (LSMs) separately for lisdexamfetamine (LDX) and placebo conditions during maintenance (Stage 2) treatment separately by LDX or cognitive-behavioral therapy plus LDX (i.e. CBT + LDX) during initial acute (Stage 1) treatments. St1 = Stage 1 (acute treatment); St2 = Stage 2 (maintenance treatment).
Mixed models of the EDE-Q global score monthly during the maintenance trial revealed a significant interaction between maintenance treatment and time (F(3154) = 10.83, p < 0.0001) and a significant main effect of time (F(3154) = 6.58, p = 0.0003); additionally, a statistically significant main effect of acute Stage 1 treatment was observed (F(1,58) = 4.25, p = 0.04) reflecting higher scores on average for those who received LDX than CBT + LDX during initial acute treatment. Throughout the maintenance trial, placebo had significant increase in EDE-Q Global scores (F(3154) = 17.28, p < 0.0001) whereas no significant changes occurred with LDX (F(3155) = 0.27, p = 0.85).
Discussion
This prospective controlled maintenance treatment study of adults with BED with comorbid obesity – only the third controlled maintenance trial performed to date with BED – tested the effectiveness of LDX for maintenance following successful acute treatments with LDX (i.e. LDX-alone or combined with CBT). Participants with good outcomes following a controlled acute trial (Grilo et al., Reference Grilo, Ivezaj, Tek, Yurkow, Wiedemann and Gueorguievain press) were re-randomized, in double-blind fashion, to either LDX or to placebo for 12 weeks. Findings indicate that adults with BED with obesity with good responses to initial acute treatment with LDX – regardless of whether also having received CBT – showed good maintenance during the subsequent 12-weeks. Relapse (to diagnosis-level binge-eating frequency) was infrequent and did not differ between LDX and placebo. Intention-to-treat binge-eating remission rates following the maintenance trial were 59.4% for LDX and 65.5% for placebo. Importantly, although maintenance LDX, relative to placebo, did not provide further benefit for most measures of binge-eating, it was associated with significantly with better eating-disorder psychopathology outcomes and with greater further weight-loss.
Our findings supporting LDX maintenance treatment add to those previously reported by Hudson et al. (Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017). Whereas Hudson et al. (Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017) found that maintenance LDX reduced relapse risk relative to placebo (3.7% v. 32.1%), we did not find a significant difference (i.e. 10.0% v. 17.9%, respectively). The maintenance trial by Hudson et al. (Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017) was 26 weeks (v. 12 weeks in the present trial) and that longer time-period could have allowed for more relapses to occur over time. However, while our shorter maintenance trial found no statistical advantage for maintenance LDX v. placebo for binge-eating (various categorical and continuous measures), LDX was significantly superior to placebo for maintaining improvements in associated eating-disorder psychopathology. Specifically, eating-disorder psychopathology remained unchanged with maintenance LDX but increased significantly with placebo. This finding may be clinically important in light of recent findings that certain aspects of associated eating-disorder psychopathology at posttreatment prospectively predicted relapse one year later in patients who had achieved remission from binge eating with behaviorally based treatments for BED (Grilo et al., Reference Grilo, Ivezaj and Gueorguieva2024).
Contrary to hypothesis, having received CBT during the initial acute treatment (i.e. in addition to LDX) did not significantly moderate LDX maintenance treatment outcomes. Broadly, these findings contrast with those reported by Grilo et al. (Reference Grilo, Ivezaj and Gueorguieva2024) in a maintenance trial testing naltrexone/bupropion; in that trial, placebo was associated with worsening outcomes amongst those treated initially with naltrexone/bupropion but not amongst those who received behavioral therapy. The reasons for this are uncertain; we can speculate that the relatively brief duration of the maintenance trial might not have allowed enough time for additional binge-eating relapse events to occur following the high response and remission rates attained with the acute treatments.
Maintenance LDX was associated with significantly greater weight-loss (2.3% loss) than placebo (which had 2.2% gain); 18.8% of the LDX group attained 5≥% weight-loss during the maintenance trial. LDX treatments resulted in significant weight-loss during the initial acute trial of roughly 5% and the maintenance LDX treatment resulted in a further weight-loss (roughly 2.3%). These findings are notable as most psychological and pharmacological (except topiramate) trials for BED have reported little-to-no weight loss (Grilo et al., Reference Grilo, Ivezaj and Gueorguieva2024) and behavioral weight-loss trials for BED have reported attenuated weight-losses in BED compared those reported for patients with obesity without BED (Forman et al., Reference Forman, Evans, Berry, Lampe, Chwyl and Zhang2023). BED is associated with subsequent onsets of chronic cardiometabolic conditions (Hudson et al., Reference Hudson, Lalondie, Coit, Tsunang, McElroy, Crow and Pope2010; Kessler et al., Reference Kessler, Berglund, Chiu, Deitz, Hudson, Shahly and Xavier2013) and research has found that modest weight losses (>5%) are associated with significantly improved glycemic control and reduced cardiometabolic abnormalities in patients with obesity (Wing et al., Reference Wing, Lang, Wadden, Safford, Knowler, Bertoni and Wagenknecht2011) including those specifically with BED (Yurkow et al., Reference Yurkow, Ivezaj and Grilo2023).
As context for the findings, we note methodological strengths and limitations. Methodological strengths include the randomized double-blind design, pharmacotherapy delivery by faculty-level physician without any additional psychotherapeutic or nutrition interventions, assessments performed by independent and blinded doctoral evaluators using well-validated measures, and high retention rates. Further, the study design allowed for prospective examination of potential moderating effects of CBT received during initial acute pharmacotherapy treatment. Several potential limitations are noteworthy. The sample size had limited power to detect smaller magnitude main/interactive effects of treatments. Generalizability of findings to different clinical settings, different providers, and to persons with BED without obesity, different sociodemographic and clinical profiles is uncertain. The participant group was 84% female, well-educated (roughly 72% attended some college), and had reasonable racial/ethnic representation (15% Black and 20% Hispanic).
Lastly, we note this maintenance trial was brief (only 12 weeks) and longer-term outcomes are unknown. With this important caveat highlighted, we cautiously comment on what practitioners and their patients can do after 12-week LDX maintenance treatment. Hudson et al. (Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017), in their controlled maintenance trial for BED (double-blind randomized withdrawal design), found that LDX was associated with significantly decreased risk of relapse over a 26-week period alongside a well-tolerated and well-known side-effect profile for this medication. Thus, it seems reasonable for practitioners to discuss with their patients with BED with co-existing obesity who have benefitted from acute and maintenance trials with LDX and have no side-effects that continuing with the medication longer may be worth considering. In addition to reducing risk for relapse (Hudson et al., Reference Hudson, McElroy, Ferreira-Cornwell, Radewonuk and Gasior2017), our study found that continued LDX was associated with significantly better eating-disorder psychopathology outcomes and greater weight-loss. We emphasize that continued or longer-term maintenance treatment with LDX should include careful on-going medical monitoring to ensure safety, address any emergent side-effects, identify any problematic medication misuse behaviors, and to identify any excessive weight-loss or emergence of eating-disorder psychopathology. We also highlight that our study group comprised participants carefully assessed to establish the absence of histories of alcohol/substance use disorders (given the US FDA's product labelling for LDX, which includes a ‘black box’ warning regarding potential for misuse) in addition to various important contraindications, including cardiac conditions. We also note that the US FDA labeling includes a ‘limitation of use’ that LDX is not indicated for weight loss and that its safety and efficacy for obesity are unknown. Thus, the discussion with patients regarding potential benefits/risks of LDX should also consider carefully the US FDA's product labelling alongside our study's outcomes and other medical data as they become available.
With these methodological considerations as context, we conclude that adults with BED with obesity who respond to acute (12-week) LDX (regardless of having additionally received CBT) could be offered maintenance treatment with LDX. Maintenance LDX, relative to placebo, did not provide further benefit for binge-eating but was associated with significantly better eating-disorder psychopathology outcomes and further weight-loss.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S003329172400148X.
Data availability statement
De-identified data will be provided in response to reasonable written request to achieve goals in an approved written proposal.
Funding statement
This study was funded by the National Institutes of Health (NIH) grant R01 DK114075 (Grilo). This research was also supported, in part, by the NIH and National Center for Advancing Translational Science CTSA grant UL1 TR001863. The NIH played no role in the content of this paper. Takeda provided lisdexamfetamine medication in response to investigator-initiated request but played no role in any aspect of the study design or study conduct and played no role in the content of this paper.
Clinicaltrials.gov registration: NCT03926052
(Cognitive-Behavioral and Pharmacologic (LDX) Treatment of Binge-Eating Disorder and Obesity: Maintenance Treatment)
Competing interests
The authors declare no conflicts of interest. Dr Grilo reports broader interests, which did not influence this research, including Honoraria for lectures, and Royalties from Guilford Press and Taylor & Francis Publishers for academic books. Dr Gueorguieva reports royalties from Taylor & Francis Publishers for an academic book.
Open access
