Semaglutide 2.4 Mg for the Management of Overweight and Obesity: Systematic Literature Review and Meta-Analysis

Introduction

Obesity is a major global public health issue; the World Health Organisation estimated that in 2016 there were almost 2 billion adults worldwide with overweight and 650 million with obesity.¹ Obesity is associated with the development of serious comorbidities, including type 2 diabetes mellitus (T2DM), cardiovascular disease, obstructive sleep apnea, and osteoarthritis.^2–4

Lifestyle interventions, such as improved diet and increased physical activity, are the cornerstone of obesity management.⁵ However, lifestyle intervention alone may not be a durable management method for overweight and obesity as it does not appear to have a long-term impact on morbidity or mortality, as found in the Look AHEAD study.^6,7 Pharmacological options for obesity include orlistat (lipase inhibitor), liraglutide (glucagon-like peptide 1 [GLP-1] agonist), naltrexone/bupropion (opioid antagonist/dual norepinephrine and dopamine reuptake inhibitor), and phentermine/topiramate (adrenergic agonist/gamma-aminobutyric acid receptor modulator). However, despite the substantial burden of obesity, pharmacological therapy has not been widely adopted as a management approach.⁸ This may be because these therapies only result in modest additional weight loss when used in conjunction with lifestyle intervention.⁸ In addition, there are safety concerns with some pharmacological weight loss therapies, such as phentermine/topiramate, that have led to negative opinions from regulatory agencies.⁹ Therefore, newer, more effective, and safer pharmacological interventions are needed to augment lifestyle interventions in overweight and obesity. Newer pharmacological interventions include tirzepatide (glucose-dependent insulinotropic polypeptide and GLP-1 agonist), which has demonstrated efficacy in people with T2DM in the SURPASS-2 trial,¹⁰ as well as sodium glucose cotransporter 2 (SGLT-2) inhibitors (eg, dapagliflozin, canagliflozin) that are used to manage T2DM but can also induce clinically significant weight loss.¹¹

Another newer intervention is semaglutide, a long-acting GLP-1 analogue that mimics the effects of native GLP-1, which regulates appetite and blood sugar levels.^12,13 Weight loss with semaglutide occurs through a reduction in energy intake, increased satiety and satiation, and reduced hunger, in conjunction with enhanced glycemic control; the latter occurs through semaglutide’s action on GLP-1, which stimulates insulin secretion and suppresses glucagon secretion when blood glucose levels are high.^12–14 The safety and efficacy of semaglutide in overweight and obesity has been demonstrated in the Semaglutide Treatment Effect in People with obesity (STEP) clinical trial program. The STEP program (currently comprising STEP 1 to STEP 8) is a collection of Phase 3 trials of 2.4 mg semaglutide administered subcutaneously once weekly in conjunction with different intensities of lifestyle interventions. The STEP study participants have overweight or obesity, with or without T2DM. Compared with placebo, semaglutide 2.4 mg conferred a significantly greater reduction in body weight and a higher proportion of participants achieved ≥5% weight reduction in STEP 1 and STEP 2.^12,15 Weight loss with semaglutide 2.4 mg was accompanied by greater improvements in cardiometabolic risk factors (eg, waist circumference, blood pressure, and lipid levels) and physical functioning than placebo.^12,15 The most common adverse events with semaglutide were gastrointestinal (nausea, diarrhea) but these were generally mild-to-moderate and transient.^12,15

Until the results of STEP 8 (weekly semaglutide 2.4 mg versus daily liraglutide 3.0 mg) were published in 2022,¹⁶ all randomized controlled trials (RCTs) in the STEP program have compared semaglutide with placebo; therefore, to compare semaglutide with other active comparators, indirect treatment comparison is required. The objective of this systematic literature review (SLR) and meta-analysis was to compare RCT evidence for weekly semaglutide 2.4 mg with that of relevant pharmacological comparators for weight management in people who have overweight or obesity. The outcomes of interest were weight change from baseline (CFB, %) at 52 weeks and proportion of participants losing ≥5% baseline fasting body weight at 12 weeks at full therapeutic dose (ie, as per the stopping rule for certain weight loss drugs, such as orlistat and liraglutide, when participants fail to achieve ≥5% baseline fasting body weight at 12 weeks at full therapeutic dose).

Materials and Methods

Systematic Literature Review

A systematic literature review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) checklist.¹⁷

To identify RCTs reporting on semaglutide 2.4 mg and comparators (to include pharmacological agents, surgical intervention, and diet) in people with overweight or obesity, searches of Medline^®, Medline^® Epub Ahead of Print (In-Process & Other Non-Indexed Citations), Embase, and EBM Reviews were performed via Ovid on 8th September 2020. The search strategy used to interrogate the EMBASE database is provided in Supplementary Table 1. Additional searches of conference proceedings (from the last 3 years), health technology assessment (HTA) body websites, clinical trial registries (ClinicalTrials.gov and World Health Organization International Clinical Trials Registry Platform), and reference lists of included studies were performed to identify other relevant evidence. Eligibility criteria included RCTs, SLR, and meta-analysis publications reporting efficacy and safety data for semaglutide 2.4 mg and relevant comparators. Full eligibility criteria are presented in Table 1.

Table 1 Eligibility Criteria for the Systematic Literature Review

Citations of interest were identified by a member of the team (authors EH or SM) and verified by an independent reviewer (authors SB or SM), based on title and abstract. Full publications were obtained for all citations of interest and were assessed by one reviewer and verified by a second reviewer. Any uncertainties were resolved through discussion. Data were extracted into an Excel spreadsheet by one reviewer and checked against the original publication by a second reviewer.

Quality Assessment

Quality (risk of bias) assessment of eligible RCTs was conducted using the criteria provided in the National Institute for Health and Care Excellence (NICE) single technology appraisal user guide.¹⁸ This approach is in line with guidance provided by the Centre for Reviews and Dissemination for assessing the quality of studies in SLRs.¹⁹

Network Meta-Analysis

A network meta-analysis (NMA) was performed to compare weight change for semaglutide 2.4 mg with comparators. Relevant data were analyzed from publications identified in the SLR (considering total trial population data or subgroup data). The populations of interest were total population, normal glucose tolerance (NGT), non-T2DM, pre-diabetes, and T2DM; separate NMA analyses were conducted for each of these populations.

The outcomes of interest were weight CFB (%) at 52 weeks and proportion of participants losing ≥5% baseline fasting body weight after 12 weeks at full therapeutic dose. The latter outcome included 16-week data for liraglutide, naltrexone/bupropion and phentermine/topiramate (4 weeks titration and 12 weeks at full dose), 12-week data for orlistat (no titration), and 28-week data for semaglutide (16 weeks titration and 12 weeks full dose).

Bayesian framework and Markov Chain Monte Carlo (MCMC) simulation were used for modelling, with the inclusion of vague prior distributions in line with guidance on evidence synthesis from the NICE Decision Support Unit.^20–22 All NMA models were fitted using WinBUGS software (MRC Biostatistics Unit, Cambridge, UK); normal likelihood, identity link (CFB weight) and binomial, logit link (proportion of participants losing ≥5% baseline fasting body weight). The models were run with three chains for a burn-in of 50,000 iterations and inferences were based on a further 20,000 iterations. The point estimate of the results represented the median of the posterior distribution with associated 95% credible intervals (CrIs). As there are no closed loops of evidence within the current evidence base (other than loops of evidence comprised multi-arm trials) it was not necessary to assess inconsistency in the networks.

Both fixed effect (FE) and random effect (RE) models were conducted, and model fit compared in terms of deviance information criterion (DIC) and residual deviance; the model with the lowest DIC and residual deviance closest to the number of data points was selected as the model of choice.

Results

Systematic Literature Review

The electronic database searches identified a total of 13,287 publications. Following deduplication, 11,352 publications were screened by title and abstract. Of these, 318 articles were deemed potentially relevant and were screened based on the full publication. A further 112 publications were excluded at this stage. Handsearching yielded an additional seven citations, giving a total of 213 publications covering 130 unique studies for inclusion. The flow of studies through the review is summarized in the Preferred Reporting Items for Systematic Reviews and Meta-analysis flow diagram (PRISMA) in the Supplementary Figure 1.

Non-surgical interventions were of primary interest for the NMAs. The potential inclusion of bariatric surgery in the NMA was explored but was not deemed appropriate because the populations included in the bariatric surgery trials (comparing surgery with medical treatment or diet) were highly selective in comparison with the studies investigating pharmacological agents. For example, populations in the surgical RCTs had a higher mean body mass index (BMI), included participants with comorbidities, often required weight loss of ≥5% prior to enrolment, included small numbers of participants, and had issues associated with participant retention. These factors may have biased the trial results in favor of bariatric surgery. Of the trials included in the SLR, 108 RCTs (164 publications) investigated non-surgical interventions. Of these, 67 RCTs (115 publications) were not considered for the meta-analysis for the following reasons: trial duration <9 months (STEP trials reported outcome data at a minimum of 9 months’ follow-up and therefore a trial duration of <9 months did not align with the STEP program and the outcomes considered in the analyses); no appropriate comparator for inclusion in evidence network; no relevant data reported; and subpopulation not of interest. The remaining 41 RCTs (49 publications) were considered for the NMAs (Table 2). Table 3 shows the active comparators with available data for each outcome-specific network (placebo/control was the common comparator across all networks) and the evidence networks are presented in Supplementary Figure 1 and Supplementary Figure 2. For the percent weight CFB to 52 weeks, data were available for all comparators in the total, non-T2DM, and T2DM populations. Liraglutide was the only comparator with available data in the NGT population. In the pre-diabetic population, data were available for liraglutide and phentermine/topiramate.

Table 2 Summary of Eligible Studies Investigating Pharmacological Agents Identified in Systematic Literature Review

Table 3 Comparators for Which Data are Available Across the Populations of Interest

For the proportion of participants losing ≥5% baseline fasting body weight at 12 weeks of full therapeutic dose, no comparisons were feasible for semaglutide 2.4 mg versus phentermine/topiramate or naltrexone/bupropion. Data on liraglutide and orlistat were available for the total and non-T2DM populations, while liraglutide was the only comparator for the NGT, pre-diabetic and T2DM populations.

Variability was noted across the trials in terms of the study design and populations (eg, age, sex, weight, and BMI); however, the studies in the evidence networks were considered sufficiently homogenous to combine in NMAs. Potential outlier studies included the Light study, which enrolled women aged ≥50 years and men aged >45 years and included a population with a higher mean age than other trials. The EQUIP trial enrolled people with BMI ≥35 kg/m² and therefore the study population had a higher mean BMI than other trials. There were also differences in the proportion of participants with comorbidities, particularly T2DM. A total of 13 studies included in the SLR exclusively enrolled people with T2DM^23–31 or a proportion of participants with T2DM.^32–34 As relative treatment effects may be different in people with or without T2DM, the current analysis considered subpopulations according to glucose tolerance to account for any differences conferred by T2DM.

Intensive behavioral therapy (IBT) is a form of obesity management that aims to change eating and exercise habits.³⁵ Inclusion of IBT as an adjunctive treatment in studies may be disadvantageous to pharmacological therapies; the relative treatment effects versus control may be lower in trials that include IBT than in those that exclude IBT.³⁶ Five trials (COR-BMOD,³⁷ NCT02911818,³⁸ STEP 3,³⁶ SCALE IBT,³⁹ and SCALE Insulin trial)³¹ specified IBT as a concomitant lifestyle therapy (comprising counselling, dietary advice, and increased physical activity). To account for the potential impact of IBT on the outcomes of the NMA, these five trials were excluded from the networks (although it is noted that participants in the remaining studies in the network may also have received dietary and exercise advice). Thus, up to 20 studies were included in the evidence networks (20 in the total population, nine in the TD2M network, 12 in the non-T2DM network, three in the pre-diabetes network, and two in the NGT network).

Results of the quality assessment are presented in Supplementary Table 2 for all 41 RCTs (49 publications) considered for the NMAs. In general, the trials included in the NMAs were high quality with adequate randomization and concealment of treatment allocation. Across all studies, baseline characteristics were well balanced between treatment groups and all measured outcomes were reported. All trials were blinded except one that was not blinded and the authors acknowledged that this prevented them from determining the independent effect of orlistat.⁴⁰ In a second study, the extent of blinding was unclear but the potential impact on results was not discussed.⁴¹ All studies used an intention-to-treat (ITT) analysis except for two in which this was unclear^42,43 and two that did not use an ITT analysis. Of the latter studies, one analyzed completers at 2 years but the impact of the non-ITT analysis was not discussed in the publication.⁴⁴ The second study was excluded from the NMAs as it included IBT.³⁷

Network Meta-Analysis

Percent Weight CFB to 52 Weeks

In all populations studied, semaglutide 2.4 mg was associated with a greater percentage weight CFB with 52 weeks of treatment versus all available comparators, except for phentermine 15 mg/topiramate 92 mg in the T2DM population.

In the total population, the differences with semaglutide 2.4 mg versus comparators ranged from –0.40% (CrI –4.22, 3.46) for phentermine 15 mg/topiramate 92 mg to –9.36% (CrI –12.06, –6.62) for placebo/control. In the T2DM population, the differences with semaglutide 2.4 mg versus comparators ranged from 0.67% (CrI –6.65, 7.86) for phentermine 15 mg/topiramate 92 mg to –6.22% (CrI –11.29, –1.19) for placebo. In the non-T2DM population, the differences with semaglutide 2.4 mg versus comparators ranged from –3.35% (CrI –5.85, –0.80) with phentermine 15 mg/topiramate 92 mg to –12.43% (CrI –14.51, –10.38) with placebo/control. In the pre-diabetes population, the differences with semaglutide 2.4 mg versus comparators ranged from –3.06% (CrI –4.55, –1.59) with phentermine 15 mg/topiramate 92 mg to –11.26% (–12.52, –10.01) with placebo/control. In the NGT population, data were available for comparison of semaglutide 2.4 mg with placebo/control (difference –13.42% [CrI –14.56, –12.28]) and liraglutide 3.0 mg (–8.13% [CrI –9.47, –6.80]). Full results of the NMAs are shown in Table 4.

Table 4 NMA Results: Estimates of Difference in Percent Weight CFB (Semaglutide versus Comparators, Excluding Trials That Included IBT)

Proportion of Participants Losing ≥5% Fasting Body Weight at 12 Weeks of Full Therapeutic Dose

In all populations, semaglutide 2.4 mg was associated with a higher likelihood of participants losing ≥5% of baseline fasting body weight at 12 weeks versus all available comparators (liraglutide 3.0 mg and orlistat 120 mg). The ORs for semaglutide versus placebo ranged from 8.86 (95% CI 6.24, 12.00 [T2DM]) to 14.81 (95% CI 11.02, 20.17 [NGT]). When considered against active comparators, semaglutide also increased the likelihood of participants losing ≥5% fasting body weight, with ORs ranging from 1.29 (95% CI 0.70, 2.13 [T2DM]) to 2.17 (95% CI 1.44, 3.27 [NGT]) with liraglutide 3.0 mg and from 6.09 (95% CI 2.86, 11.95 [total population]) to 7.75 (95% CI 5.29, 11.37 [non-T2DM]) with orlistat 120 mg three times daily (TID). Full results of the NMA for this outcome are presented in Table 5.

Table 5 NMA Results: OR (95% CrI) for Proportion of Participants Losing ≥5% of Baseline Fasting Body Weight at 12 Weeks Full Therapeutic Dose (Semaglutide versus Comparators, Excluding Trials That Included IBT)

Discussion

Obesity is a substantial clinical and economic burden¹ that has grown rapidly in recent decades⁴⁵ and is expected to continue to increase globally in the coming years.⁴⁶ Despite overweight and obesity affecting >2 billion people worldwide,¹ uptake of pharmacological treatments for the management of these conditions is generally low.⁸ Low uptake has been attributed to several factors, including only modest additional weight loss with the available therapies, safety concerns, lack of health-care professional experience with such therapies, and expectations that overweight/obesity should be managed with behavioral measures.^8,47 There is therefore an unmet need for an effective and safe pharmacological therapy for the management of overweight and obesity.

Randomized controlled trials of semaglutide in conjunction with lifestyle modifications have shown that semaglutide 2.4 mg represents an effective and safe weight loss reduction therapy in people with overweight and obesity with and without T2DM.^12,15 Because the trials compared semaglutide with placebo, no head-to-head data were available to compare semaglutide with active treatments used in the management of overweight and obesity. The current SLR and meta-analysis was performed to identify and compare RCT evidence for weekly semaglutide 2.4 mg with that of relevant pharmacological comparators for weight management in overweight or obesity.

The SLR identified 20 RCTs that were ultimately included in the NMA. In the NMA, semaglutide 2.4 mg was associated with a greater percent weight CFB with 52 weeks of treatment versus all available comparators in all populations, except phentermine 15 mg/topiramate 92 mg in the T2DM population. A single study, CONQUER,³⁴ contributed subgroup data to this comparison. It is noteworthy that in the phentermine 15 mg/topiramate 92 mg arm of CONQUER, 192 of 995 participants in the total population discontinued due to adverse events (it was unclear how many of the 664 participants with T2DM in this arm discontinued), which was notably higher than the placebo and phentermine 7.5 mg/topiramate 46 mg treatment arms (89 of 994 and 58 of 498 discontinuations due to adverse events, respectively). The results of this trial confirm the safety concerns associated with phentermine/topiramate. Further, the analyses in this trial were conducted on the ITT sample and applied the last observation carried forward (LOCF) principle which may have biased the trial level results in favor of phentermine 15 mg/topiramate 92 mg. Whilst LOCF was the most commonly used method to address missing data across the trials included in the NMA, LOCF-based statistical approaches are no longer recommended due to concerns regarding the plausibility of the assumptions (ie, body weight would be unaffected by discontinuations for trial participants lost to follow-up) and the potential for bias.⁴⁸ In contrast, the STEP trials used treatment policy estimands in the primary analyses, which assess all participants who were randomly assigned to treatment regardless of adherence to treatment and regardless of initiation of other therapies; missing data were imputed using a multiple imputation approach. Thus, the use of treatment policy estimands and the difference in approach to addressing missing data in the STEP trials versus the comparator trials in the NMA will have likely biased the treatment effect against semaglutide.

Semaglutide 2.4 mg was also associated with a higher chance of losing ≥5% baseline fasting body weight at 12 weeks (at full therapeutic dose) versus all available comparators. When considering the outcomes associated with semaglutide, it is important to note that semaglutide has a longer dose escalation period (at least 4 weeks) than other therapies (eg, at least one week for liraglutide); therefore, patients receiving semaglutide have a longer treatment exposure duration compared with those receiving other therapies.

While head-to-head comparison data are preferable where available, NMA is a robust statistical approach that facilitates indirect comparison of multiple treatment options when direct comparative data are not available. NMAs are widely used in healthcare decision-making, including in health technology assessment. Compared with head-to-head data from RCTs, in which there is expected to be minimal confounding between the treatment groups, NMA will inevitably be based on data with a degree of heterogeneity between studies. Although the studies in the current evidence networks were sufficiently homogeneous to perform NMAs, some variability was observed between the trials in terms of study designs and patient populations. Differences in the proportion of participants with abnormal glucose tolerance or T2DM were addressed through the analysis of subpopulations according to glucose tolerance. However, other potential effect modifiers were observed, including one study that only included women⁴⁰ and variation in the age categories included (eg, all participants aged ≥18 years or a narrower age range, such as 25–55 years). It is also important to recognize that publications cover a wide timescale, ranging from 1998 to present. The older studies (which primarily reported on orlistat) had significant amounts of missing outcome data, particularly for adverse events, which may have resulted in reporting bias. Underreporting of adverse events in orlistat trials was also described by Schroll et al who found considerable disparities in adverse event reporting between clinical study protocols, reports, and publications. It was estimated that only 3–33% of all investigator-reported adverse events were eventually presented in the relevant publications, even though most publications claimed that all adverse events were recorded.⁴⁹

The outcomes reported here were part of a larger analysis that included additional outcomes, such as glucose control, cholesterol, blood pressure and serious adverse events (SAEs). The safety of semaglutide compared with other pharmacological therapies is an important consideration as many of the other therapies have an adverse event profile that contributes to their low uptake.^47,50 Although phentermine/topiramate is licensed in the US, it received a negative opinion for the European Committee for Medicinal Products for Human Use (CHMP) due to concerns about the phentermine’s long-term impact on the heart and blood vessels, as well as the potential for psychiatric and cognitive effects.⁹ The CHMP also noted methodological concerns because the data showed a study dropout rate of approximately 40% and a lost to follow-up rate of more than 10%.⁹ An NMA conducted for SAEs associated with semaglutide 2.4 mg and comparators showed that semaglutide 2.4 mg was associated with a higher proportion of SAEs than comparators in the populations studied. The exception to this was in the T2DM population where semaglutide 2.4 mg was associated with fewer SAEs than all available comparators, except naltrexone 32 mg/bupropion 360 mg. However, it is important to note that the credible intervals associated with the results were very wide, indicating a high degree of uncertainty. Furthermore, it has been reported elsewhere that the side effects of semaglutide 2.4 mg are mainly nausea, diarrhea, and cholelithiasis, which are typical for this drug class.⁵⁰ Finally, as described earlier, there is evidence to suggest that adverse events associated with other weight loss drugs, particularly orlistat, are considerably underreported in the published literature.⁴⁹

Since the current SLR and NMA were conducted, the results of several other studies of semaglutide have been published, as well as a SLR and NMA of pharmacotherapy for overweight and obesity. Published trials include STEP 8, which compared once-weekly subcutaneous semaglutide 2.4 mg with once-daily subcutaneous liraglutide 3.0 mg in conjunction with counselling for diet and physical activity for people with overweight or obesity.¹⁶ Semaglutide demonstrated a significantly greater weight CFB compared with liraglutide and participants receiving semaglutide were significantly more likely to achieve ≥10%, ≥15%, and ≥20% weight loss versus those receiving liraglutide at Week 68.¹⁶ In STEP 8, the weight CFB reported with semaglutide versus liraglutide was higher than that in the current NMA (–9.4% at Week 68 versus –4.5% at Week 52). Although reported at different timepoints, these data suggest that the NMA results validate (or even somewhat underestimate) the benefits of semaglutide in comparison with liraglutide that have been reported directly in the STEP 8 RCT.¹⁶ STEP 8 also showed that the rates of discontinuation due to AEs were lower with semaglutide (3.2%) than with liraglutide (12.6%), with the semaglutide discontinuation rate in line with that of the placebo group (3.5%).¹⁶ Additional data from recent reviews and/or meta-analyses show that semaglutide is effective in people with overweight or obesity with or without T2DM.^51–53 These findings support those of the present analysis in demonstrating the benefits of semaglutide 2.4 mg versus liraglutide.

Shi et al published the findings of a SLR and NMA of RCTs and pharmacotherapy for adults with overweight and obesity.⁵⁴ The analysis included data from trials that were inclusive of IBT (whereas they were excluded from the current NMAs) and had a mean follow-up of 24 weeks compared with 12 months in the current analysis. The main analyses by Shi et al included grouped treatment nodes for GLP-1 antagonists and sodium glucose cotransporter inhibitors and the results showed that phentermine/topiramate provided the most effective weight loss; however, it is important to consider this in light of the safety concerns associated with phentermine/topiramate that are reported earlier. The second most effective therapies were GLP-1 receptor agonists, including semaglutide, which was the most effective drug in its class (and versus all other comparators) when sensitivity analyses were conducted. Furthermore, although semaglutide and several other therapies were associated with an increased risk of adverse events, semaglutide showed substantially larger weight loss benefits than other therapies with a similar risk of adverse events.⁵⁴

Limitations of the current analysis include those that are inherent to indirect treatment comparison and NMA. The creation of evidence networks requires assumptions around study homogeneity with regard to multiple aspects, including study populations and outcome definitions. A heterogeneity assessment and subgroup analyses by glucose tolerance were performed in the current study; however, as noted previously, there was still heterogeneity between studies in factors such as participant age, sex, and BMI. Some of the older studies also had significant amounts of missing data, which limited the analyses that could be conducted and may have resulted in reporting bias. Nevertheless, in the populations for which data were available, semaglutide mainly showed clinical benefits versus the available comparators. Finally, as noted in the methods, trials of bariatric surgery were not included in the NMAs because these trials were highly selective in comparison with studies investigating pharmacological agents. Therefore, the current analysis was limited to pharmacological agents; however, there is potential for future exploration of weight loss indications where a pharmacological agent is used prior to bariatric surgery or following bariatric surgery if weight loss is insufficient.

Conclusion

In summary, high-quality, RCT-based evidence has shown that semaglutide 2.4 mg is an effective weight management therapy in conjunction with lifestyle interventions for overweight and obesity in people with and without T2DM. In NMA, semaglutide 2.4 mg was compared against active comparators, including orlistat and liraglutide, and demonstrated effective weight loss in the total population and nearly all subpopulations of glucose tolerance; this differentiates the present NMA from others conducted previously, which have focused on specific populations of glucose tolerance (eg, only people with T2DM) and shows that semaglutide is effective across different levels of glucose tolerance. The present study also uses longer term data than that of previous studies and excludes IBT to enable more focused comparison between pharmacological therapies. Current pharmacological therapies for obesity have low uptake rates due to limited efficacy but semaglutide represents a new and effective treatment that may address this unmet need.