Article Text
PDF
Abstract
Semaglutide (▼Ozempic solution for injection, ▼Rybelsus tablets—Novo Nordisk) was initially granted market authorisation for the treatment of type 2 diabetes as an adjunct to diet and exercise. In 2021 and 2022, regulatory agencies in the USA and Europe licensed semaglutide (▼Wegovy solution for injection—Novo Nordisk) for the treatment of individuals who are obese, or overweight and who have at least one weight-related comorbidity. Manufacturer-sponsored randomised controlled trials have shown a loss of almost 12% of body weight over a 68-week period, however, once the medication is stopped people regain most of their pretreatment weight. Gastrointestinal adverse events occur commonly with semaglutide, and pancreatitis, diabetic retinopathy and severe allergic reactions have also been reported. Extensive hype in social and general media has resulted in increased demand for semaglutide leading to supply problems across the various licensed products including those used for treatment of diabetes. In the UK, the National Institute for Health and Care Excellence has recommended semaglutide as an option for weight management for a maximum treatment duration of 2 years. Further studies are underway to assess the effect of semaglutide on longer-term health benefits.
- Drug-Related Side Effects and Adverse Reactions
- Obesity
Statistics from Altmetric.com
Key messages
Semaglutide has been licensed for the treatment of adults with a body mass index (BMI) of 30 kg/m2, or a BMI of 27 kg/m2 and at least one weight-related health problem.
In placebo-controlled clinical trials that assessed the effect on weight loss, participants who received semaglutide over a 68-week period achieved a loss of almost 12% of body weight.
Weight regain is common once semaglutide is stopped.
Most participants in the trials were white and female.
Very common adverse effects include headache, vomiting, diarrhoea, nausea and constipation. Common adverse events include injection site reactions, hair loss, dyspepsia, flatulence and dizziness. Uncommon adverse effects include acute pancreatitis and increased heart rate.
Background
Obesity, defined as excessive weight presenting health risks because of the high proportion of body fat, is a well-recognised growing global problem.1 Based on the WHO classification, adults with a body mass index (BMI)≥30 kg/m2 are defined as obese. Across Europe, the prevalence of obesity in adults has increased from 11% in 2000 to 17% in 2018 and in the UK it is estimated that 28% of adults are obese.2 Obesity has been linked to numerous diseases, including hypertension, diabetes, cardiovascular diseases and some forms of cancer.1
There is an ongoing debate about whether to classify obesity as a disease. The World Obesity Federation (WOF), which receives funding from various companies including Pfizer, Lilly, Medtronic, Johnson & Johnson, Rhythm Pharmaceutical, Vivus, IFA, Boehringer Ingelheim, and Novo Nordisk (the latter having made a funding commitment to WOF of £900k in 2023) published a consensus statement in one of its journals favouring the definition of obesity as a disease.3 4 The Federation presents an epidemiological model for obesity as a chronic progressive relapsing disease process with food as the main environmental agent for obesity, with a decline in the level of physical activity being second and genetic factors contributing to the host response to the environmental agents. In an article in The BMJ in which he argues against classifying obesity as a disease, Richard Pile, a general practitioner specialising in cardiovascular and lifestyle medicine, counters that ‘labelling obesity as a disease risks reducing autonomy, disempowering and robbing people of the intrinsic motivation that is such an important enabler of change. It encourages fatalism, promoting the fallacy that genetics are destiny…Making obesity a disease may not benefit patients, but it will benefit healthcare providers and the pharmaceutical industry when health insurance and clinical guidelines promote treatment with drugs and surgery.’5
Regardless of the position taken on the debate, we need to recognise the commercial determinants of obesity and the need to address those causes. At a societal level, research has identified the marketing and consumption of ultraprocessed foods and sugar sweetened beverages as the main causes of obesity.6 7 Both are heavily promoted to the public, especially in low socioeconomic areas.8 WHO has recently called for taxation of sugar sweetened beverages to reduce consumption.9 Over 50 countries have introduced such taxes, including the UK. The UK’s soft drinks levy has two levels based on sugar content (≥5 g/100 mL and ≥8 g/100 mL), and has led to reformulated products containing less sugar.10 11 Additional policy initiatives are needed to restrict population exposure to advertising of obesogenic products, to increase access to and affordability of healthier foods, and to support environmental changes that encourage and facilitate physical activity.
For individuals living with obesity, the primary strategy for achieving and maintaining weight loss is a comprehensive approach that includes, in particular, nutritionally balanced dietary measures, increased physical activity and personalised support. Bariatric surgery is typically reserved for adults with a BMI ≥40 kg/m2 or a BMI between 35 and 39.9 kg/m2 and a significant health problem that could be improved with weight loss.12 Up until now, the role of medications in the management of obesity has been minimal with nearly all the drugs approved for weight loss having been later removed from the market because of safety concerns.13
Regulatory standards for weight loss drugs
European Medicines Agency (EMA) guidelines for drugs for weight reduction require pivotal trials to demonstrate a statistically significant, placebo-corrected weight loss of at least 5% of baseline weight after 12 months of treatment.14 Reduction of body weight is the recommended primary efficacy endpoint. Sponsors are asked to document weight change in all randomised patients, regardless of treatment adherence. The EMA recommends classifying dropouts as non-responders, as obesity trials often have had high drop-out rates. The US Food and Drug Administration (FDA) recommends a joint primary outcome measure over 1 year of the mean difference in percentage body weight lost between drug and placebo, and the difference between the drug and placebo populations in the proportion of subjects losing at least 5% of body weight.15
What is semaglutide?
Semaglutide is an injectable glucagon-like peptide-1 (GLP-1) receptor agonist marketed by Novo Nordisk (see table 1). Activation of brain GLP-1 receptors potentiates glucose-dependent secretion of insulin, suppresses glucagon secretion, slows gastric emptying and promotes satiety.16 The Medicines and Healthcare products Regulatory Agency (MHRA) approved semaglutide for weight loss and weight maintenance as an adjunct to a reduced-calorie diet and increased physical activity in adults who have an initial BMI≥30 kg/m2 or BMI≥27 kg/m2 and <30 kg/m2 and at least one weight-related health problem.17 The drug is also licensed for adolescents aged ≥12 years with obesity and bodyweight >60 kg. The FDA and the EMA have adopted similar criteria.
Licensed formulations of semaglutide18 45 49
For weight management, patients self-inject semaglutide on a weekly basis, starting at a dose of 0.25 mg/week, gradually increasing over a 16-week period to a maintenance dose of 2.4 mg/week.18 If a patient has not lost ≥5% of their original bodyweight after 6 months, treatment should be reviewed.
Clinical trial results: weight loss in adults without diabetes
The MHRA’s assessment of semaglutide for obese and overweight patients without diabetes included three double-blind, randomised, placebo-controlled trials that formed part of the semaglutide treatment effect in people with obesity (STEP) series of studies (STEP 1, STEP 3, STEP 4) mostly in white women who were highly motivated to lose weight, all of whom received lifestyle and dietary counselling or an intensive support programme.17 19–21 The MHRA’s assessment also included the STEP 2 study that involved patients with diabetes and used two different doses of semaglutide, but we have excluded it from this review.17 22
The STEP trials had a similar design and all were 68-week studies that involved adults who were obese or overweight with at least one weight-related comorbidity.17 All participants had to have tried at least once to lose weight with lifestyle intervention alone. In addition to semaglutide or placebo, the studies involved a reduced calorie diet, counselling and increased physical activity (see table 2). The primary outcome of all the studies was percentage change in bodyweight from baseline. STEP 1 and STEP 3 had a coprimary outcome of the number of people with ≥5% reduction in weight from baseline at week 68.20 21 Studies included an additional 7-week extension period of follow-up after treatment discontinuation during which participants did not receive semaglutide or placebo or lifestyle intervention.17 Details of the studies are summarised in table 2.
Double blind randomised placebo-controlled trials of semaglutide 2.4 mg weekly in adult patients with obesity without type 2 diabetes19–21 25
In STEP 1 (1961 participants) and STEP 3 (611 participants), mean weight loss up to week 68 was −15.3 kg (−14.9%) with semaglutide vs −2.6 kg (−2.4%) for placebo and −16.8 kg (−16.0%) with semaglutide vs −6.2 kg (−5.7%) for placebo, respectively.20 21 The STEP 3 trial had a more intense diet and exercise intervention, resulting in greater weight loss in both drug and placebo groups and a smaller difference between drug and placebo compared with STEP 1 (−10.3% vs −12.4%). Patients enrolled in these trials had an average BMI of 38 kg/m2 at baseline. In both trials, there were no clinically meaningful changes in quality of life.23
The STEP 1 trial extension followed 327 participants for a year following cessation of treatment.24 Patients who had taken semaglutide regained 11.5 kg (11.6%) or 75% of their prior weight loss, compared with 2.3 kg (1.9%) or 85% of their prior weight loss, in the placebo group. Overall, after 68 weeks of treatment and 52 weeks without treatment the mean net change in weight was −5.6% in the semaglutide group and −0.1% in the placebo group.
In the STEP 4 double-blind randomised placebo-controlled trial, 902 participants were titrated to a semaglutide dose of 2.4 mg per week by week 20 and then 803 were randomised to either continue with semaglutide or switch to placebo for 48 weeks.19 All participants received a lifestyle intervention from week 0 to week 68, including monthly counselling by qualified healthcare professionals. At 68 weeks, there was a statistically significant difference in weight loss: −7.1 kg (−7.9%) in the semaglutide group versus a weight gain of 6.1 kg (6.9%) in those switched to placebo. Secondary outcomes included physical functioning assessed using the Short Form 36 (SF36; Version 2 Health Survey, Acute Version), which has a range from 19.03 to 57.60 for the physical functioning domain. Results favoured the semaglutide group, but average drug minus placebo differences were below the threshold for clinically meaningful change.
Longer-term treatment
The longest clinical trial to date, the STEP 5 study, followed 304 patients for a 2-year period.25 Two years of semaglutide treatment resulted in similar weight loss compared with 1 year of treatment (see table 2).
Comparison with liraglutide
In the STEP 8 open-label study, 338 participants without diabetes with a BMI >30 kg/m2 or BMI 27–30 kg/m2 with at least one weight-related comorbidity were randomised to weekly semaglutide injections or placebo, or daily liraglutide injections or placebo.26 Participants also received counselling every 4–6 weeks, were advised to adhere to a 500 kcal/day deficit diet and to exercise for ≥150 min/week. After 68 weeks, weight loss was −16% with semaglutide, −6% with liraglutide and −2% in the pooled placebo groups.
Cardiovascular outcomes
Importantly, none of the studies in table 2 assessed clinically important cardiovascular events as part of the primary or secondary outcomes. The SELECT placebo-controlled study has assessed the effect of semaglutide on the incidence of major cardiovascular outcomes in 17 600 adults aged ≥45 years.27 The full results have not yet been published in a peer reviewed journal. A press release issued by the company stated that in people treated with semaglutide for up to 5 years there was a relative risk reduction of 20% in a composite outcome of major adverse cardiovascular events (cardiovascular death, non-fatal myocardial infarction or non-fatal stroke).28 Based on the total of 1270 events cited in the press release and a 20% relative risk reduction, around 565 events would have occurred in the treatment arm (6.4%) and around 705 in the control arm (8.0%) for an absolute risk reduction of 1.6% and a number needed to treat of 63 to prevent one major adverse cardiovascular event.
Issues to consider
There was heterogeneity in the outcomes of the three weight-loss trials possibly due to differences in diet and exercise intervention. In STEP 3, participants received a very low-calorie diet of 1000–1200 kcal/day for the first 8 weeks followed by 1200–1800 kcal/day for the remainder of the study.20 They were also prescribed physical activity of 100 min per week that was slowly increased to reach 200 min per week. The other studies also included a reduced calorie diet and increased physical activity, but only a 500 kcal deficit per day and 150 min of physical activity per week. There was no information about adherence to non-drug interventions in any of the trials. A subsequent trial, reported after semaglutide was approved for weight-loss, followed patients without diabetes over a 2-year period.25 The mean change in body weight from baseline to week 104 was −15.2% in the semaglutide group (n=152) vs −2.6% with placebo (n=152). Adjunctive measures were similar to those offered to patients in other trials. All the primary trials were funded by Novo Nordisk.
Average differences in self-reported physical functioning (SF-36 physical functioning score) between patients on semaglutide and placebo were generally lower than thresholds set by trialists a priori as representing a clinically meaningful difference (table 2). The cited thresholds in STEP 1 and STEP 4 were 3.7 points and 4.3 points, respectively.19 21 The largest reported difference of 2.5 points (95% CI 1.6 to 3.3) was documented in STEP 4, which compared discontinuation of semaglutide (ie, patients switched to placebo) after 20 weeks with ongoing use of semaglutide.19 This trial also found that 82% of patients experienced weight regain during 48 weeks of follow-up postdiscontinuation, as compared with 15% of patients who continued on semaglutide.19 29
The European Public Assessment Report points out a number of uncertainties in the clinical trial results.30 Most trials included mostly women and subgroup analyses showed greater effectiveness in women than men, suggesting a possible overestimate of joint benefit. Effectiveness was also lower in those weighing over 115 kg at baseline, although it remained clinically relevant. Few older people participated in the STEP 1, STEP 3 and STEP 4 trials with a total of 233 (7%) aged 65–74 years and 18 (0.5%) aged ≥75 years.23 Obesity is prevalent in older adults and the limited trial data raise concerns about clinical uncertainty and potential greater vulnerability to serious gastrointestinal adverse effects.31
Clinical trial results: adverse events
Combining the results of STEP 1, STEP 3 and STEP4,19–21 (table 2) with results from a phase II trial,32 which compared semaglutide against liraglutide, patients given semaglutide were twice as likely as those on placebo to discontinue treatment due to adverse events (RR 2.19, 95% CI 1.36 to 3.55; p=0.001).33 More patients treated with semaglutide experienced acute pancreatitis than those treated with placebo (4 [0.2%] vs 1 [<0.1%]).23
The most frequent adverse events for semaglutide compared with placebo were gastrointestinal.17 These included nausea (38% vs 14%), diarrhoea (27% vs 14%), constipation (22% vs 10%), vomiting (22% vs 6%), abdominal pain (8% vs 4%), dyspepsia (8% vs 3%). Headache (13% vs 9%) and dizziness (7% vs 3%) also occurred more often on semaglutide than placebo.23 As described in table 2, although the number of patients experiencing any adverse event did not differ substantially between semaglutide and placebo, adverse event rates over the course of the trials were higher on semaglutide, with differences in adverse event rate per 100 patient years on drug versus placebo ranging from 53 to 260 (table 2). The EMA concluded that there was no difference in death rate on semaglutide versus placebo.23 The FDA and MHRA reported that there were too few deaths to identify meaningful trends.17 34
In July 2023, the EMA announced that it was reviewing data on the risk of thoughts of suicide and self-harm associated with GLP-1 medicines.35 Authorities were analysing about 150 reports. The EMA had previously reported suicidality as an area of uncertainty based on FDA information.
The EMA’s risk management plan highlights harmful effects for which there is suggestive evidence, but uncertainty about causation: diabetic retinopathy complications, pancreatic cancer and medullary thyroid cancer. Effects in pregnancy and lactation and in patients with severe hepatic impairment also remain unknown.30 The EMA also states that serious potential risks, such as acute pancreatitis, neoplasms, diabetic retinopathy complications and suicidal behaviour are unconfirmed and have occurred only rarely in patients exposed to semaglutide.34
Meta-analyses/systematic reviews
There have been three meta-analyses/systematic reviews of semaglutide use for weight loss in patients without diabetes (table 3).33 36 37 All three reviews included a phase 2 study,32 the STEP 320 and STEP 1 studies.21 In addition, two of the reviews included a trial that assessed the effect of semaglutide on energy intake, appetite, control of eating and gastric emptying38 while the other review33 included the STEP 4 study.19 One review had one author with multiple conflicts of interest33 and another review had non-industry funding.37 All three reviews reported that semaglutide’s effects on weight loss ranged from −11.6% to −11.9% compared with placebo. Adverse events were not comprehensively meta-analysed. One review reported that the risk of developing gastrointestinal adverse events was 1.6 times more likely with semaglutide, the risk of discontinuation due to adverse events was 2.2 times greater with semaglutide and the risk for serious adverse events was 1.6 times more likely with semaglutide.33 Another review reported no difference in deaths due to adverse events (risk ratio 0.50).37 There are no Cochrane reviews either of semaglutide or GLP-1 receptor agonists for weight loss.
Meta-analyses/systematic reviews of trials of semaglutide in people with obesity without diabetes
National recommendations
The National Institute for Health and Care Excellence (NICE) has recommended semaglutide for a maximum 2-year period as an option for adults with at least one weight-related comorbidity and BMI of at least 35 kg/m2 or for people with a BMI of 30.0 kg/m2 to 34.9 kg/m2 who meet NICE’s criteria for referral to specialist weight management services.39 The 2-year maximum was chosen to coincide with the typical withdrawal of specialist services at that time. Semaglutide will only be available on the National Health Service (NHS) as part of a specialist weight management service with multidisciplinary input (such as a tier 3 weight management programme or tier 4 specialist obesity services including surgery service). If less than 5% of initial weight is lost in the first 6 months, stopping semaglutide is recommended.
The Scottish Medicines Consortium is expected to issue a recommendation about the use of semaglutide for the treatment of obesity in NHS Scotland in October 2023.40
Extensive promotion, a media fanfare and controversies
Many UK media commentators have been enthusiastic about semaglutide. As reported in the BMJ, BBC Science Focus magazine called it a ‘weight loss game changer’; The Times said its arrival ‘could trim benefits bill’ and The Economist’s cover story proclaimed: ‘Eat, inject, repeat—curing obesity, worldwide: The long-term effects must be carefully studied. But the excitement is justified’.41 The president of the European Association for the Study of Obesity (ASO) which received £3.4 million from Novo Nordisk between 2019 and 2021 did not declare his conflict-of-interest when he wrote a column about semaglutide for The Guardian.41 (The Guardian has since added a footnote about this omission.) The Guardian reported that Novo Nordisk has paid £21.7 million to UK health organisations and professionals,42 although it is not known how much of that total was directly related to the promotion of semaglutide for the treatment of obesity.
Novo Nordisk has recently had its membership of the Association of the British Pharmaceutical Industry suspended because of a free course on weight management offered on LinkedIn to healthcare professionals for another GLP-1 receptor agonist, liraglutide (Saxenda), also approved for the treatment of obesity. The event sponsorship was ruled a ‘disguised promotional campaign’ and amounted to a prohibited ‘inducement to prescribe, supply, administer and/or recommend’ the drug.43 Following the suspension, the Royal College of Physicians severed its relationship with Novo Nordisk. In the USA, the FDA is investigating a story on the television show 60 Minutes that is alleged to have been disguised advertising for semaglutide. CBS’s 60 Minutes had received payment from Novo Nordisk to air a 13 min promotional segment on Wegovy on 1 January 2023, which was presented as a news story.44
The UK ASO, which gave positive evidence about semaglutide to NICE, received a £100 000 donation from Novo Nordisk in 2021 as did the Royal College of Physicians, which did not declare its interests to NICE. The ASO said its interests were appropriately declared.42
Semaglutide is licensed for the treatment of type 2 diabetes under the brand name Ozempic.45 Demand for use of semaglutide as a weight loss drug has affected its availability for people with diabetes. In the USA, the medication was in short supply for a number of months until mid-March 2023.46 In Canada, semaglutide is significantly less expensive than in the USA, leading Americans to try and access the drug from Canada. Over the course of several months one physician with a licence in the Canadian province of Nova Scotia wrote 17 000 prescriptions for semaglutide which were then mailed to the USA by a Canadian internet pharmacy. (The physician’s license was later suspended.)47 In the UK, a national alert has been issued regarding the availability of the entire GLP-1 class of drugs largely due to its off-label use for the treatment of obesity.48
Conclusion
Obesity is a growing problem in the UK with the latest figures showing over 25% of men and close to 30% of women are obese. Being obese carries morbidity and mortality complications including hypertension, diabetes, cardiovascular diseases and some forms of cancer.
For individuals living with obesity, the primary strategy for achieving and maintaining weight loss is a comprehensive approach that includes, in particular, nutritionally balanced dietary measures, increased physical activity and personalised support.
Semaglutide, an injectable GLP-1 agonist originally licensed for glycaemic control in diabetes, is now also licensed for the treatment of obesity as an option for adults with a BMI≥30 kg/m2 or adults with a BMI≥27 kg/m2 to <30 kg/m2 and at least one weight-related comorbidity. However, in England, NICE has recommended semaglutide as an option only for people with a BMI of ≥30 kg/m2 as part of a specialist weight management service with multidisciplinary input and for a maximum of 2 years.
In randomised controlled trials, over a 68-week course of treatment, patients who received semaglutide lost almost 12% of body weight compared with placebo but the majority of participants were female and white and semaglutide might not be as effective in males. In one trial, 1 year after treatment was stopped participants regained two-thirds of the weight they had lost.
The effect of semaglutide on cardiovascular outcomes associated with obesity will be known when the final results of the SELECT study are published.
Up to 7% of participants withdrew from trials due to adverse events associated with semaglutide. Gastrointestinal adverse effects such as nausea, vomiting, diarrhoea and constipation were common, with nausea and diarrhoea occurring in up to 38% and over 26% of participants, respectively. Cases of pancreatitis have been reported. There is no evidence of increased deaths due to use of semaglutide compared with placebo.
Although use of semaglutide can result in substantial weight loss, the impact of weight regain on stopping treatment is of concern; continuous use is needed to maintain that loss. We believe that more data on the safety, efficacy and cost-effectiveness of semaglutide, including longer-term trials, evidence of its effect in currently under-represented populations as well as its effect on long-term weight loss and clinical complications of obesity, are needed before semaglutide can be recommended for routine use for treatment of obesity. It is important that society addresses the complex factors that contribute to the development of obesity rather than relying on medication as a damage limitation exercise.
Information for patients
Weekly subcutaneous injections of semaglutide in conjunction with lifestyle and dietary support can help obese or overweight adults lose 12% of their body weight over a 68-week period but people regain the majority of the weight they lost after the medication is stopped. Gastrointestinal adverse events such as nausea, vomiting, diarrhoea and abdominal pain are very common and serious complications including acute pancreatitis and anaphylactic reactions have been reported. Further research will assess whether semaglutide will help in improving cardiovascular health and other health problems linked to obesity.
References
Footnotes
Competing interests None declared. Refer to the online supplementary files to view the ICMJE form(s).
Provenance and peer review Commissioned; externally peer reviewed.