Key points are not available for this paper at this time.
Both semaglutide, a GLP-1 receptor agonist, and tirzepatide, a dual GIP/GLP-1 receptor agonist, are changing the way we treat type 2 diabetes mellitus, obesity, as well as an increasing number of other cardiometabolic diseases. Major adverse cardiovascular events (MACE) were reduced by 20% in overweight/obese individuals with pre-existing cardiovascular disease without diabetes treated with semaglutide 2.4 mg compared to placebo (6.5% vs. 8.0%; absolute risk reduction ARR 1.5%, number needed to treat NNT ≥ 67 over ∼4 years)1. Semaglutide also improved renal outcomes, reducing major kidney disease events in patients with type 2 diabetes and chronic kidney disease by 24% hazard ratio (HR) 0.76; 95% confidence interval (CI) 0.66–0.88; P = 0.00032. Additionally, the ESSENCE study demonstrated that semaglutide 2.4 mg provided MASH resolution without worsening fibrosis in 62.9% of participants at week 72, resulting in FDA accelerated approval in August 2025 (awaiting confirmation results from the ongoing ESSENCE Part 2 Study due in 2029)3. There is emerging evidence supporting benefits in treating substance use disorder as well, through modulating central reward pathways. It is clear that there is real potential for this class of drugs; however, the increased use of these agents as off-label treatments for cosmetic weight loss, especially when used inappropriately under no clinical supervision, presents many serious concerns regarding both ethics and public health, which need to be addressed immediately. Unregulated compounded product use (as opposed to regulated pharmaceutical product use) has created a significant increase in the risk to patient safety. The FDA has repeatedly expressed concern about products that can legally circumvent regulatory review for their safety, efficacy, and quality. By 31 July 2025, more than 1150 adverse event reports had been filed by patients using compounded semaglutide and tirzepatide in combination – a number that undoubtedly represents a conservative estimate due to the lack of mandatory reporting requirements for all state-licensed compounders. By 31 December 2024, at least 17 of those cases were deaths reported to the FAERS (according to Senator Banks, quoting FAERS, NAM, August 2025)4,5. Dosing errors have proven to be especially worrisome, with patient misadministration due to unfamiliarity with multi-dose vials leading to five to twenty times the prescribed dose, causing adverse reactions ranging from nausea, vomiting, and dehydration to acute pancreatitis, gallstones, and syncope4. In an analysis published by the Brookings Institution in 2025, Chinese manufacturers representing about 20% of semaglutide compounds imported for compounding had never been inspected by the FDA, whereas companies responsible for more than 44% of imports had been found guilty of violations during inspection5. Unsupervised patients might also take these drugs when there are definitive contraindications present, such as a family history of medullary thyroid carcinoma or multiple endocrine neoplasia type 2, which can only be determined with the aid of a physician. Despite being well indicated for use, GLP-1RAs have a distinct adverse effect profile, which is commonly underestimated in society. Delayed gastric emptying has proven to be one of their most relevant perioperative risks: case reports of pulmonary aspiration in fasted individuals under anesthesia, while taking semaglutide, prompted the FDA to amend its labeling despite standard preoperative fasting and not knowing how long to stop using semaglutide prior to surgery. In terms of pancreatitis, we must remember the need to properly contextualize these data. A recently updated meta-analysis from 2024, which considered 21 placebo-controlled randomized trials of semaglutide use (totaling 34 721 patients), failed to show any increased risk of acute pancreatitis compared to placebo (odd ratio OR 0.7; 95% CI 0.5–1.2; I2 = 0%)6. An observational study in 2023 from the Journal of the American Medical Association, examining adults without diabetes and taking GLP-1RAs for obesity treatment, showed a significantly increased pancreatitis risk compared to patients taking naltrexone-bupropion (HR 9.09; 95% CI 1.25–66.00). It is important to note that obesity and type 2 diabetes are independent pancreatitis risks7. A 2025 PRISMA-based meta-analysis of 62 randomized clinical trials using all GLP-1 RA drugs (N = 66 232) identified a significant class effect with regards to an increased risk of acute pancreatitis (risk ratio 1.44, 95% CI 1.09–1.89, P = 0.009), which was reduced to statistical insignificance following adjustment for confounders based on co-medication8. Semaglutide’s link with nonarteritic anterior ischemic optic neuropathy (NAION) has been officially established. As of June 2025, the PRAC of the European Medicines Agency confirmed NAION to be a very rare side effect (up to 1 in 10 000 individuals) and updated the relevant European labeling, with the WHO publishing a similar global safety warning during the same period9. Patients experiencing sudden painless vision loss during treatment should stop using semaglutide and have themselves examined by a specialist. Data from the STEP 1 trial extension show that 1 year following discontinuation of semaglutide, patients regained a mean of 11.6 out of the 17.3 percentage points of weight lost while on the drug (about two-thirds of the previous weight loss). Improvements in cardiometabolic parameters partially revert back to, but do not fall below, their pretreatment levels10. Body-composition evidence from semaglutide clinical trials indicates that weight reduction is driven predominantly by fat-mass loss, while measured lean-mass changes are smaller and should be interpreted cautiously11. Therefore, weight regain after discontinuation should be discussed within a broader clinical framework that includes nutrition, resistance training, functional monitoring, and avoidance of unsupervised or cosmetic use. Regarding the psychological–social aspects, a 2023 FAERS analysis uncovered a four-fold higher proportional reporting rate (PRR) for semaglutide misuse compared to other GLP-1 receptor agonists (PRR ≈ 4.05 for drug abuse and withdrawal syndrome), with no statistical differences between semaglutide and the Schedule IV weight-loss compound phentermine-topiramate (phentermine being the sympathomimetic; topiramate the anticonvulsant)12. Social media promotion has been especially dangerous amongst the youth population: clinicians and advocacy groups have reported cases of adolescents or young adults with active or dormant anorexia nervosa acquiring semaglutide via online questionnaire manipulation, in some cases eliciting recurrent restrictive and obsessive-compulsive symptoms. A 2024 systematic review demonstrated the absence of any randomized trial evidence in patients with anorexia or bulimia nervosa and found that GLP-1RAs can worsen eating disorder pathology in predisposed subjects13. The portrayal of GLP-1 treatment as an easy fix rather than a form of comprehensive obesity treatment is even more dangerous. Metabolic bariatric surgery has proven to produce even more effective results when it comes to achieving weight loss, inducing complete remission of type 2 diabetes, and decreasing obesity-related mortality among individuals suffering from severe obesity (n = 174 772; HR of all-cause mortality decreased by 49.2%; median life expectancy (LE) increased by 6.1 years in comparison with the usual standard of care)14. For instance, the SURMOUNT-5 trial has confirmed that, in spite of significant improvements in the efficacy of pharmacological treatments (e.g., the 15-mg dose of tirzepatide showed 20.2% mean weight loss at 72 weeks), the mortality benefits of bariatric surgery, as well as its superiority to pharmacology with regard to achieving diabetes remission, are currently unmatched15. This distinction is clinically important because lean-mass changes during GLP-1RA therapy should not be interpreted in isolation; rather, they require assessment alongside measurement modality, muscle function, nutritional adequacy, and resistance-training support16. The perception of GLP-1RAs and GIP/GLP-1RAs as a way of quickly solving one’s appearance issues may cause patients to avoid undergoing comprehensive testing of their condition, which includes diet, mental health, and the use of surgery. Moreover, high discontinuation rates (within a year) due to high costs, side effects, and unrealistic expectations may promote medication dependence. Indeed, a comprehensive strategy is necessary. Enforcement actions against unauthorized formulations and off-label compounding are required. Education for clinicians should emphasize pre-prescription screening of eating disorder history using the SCOFF questionnaire and the EDE-Q questionnaire, both brief and suitable for routine use in a clinical setting. The SCOFF questionnaire may have greater sensitivity in specialized eating-disorder settings than in community settings; therefore, a positive screening result should prompt referral for specialist evaluation rather than serve as an absolute exclusion criterion. However, at present, no existing national guidelines require the use of either test prior to the prescription of GLP-1 receptor agonists13. Information directed to patients should emphasize indications, the newly labeled risk of NAION confirmed in June 202513, and risks of unauthorized self-medication, including unknown risks associated with unauthorized API procurement4,5. Longitudinal studies regarding the patterns of misuse, neurobiology behind appetite-suppressing and reward-modulating effects of GLP-1 receptor agonists, and withdrawal approaches, including gradual dose reduction, which requires further randomized evaluation, are highly desirable to facilitate evidence-based guidelines for physicians. GLP-1 receptor agonists are indeed a revolutionary breakthrough in the management of metabolic disorders; such improper use must not detract from their benefits.
Maqsood et al. (Mon,) studied this question.