Published in Diabetes, Obesity and Metabolism in 2022, the STEP 1 trial extension followed participants for an additional year after the 68-week treatment period ended. This extension directly tested what happens when semaglutide is discontinued.

After stopping semaglutide, participants regained 11.6 percentage points of the 17.3% they had lost, leaving a net loss of only 5.6% at week 120 (one year after stopping). Two-thirds of the weight loss was reversed. Cardiometabolic improvements in blood pressure, lipids, and inflammatory markers also partially reversed.

This finding has major implications for long-term safety because it means semaglutide must be taken indefinitely to maintain weight loss. Lifetime use exposes patients to cumulative risk from side effects and requires permanent pharmaceutical dependence. The drug functions more like a chronic suppressive therapy than a cure, raising questions about whether decades of continuous exposure will reveal safety problems not apparent in 2-3 year trials.

Published in Diabetes, Obesity and Metabolism in 2022, an exploratory analysis of the STEP 1 trial examined body composition changes using dual-energy X-ray absorptiometry (DEXA) scans in a subset of participants.

Total lean body mass (muscle, bone, organ tissue) decreased by 9.7% from baseline in the semaglutide group. While the proportion of lean mass relative to total body weight actually increased (because fat was lost faster than muscle), the absolute loss of lean tissue is clinically concerning. Approximately 40% of total weight lost was lean mass rather than fat.

This muscle loss is particularly worrying for several reasons: muscle mass is protective against falls and fractures in aging; lower muscle mass reduces resting metabolic rate (potentially worsening weight regain if the drug is stopped); and sarcopenia (age-related muscle wasting) is already a major health concern in older adults. If semaglutide is taken for decades, cumulative muscle loss could produce significant functional impairment.

The FDA prescribing information for both Ozempic and Wegovy carries a black box warning, the most serious safety warning the FDA can require. The warning states that semaglutide causes dose-dependent and treatment-duration-dependent thyroid C-cell tumors (both adenomas and carcinomas) in rats and mice at clinically relevant drug exposures.

Semaglutide is contraindicated in patients with a personal or family history of medullary thyroid carcinoma (MTC) or in patients with Multiple Endocrine Neoplasia syndrome type 2 (MEN2). The human relevance of the rodent thyroid tumor findings has not been determined despite over a decade of GLP-1 receptor agonist class use.

While no increased signal of medullary thyroid cancer has been confirmed in human post-marketing surveillance, the C-cell tumor mechanism in rodents is biologically plausible in humans (GLP-1 receptors exist on human thyroid C-cells). The absence of a confirmed human signal may reflect the long latency period for thyroid cancer development and the relatively short duration of widespread use.

Analysis of pooled data from the STEP 1 through STEP 4 trials revealed that gallbladder-related disorders occurred in 2.6% of semaglutide-treated participants compared to 1.2% of those on placebo, representing more than a doubling of risk. Cholelithiasis (gallstones) specifically occurred in 1.6% versus 0.7%.

A meta-analysis published in JAMA Internal Medicine in 2022 confirmed the increased risk across the entire GLP-1 receptor agonist drug class, not just semaglutide. The risk was dose-dependent and duration-dependent, with higher rates observed in patients treated for more than 26 weeks.

Rapid weight loss from any cause increases gallstone risk because cholesterol supersaturation in bile increases as the liver metabolizes released fat. However, the rate with semaglutide exceeds what would be expected from weight loss alone, suggesting a possible direct drug effect on gallbladder motility (GLP-1 receptors are present in the gallbladder). Some patients required cholecystectomy (surgical gallbladder removal).

In September 2023, the FDA required an update to the Ozempic prescribing information to add a warning about ileus (intestinal obstruction). Approximately 20 cases had been reported to the FDA, including 2 deaths. This rare but serious complication occurs when the drug''s effect on slowing gut motility becomes extreme enough to cause a complete bowel blockage.

Gastroparesis (severely delayed stomach emptying) was reported at a rate of 7.2 per 1,000 person-years in injectable semaglutide users in pharmacoepidemiological studies. While most cases resolve with dose reduction or discontinuation, some patients developed persistent gastroparesis that continued after stopping the medication.

The FDA Adverse Event Reporting System (FAERS) database contained 187,757 total adverse event reports for GLP-1 receptor agonists, with 16,568 classified as gastrointestinal. While most GI events are mild nausea that resolves, the rare cases of ileus and persistent gastroparesis represent serious, potentially life-threatening complications that were not fully characterized in the clinical trial populations.

Published in eClinicalMedicine (The Lancet) in 2024, Danish researchers conducted a randomized, double-blind, placebo-controlled phase 2 trial specifically in adults with increased fracture risk to assess semaglutide''s effects on bone health.

Over 52 weeks, semaglutide reduced hip bone mineral density (BMD) by 2.6% and lumbar spine BMD by 2.1% compared to placebo. Greater weight loss correlated with greater hip BMD decline. Bone resorption markers increased in the semaglutide group, indicating accelerated bone breakdown. The rate of bone loss exceeded typical annual postmenopausal loss of 1-2%.

This finding is particularly concerning for long-term use because bone density loss is cumulative and largely irreversible. If semaglutide is taken for 5-10 years, the cumulative bone loss could significantly increase fracture risk, especially in postmenopausal women and older adults who are already losing bone. The study highlights a potential long-term safety concern that would not be apparent in the 1-2 year clinical trials.

A meta-analysis published in Endocrinologia, Diabetes y Nutricion in 2024 examined pancreatitis risk across 21 randomized controlled trials involving 34,721 participants taking various semaglutide regimens. The overall odds ratio was 0.7 (95% CI 0.5-1.2), not reaching statistical significance.

However, in the STEP weight management trials specifically, pancreatitis occurred in 3 of 1,306 semaglutide participants (0.23%) compared to 0 of 655 on placebo. All clinical trials systematically excluded patients with a history of pancreatitis, limiting the ability to assess risk in vulnerable patients who may be most susceptible. Fatal cases of pancreatitis have been reported in post-marketing surveillance.

The concern about pancreatitis with GLP-1 drugs has persisted since the early days of this drug class. While the absolute risk appears low in trial populations, the exclusion of at-risk patients from trials means the real-world risk in the broader prescribing population may be higher. The mechanism involves GLP-1-mediated stimulation of pancreatic exocrine secretion.

Published in JAMA Network Open in 2024, Swiss researchers conducted a disproportionality analysis using the WHO''s VigiBase pharmacovigilance database, which collects adverse drug reaction reports from over 130 countries.

The analysis found disproportionate reporting of suicidal ideation associated with semaglutide and liraglutide compared to other drugs in the database. While disproportionality signals do not prove causation (reporting biases, media attention, and the underlying health conditions of users all contribute to reporting patterns), the authors concluded that continued pharmacovigilance is warranted.

This finding exists in tension with the EMA''s April 2024 conclusion of "no causal link established" and the Nature Medicine study showing lower suicidal ideation with semaglutide versus other weight loss drugs. The discrepancy highlights the uncertainty inherent in pharmacovigilance and the fact that rare psychiatric effects may take years of widespread use to characterize definitively.