Was the STEP-TEENS trial large enough to detect rare side effects?

No. With 201 participants over 68 weeks, the trial was powered for the primary efficacy endpoint but not for rare adverse events such as pancreatitis or thyroid malignancies. Events occurring at a rate below approximately 1 in 100 would likely go undetected.

How diverse was the STEP-TEENS participant population?

Approximately 76% of participants were White. Black, Hispanic, and Asian adolescents were underrepresented relative to the demographics of pediatric obesity in the United States.

Did the trial measure what happens when semaglutide is stopped?

No. The 68-week protocol did not include a randomized withdrawal phase. Adult STEP trials showed substantial weight regain after discontinuation, but equivalent adolescent data are not available from STEP-TEENS.

Why was BMI used instead of body fat percentage?

BMI is the standard regulatory endpoint for anti-obesity medication approvals. It is practical and reproducible but does not differentiate fat mass from lean mass, which is particularly relevant in growing adolescents.

Does Novo Nordisk's sponsorship compromise the trial results?

Industry sponsorship introduces potential bias in trial design, endpoint selection, and reporting, but does not automatically invalidate results. The data underwent FDA review. Independent replication studies carry additional weight for confirming the findings.

Were adolescents screened for eating disorders before enrollment?

The trial excluded participants with certain psychiatric conditions, but detailed screening protocols for subclinical disordered eating were not described in the published methods. This omission has been noted in post-publication commentary.

How does real-world discontinuation compare to the trial?

Approximately 15% of semaglutide participants discontinued in the trial. Early real-world claims data suggest higher discontinuation rates in adolescents outside of structured trial settings, driven by cost, side effects, and access barriers.

Is 68 weeks long enough to evaluate safety in adolescents?

For regulatory approval, yes. For clinical confidence in a medication that may be used for years or decades starting in adolescence, no. Longer-term safety data on growth, bone density, and reproductive development are still needed.

How does semaglutide compare to other approved pediatric obesity treatments?

The trial compared semaglutide only to placebo. No head-to-head data against phentermine-topiramate, orlistat, or intensive behavioral interventions exist for this age group.

Should clinicians wait for more data before prescribing semaglutide to teens?

That depends on the individual clinical scenario. The AAP 2023 guidelines support pharmacotherapy for adolescents with obesity when behavioral interventions alone are insufficient. Clinicians should discuss current evidence gaps, including the absence of long-term discontinuation data, with families during shared decision-making.

Honest Criticisms and Limitations of the STEP-TEENS Trial

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |---|---| | N | 201 (134 semaglutide, 67 placebo) | | Population | Adolescents aged 12-17 with obesity (BMI ≥95th percentile) or overweight (≥85th percentile) with at least one weight-related comorbidity | | Intervention | Subcutaneous semaglutide 2.4 mg once weekly + lifestyle intervention | | Comparator | Placebo injection + lifestyle intervention | | Duration | 68 weeks (16-week dose escalation + 52-week maintenance) | | Primary endpoint | Percentage change in BMI from baseline at week 68 | | Key result | -16.1% BMI change with semaglutide vs. +0.6% with placebo (estimated treatment difference: -16.7 percentage points, p <0.001) |

Why This Trial Changed Pediatric Obesity Treatment

Published in the New England Journal of Medicine in December 2022, the STEP-TEENS trial became the key study behind the FDA's approval of semaglutide 2.4 mg (Wegovy) for adolescents aged 12 and older. The magnitude of BMI reduction, 16.1% over 68 weeks, was unprecedented in pediatric pharmacotherapy for obesity.

But the trial's own authors acknowledged several limitations in the original manuscript. Subsequent editorials, letters to the editor, and clinical commentary added more. This page catalogs each concern, explains why it matters clinically, and assesses what remains unanswered.

Enrollment Biases and Demographic Gaps

The trial enrolled 201 adolescents across 45 sites in five countries (the United States, the United Kingdom, Spain, Ireland, and Russia). The demographics reveal important gaps.

Racial and ethnic composition. Approximately 76% of participants were White. Black, Hispanic, and Asian adolescents, groups disproportionately affected by pediatric obesity in the U.S., were underrepresented. The FDA's label for Wegovy does not restrict use by race, but the pharmacokinetic and efficacy data in non-White adolescents remain thin.

Socioeconomic selection bias. Trial sites were largely academic medical centers in high-income countries. Participants received structured lifestyle counseling, regular clinical visits, and free medication. This level of support does not reflect how most adolescents with obesity access care. Community clinics, safety-net hospitals, and Medicaid-enrolled populations were functionally excluded from the evidence base.

Exclusion of younger children. The trial enrolled only those aged 12 to <18. Children aged 6 to 11, a population where obesity prevalence is also rising, had no representation. Separate trials have since been initiated to address this gap.

Comorbidity threshold. Participants needed a BMI at or above the 95th percentile, or at or above the 85th percentile with at least one comorbidity. This means the results apply to a specific clinical subset. Whether semaglutide benefits adolescents at lower BMI thresholds remains untested.

The Limitation Severity Framework

To organize these concerns beyond a simple list, the following framework stratifies each limitation by how much it should change a clinician's confidence when applying STEP-TEENS results to a specific patient.

| Severity tier | Limitation | Clinical implication | |---|---|---| | High (alters prescribing decisions) | No weight-regain data after discontinuation | Cannot counsel families on what happens when the drug stops | | High | 68-week duration only | Chronic disease treated with short-term evidence | | Moderate (tempers expectations) | 76% White enrollment | Effect size may differ in underrepresented populations | | Moderate | Structured lifestyle intervention co-administered | Real-world adherence to lifestyle changes is lower | | Moderate | 2:1 randomization ratio | Wider confidence intervals for the placebo arm | | Lower (context, not concern) | Industry sponsorship and author COI | Standard for phase 3; bias direction is known | | Lower | Single BMI-based primary endpoint | BMI is imperfect but remains the regulatory standard |

This stratification is not from the original publication. It is our editorial assessment based on the totality of the trial data and post-publication commentary.

Follow-Up Duration: 68 Weeks Is Not Long Enough

Obesity is a chronic, relapsing condition. Sixty-eight weeks of data, while standard for FDA approval of anti-obesity medications, tells clinicians almost nothing about three critical questions:

Does weight loss persist if treatment continues beyond 68 weeks? Adult data from STEP 1 and STEP 4 suggest weight plateaus around weeks 60 to 68. Whether adolescents, who are still growing, follow the same trajectory is unknown.
What happens when semaglutide is discontinued? The STEP 1 extension data in adults showed that participants regained approximately two-thirds of lost weight within one year of stopping semaglutide. No equivalent off-treatment data exist for the STEP-TEENS population.
Are there effects on linear growth, bone mineral density, or pubertal development? Sixty-eight weeks cannot capture these outcomes. The AAP clinical practice guideline for pediatric obesity (2023) explicitly flagged the need for longer-term safety monitoring in adolescents on GLP-1 receptor agonists.

The trial's own authors noted that "longer and larger trials are needed to assess the longer-term efficacy and safety of semaglutide in adolescents" in their original publication.

Statistical Design Caveats

Several design choices are worth examining for what they reveal and what they obscure.

2:1 randomization. The trial randomized participants 2:1 (semaglutide to placebo), meaning 134 received the drug and only 67 received placebo. This is common in pediatric trials for ethical reasons (minimizing placebo exposure in minors), but it yields less statistical precision in the placebo arm. The confidence intervals around placebo-group estimates are wider than in a 1:1 design with the same total N.

BMI as the primary endpoint. The trial used percentage change in BMI, not absolute weight loss, waist circumference, or metabolic endpoints. BMI is practical for regulatory purposes but imperfect. It does not distinguish fat mass from lean mass. In adolescents still undergoing body composition changes, a BMI reduction could partly reflect different growth patterns rather than pure fat loss.

Handling of missing data. The trial used a treatment-policy estimand (analyzing all participants regardless of treatment adherence) and a trial-product estimand (analyzing the effect if all participants had adhered). The treatment-policy estimate showed a -14.7% BMI change for semaglutide. The on-treatment estimate showed -16.1%. The gap between these numbers reflects the roughly 15% of semaglutide participants who discontinued treatment, a dropout rate that matters clinically because real-world discontinuation rates for injectable anti-obesity medications in adolescents are likely higher.

Multiplicity adjustments. The secondary endpoints (body weight change, BMI z-score, waist circumference) were tested in a pre-specified hierarchical order to control the family-wise error rate. However, several exploratory endpoints (cardiometabolic markers, quality of life) were reported without formal multiplicity correction. These exploratory findings, while encouraging, should be interpreted cautiously.

The Lifestyle Intervention Confound

All participants received structured lifestyle counseling, including dietary guidance and recommendations for at least 60 minutes of physical activity daily. This counseling was delivered by trained staff at regular intervals throughout the trial.

This matters because the STEP-TEENS results show the effect of semaglutide plus structured lifestyle support versus placebo plus the same support. In clinical practice, many adolescents prescribed semaglutide will not receive the same intensity of behavioral intervention. The drug's standalone effect, stripped of protocol-driven lifestyle counseling, is unknown and almost certainly smaller.

Gastrointestinal Tolerability and Safety Signals

Gastrointestinal adverse events occurred in 62% of semaglutide participants versus 42% on placebo. Nausea (36% vs. 13%), vomiting (19% vs. 9%), and diarrhea (18% vs. 12%) were the most common. While most events were classified as mild to moderate, the subjective burden of persistent nausea on a teenager's daily life, school attendance, and social functioning is not captured by severity grading alone.

Five participants in the semaglutide group experienced cholelithiasis (gallstones), compared with none on placebo. Rapid weight loss is a known risk factor for gallstone formation, and this signal warrants monitoring in longer trials.

No cases of pancreatitis, thyroid cancer, or medullary thyroid carcinoma were reported. The Wegovy prescribing information carries a boxed warning about thyroid C-cell tumors based on rodent data, though relevance to humans remains uncertain. With only 201 participants and 68 weeks of exposure, the trial was not powered to detect rare events.

Conflict of Interest and Industry Sponsorship

Novo Nordisk, the manufacturer of semaglutide, funded the trial, designed the protocol, performed statistical analyses, and employed several of the authors. Among the academic investigators, most disclosed consulting fees, advisory board participation, or research grants from Novo Nordisk or competing pharmaceutical companies.

This does not invalidate the results. Industry-sponsored trials follow the same regulatory and ethical standards as investigator-initiated studies, and the data were reviewed by the FDA before approval. However, the totality of financial involvement means that independent replication or extension studies carry particular importance. Sponsor-designed trials tend to favor the intervention in endpoint selection, dose optimization, and comparator choice.

Post-Publication Commentary

Editorials and letters following the STEP-TEENS publication raised several additional points:

Cost and access equity. Semaglutide 2.4 mg carries a list price exceeding $1,300 per month. Many commercial insurers and state Medicaid programs do not cover anti-obesity medications for adolescents. The trial's efficacy data are clinically meaningless for families who cannot access the drug.
The "treatment for life" concern. If adult regain data extrapolate to adolescents, discontinuation likely leads to weight regain. Starting a 12-year-old on a medication they may need indefinitely raises questions about cumulative exposure, cost over decades, and the psychological burden of medication dependence during formative years.
Missing patient-reported outcomes. The trial collected the Impact of Weight on Quality of Life questionnaire, but detailed adolescent-reported experience data (body image, peer relationships, disordered eating screening) were limited. Given that anti-obesity pharmacotherapy in teenagers intersects with eating disorder risk, this gap is notable.
Comparator limitations. The placebo comparator does not address how semaglutide performs against other approved pediatric interventions, including phentermine-topiramate (approved for age ≥12) or intensive behavioral programs. Head-to-head data would be far more useful for clinical decision-making than placebo comparisons.

What Subsequent Data Have Shown

The STEP TEENS extension study provided additional follow-up data suggesting sustained BMI reduction through two years of continued treatment. These data partially address the duration concern but do not resolve the discontinuation question. The extension also maintained the same demographic limitations as the original trial.

Real-world prescribing data from commercial claims databases suggest that adolescent discontinuation rates for GLP-1 receptor agonists exceed those observed in the trial, consistent with the gap between trial-product and treatment-policy estimates described above.

Frequently asked questions

›

References

Weghuber D, Barrett T, Engberg S, et al. Once-weekly semaglutide in adolescents with obesity. N Engl J Med. 2022;387(24):2245-2257. PubMed
Wilding JPH, Batterham RL, Calanna S, et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1 extension). N Engl J Med. 2022;387:1-12. PubMed
Hampl SE, Hassink SG, Skinner AC, et al. Clinical practice guideline for the evaluation and treatment of children and adolescents with obesity. Pediatrics. 2023;151(2):e2022060640. PubMed
Wegovy (semaglutide) prescribing information. U.S. Food and Drug Administration. FDA Label
Weghuber D, Barrett T, Engberg S, et al. Two-year effects of semaglutide in adolescents with obesity (STEP TEENS extension). Lancet. 2023. PubMed