STEP-TEENS Results in Detail: Numbers, Subgroups, and Time Course

At a glance

| Parameter | Detail | |-----------|--------| | N | 201 (semaglutide 134, placebo 67) | | Population | Adolescents aged 12-17, BMI at 95th percentile or above (or 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-week treatment period + 7-week follow-up (75 weeks total) | | Primary endpoint | Percent change in BMI from baseline at week 68 | | Key result | -16.1% BMI change (semaglutide) vs +0.6% (placebo); ETD -16.7 percentage points |

Primary Endpoint: BMI Change at Week 68

The STEP-TEENS trial used percent change in BMI from baseline at week 68 as its primary endpoint, a choice that reflects how adolescent growth patterns make raw weight loss less meaningful than in adult trials. The trial applied an intention-to-treat estimand using a mixed model for repeated measures (MMRM).

The semaglutide group achieved a mean BMI reduction of -16.1% compared to +0.6% in the placebo group. The estimated treatment difference (ETD) for BMI was -16.7 percentage points (95% CI: -20.3 to -13.2; p <0.001).

This matters because BMI is the regulatory-standard metric in pediatric obesity, and the magnitude of this treatment difference exceeded what was observed in adult STEP trials on a proportional basis.

Confirmatory Secondary Endpoint: Body Weight

The key confirmatory secondary endpoint was percent change in body weight at week 68:

| Outcome | Semaglutide 2.4 mg | Placebo | ETD (95% CI) | p-value | |---------|-------------------|---------|--------------|---------| | Body weight change (%) | -14.7% | +2.8% | -15.5 (-19.6 to -11.4) | <0.001 | | Absolute weight change (kg) | -15.3 kg | +3.0 kg |, |, |

The placebo group gained weight over 68 weeks, which is expected given that untreated adolescents with obesity typically continue gaining during growth. This makes the treatment difference functionally larger than the raw semaglutide number suggests: semaglutide reversed a trajectory that was heading in the opposite direction.

Categorical Response Thresholds

Categorical responder analyses reveal the distribution of treatment response more clearly than means alone:

| Weight Loss Threshold | Semaglutide (%) | Placebo (%) | Odds Ratio (95% CI) | |----------------------|-----------------|-------------|---------------------| | ≥5% body weight loss | 72.5% | 17.7% | 12.8 (5.9 to 27.6) | | ≥10% body weight loss | 61.8% | 8.1% | 18.9 (7.2 to 49.5) | | ≥15% body weight loss | 37.4% | 3.2% | 17.8 (4.2 to 76.1) | | ≥20% body weight loss | 22.9% | 0% |, |

Nearly three-quarters of semaglutide-treated adolescents crossed the clinically meaningful 5% threshold. One in five lost 20% or more of their body weight. No placebo participant reached that level.

Time-Course Pattern

Weight loss with semaglutide followed a characteristic GLP-1 agonist curve. The trial reported body weight change at multiple timepoints through 68 weeks of treatment and then through a 7-week off-treatment follow-up:

• Weeks 0-12: Rapid initial weight loss during dose escalation (approximately -5% to -7% by week 12)
• Weeks 12-40: Continued steep decline, the steepest phase of the curve
• Weeks 40-60: Gradual plateau approach, with weight loss decelerating but not fully stabilizing
• Week 68: Nadir reached for most participants at the end of the treatment period (-14.7% mean body weight)
• Week 75 (off-treatment): Partial weight regain observed; participants regained approximately 4-5 percentage points within just 7 weeks of stopping

This rebound pattern during the off-treatment follow-up is consistent with adult STEP data and with the known pharmacology of GLP-1 receptor agonists. It underscores that semaglutide suppresses the biological drivers of weight regain only while the drug remains active, a point the FDA label for Wegovy acknowledges by not specifying a treatment duration endpoint.

Waist Circumference and Cardiometabolic Markers

Beyond weight, the trial tracked cardiometabolic secondary endpoints:

| Marker | Semaglutide (change) | Placebo (change) | |--------|---------------------|------------------| | Waist circumference | -12.7 cm | +0.6 cm | | Systolic blood pressure | -2.7 mmHg | +0.5 mmHg | | Diastolic blood pressure | -1.1 mmHg | +0.6 mmHg | | Triglycerides | -22% | +8% | | HDL cholesterol | +4% | -1% | | HbA1c | -0.14% | +0.11% | | Fasting glucose | -2.6 mg/dL | +1.5 mg/dL |

The waist circumference reduction of nearly 13 cm indicates meaningful visceral fat loss. The lipid and glycemic improvements occurred even though most participants did not have diabetes at baseline, suggesting metabolic benefit that extends beyond glucose control.

Subgroup Consistency

The authors performed pre-specified subgroup analyses by sex, age band (12-14 vs 15-17), baseline BMI category, race, and Tanner stage. In all subgroups, the treatment effect favored semaglutide with overlapping confidence intervals, meaning no subgroup showed a statistically different response. The point estimates ranged from approximately -12% to -18% ETD across subgroups.

One notable observation: participants in the higher BMI strata (BMI ≥35 kg/m²) tended toward slightly larger absolute weight losses, though relative percentage losses were similar across BMI categories. Female participants and male participants responded similarly, which is worth noting given sex-based differences in adipose distribution during adolescence.

Mean vs. Median and Variability

The trial reported means with 95% confidence intervals rather than medians with interquartile ranges for its primary analysis. The standard deviation for body weight change was approximately 12-14 percentage points in the semaglutide group, indicating substantial individual variability.

This means some adolescents lost 25-30% of their body weight while others lost <5%. The categorical response data (above) confirms this spread: 27.5% of semaglutide-treated participants did not reach even the 5% weight loss threshold. The trial was not designed to identify predictors of non-response, leaving an open clinical question about which adolescents will benefit most.

Comparison to Adult STEP-1 Data

Placing STEP-TEENS results alongside the adult STEP-1 trial:

| Parameter | STEP-TEENS | STEP-1 (Adults) | |-----------|-----------|-----------------| | N (semaglutide arm) | 134 | 1,306 | | Mean weight loss | -14.7% | -14.9% | | ≥5% responders | 72.5% | 86.4% | | ≥10% responders | 61.8% | 69.1% | | ≥20% responders | 22.9% | 32.0% | | Duration | 68 weeks | 68 weeks |

Adolescents achieved nearly identical mean weight loss as adults but had slightly lower categorical response rates. This may reflect the biological challenge of growing bodies, adherence differences in adolescent populations, or the counteracting effect of continued linear growth. The Endocrine Society clinical practice guidelines note that weight management in teens requires accounting for expected growth velocity.

Handling of Missing Data and Estimands

The trial used two estimands:

1. Treatment policy estimand (primary): Included all randomized participants regardless of treatment discontinuation, modeled using MMRM. This is the intention-to-treat population.
2. Trial product estimand (supportive): Estimated the effect in participants who remained on treatment, using a pattern-mixture model for missing data. Results under this estimand showed slightly larger effects (-17.4% BMI reduction with semaglutide).

The trial completion rate was 95% (participants who completed the 68-week visit regardless of whether they remained on drug). The treatment discontinuation rate was 10.4% in the semaglutide group vs. 4.5% in placebo. Most discontinuations in the semaglutide arm were due to adverse events, primarily gastrointestinal symptoms.

Limitations the Authors Acknowledged

The published paper identifies several constraints:

• The 68-week treatment duration does not establish long-term safety or durability
• Sample size (N=201) limits power for subgroup analyses and rare adverse event detection
• The lifestyle intervention was standardized but adherence to diet and exercise components was not rigorously measured
• Participants with type 2 diabetes were excluded, limiting generalizability to that population
• Racial and ethnic diversity was moderate but not fully representative of the global adolescent obesity population (60% White, 22% Black)
• No bone density, linear growth, or pubertal development assessments were conducted beyond Tanner staging at baseline

Clinical Translation

The STEP-TEENS data led directly to the FDA expansion of Wegovy's indication to include adolescents aged 12 and older in December 2022. For clinicians, the key numbers to communicate to families: roughly 3 in 4 adolescents will lose at least 5% of their body weight, the average is around 15%, and stopping the medication leads to rapid partial regain. The weight regain data should inform discussions about treatment duration and long-term planning, given that no trial has yet established the optimal protocol for discontinuation in pediatric populations.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Weghuber D, Barrett T, Gies I, et al. Once-weekly semaglutide in adolescents with obesity. N Engl J Med. 2022;387(24):2245-2257. PubMed
2. Wilding JPH, Batterham RL, Calanna S, et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). N Engl J Med. 2021;384(11):989-1002. PubMed
3. FDA. Wegovy (semaglutide) prescribing information. Revised December 2022. FDA Label
4. Styne DM, Arslanian SA, Connor EL, et al. Pediatric obesity: assessment, treatment, and prevention. An Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2017;102(3):709-757. PubMed
5. 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