Early weight loss predicts six-month liraglutide response

One study followed 41 adolescents with obesity across 14 hospitals and clinics in Taiwan from February 2024 to March 2025. It measured body mass index (BMI), a way to compare weight to height, and BMI standard deviation score, a way to compare a young person's BMI with others of the same age and sex. Assessments happened at baseline, week 13, and week 26 ². A second study looked back at records for 112 adults with obesity and no diabetes who began liraglutide between 2019 and 2021 and had complete body-weight observations through six months ¹. The third study used 24 male C57BL/6J mice, a common laboratory strain, on high-fat or low-fat diets. A high-fat subgroup received daily liraglutide for 14 days followed by a 7-day break ³.

In the adult record review, losing more weight in the first month was linked to losing more by six months ¹. Among the 112 adults, median total body weight loss rose from 3.88% at one month to 13.27% at six months ¹. The relationship between month-1 and month-6 percentage total weight loss was strong. The correlation measure rho was 0.720, and the p-value was below 0.001, meaning the result is unlikely to be due to chance ¹. In the adjusted analysis, which accounts for other factors, each additional percentage point of month-1 weight loss corresponded to 1.85 percentage points more total weight loss at six months (95% confidence interval 1.37 to 2.32, a range likely to contain the true value; p<0.001) ¹.

At month 6, 69 patients (61.6%) had lost at least 10% of baseline weight, and 36 patients (32.1%) had lost at least 15% ¹. For each additional percentage point lost in month one, the adjusted odds of reaching at least 10% loss were 8.89 times higher (95% confidence interval 3.38 to 23.36) ¹. The adjusted odds of reaching at least 15% loss were 1.66 times higher (95% confidence interval 1.20 to 2.29) ¹. An exploratory analysis produced an area under the curve of 0.910 for predicting at least 10% loss. Area under the curve is a measure of how well a test separates people who reach a benchmark from those who do not, and 0.910 is high ¹. The authors state that the internally identified threshold should be validated in independent prospective cohorts before clinical use ¹.

The six-month median loss of 13.27% sits above the 5–10% average weight loss reported in clinical trials and real-world data for liraglutide at 3.0 mg daily ⁴. This does not establish that real-world response is universally larger. The retrospective design can select for people who stayed on therapy and had complete follow-up data ¹.

Adolescents in the Taiwan study lost weight and body mass index points, and the lower average dose was well tolerated ². At week 13, mean body weight fell 2.7 kg, plus or minus 3.8 kg (p=0.0003). BMI fell 1.3, plus or minus 1.4 kg/m² (p<0.0001), and BMI standard deviation score fell 0.2, plus or minus 0.2 (p<0.0001) ². At week 26, mean body weight fell 4.4 kg, plus or minus 5.7 kg (p=0.0015). BMI fell 1.9, plus or minus 2.1 kg/m² (p=0.0004), and BMI standard deviation score fell 0.4, plus or minus 0.3 (p<0.0001) ². The mean daily dose was 1.4 mg, plus or minus 0.5 mg. The mean maximum dose was 1.8 mg, plus or minus 0.7 mg, and only 14.4% escalated to 3.0 mg daily ².

Adverse events occurred in 19.5% of participants (8 of 41) and were all mild. Adverse drug reactions occurred in 9.8% (4 of 41) ². Gastrointestinal events were most frequent, followed by injection site events, and no participant discontinued because of intolerance ².

Earlier adolescent trials reported around a 4.3% BMI reduction over 56 weeks with liraglutide 3.0 mg, compared with lifestyle changes alone ⁷. The Taiwan cohort used a lower mean dose and a shorter timeline, so the absolute BMI change of 1.9 kg/m² over 26 weeks is not a direct comparison. It adds a real-world finding that lower doses were tolerated in this sample, but the evidence does not establish that lower doses produce equivalent BMI reduction ².

In mice, liraglutide was linked to weight loss and changes in gut bacteria, and most bacterial changes reversed after a break ³. The mice lost significant weight by day 4, and the loss persisted during treatment and partially reversed after the 7-day break ³. The bacterial community structure shifted during treatment.

Nine ASVs, short genetic tags for bacterial species, related to Lactobacillus gasseri, L. paragasseri, L. johnsonii, and Leptogranulimonas caecicola significantly increased during liraglutide treatment and declined after the break ³. Twelve ASVs associated with bacteria that ferment protein and carbohydrates, including Romboutsia, Faecalicatena, and Oscillibacter, decreased ³. Most changes returned to baseline after the 7-day break ³. This was a mouse study, so it does not establish a human gut effect or a weight-loss mechanism ³.

The adult record review fits with earlier research showing that early response can help identify who is likely to respond, but it does not replace the weight-loss expectations from prior studies for liraglutide. A network meta-analysis, a study that compares many trials through a common statistical model, ranked liraglutide effective but less potent than newer agents such as semaglutide or tirzepatide ⁹. Clinical trials and real-world data show liraglutide 3.0 mg leads to average weight loss of 5–10% over 26–52 weeks ⁴. The adult chart review's 13.27% median loss at six months may reflect a real-world population that remained on treatment, but retrospective data cannot control for all selection factors ¹.

The adolescent findings add to an evidence base that includes clinical trials and a Swiss single-center analysis ⁷. The Taiwan cohort's safety results are consistent with the known gastrointestinal side effect profile, with no discontinuations from intolerance at lower mean doses ².

The evidence does not establish a validated early time point for treatment decisions in clinical practice. The adult study found a strong association between one-month and six-month weight loss, but the authors explicitly state that the internally identified threshold should be validated in independent prospective cohorts before clinical use ¹. The adolescent study does not answer whether lower doses are as effective as 3.0 mg over longer follow-up, because only 14.4% escalated to the full dose and the mean maximum dose was 1.8 mg ². The mouse gut findings do not clarify whether the bacterial changes cause weight loss or simply accompany it ³.