Cagrilintide

At week 32, HbA1c had fallen by 2.2 percentage points with CagriSema, 1.8 with semaglutide alone and 0.9 with cagrilintide alone (journal averages). CagriSema was better than cagrilintide alone (difference 1.3 points) but not clearly different from semaglutide alone (difference 0.4 points, P=0.075). Body weight fell 15.6% with CagriSema, 5.1% with semaglutide and 8.1% with cagrilintide; the CagriSema difference from each single medicine was statistically significant. No level 2 or 3 low-blood-sugar episodes and no deaths were reported.

• A Phase 2 trial with 92 participants over 32 weeks, so it cannot show long-term benefit or rare side effects.
• It tests a combination, so any effect belongs to cagrilintide plus semaglutide together; the study cannot isolate what cagrilintide adds beyond semaglutide for blood sugar.
• The blood-sugar difference between CagriSema and semaglutide alone was not statistically clear (P=0.075 in the journal; P=0.2284 in the registry), and the two sources report slightly different figures.
• Only the PubMed abstract of the journal report was reviewed, not the full paper.
• All participants were adults with type 2 diabetes and BMI of 27 or higher treated at 17 US sites, so results may not apply to other groups.

• Blood-sugar (HbA1c) difference, CagriSema versus semaglutide alone, week 32: Lancet abstract reports -0.4 percentage points (95% CI -0.8 to 0.0; P=0.075); ClinicalTrials.gov posted results reports -0.30 (95% CI -0.79 to 0.19; P=0.2284). The two analyses give a different difference and P value; both are not statistically clear. The reason for the difference is not established from the sources reviewed.
• Average weight change, CagriSema versus semaglutide alone, week 32: Lancet abstract (model-based average) reports -15.6% vs -5.1%; ClinicalTrials.gov posted results (mean of observed values) reports -14.7% vs -5.0%. The registry averages are 0.9 and 0.1 percentage points smaller in magnitude. The reason for the difference is not established from the sources reviewed.

Average body weight fell 6.1% (0.3 mg), 6.9% (0.6 mg), 8.5% (1.2 mg), 9.5% (2.4 mg) and 10.8% (4.5 mg) with cagrilintide. It fell 8.5% with liraglutide 3.0 mg and 3.0% with placebo. These are registry means for people with data at week 26. Every cagrilintide dose lost more weight than placebo (treatment differences from -2.99 to -7.79 percentage points). Compared with the liraglutide comparator, only the 4.5 mg dose lost significantly more (difference -1.81 percentage points, 95% CI -3.46 to -0.16, P=0.03); the 0.3 and 0.6 mg doses lost significantly less, and the 1.2 and 2.4 mg doses did not differ. Model-based estimates of the share of people losing at least 10% of body weight ranged from 15.3% (0.3 mg) to 53.5% (4.5 mg), versus 39.4% on liraglutide and 10.4% on placebo. Blood sugar (HbA1c) changed little in people without diabetes.

• A Phase 2 dose-finding trial with 26 weeks of treatment: it does not show how long weight loss lasts or whether it lowers the risk of disease.
• Participants did not have diabetes, and 543 of 706 were White, so results may not apply to other groups.
• Participants and investigators were masked only for active versus placebo, not between the different active treatments, so cagrilintide-versus-liraglutide comparisons were not blinded.
• The six placebo arms were pooled for the main analyses.
• The shares of people losing 5% or 10% of body weight are model-based predictions, not simple counts.
• The registry's descriptive means and the journal's estimand-based figures differ slightly (for example 8.5% versus 9.0% for liraglutide).
• The trial was funded by the drug's developer, and only the journal abstract was reviewed, not the full paper.

• Average weight change with the liraglutide 3.0 mg comparator at week 26: ClinicalTrials.gov results (in-trial descriptive mean) reports -8.5%; The Lancet abstract (trial product estimand) reports -9.0%. 0.5 percentage points: the registry describes its means for people with data at week 26 while the abstract reports the estimand that assumes everyone stayed on treatment. The registry states that its descriptive figures use the in-trial period and its statistics use the treatment-adherent period; the abstract names the trial product estimand as the basis for its figures.

At week 68, average body weight had fallen by 20.4% with CagriSema versus 3.0% with placebo, a difference of 17.3 percentage points (confidence interval 16.6 to 18.1 points). CagriSema recipients were also more likely than placebo recipients to lose at least 5%, 20%, 25% and 30% of body weight. Digestive side effects affected 79.6% of the CagriSema group and 39.9% of the placebo group.

• Only the PubMed abstract of the journal report was reviewed, not the full paper, so results for the semaglutide-alone and cagrilintide-alone groups, discontinuations and serious events are not covered.
• The result applies to the combination; it cannot be used to attribute the weight loss to cagrilintide alone or semaglutide alone.
• The semaglutide-alone and cagrilintide-alone groups were small (302 people each) compared with the CagriSema group (2,108).
• The comparison reported here is with placebo; it is not a comparison with other approved weight-loss medicines.
• The trial excluded people with diabetes, so the results do not describe people with type 2 diabetes.
• The registry still lists the trial as active, not recruiting, with an extension period, so the record is not final and the extension findings are not part of this summary.
• The study was funded by Novo Nordisk.

• Number of participants: ClinicalTrials.gov reports 3,400 (estimated enrollment); NEJM abstract reports 3,417 randomised. The paper reports 17 more participants than the registry figure. The registry labels its 3,400 figure as an estimate, whereas the paper reports the number actually randomised. The registry record has not been updated to an actual count.