CagriSema in type 2 diabetes: read the HbA1c result first

REIMAGINE 2 is often part of a weight-loss conversation, but its primary comparison was glycaemic control. In the published study abstract, CagriSema 2.4/2.4 mg reduced HbA1c more than semaglutide 2.4 mg at week 68 under the efficacy estimand. The size, setting and meaning of that difference deserve to stay together.

The paper was published online June 7, 2026, and appeared in the August issue of The Lancet Diabetes & Endocrinology, DOI 10.1016/S2213-8587(26)00125-7. This review uses its complete PubMed abstract, not the paywalled full text or supplements. Novo Nordisk funded the study.

The population and the primary comparison

The randomized, double-blind, placebo- and active-controlled study ran in 30 countries. Its 2,713 participants had inadequately controlled type 2 diabetes while receiving metformin with or without an SGLT2 inhibitor, with HbA1c 7.0–10.5% and BMI at least 25. They were not a general medspa wellness population.

There were several dose and comparator arms. For the primary comparison, 603 participants were randomized to cagrilintide–semaglutide 2.4/2.4 mg and 605 to semaglutide 2.4 mg. Mean baseline HbA1c across the study was 8.2%. The primary endpoint was change in HbA1c at 68 weeks in the full analysis set; safety was assessed in participants who received at least one dose.

A percentage point is not a percent

Using the efficacy estimand, mean HbA1c changes were −1.91 percentage points with CagriSema and −1.75 with semaglutide. The estimated treatment difference was −0.16 percentage points, with a 95% confidence interval of −0.27 to −0.05 and p=0.0035. This is a statistically supported average difference in the specified population—not a 16% HbA1c improvement, not a promised individual response, and not a directly measured reduction in heart attacks or other long-term complications.

Novo's same-day source release separately reports efficacy-estimand relative weight changes of −14.2% versus −10.2% for these arms. Those figures come from the sponsor's presentation of the trial, while the primary HbA1c contrast above is directly reported in the journal abstract. The 4.0-percentage-point difference between rounded weight changes is descriptive arithmetic, not a full independent secondary-endpoint appraisal.

Count any adverse event separately from treatment-related harm

The abstract reports adverse events in 524 of 603 participants (86.9%) in the higher-dose CagriSema group and 491 of 605 (81.2%) in the semaglutide 2.4-mg group. Gastrointestinal disorders were the most common events in the active-treatment groups. These are any-adverse-event counts, not counts of proven drug-caused injury, serious events or discontinuations. Do not relabel them as such.

Across all randomized arms, 95.7% completed the study and 87.6% were on treatment at week 68. Those are whole-study figures, not arm-specific persistence rates. This abstract-based review cannot supply a complete serious-event profile or a comparative discontinuation analysis. The full paper and supplements are the next evidence step.

Why different CagriSema headlines are not interchangeable

REIMAGINE 2 asks a glycaemic question in people with type 2 diabetes receiving established background therapy. REDEFINE 4 asks a different comparative weight-loss question in obesity over 84 weeks. A favorable result in one does not cancel a missed endpoint in the other, and cross-trial percentage rankings ignore their different populations, comparators, durations and analysis choices.

CagriSema is described as investigational in the same-day sponsor record. A peer-reviewed study and a marketed approved product are different evidence stages. For an RN or practice owner, the useful distinction is between a measured glycaemic advantage, a secondary weight result, and outcomes that were not directly established—not a new indication inferred from the headline.

Evidence record

Event date: 2026-06-07. Source review: 2026-10-29. Editorial publication: 2026-10-29. Original source links below retain their publication dates. These dates are not interchangeable. This is targeted source synthesis, not an exhaustive archive or a qualified clinical/legal review.

Revision history

29 October 2026: original analysis added from the specific source passages identified in the evidence audit. Journal abstracts and sponsor reporting are labeled separately; full-paper and supplementary safety assessment remain outstanding. Report a proposed correction through Contact us; do not include patient data.