A Research Study to See How a New Weekly Insulin, Insulin Icodec When Given Along With Semaglutide Helps in Reducing the Blood Sugar Level in Patients With Type 2 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Research Study to See How a New Weekly Insulin, Insulin Icodec When Given Along With Semaglutide Helps in Reducing the Blood Sugar Level in Patients With Type 2 Diabetes (NCT05813912), a na phase 3 study. The primary outcome is Change in Glycated Haemoglobin (HbA1c).

2. Clinical Question

The question is therefore about the observed change over time in a selected group that proceeded to treatment intensification, not about a randomized comparison of combination therapy against another treatment. Selection into the intensification phase depended on meeting the study criteria after the run-in. That distinction matters: results for the intensification group cannot automatically be generalized to everyone who entered the study, or interpreted as the effect that would have occurred relative to an untreated or alternative-treatment group.

3. Trial Design

This phase 3 study is registered as a single-group, unmasked study with no randomized allocation. It enrolled 148 participants with type 2 diabetes and was listed as completed. Participants first received insulin icodec once weekly for a 26-week run-in period, during which the dose was optimized. The registry summary says participants whose blood sugar levels were not normal after this period could proceed to a further 26-week intensification period with weekly semaglutide added to weekly insulin icodec. Participants whose blood sugar levels were normal after the run-in continued into a 5-week follow-up period instead.

The primary endpoint compares HbA1c at baseline, defined as week 26, with HbA1c at week 52. Its analysis concerns the on-intensification phase. The registry describes the full analysis set as all participants allocated to that phase, with the number analyzed determined by data availability. It also states that observations after permanent discontinuation of either treatment were not included. The reported enrollment is 148, while 94 participants were at risk in the treatment-period safety group; these counts refer to different registry contexts and do not by themselves explain why participants did or did not enter that period.

The registry lists several other outcomes, including glucose profiles, fasting plasma glucose, hypoglycaemic episodes, body weight, and weekly insulin dose. It does not report a separate comparator arm. Although participants moved through different study periods, those periods do not provide a randomized control comparison: entry into intensification depended on the post-run-in criteria. The statistical plan, including any detailed model specification or sample-size justification, is not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: The trial included a 26-week run-in period during which all participants were switched from their previous basal insulin regimen to insulin icodec. This was followed by a 26-week treatment phase with insulin icodec combined with semaglutide.

Run-in Period (Week 0 to Week 26)

Reasons for not completing this period:

Treatment Period (Week 26 to Week 52)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS) included all participants allocated to the intensification phase. Overall number of participants analyzed = participants with available data.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Safety analysis set (SAS) included all participants who were exposed to investigational intervention (s).

6. Primary Endpoint

The registry lists the primary outcome as Change in Glycated Haemoglobin (HbA1c), measured over Baseline (Week 26), Week 52.

The primary endpoint is change in glycated haemoglobin, or HbA1c, expressed as a percentage. It measures the difference between the week 26 baseline and the week 52 value for participants in the intensification phase. The registry characterizes the endpoint as a continuous measure summarized by a mean and standard deviation.

General statistical context: a change endpoint focuses on how measurements differ within the defined study period, while a mean describes the average change across participants. The standard deviation describes how dispersed individual changes are around that average; it is not a measure of the precision of the mean. Precision would ordinarily be assessed using an uncertainty measure such as a confidence interval, but a confidence interval is not reported in the registry record.

HbA1c is used in this study as the primary measure of glycaemic control, consistent with the stated study objective. The timing is important: the comparison begins at week 26, after the insulin icodec run-in, rather than at initial study entry. Thus the endpoint concerns change during the intensification phase, not change over the entire study. Because numerical endpoint results are not reported in the registry record, the mean change, its variability, and the precision of the estimate cannot be described.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes the primary outcome as a mean change in HbA1c, with standard deviation, from baseline at week 26 to week 52. It specifies that the analysis uses observed data from planned visits during the intensification phase. For participants who permanently discontinued either study treatment, post-discontinuation observations were excluded. The full analysis set included participants allocated to the intensification phase, while the number analyzed was limited to those with available data.

What does a mean change describe?

General statistical context: the mean is the arithmetic average of participant-level changes. It can summarize the overall direction and size of change in a group, but it does not show whether individual participants had similar changes or widely differing ones. The standard deviation describes the spread of those changes around the mean. It does not, on its own, show how precisely the mean estimates the average change in a broader population.

How should precision be assessed?

General statistical context: a confidence interval is an estimate range that conveys uncertainty around a statistical estimate under the analysis assumptions. A narrower interval indicates greater precision than a wider interval, while the interval’s position relative to a null value can inform interpretation in a hypothesis test. The registry record does not report a confidence interval or p-value for the primary endpoint. Since it also does not report the numerical mean change or standard deviation, neither effect size nor precision can be evaluated from the posted record.

What is the analysis population?

The registry defines the full analysis set as participants allocated to the intensification phase. The number analyzed is described as participants with available data. These definitions matter because the primary endpoint is not described as an analysis of every person enrolled at study entry. Participants first completed a run-in, and only those meeting intensification criteria proceeded to the treatment period. The record does not give the exact number with available primary endpoint data.

The stated handling of discontinuation also affects interpretation. Observations after permanent discontinuation of either treatment were not included in the on-intensification outcome data. This is a specific description of which observations were used; it is not a complete account of how all missing values were handled. The statistical method for any other missing data, and any sensitivity analyses, are not reported in the registry record.

Why does the study design matter?

General statistical context: a single-group study can estimate change over time in its observed participants, but it cannot directly estimate a between-group treatment contrast. Without a concurrent control group, the observed change may reflect more than the study regimen. The design also makes it important to distinguish the initial enrolled population from the subgroup selected for intensification. The registry describes no random allocation and no masking.

The registry lists one analysis for a set of nine outcome measures. It does not describe whether or how multiplicity was addressed, how the primary endpoint model was specified, or whether adjustment was made for baseline covariates. No detailed statistical analysis plan is reported in the registry record. It is therefore not possible to establish from the record whether any particular test, model, or adjustment method was prespecified.

General statistical context: over-dispersion refers to variation greater than expected under a particular probability model and can matter when analyzing event counts. The registry includes hypoglycaemic episode outcomes as secondary measures, but it does not report their numerical results or the statistical methods used to analyze them. No conclusion about over-dispersion or a count-data model can be drawn from the registry record.

The record also describes a planning illustration based on a fraction of an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. General statistical context: planning calculations estimate how many participants might be needed under specified assumptions; they are not observed outcomes. In this case, the assumptions describe a hypothetical two-arm allocation even though the registered trial is single-group. The numerical illustrations and the trial’s own sample-size justification are not reported in the registry record. Planning quantities should not be described as observed power.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Insulin Icodec + Semaglutide")
n   <- c(91)
m   <- c(-1.23)
sp  <- c(0.86)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Insulin Icodec + Semaglutide  91  -1.230  0.860  -1.409  -1.051

11. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary HbA1c outcome and multiple secondary outcomes, including glucose profiles, post-prandial glucose increment, fasting plasma glucose, hypoglycaemic episodes, body weight, and insulin dose. It reports nine outcome measures and one analysis, but does not explain how the risk of false-positive findings across multiple outcomes was handled. Whether a hierarchy, adjustment, or other multiplicity strategy was prespecified is not reported in the registry record. The missing numerical results also prevent assessment of how any multiplicity approach might affect interpretation.

For the primary outcome, the record defines the analysis population as participants allocated to the intensification phase and says the analyzed number consists of participants with available data. It further states that observations after permanent discontinuation of either treatment were excluded from the on-intensification analysis. The exact number contributing to the primary analysis is not reported in the registry record.

The registry does not describe the method used to address missing endpoint values apart from the available-data definition and exclusion of post-discontinuation observations. It does not report imputation, sensitivity analyses, or assumptions about why data may be missing. General statistical context: missing-data choices can affect both the population represented by an estimate and the estimate’s uncertainty. Without details on the reasons for missingness and the analysis method, the potential impact cannot be assessed.

The record’s description is therefore sufficient to identify the stated analysis population and one exclusion rule, but not to reconstruct the complete missing-data strategy or determine whether multiplicity was formally controlled.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hydrocele (1.1%), intervertebral disc protrusion (1.1%), osteoarthritis (1.1%), spinal osteoarthritis (1.1%), phimosis (0.7%).

The registry reports safety counts separately for the run-in and treatment periods. In the insulin icodec run-in group, 148 participants were at risk; 7 were affected by serious adverse events, 14 by other adverse events, and 1 death was reported. In the insulin icodec plus semaglutide treatment-period group, 94 participants were at risk; 3 were affected by serious adverse events, 50 by other adverse events, and no deaths were reported.

These are counts from different study periods, not results from randomized treatment groups. The treatment-period group was formed after the run-in and intensification criteria, so its counts should not be compared with the run-in counts as though the groups were equivalent. The registry’s listed serious adverse-event terms include Hydrocele, Intervertebral disc protrusion, Osteoarthritis, and Spinal osteoarthritis at 1.1% each, and Phimosis and Angina Pectoris at 0.7% each. The record does not give enough information in the registry record to assign each listed term to a study period or infer a cause.

The counts describe reported events or affected participants as recorded, but do not establish whether an event was caused by either medicine. Event relatedness, severity detail beyond the serious-adverse-event classification, and a comparative risk estimate are not reported in the registry record. These figures should be interpreted descriptively.

13. What the Result Does, and Does Not, Mean

The registry record establishes the study’s structure and the outcome it was designed to assess: HbA1c change from week 26 to week 52 among participants entering the intensification phase. It also reports safety counts for the run-in and treatment periods. However, it does not report the numerical primary outcome, uncertainty around that outcome, or the numerical results for the listed secondary outcomes. The posted record therefore does not permit a quantitative assessment of the primary change or its precision.

Even with a numerical before-and-after result, this design would describe change in the selected intensification group rather than a randomized treatment effect. General statistical context: a treatment effect is a contrast between outcomes under different treatment conditions, and a single observed group does not supply the concurrent counterfactual comparison needed to isolate that contrast. The run-in period and the intensification phase are sequential study periods with different eligibility pathways, not interchangeable treatment arms.

The safety counts likewise do not establish that an event was caused by treatment, nor do they provide a like-for-like comparison between regimens. The record does not report event attribution or a randomized control group. A difference in counts across these periods could not be interpreted as a causal comparison from the registry information alone.

Thus, the record supports a description of the study design, population pathway, endpoint definition, and reported event counts. It does not support a conclusion about the magnitude or precision of the primary HbA1c change, comparative efficacy, or treatment causation for adverse events. Those conclusions would require numerical results and, for comparative claims, an appropriate comparison design and analysis.

14. Why This Trial Matters Statistically

This trial illustrates why the analysis population and study pathway are essential parts of statistical interpretation. The total enrollment was 148, but the primary endpoint concerns participants who entered the intensification phase after a 26-week run-in and met the intensification criteria. The treatment-period safety group lists 94 participants at risk. These counts describe distinct stages and populations; the difference between them should not be treated as a simple measure of dropout because eligibility determined who proceeded.

The endpoint is also tied to a specific time origin. HbA1c change is defined from week 26 to week 52, so it addresses the intensification period rather than the entire study from initial entry. This makes the time window and selection process central to understanding what any estimate would represent.

General statistical context: a mean change can be informative about observed experience in a group, but precision and causal interpretation require more than a point estimate. Measures of uncertainty help readers judge how much the estimate could vary, while a suitable comparison group helps separate treatment-associated change from time trends and other influences. Here, the registry record does not report the primary numerical estimate or its uncertainty, and the study is single-group without random allocation.

The record also distinguishes treatment exposure periods for safety reporting. Treating sequential periods as if they were parallel, comparable arms would ignore the way participants were selected into intensification. This is a useful reminder that counts and outcomes must be interpreted in light of the population and design that generated them, rather than compared solely by their labels.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

A central limitation for interpretation is the absence of a separate randomized comparison group. The study is single-group and unmasked, and participants were selected for intensification after the run-in. Therefore, the design does not provide a direct randomized contrast between combination treatment and an alternative regimen or control condition. Changes over time cannot be attributed to the added medicine solely on the basis of this design.

The primary analysis population is narrower than the enrollment population: it consists of participants allocated to the intensification phase, with the analyzed number depending on data availability. The record does not report the exact primary endpoint sample size. Because progression depended on intensification criteria, participants entering the treatment period may differ from those who did not proceed. The record does not provide enough information to characterize those differences.

The stated exclusion of post-discontinuation observations limits the outcome analysis to observations before permanent discontinuation. Other details about missing data handling, sensitivity analyses, and the statistical model are not reported in the registry record. The numerical primary result and its uncertainty are also not reported, preventing assessment of magnitude and precision.

For safety, the registry presents counts from sequential periods with different numbers at risk rather than a like-for-like randomized comparison. The record does not report event attribution for the listed adverse events. These limitations are specific to interpreting the design and reported data; they do not establish what unreported analyses or results might show.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Research Study to See How a New Weekly Insulin, Insulin Icodec When Given Along With Semaglutide Helps in Reducing the Blood Sugar Level in Patients With Type 2 Diabetes — NCT05813912.
• Registry record last updated 2026-06-03; status last verified 2026-05.
• Results first posted 2026-06-03.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

This completed phase 3 study enrolled 148 participants with type 2 diabetes and used a single-group, unmasked design without randomized allocation. Participants received weekly insulin icodec during a 26-week run-in. Those meeting the intensification criteria proceeded to 26 weeks of weekly insulin icodec plus semaglutide; the registry describes a 5-week follow-up for participants after their applicable treatment period.

The primary endpoint was mean change in HbA1c from week 26 to week 52 among participants in the intensification phase. The analysis used observed data at planned visits and excluded post-discontinuation observations after permanent discontinuation of either treatment. The exact analyzed number and the primary numerical result are not reported in the registry record.

The record lists secondary glucose, hypoglycaemia, body-weight, and insulin-dose outcomes, along with safety counts for the run-in and treatment periods. It does not report the numerical secondary outcome results or details of the complete statistical plan, missing-data methods, or multiplicity strategy.