This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Phase 3 28-Week Study With 24-Week and 52-week Extension Phases to Evaluate Efficacy and Safety of Exenatide Once Weekly and Dapagliflozin Versus Exenatide and Dapagliflozin Matching Placebo (NCT02229396), a randomized phase 3 study. The primary outcome is Change in HbA1c From Baseline to Week 28.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02229396 |
| Phase | Phase 3 |
| Condition | Diabetes Mellitus |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 695 |
| Arms in the primary outcome | 3 |
| Lead sponsor | AstraZeneca |
| Results posted | Yes |
2. Clinical Question
Population
People with type 2 diabetes who had inadequate glycemic control on metformin. The primary analysis set included randomized patients who received at least one dose of study medication and had at least one post-baseline HbA1c assessment. The overall enrollment was 695.
Intervention
Simultaneous exenatide once weekly 2 mg and dapagliflozin once daily 10 mg. The trial evaluated this combination during the 28-week primary study period, with extension phases also described in the registry record.
Comparator
The combination was compared with exenatide once weekly 2 mg alone and dapagliflozin once daily 10 mg alone. The posted arm labels describe the single-medicine groups as receiving the active medicine with matching placebo.
Primary question
How did HbA1c change from baseline to Week 28 in the combination group compared with each single-medicine group? HbA1c is a measure of longer-term glycemic status. The registry reports estimated mean changes and mixed-model comparisons for the primary endpoint.
3. Trial Design
This was a randomized, double-blind, parallel-group, active-controlled Phase 3 study. It was multicenter and enrolled 695 participants. The registry describes a 28-week primary study period followed by a 24-week extension phase and then a 52-week extension phase. The primary endpoint assessed change from baseline to Week 28, so the primary comparisons concern the initial study period, not the extension periods.
Participants were assigned to simultaneous exenatide once weekly 2 mg and dapagliflozin once daily 10 mg, exenatide alone, or dapagliflozin alone. The registry describes the single-medicine arms as exenatide plus placebo and dapagliflozin plus placebo. Double masking means that treatment assignment was intended to be concealed from participants and relevant study personnel, though details of masking procedures are not reported in the registry record.
The primary analysis set was not simply everyone enrolled. It included randomized patients who received at least one dose of study medication and had at least one post-baseline HbA1c assessment. The endpoint-specific analysis counts were 228 for the combination, 227 for exenatide plus placebo, and 230 for dapagliflozin plus placebo. The registry record does not report a detailed account of participant flow, exclusions from this analysis set, or the number completing the extension phases.
The registry record does not report the sample-size justification or the planned statistical analysis in enough detail to establish whether the posted analyses follow a prespecified hierarchy. It also does not report interim analyses. These absences limit conclusions about how the study was planned and how the reported comparisons relate to any prespecified decision rules.
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 study had a Screening Visit, a 1-week placebo Lead-in Period, a Randomization Visit, and 9 further visits at 1- to 4-week intervals during a 28-week Treatment Period. Patients then entered a 24-week Extension Period 1 and subsequent 52-week Extension Period 2. A follow-up visit occurred 10 weeks after last dose of study medication.
Overall Study
| Stage | Dapagliflozin + Placebo | Exenatide + Dapagliflozin | Exenatide + Placebo |
|---|---|---|---|
| STARTED | 233 | 231 | 231 |
| Randomization Code Allocated | 233 | 231 | 231 |
| Safety Analysis Set | 233 | 231 | 230 |
| Intent-to-Treat (ITT) Analysis Set | 230 | 228 | 227 |
| Completed 28-Week Study Period | 208 | 202 | 193 |
| Completed 52-Week Study Period | 194 | 193 | 177 |
| COMPLETED | 155 | 154 | 136 |
| NOT COMPLETED | 78 | 77 | 95 |
Reasons for not completing this period:
| Reason | Dapagliflozin + Placebo | Exenatide + Dapagliflozin | Exenatide + Placebo |
|---|---|---|---|
| Adverse Event | 6 | 12 | 12 |
| Death | 2 | 3 | 1 |
| Lost to Follow-up | 14 | 14 | 21 |
| Protocol Violation | 2 | 1 | 0 |
| Patient Incorrectly Randomized | 0 | 0 | 1 |
| Other | 10 | 18 | 15 |
| Withdrawal by Subject | 42 | 28 | 41 |
| Study-Specific Withdrawal Criteria | 0 | 1 | 3 |
| Withdrew; No Record on Termination Page | 2 | 0 | 1 |
Analysis population for the primary outcome, as defined in the registry: The ITT analysis set included all randomized patients who received at least 1 dose of study medication and had at least 1 post-baseline HbA1c assessment.
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: The overall number of baseline participants is comprised of the ITT analysis set, defined as all randomized patients who received at least 1 dose of study medication and had at least 1 post-baseline hemoglobin A1c (HbA1c) assessment.
| Characteristic | Unit | Dapagliflozin + Placebo | Exenatide + Dapagliflozin | Exenatide + Placebo | Total |
|---|---|---|---|---|---|
| Participants analysed | 230 | 228 | 227 | 685 | |
| Age, Continuous | Years | 54.5 (SD 9.16) | 53.8 (SD 9.82) | 54.2 (SD 9.62) | 54.2 (SD 9.53) |
| Sex: Female, Male: Female | Participants | 120 | 126 | 111 | 357 |
| Sex: Female, Male: Male | Participants | 110 | 102 | 116 | 328 |
| Race/Ethnicity, Customized: American Indian Or Alaska Native | Participants | 0 | 0 | 2 | 2 |
| Race/Ethnicity, Customized: Asian | Participants | 1 | 3 | 1 | 5 |
| Race/Ethnicity, Customized: Black Or African American | Participants | 33 | 34 | 27 | 94 |
| Race/Ethnicity, Customized: Other | Participants | 7 | 1 | 3 | 11 |
| Race/Ethnicity, Customized: White | Participants | 189 | 190 | 194 | 573 |
6. Primary Endpoint
The registry lists the primary outcome as Change in HbA1c From Baseline to Week 28, measured over Baseline to Week 28.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | % HbA1c |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was change in HbA1c from baseline to Week 28, expressed as a percentage of HbA1c. HbA1c reflects glycemic status over a period of time rather than a single point-in-time glucose reading. Measuring its change lets investigators compare the extent of change during the trial across the randomized groups.
The registry reports least-squares mean changes with 95% confidence intervals. A least-squares mean is a model-based estimate, rather than simply the arithmetic average of the observed values. The registry identifies the primary analysis population as randomized patients who received at least one dose of study medication and had at least one post-baseline HbA1c assessment.
The outcome is clinically relevant to the study question because the enrolled population was described as having inadequate glycemic control on metformin. A group comparison estimates how the average HbA1c change differed between assigned regimens under the study analysis. It does not, by itself, describe each participant’s experience, establish the size of an individual response, or settle questions about other outcomes. The registry record does not report a threshold for a clinically important difference in HbA1c, so statistical estimates should not be treated as a direct measure of clinical importance.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
At Week 28, the least-squares mean change in HbA1c was -1.98 in the exenatide-plus-dapagliflozin group, -1.60 in the exenatide-plus-placebo group, and -1.39 in the dapagliflozin-plus-placebo group. The corresponding 95% confidence intervals were -2.16 to -1.79, -1.79 to -1.41, and -1.57 to -1.21. Each estimate describes change from baseline within an arm; the contrasts between arms address the comparative question.
The sponsor-reported mixed-model comparison of the combination with exenatide alone estimated a mean difference of -0.38, with a 95% confidence interval from -0.63 to -0.13 and p = 0.003. The comparison with dapagliflozin alone estimated a mean difference of -0.59, with a 95% confidence interval from -0.84 to -0.34 and p < 0.001. The negative differences indicate that the estimated HbA1c change was lower in the combination group than in each comparator group, under the reported analysis.
Both confidence intervals for the sponsor-reported comparisons exclude zero. That is consistent with evidence of a difference under the model and test used. A confidence interval describes uncertainty around an estimate: its width conveys the range of values reasonably compatible with the data and analysis, while its position indicates the direction of the estimated contrast. It is not a guarantee that the true value lies within the particular interval.
The posted results therefore show a greater estimated reduction in HbA1c for the combination group than for either single-medicine group over the primary study period. This statement describes the reported group-level estimates, not an individual outcome or a conclusion about every aspect of benefit and risk. The registry record does not report a clinical-importance threshold for these contrasts, nor does it provide enough detail to assess all modeling assumptions or the handling of missing outcome values.
The sponsor’s analyses are adjusted model-based comparisons, while the arm-level values are estimated changes. Their precision and interpretation depend on the analysis approach. The confidence intervals and p-values are evidence about the specified comparisons, but they do not quantify the probability that a treatment effect is true, and they do not resolve the effects of multiple outcome testing or unreported analysis choices.
9. Statistical Methods Explained
Understanding these results requires distinguishing an estimated change within a group from a contrast between groups. The arm-level HbA1c estimates describe change from baseline to Week 28. A between-group mean difference asks how much the estimated change differed between two assigned regimens. These are related quantities, but they answer different questions: a change within one arm alone does not establish a comparative effect.
What does a confidence interval convey?
A confidence interval presents an estimate alongside a range reflecting sampling uncertainty under the analysis method. A narrower interval indicates greater precision than a wider interval, though precision does not establish that the method is unbiased or that the outcome is clinically important. For the primary comparisons, the reported 95% confidence intervals lie below zero. The negative estimates are therefore compatible with a lower change value in the combination group than in either comparator, under the specified contrasts and model.
A confidence interval is not a range containing every participant’s outcome. Nor is it a probability statement about the true value after the data have been observed. Its interpretation relies on the model and assumptions used to calculate it. The registry names a mixed-model analysis for the sponsor’s primary comparisons but does not provide the model specification, covariates, covariance structure, or diagnostics in the registry record.
How should the p-values be read?
A p-value measures how inconsistent the observed data, or more extreme data, would be with a specified null hypothesis under the test assumptions. The reported p-values are 0.003 and < 0.001 for the primary comparisons. They are not the probability that the null hypothesis is true, and they do not measure the size or importance of a difference. The estimated differences and their confidence intervals are needed to describe direction and precision.
The registry does not report whether the primary comparisons were adjusted for multiple testing or arranged in a testing hierarchy. That matters because interpreting several hypotheses as if each were the sole planned test can overstate the strength of evidence. The p-values should therefore be read as results of the reported comparisons, without assuming an unreported multiplicity procedure.
Why use a mixed model?
Mixed models can use repeated outcome measurements and account for the correlation among measurements from the same participant. They can also estimate group contrasts while incorporating terms specified in the model. These are general properties of the method, not confirmation of exactly how this study’s model was implemented. The registry does not report the model’s detailed terms or the assumptions used for incomplete follow-up.
Methods for missing data matter because participants may lack a Week 28 assessment or other measurements. A model’s estimates can depend on assumptions about why values are missing and how observed measurements relate to unobserved ones. The primary analysis population required at least one post-baseline HbA1c assessment, but the specific missing-data method and sensitivity analyses are not reported in the registry record.
What do the secondary binary analyses add?
The registry also reports binary outcomes as percentages meeting a stated criterion. A percentage is an arm-level proportion; it does not indicate how far an individual was from the criterion. The posted results include sponsor-reported Cochran-Mantel-Haenszel comparisons and simple unadjusted recalculations from the posted summaries. These methods need not produce identical estimates because they can use different analysis approaches. The recalculations are descriptive checks, not substitutes for the sponsor’s analysis.
Confidence intervals for a binary comparison describe uncertainty in the estimated contrast on its chosen scale. If an interval includes no difference, the data are compatible with values on either side of no difference under that calculation. The registry’s sponsor-reported secondary tests do not include confidence intervals in the listed analyses, so those p-values alone do not communicate the precision or range of the estimated difference.
How should sample-size illustrations be understood?
Planning illustrations based on observed effects are conditional examples, not observed power and not evidence that the trial was designed around those assumptions. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect set to a stated fraction of the observed effect. Such calculations can show how required group sizes would change when the target effect is smaller or larger, but they do not establish the trial’s original sample-size rationale. That rationale is not reported in the registry record.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Mixed Models Analysis | Mean Difference (Final Values) -0.38 | -0.63 to -0.13 | 0.003 |
| Mixed Models Analysis | Mean Difference (Final Values) -0.59 | -0.84 to -0.34 | < 0.001 |
The sponsor reported the primary endpoint comparisons using a mixed-model analysis. For the combination versus exenatide alone, the reported mean difference in final values was -0.38, with a 95% confidence interval from -0.63 to -0.13 and p = 0.003. For the combination versus dapagliflozin alone, the reported difference was -0.59, with a 95% confidence interval from -0.84 to -0.34 and p < 0.001.
These model-based comparisons are the sponsor’s own analyses of the primary question. They complement the arm-level least-squares mean changes by directly comparing groups. The posted results also include recalculated unadjusted comparisons for selected secondary binary outcomes; those calculations are not the same as the sponsor’s Cochran-Mantel-Haenszel analyses. For the primary endpoint, the registry record does not identify a discrepancy between a recalculation and the sponsor’s analysis.
The sponsor’s estimates and uncertainty intervals are the appropriate reported results to describe for the primary endpoint. Their interpretation remains conditional on the model and analysis assumptions, which are not fully detailed in the registry record. The registry does not report enough information to assess all prespecified analysis decisions, including multiplicity handling and missing-data procedures.
11. Multiplicity, Missing Data and Analysis Populations
The registry record does not report whether the primary comparisons were protected by a prespecified testing hierarchy or adjusted for multiplicity. The primary endpoint was compared across the combination and each single-medicine group, and the record also lists multiple secondary outcomes. When several hypotheses are assessed, the chance of at least one apparently unusual result can increase if each test is interpreted independently. An adjustment or hierarchy can address this issue, but the registry record does not describe one, so it would be inappropriate to assume either that adjustment occurred or that it was absent from the full study analysis plan.
The primary analysis set included randomized patients who received at least one dose of study medication and had at least one post-baseline HbA1c assessment. This is the population definition reported for the primary endpoint; it is not the same as saying that every randomized participant contributed a Week 28 value. The registry record does not report the number of missing Week 28 outcomes, the reasons for missingness, the imputation or modeling strategy for missing values, or sensitivity analyses addressing alternative missing-data assumptions.
Those details affect interpretation because estimates can depend on which participants contribute information and how incomplete observations are handled. Randomization supports comparisons between assigned groups, but exclusions or differential missingness after randomization can complicate the interpretation of an analysis population. The available description does not allow a full assessment of that possibility.
The registry record also does not report the sample-size justification or interim analyses. It names the primary analysis method but does not give the model specification or establish how the sponsor’s analysis plan treated all outcomes. The planning illustrations based on observed effects should not be confused with the trial’s original design rationale or with observed power. These limitations concern what can be established from the registry record; they do not demonstrate that a particular procedure was or was not used in the full study.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Dapagliflozin + Placebo | 18 | 233 | 7.7% |
| Exenatide + Dapagliflozin | 17 | 231 | 7.4% |
| Exenatide + Placebo | 18 | 230 | 7.8% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Dapagliflozin + Placebo | 2 of 233 |
| Exenatide + Dapagliflozin | 3 of 231 |
| Exenatide + Placebo | 1 of 230 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Dapagliflozin + Placebo | 73 of 233 |
| Exenatide + Dapagliflozin | 87 of 231 |
| Exenatide + Placebo | 78 of 230 |
Most frequent serious adverse event terms
| Event term | Organ system | Dapagliflozin + Placebo | Exenatide + Dapagliflozin | Exenatide + Placebo |
|---|---|---|---|---|
| Acute myocardial infarction | Cardiac disorders | 2 of 233 | 0 of 231 | 1 of 230 |
| Atrial fibrillation | Cardiac disorders | 1 of 233 | 0 of 231 | 2 of 230 |
| Cholelithiasis | Hepatobiliary disorders | 0 of 233 | 2 of 231 | 1 of 230 |
| Myocardial infarction | Cardiac disorders | 0 of 233 | 0 of 231 | 2 of 230 |
| Chest pain | General disorders | 1 of 233 | 0 of 231 | 1 of 230 |
| Anaphylactic reaction | Immune system disorders | 1 of 233 | 0 of 231 | 1 of 230 |
| Pneumonia | Infections and infestations | 0 of 233 | 1 of 231 | 1 of 230 |
| Hepatic enzyme increased | Investigations | 0 of 233 | 0 of 231 | 2 of 230 |
| Ischaemic stroke | Nervous system disorders | 1 of 233 | 0 of 231 | 1 of 230 |
| Abdominal pain lower | Gastrointestinal disorders | 0 of 233 | 1 of 231 | 0 of 230 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Dapagliflozin + Placebo | Exenatide + Dapagliflozin | Exenatide + Placebo |
|---|---|---|---|---|
| Upper respiratory tract infection | Infections and infestations | 22 of 233 | 15 of 231 | 17 of 230 |
| Urinary tract infection | Infections and infestations | 16 of 233 | 19 of 231 | 15 of 230 |
| Nausea | Gastrointestinal disorders | 10 of 233 | 13 of 231 | 26 of 230 |
| Injection site nodule | General disorders | 13 of 233 | 20 of 231 | 14 of 230 |
| Diarrhoea | Gastrointestinal disorders | 11 of 233 | 13 of 231 | 17 of 230 |
| Headache | Nervous system disorders | 12 of 233 | 16 of 231 | 12 of 230 |
| Nasopharyngitis | Infections and infestations | 12 of 233 | 15 of 231 | 8 of 230 |
| Vomiting | Gastrointestinal disorders | 7 of 233 | 8 of 231 | 12 of 230 |
Serious events with the highest arm-level rate: atrial fibrillation (0.9%), myocardial infarction (0.9%), hepatic enzyme increased (0.9%), cholelithiasis (0.9%), acute myocardial infarction (0.9%).
The registry reports serious adverse events for each arm using a number at risk and a number affected. Serious adverse events affected 18 of 233 participants in the dapagliflozin-plus-placebo group, 17 of 231 in the exenatide-plus-dapagliflozin group, and 18 of 230 in the exenatide-plus-placebo group. These are counts of participants affected, not counts of every event, and the registry excerpt does not provide a comparative statistical analysis of these figures.
The number reported with other adverse events was 73 of 233, 87 of 231, and 78 of 230 in those same respective groups. Deaths were reported as 2, 3, and 1. The registry record does not state in these figures whether any death was considered related to study medication, nor does it give a comparative analysis. The listed serious-event terms include atrial fibrillation, myocardial infarction, hepatic enzyme increased, cholelithiasis, acute myocardial infarction, and rib fracture, with percentages as recorded. The record does not supply an arm-specific breakdown for those listed terms.
Counts alone do not establish whether groups differ in safety. They also do not show timing, duration, severity beyond the serious-event classification, or causal attribution. The figures describe what was recorded in the registry and should be interpreted alongside the limits of the safety information reported there.
13. What the Result Does, and Does Not, Mean
The primary result describes an estimated group-level difference in HbA1c change through Week 28. In the sponsor’s mixed-model analyses, the combination group had a more negative estimated change than either single-medicine group, and both reported confidence intervals excluded zero. This is evidence of a difference under the stated analyses, with uncertainty represented by the intervals.
It does not establish that every participant experienced a larger change, or that the estimated difference has a particular clinical importance. The registry record does not report a threshold for clinical importance. A group mean summarizes outcomes across participants and cannot describe the full distribution of individual responses.
The result also does not answer every question about the treatment regimens. The primary endpoint addresses HbA1c at Week 28; it does not by itself establish effects on every secondary outcome, long-term outcomes, or safety. Although extension phases are described, the primary endpoint result is tied to the initial study period. The registry record does not provide extension-phase results in the registry record.
Statistical evidence is conditional on the population, outcome definition, model, and assumptions used. The primary analysis set required a post-baseline HbA1c assessment, while the registry record does not describe missing-data handling in detail. Multiplicity procedures and a clinical-importance threshold are also not reported. These gaps limit the strength and scope of conclusions that can be drawn from the posted results.
The most defensible interpretation is therefore specific: the reported Week 28 analysis estimated a greater reduction in HbA1c for the combination group than for either single-medicine group. That is a description of the posted estimates, not a recommendation about treatment or a complete judgment of benefits and harms.
14. Why This Trial Matters Statistically
This trial illustrates why a continuous primary endpoint and direct between-group contrasts matter. The arm-level estimates describe changes, while the mixed-model comparisons quantify how those changes differed across randomized groups. Reporting both makes it possible to distinguish within-group change from the comparative question the trial set out to address.
It also demonstrates the role of uncertainty intervals. The point estimates communicate direction and magnitude on the reported scale; confidence intervals show the precision of those estimates. P-values add evidence relative to a null hypothesis, but they do not replace the estimated contrast or its interval and do not establish clinical importance.
The secondary outcomes show why outcome type and multiplicity deserve attention. Continuous measures and threshold-based outcomes answer different questions, and a collection of secondary tests is not equivalent to a single confirmatory endpoint. Without a reported hierarchy or adjustment, secondary findings are more appropriately treated as contextual evidence than as a set of independent confirmations.
Finally, the record highlights the distinction between a trial’s design rationale and later planning illustrations. Sample-size examples based on observed effects can demonstrate assumptions, but they do not reveal the original justification or observed power. Missing-data procedures, model details, and multiplicity plans are also important for evaluating how robust the reported estimates are. Their absence from the registry record narrows what can be concluded, even when the main estimates and confidence intervals are clearly reported.
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.
Several limitations are specific to what the registry record reports for this trial. The primary endpoint is defined at Week 28, while the study description includes extension phases. Results for those phases are not included in the posted outcome information, so the primary estimates do not characterize outcomes across the full described study duration.
The record identifies the primary analysis population but does not report participant flow in detail, reasons for missing HbA1c assessments, or the method used to address missing data. It also does not provide the mixed model’s full specification, such as its covariates or covariance structure. These omissions make it difficult to independently assess the model assumptions and the sensitivity of estimates to incomplete follow-up.
The registry record does not report the sample-size justification, interim analyses, a multiplicity plan, or a threshold for clinical importance. It lists multiple secondary outcomes, but numerical results are not available for all of them in the posted results. Secondary interpretations therefore cannot be considered comprehensive.
The safety information gives affected-participant counts and deaths, but does not include a comparative statistical assessment, causal attribution, or arm-specific counts for the listed serious-event terms. The record also does not provide detailed information about event timing or duration. These reporting limits do not show that a particular analysis or safety procedure was omitted from the full study; they restrict what can be evaluated from the registry record.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Phase 3 28-Week Study With 24-Week and 52-week Extension Phases to Evaluate Efficacy and Safety of Exenatide Once Weekly and Dapagliflozin Versus Exenatide and Dapagliflozin Matching Placebo — NCT02229396.
- Registry record last updated 2018-12-31; status last verified 2018-12.
- Results first posted 2017-09-08.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT02229396 is a completed randomized, double-blind, parallel-group Phase 3 trial sponsored by AstraZeneca. It enrolled 695 people with type 2 diabetes and inadequate glycemic control on metformin, comparing exenatide plus dapagliflozin with each medicine alone. The primary endpoint was change in HbA1c from baseline to Week 28.
The sponsor-reported mixed-model comparisons estimated greater HbA1c reductions in the combination group than in either single-medicine group. The registry also reports selected secondary outcomes and safety counts. Multiplicity procedures, detailed missing-data methods, and the sample-size justification are not reported in the registry record, limiting assessment beyond the posted estimates.