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 52-week add-on to Metformin Comparison of Saxagliptin and Sulphonylurea, With a 52-week Extension Period (NCT00575588), a randomized phase 3 study. The primary outcome is Hemoglobin A1c (HbA1c) Change From Baseline to Week 52.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00575588 |
| Phase | Phase 3 |
| Condition | Type 2 Diabetes |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 891 |
| Arms in the primary outcome | n/a |
| Lead sponsor | AstraZeneca |
| Results posted | Yes |
2. Clinical Question
The trial asks whether the change in HbA1c from baseline to Week 52 differs between adults with type 2 diabetes receiving saxagliptin added to metformin and those receiving glipizide added to metformin. The participants were described as having inadequate glycaemic control on metformin therapy alone. Both groups therefore received metformin alongside a second drug; the comparison is between those add-on strategies, not between either strategy and no additional treatment.
Population
Adults with type 2 diabetes who had inadequate glycaemic control on metformin alone. The primary analysis description further restricts its population to randomized participants who completed treatment through 52 weeks, had baseline and Week 52 HbA1c measurements, and had no significant protocol deviations.
Intervention
Saxagliptin added to metformin. The registry describes the study as assessing efficacy and tolerability of this combination.
Comparator
Glipizide, described as a sulphonylurea, added to metformin. The registry title uses the broader term sulphonylurea for the comparison.
Primary question
How did adjusted mean HbA1c change from baseline to Week 52 compare between the two add-on groups in the per-protocol analysis set? The numerical adjusted changes and their between-group comparison are not reported in the registry record.
This distinction matters when interpreting the endpoint. The primary outcome is a continuous measure, and the registry describes the change for each participant as the Week 52 value minus the baseline value. The question is about the groups' adjusted mean changes, not about the proportion of participants who reached a particular HbA1c threshold. No threshold-based primary outcome is described.
3. Trial Design
This was a Phase III, randomized, parallel-group, double-blind, active-controlled study with a 52-week extension period. The title describes it as international and multi-centre. The registry lists quadruple masking, but does not specify which groups of people were masked. Allocation is recorded as randomized. The study was sponsored by AstraZeneca and is marked completed.
The registry reports total enrollment of 891. It describes the two comparison groups as saxagliptin with metformin and glipizide with metformin. The intervention list also names metformin, sulphonylurea, and saxagliptin. The arm descriptions are blank, so details such as drug doses, titration, adherence procedures, and the precise treatment schedules are not reported in the registry record.
The primary outcome was HbA1c change from baseline to Week 52. Its description specifies an adjusted mean change for the per-protocol analysis set. The primary population consists of randomized participants who completed 52 weeks of treatment, had both baseline and Week 52 HbA1c measurements, and had no significant protocol deviations. That is narrower than the full group enrolled, but the registry does not report how many participants met those criteria or how many were included in the primary analysis.
The trial also lists secondary outcomes concerning at least one hypoglycaemic event over 52 weeks, body-weight change from baseline to Week 52, and the mean slope of regressions of HbA1c change from Week 24 to Week 52. The record lists eight outcome measures and eight analyses, but does not provide the numerical results for those outcomes in the available record.
The recorded start is 2007-12, and primary completion is recorded as 2009-08. Although the title includes a 52-week extension period, details of extension-period treatment, analysis, follow-up, and results are not reported in the registry record. The registry also does not report the randomization procedure, allocation concealment, site count, or a protocol-level statistical analysis plan.
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: 891 participants were enrolled in the study; 33 participants did not enter the treatment period; 858 participants were randomized and treated.
Overall Study
| Stage | Saxagliptin + Metformin | Glipizide + Metformin |
|---|---|---|
| STARTED | 428 | 430 |
| COMPLETED | 165 | 147 |
| NOT COMPLETED | 263 | 283 |
Reasons for not completing this period:
| Reason | Saxagliptin + Metformin | Glipizide + Metformin |
|---|---|---|
| Adverse Event | 11 | 13 |
| Withdrawal by Subject | 27 | 31 |
| Lost to Follow-up | 1 | 3 |
| Death | 4 | 2 |
| Incorrect enrollment | 8 | 4 |
| Study specific discontinuation criteria | 203 | 218 |
| Severe non-compliance to protocol | 5 | 7 |
| Safety reasons | 2 | 1 |
| Elevated triglyceride | 1 | 0 |
| Patient moved | 1 | 3 |
| Impossible to determine HbA1c | 0 | 1 |
Analysis population for the primary outcome, as defined in the registry: Randomized participants who completed the 52 weeks of treatment had both baseline and week 52 HbA1c measurement and had no significant protocol deviations
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.
| Characteristic | Unit | Saxagliptin + Metformin | Glipizide + Metformin | Total |
|---|---|---|---|---|
| Participants analysed | 428 | 430 | 858 | |
| Age Continuous | years | 57.50 (SD 10.26) | 57.59 (SD 10.37) | 57.55 (SD 10.31) |
| Sex: Female, Male: Female | Participants | 216 | 198 | 414 |
| Sex: Female, Male: Male | Participants | 212 | 232 | 444 |
6. Primary Endpoint
The registry lists the primary outcome as Hemoglobin A1c (HbA1c) Change From Baseline to Week 52, measured over Baseline to 52 Weeks.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Error |
| Unit | Percent |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the change in hemoglobin A1c, or HbA1c, from baseline to Week 52. HbA1c is a continuous measure. The registry defines each participant's change as the Week 52 value minus the baseline value, and identifies the reported target as an adjusted mean change. The endpoint is expressed in percent, with a standard error as the stated measure of dispersion.
In general, a continuous endpoint can retain information across the range of observed measurements rather than reducing results to whether participants did or did not cross a selected cutoff. A mean change describes the average direction and magnitude of change in the analyzed group, while an adjustment can account for specified factors in a statistical model. However, the factors included in the adjustment and the model used are not reported in the registry record.
The endpoint's timing makes it a comparison of change through the end of the initial treatment period, not a measure of change across the full extension period. The registry describes the primary analysis as per protocol and restricts it to participants who completed treatment, had both measurements, and had no significant protocol deviations. The number of participants included, their baseline HbA1c values, and the adjusted mean changes are not reported in the registry record. Consequently, the endpoint definition is available, but its observed result and precision cannot be assessed from the record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry identifies the primary comparison and defines the endpoint, but it does not report the numerical adjusted mean HbA1c changes or the estimated difference between the treatment groups. It also does not report a confidence interval, standard error value, or p-value for the primary comparison. Without those results, the direction, size, and precision of the observed primary effect cannot be described from the registry record. The endpoint description alone does not establish which group had a larger or smaller change.
The stated primary analysis uses a per-protocol population: randomized participants who completed treatment through 52 weeks, had baseline and Week 52 measurements, and had no significant protocol deviations. This is important because the estimate pertains to that defined subset, rather than automatically representing all 891 enrolled participants. The number meeting the criteria and the reasons other participants were not included are not reported in the registry record. The registry does not report a separate intention-to-treat result that could be compared with the per-protocol estimate.
In general, an estimated group difference addresses the size and direction of a contrast, while its confidence interval describes uncertainty around that estimate under the analysis assumptions. A p-value addresses compatibility of the data with a specified null hypothesis under a statistical model; it is not a measure of clinical importance. Those concepts cannot be applied to this primary result without the numerical estimate and its uncertainty. The registry states that the dispersion measure is a standard error, but gives no standard error value.
The total enrollment is 891, while the safety counts identify 428 participants at risk in the saxagliptin plus metformin group and 430 in the glipizide plus metformin group. Those safety denominators are not a substitute for the number in the primary efficacy analysis. The registry does not explain the relationship between those counts and the analyzed per-protocol population, and the numbers should not be used to infer participant flow.
Safety information is reported separately. At-risk counts for serious adverse events are 428 and 430, with 54 and 55 affected participants, respectively. For the registry's other adverse-event category, the at-risk counts are also 428 and 430, with 105 and 201 affected participants. These are descriptive counts; the record does not provide a statistical comparison or adjusted analysis of these safety outcomes. The serious adverse-event terms and percentages are also listed, but the registry does not explain their analysis method or provide a comparison between groups for each term.
Overall, the registry supports a description of the design, endpoint definition, analysis-population criteria, and reported safety counts. It does not support a conclusion about the magnitude or statistical precision of the primary efficacy contrast. That limitation is about the information in the registry record, not evidence that the result was null or that no analysis was conducted.
9. Statistical Methods Explained
The trial is described as randomized, parallel-group, and double-blind, with quadruple masking recorded in the registry. In general, random allocation is intended to make treatment groups comparable with respect to both measured and unmeasured characteristics at the start of a study. Masking is intended to limit the influence of treatment knowledge on care, outcome assessment, or reporting. The registry does not describe the randomization sequence, allocation concealment, which roles were masked, or whether masking was maintained. Those details are therefore not reported in the registry record.
What does a mean change estimate?
The primary outcome is continuous: each participant's HbA1c change is calculated as the Week 52 value minus the baseline value. The analysis description refers to an adjusted mean change for each treatment group. In general, a mean summarizes the central tendency of measurements, and a change score puts the follow-up value in relation to that participant's baseline value. The adjustment indicates that the reported mean is intended to account for specified model terms, but the covariates, model form, and estimation method are not reported in the registry record.
The registry names standard error as the dispersion measure. In general, a standard error describes the estimated sampling variability of an estimate under the statistical model. It is not the same as the spread of individual participants' values. A confidence interval, if reported, would communicate a range of values compatible with the estimate and its uncertainty under the analysis assumptions. No numerical standard errors or confidence intervals for the primary endpoint are reported in the registry record.
What does the analysis population imply?
The primary analysis is described as per protocol. It includes randomized participants who completed the 52 weeks of treatment, had both baseline and Week 52 HbA1c measurements, and had no significant protocol deviations. In general, a per-protocol analysis focuses on participants meeting specified adherence or protocol criteria. Because eligibility for this analysis depends on events after randomization, the analyzed subset may not retain all the comparability created by random allocation. This is a methodological consideration, not a claim that bias occurred in this trial.
The record does not report the number of participants in the primary analysis, the exclusions from it, the handling of missing measurements, or the definition of significant protocol deviations. It also does not report an intention-to-treat analysis or a sensitivity analysis. It is therefore not possible to assess from this record how conclusions might change under alternative population definitions or missing-data assumptions.
How should safety counts be read?
The safety record reports the number at risk and the number affected in each treatment group for serious adverse events and for another adverse-event category. These counts are descriptive and use group-specific denominators. In general, a count of participants with at least one event differs from a count of all event occurrences, because one participant may experience more than one event. The record labels the affected participants but does not state whether the displayed figures represent unique participants, how repeated events were handled, or what statistical method was used for group comparisons. The deaths field contains no numerical value, so deaths are not reported in the registry record.
The record also gives selected serious adverse-event terms with percentages: hypertension and hypertensive crisis are each 0.7%; atrial fibrillation, coronary artery disease, osteoarthritis, and pneumonia are each 0.5%. It does not give the group allocation for each term, the corresponding participant counts, or a statistical comparison. These percentages should therefore be read as the registry's descriptive entries, not as estimates of comparative risk.
What can the planning assumptions tell us?
The record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. The numerical planning outputs and the selected fraction are not reported in the registry record. These assumptions describe an illustration, not the trial's own design calculations. Such sample-size illustrations do not represent observed power. The registry does not report the original sample-size justification, target effect, or power calculation used to plan the trial.
In general, a normal approximation uses a simplified probability model for statistical planning or inference. Its suitability depends on the endpoint, design, and analysis assumptions. The record does not provide enough information to evaluate the assumptions behind its planning illustration, nor does it report an interim-analysis plan or stopping rules. Those features are not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary HbA1c outcome and several secondary outcomes, including hypoglycaemic events, body-weight change, and HbA1c slope from Week 24 to Week 52. It also states that eight outcome measures and eight analyses are present. Listing multiple outcomes raises a general statistical question: if many hypotheses are tested, the chance of at least one apparently unusual result can increase unless the testing approach accounts for multiplicity. This is a general consideration, not evidence that any particular approach was used in this trial.
The registry does not report a multiplicity adjustment, a hierarchy of hypotheses, a gatekeeping procedure, or which secondary outcomes were confirmatory versus exploratory. It also does not report the p-values or confidence intervals needed to assess the reported outcome analyses. The multiplicity strategy is therefore not reported in the registry record.
The primary analysis population requires completion of 52 weeks, both HbA1c measurements, and no significant protocol deviations. The record does not explain how missing Week 52 values were handled, whether participants with incomplete follow-up were excluded or modeled, or whether sensitivity analyses assessed assumptions about missing data. Missing-data methods are not reported in the registry record.
Likewise, the number included in each analysis, participant disposition by group, and reasons for discontinuation or exclusion are not reported in the registry record. These omissions limit assessment of how the per-protocol population relates to the randomized population. No assumptions about unreported missing-data methods, interim analyses, or multiplicity adjustments can be made from the information recorded.
11. 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 |
|---|---|---|---|
| Saxagliptin + Metformin | 54 | 428 | 12.6% |
| Glipizide + Metformin | 55 | 430 | 12.8% |
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 |
|---|---|
| Saxagliptin + Metformin | 105 of 428 |
| Glipizide + Metformin | 201 of 430 |
Most frequent serious adverse event terms
| Event term | Organ system | Saxagliptin + Metformin | Glipizide + Metformin |
|---|---|---|---|
| Atrial Fibrillation | Cardiac disorders | 2 of 428 | 2 of 430 |
| Coronary Artery Disease | Cardiac disorders | 2 of 428 | 1 of 430 |
| Osteoarthritis | Musculoskeletal and connective tissue disorders | 2 of 428 | 1 of 430 |
| Pneumonia | Infections and infestations | 2 of 428 | 1 of 430 |
| Hypertension | Vascular disorders | 0 of 428 | 3 of 430 |
| Hypertensive Crisis | Vascular disorders | 0 of 428 | 3 of 430 |
| Cardiac Failure | Cardiac disorders | 1 of 428 | 1 of 430 |
| Myocardial Infarction | Cardiac disorders | 1 of 428 | 1 of 430 |
| Angina Unstable | Cardiac disorders | 0 of 428 | 2 of 430 |
| Colon Cancer | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 2 of 428 | 0 of 430 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Saxagliptin + Metformin | Glipizide + Metformin |
|---|---|---|---|
| Hypoglycaemia | Metabolism and nutrition disorders | 15 of 428 | 165 of 430 |
| Nasopharyngitis | Infections and infestations | 46 of 428 | 41 of 430 |
| Hypertension | Vascular disorders | 19 of 428 | 27 of 430 |
| Diarrhoea | Gastrointestinal disorders | 25 of 428 | 17 of 430 |
| Upper Respiratory Tract Infection | Infections and infestations | 25 of 428 | 16 of 430 |
Serious events with the highest arm-level rate: hypertension (0.7%), hypertensive crisis (0.7%), atrial fibrillation (0.5%), coronary artery disease (0.5%), osteoarthritis (0.5%).
For serious adverse events, the registry reports 54 affected participants among 428 at risk in the saxagliptin plus metformin group and 55 among 430 at risk in the glipizide plus metformin group. For the other adverse-event category, it reports 105 affected participants among 428 at risk in the saxagliptin plus metformin group and 201 among 430 at risk in the glipizide plus metformin group. These are the registry's reported counts and denominators; they are not adjusted estimates or, by themselves, a statistical test of the difference between groups.
The record lists hypertension and hypertensive crisis at 0.7% each, and atrial fibrillation, coronary artery disease, osteoarthritis, and pneumonia at 0.5% each, among the top serious adverse-event terms. It does not specify the group for each term or give associated counts. The percentages therefore do not establish a between-group comparison. The deaths field has no numerical value, so deaths are not reported in the registry record.
In general, adverse-event counts describe events observed and recorded during a study; they do not alone establish that treatment caused an event. Interpretation depends on the definitions, follow-up, event collection, participant exposure, and analysis methods. The registry does not report those details for these summaries, nor does it provide a statistical safety comparison. The counts should be read as descriptive safety information, not as a causal or definitive comparative conclusion.
12. What the Result Does, and Does Not, Mean
This trial's registry record defines a direct active comparison: saxagliptin plus metformin versus glipizide plus metformin in adults with type 2 diabetes whose control on metformin alone was inadequate. Its primary endpoint is adjusted mean HbA1c change from baseline to Week 52, analyzed in a per-protocol population. The record also identifies secondary outcomes and provides selected safety counts.
However, the numerical primary outcome result is not reported in the registry record. The adjusted mean change in each group, the between-group estimate, and its uncertainty are absent. It is therefore not possible to describe from this record whether the observed changes favored either group, how large a difference was observed, or how precisely that difference was estimated. The absence of a reported result is not evidence that the groups had the same outcome.
In general, statistical significance and clinical importance are different questions. A p-value, when available, concerns compatibility with a statistical null hypothesis under a model; an effect estimate and its confidence interval help convey magnitude and uncertainty. Neither can be assessed for the primary comparison without the numerical results. The registry also does not provide a threshold-based primary endpoint, so the primary question should not be recast as a comparison of participants reaching a target.
The reported safety figures add descriptive information, but do not establish causation or comparative safety. The record does not report deaths numerically, nor does it provide group-specific details for each listed serious adverse-event term. Information about treatment exposure, missing data, discontinuations, and safety-analysis methods is also not reported in the registry record.
Accordingly, the record supports understanding what comparison was planned, which population was specified for the primary analysis, and which safety counts were posted. It does not, on its own, support a numerical judgment about the primary efficacy contrast or a complete evaluation of benefits and harms. Any interpretation beyond these recorded details would require results and methods not reported in the registry record.
13. Why This Trial Matters Statistically
Several statistical features make this trial useful to examine. It compares two active add-on approaches while keeping metformin in both groups. That design targets a comparative question between treatment strategies rather than the effect of adding treatment versus leaving therapy unchanged. Randomized allocation is recorded, and the parallel-group design assigns participants to one of the two group strategies for comparison.
The primary endpoint is a continuous change measure, with an adjusted mean identified as the target statistic. This makes the analysis question about average change through Week 52 rather than only about a threshold or event count. The registry's definition also makes the analysis population explicit: participants had to complete treatment, have both HbA1c measurements, and have no significant protocol deviations. This distinction is statistically important because the specified per-protocol population is not automatically identical to the randomized population.
The record also illustrates why endpoint definitions and results must be read separately. The endpoint and its measurement timing are described, but the numerical effect estimate and uncertainty are not reported. A clearly defined outcome does not by itself reveal the magnitude, direction, or precision of the observed comparison. Similarly, enrollment and safety denominators cannot replace the missing primary analysis results.
Finally, the record contains safety counts and a planning-illustration description without the corresponding numerical planning outputs. This underscores the difference between descriptive observations, inferential results, and design calculations. Safety counts do not establish causation or a tested group difference, and planning illustrations do not represent observed power. The record is therefore informative about design and reporting structure while leaving important questions about the primary estimate, missing data, multiplicity, and analysis methods unresolved.
14. 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.
The main limitation for interpreting efficacy is that the registry record does not report the numerical primary outcome results. Adjusted mean changes, the between-group contrast, a standard error value, confidence interval, and p-value are not reported in the registry record. As a result, the size, direction, and precision of the primary comparison cannot be evaluated from this record.
The primary analysis is restricted to participants who completed the 52-week treatment period, had baseline and Week 52 HbA1c measurements, and had no significant protocol deviations. The number in this analysis and the reasons for exclusion are not reported in the registry record. Without those details, it is not possible to assess the degree to which the per-protocol analysis represents the randomized participants or how selection into the analysis may affect interpretation.
Important analysis details are also absent. The adjustment factors and model are not reported in the registry record. Missing-data handling, sensitivity analyses, multiplicity procedures, interim analyses, and the original sample-size justification are not reported. The record does not describe allocation concealment or identify the roles included in quadruple masking. Arm descriptions are blank, leaving treatment schedules and doses unreported.
Safety results are limited to counts and selected serious-event terms. The record does not report the statistical method for comparing safety outcomes, group-specific counts for the listed terms, or a numerical deaths result. Details about event ascertainment and repeated events are not reported in the registry record.
Although the study title includes a 52-week extension period, extension-period methods and results are not reported in the registry record. These reporting gaps should not be mistaken for evidence that an analysis or procedure was absent; they mean it cannot be evaluated from this record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: 52-week add-on to Metformin Comparison of Saxagliptin and Sulphonylurea, With a 52-week Extension Period — NCT00575588.
- Registry record last updated 2012-03-21; status last verified 2012-03.
- Results first posted 2010-09-06.
- 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.
18. Record Summary
The registry record describes NCT00575588 as a completed Phase III randomized, parallel-group, double-blind active-controlled study, with quadruple masking recorded. It compares saxagliptin plus metformin with glipizide plus metformin in adults with type 2 diabetes and inadequate glycaemic control on metformin alone. Total enrollment is 891, and the title includes a 52-week extension period.
The primary outcome is adjusted mean HbA1c change from baseline to Week 52 in a per-protocol population. The numerical primary result and its uncertainty are not reported in the registry record. Safety counts are reported for serious and other adverse events, with selected serious-event terms; deaths are not reported in the registry record.
The record does not report the primary analysis model, missing-data methods, multiplicity strategy, or numerical planning-illustration outputs. It supports a description of the study question, design, endpoint definition, and posted safety counts, but not a numerical assessment of the primary treatment contrast.