A Study of Saxagliptin in Subjects With Type 2 Diabetes Who Have Inadequate Blood Sugar Control With Sulfonylureas: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Saxagliptin in Subjects With Type 2 Diabetes Who Have Inadequate Blood Sugar Control With Sulfonylureas (NCT00313313), a randomized phase 3 study. The primary outcome is Change From Baseline in Hemoglobin A1c (A1C) at Week 24.

2. Clinical Question

The trial addresses whether adding saxagliptin to glyburide changes glycemic control in people with type 2 diabetes whose control with glyburide alone is inadequate. The primary outcome is the mean change from baseline in hemoglobin A1C at Week 24. The record also lists outcomes for fasting plasma glucose, the percentage of participants achieving A1C < 7%, and postprandial glucose response.

The trial’s brief purpose describes a question about adding saxagliptin versus increasing sulfonylurea. However, the listed arms use glyburide 7.5 mg in each group and do not show a higher-dose glyburide comparison. The recorded arm structure therefore supports a comparison of saxagliptin addition with placebo addition on the same listed glyburide dose; it does not establish a direct comparison with glyburide dose escalation.

3. Trial Design

This was a multicenter, randomized, double-blind, placebo-controlled, parallel Phase 3 trial. The registry record lists enrollment of 768 and a completed status. The recorded start is 2006-04, and primary completion is recorded as 2007-09. AstraZeneca is listed as the sponsor.

The three listed treatment groups were saxagliptin 2.5 mg plus glyburide 7.5 mg, saxagliptin 5 mg plus glyburide 7.5 mg, and placebo plus glyburide 7.5 mg. Each arm description also permits metformin 500-2500 mg as needed. The registry record does not report how participants were allocated among arms, whether metformin use differed across groups, or how use of metformin was handled in the analysis.

Randomization is intended to make treatment groups comparable, on average, with respect to measured and unmeasured baseline characteristics. Double masking is intended to limit the influence of treatment knowledge on participant experience, study conduct, and outcome assessment. These are general features of randomized, masked trials; they do not establish how masking was maintained or whether it was successful in this study. Those details are not reported in the registry record.

The primary outcome was change from baseline in A1C at Week 24, adjusted for the baseline value. The outcome description specifies a mean change and standard error. The registry record does not report the method used for baseline adjustment, the planned contrast or contrasts among arms, or the statistical model. Although the record lists four outcome measures, it does not provide analysis results for them. The trial’s sample-size justification and any interim-analysis plan are 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomized participants who took at least 1 dose of double-blind treatment. To be included in analysis of change from baseline to Week 24 Last Observation Carried Forward (LOCF), participants must have had a baseline and at least 1 post-baseline measurement. If participant received rescue medication, measurement must have been taken before rescue.

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.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Hemoglobin A1c (A1C) at Week 24, measured over Baseline, Week 24.

The primary endpoint measures change in hemoglobin A1C from baseline to Week 24. The registry record describes the outcome as a mean change adjusted for the baseline value and gives the unit as percent. A1C is used as a measure of glycemic control because, in general, it reflects blood glucose over time rather than capturing a single moment. The record identifies the measurement times as baseline and Week 24.

A mean change summarizes the average direction and magnitude of change among the participants included in the analysis. Comparing mean changes between randomized groups can estimate how outcomes differed under the assigned treatment strategies, subject to the analysis methods and data completeness. The record does not provide the observed group means, adjusted comparison, confidence interval, or p-value, so the size and precision of any difference cannot be described.

The outcome description specifies standard error as the dispersion measure. In general, a standard error describes the precision of an estimated mean; it is not the same as the spread of individual participants’ measurements. The registry record does not report the actual standard errors. It also does not fully state how participants receiving rescue medication were handled: the population description ends partway through that condition. That detail is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies a continuous primary outcome: mean change from baseline in A1C at Week 24, with standard error as the stated measure of dispersion. It says the mean change was adjusted for baseline value. It does not report the statistical model, the exact treatment contrasts, or the procedure used to calculate the adjusted estimates. The explanations below describe general statistical principles and do not assert that any unreported method was used in this trial.

What does a mean change represent?

In general, a change score is calculated from a participant’s follow-up measurement relative to that participant’s baseline measurement. A group mean change then summarizes the average change among participants included in the analysis. A comparison between randomized groups asks whether those average changes differ. An adjusted analysis may account for baseline measurement values to improve the comparison’s precision or address baseline variation. The registry record states that the outcome was adjusted for baseline, but the adjustment method is not reported in the registry record.

The reported unit is percent. The registry record does not supply the observed mean changes or any group contrast, so no result can be interpreted as a particular increase or decrease. In particular, the endpoint description alone does not establish the direction or size of an effect.

What does the standard error tell the reader?

In general, a standard error describes uncertainty in an estimated mean. Smaller standard errors indicate more precise estimates under the statistical model and sampling assumptions. A standard error does not describe the full range of individual participant measurements; a standard deviation is commonly used for that purpose. Because the actual standard errors are not reported in the registry record, the precision of the group means cannot be assessed.

A confidence interval is a general way to express uncertainty around an estimate. It gives a range of values compatible with the data under a specified method and its assumptions. The confidence level and interval for this primary outcome are not reported in the registry record. A p-value, also a general inferential measure, evaluates data compatibility with a null hypothesis under specified assumptions; it does not quantify the size or clinical importance of an effect. No p-value is reported for the primary comparison.

Why does the analysis population matter?

The stated population includes randomized participants who took at least one dose of double-blind treatment. For the change-from-baseline analysis, participants also had to have a baseline and at least one post-baseline measurement. This means the described population is not simply all enrolled participants. The number of participants analyzed in each arm and the number excluded for missing measurements are not reported in the registry record.

The outcome description mentions Last Observation Carried Forward, or LOCF, in connection with the Week 24 analysis. In general, LOCF substitutes a participant’s most recent available measurement for a later missing measurement. This can retain participants with incomplete follow-up, but it relies on an assumption that the carried-forward value is an appropriate stand-in for the missing value. If outcomes would have changed after the last measurement, that assumption may not hold. The registry record does not report how many observations were carried forward or provide sensitivity analyses using other missing-data approaches.

The population description also begins a condition about participants who received rescue medication, but the text ends before specifying the rule. The handling of rescue medication is therefore not reported in the registry record. Rescue medication can affect outcome interpretation because it may alter measurements after its use; without the analysis rule, the impact of such use cannot be assessed.

What can be said about design and analysis?

Randomization, double masking, a placebo comparator, and a parallel-group design are recorded. In general, randomization supports a fair comparison by reducing systematic differences between assigned groups, while masking can reduce bias linked to knowledge of treatment. These design features do not remove the need to examine missing data, adherence, outcome measurement, and analysis choices. Details about allocation procedures, masking procedures, adherence, and the statistical model are not reported in the registry record.

The registry record lists four outcome measures and no analysis entries. It also indicates that the primary outcome was not reduced to a single comparison because it has several categories or time points. Thus, it does not provide a single inferential result that can be summarized as the primary treatment contrast. The statistical analysis plan, including any approach to multiple comparisons, is not reported in the registry record.

How should planning illustrations be read?

In general, a sample-size illustration based on an observed effect asks how many participants per arm would be needed under stated assumptions to detect an effect of a specified size. Such an illustration is not the trial’s original design justification and is not observed power. The record describes a planning approach using a two-sided significance level of 0.05, equal allocation, a normal approximation, and fractions of an observed effect. However, the registry record does not report the primary observed effect needed to evaluate those illustrations, nor does it report the trial’s own sample-size justification. Any planning values should be understood as hypothetical rather than as evidence about the trial’s achieved power.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists the primary A1C change outcome and secondary outcomes for fasting plasma glucose, achievement of A1C < 7%, and postprandial glucose response. It does not report how multiplicity was handled. In general, testing multiple outcomes or contrasts can increase the chance of obtaining at least one apparently notable result by chance. Adjustment procedures, a prespecified testing order, or a distinction between confirmatory and exploratory analyses can address that issue, but no such approach is reported in the registry record. No outcome-specific analysis results are reported either.

Missing-data handling is only partly described. The primary outcome population required baseline and at least one post-baseline measurement, and the description refers to Week 24 LOCF. In general, LOCF carries the most recent available observation forward, but that choice depends on assumptions about what would have happened after the last observed value. The record does not report the amount or pattern of missing data, how many participants had Week 24 measurements, or whether results were checked using alternative missing-data methods.

The primary-outcome population includes randomized participants who took at least one dose of double-blind treatment. That is the recorded analysis-population definition; it should not be assumed to be a full intention-to-treat analysis of every randomized participant. The record does not report analysis counts by group. Its description of rescue-medication handling is incomplete, so whether and how those observations were included is not reported in the registry record.

The registry record lists no analysis entries and does not provide a statistical analysis plan. The primary outcome is also marked as not reduced to a single comparison because it has several categories or time points. As a result, the multiplicity strategy, final analysis population, and missing-data sensitivity analyses cannot be reconstructed from the registry record.

11. Safety Results

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

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: pneumonia (1.1%), nephrolithiasis (1.1%), myocardial ischaemia (0.8%), appendicitis (0.8%), chest pain (0.8%).

The registry record lists serious adverse-event counts for placebo plus glyburide 7.5 mg, saxagliptin 2.5 mg plus glyburide 7.5 mg, and saxagliptin 5 mg plus glyburide 7.5 mg. The numbers at risk are 267, 248, and 253, respectively; the numbers affected by serious adverse events are 23, 18, and 12, respectively. These are counts as recorded, not adjusted comparisons between groups.

The record also lists other adverse-event counts: 163 of 267 in the placebo-plus-glyburide group, 144 of 248 in the saxagliptin 2.5 mg group, and 154 of 253 in the saxagliptin 5 mg group. The record does not provide a statistical comparison of these counts. Counts alone do not account for differences in exposure or follow-up and should not be read as evidence that one treatment caused more or fewer events.

The serious-event terms listed with percentages are pneumonia and nephrolithiasis, each at 1.1%, and myocardial ischaemia, appendicitis, chest pain, and acute cholecystitis, each at 0.8%. The list does not attach these terms to specific treatment groups. Death counts are not reported in the registry record. The event figures are descriptive; the registry record does not provide enough information to infer causation or compare safety statistically.

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

The trial record documents a randomized, double-blind comparison of saxagliptin added to glyburide against placebo added to glyburide, with A1C change at Week 24 as the primary endpoint. That design identifies the comparison the study was intended to address. It does not, by itself, establish what the results were. The registry record does not report the group estimates or the adjusted treatment comparison for the primary endpoint.

Accordingly, the record does not show whether mean A1C change differed between groups, how large any difference was, or how precisely it was estimated. It also does not provide a basis for judging whether a difference was statistically significant or clinically important. The same limits apply to the listed secondary outcomes, for which numerical results are not reported in the registry record.

The adverse-event information describes counts of participants affected and a list of serious-event terms. These observations can be reported as recorded, but they do not establish that an event was caused by a study treatment. Nor do the counts alone establish that risks differed between groups. The registry record does not report death counts or a statistical safety comparison.

The listed arms hold glyburide at 7.5 mg and compare saxagliptin doses with placebo. Although the brief purpose describes a comparison with increasing the amount of sulfonylurea, no higher-dose glyburide arm is listed. The posted design therefore does not document a direct comparison against glyburide dose escalation.

In general, a completed trial can contribute evidence only to the extent that its methods and results are available for evaluation. Here, the design and outcome definitions are documented, but the primary and secondary efficacy estimates, analysis details, and several population and missing-data details are not reported in the registry record. Conclusions about treatment effects cannot be drawn from the registry information alone.

13. Why This Trial Matters Statistically

This trial is statistically notable for the distinction between a clearly stated randomized comparison and the absence of reported efficacy estimates. Randomization and double masking describe design features intended to support a credible comparison, but readers still need the group estimates and uncertainty measures to evaluate the outcome. The registry record does not report those results for the Week 24 A1C endpoint.

The endpoint is a mean change adjusted for baseline, which raises practical interpretation questions about the adjustment method, the analysis population, and how incomplete follow-up was handled. The record identifies a population requiring treatment exposure and at least one post-baseline measure, and it refers to LOCF. However, it does not report participant counts contributing to the analysis or the extent of missing data. These details matter because they help readers assess how closely the analyzed participants represent those randomized and how sensitive results may be to assumptions about missing measurements.

The trial also lists multiple outcomes while leaving its multiplicity strategy unreported. In general, multiple outcome tests can complicate interpretation, especially when readers do not know which outcomes were prioritized or how the chance of false-positive findings was addressed. Without reported analyses, neither the outcome-specific evidence nor the relationship among the tests can be assessed.

Finally, the stated brief purpose invokes increasing sulfonylurea dose, but the recorded treatment arms list glyburide 7.5 mg in each group. Making that distinction explicit matters statistically: the listed comparison evaluates adding saxagliptin versus placebo on the recorded glyburide regimen, not a documented higher-dose glyburide strategy. The design question and the comparison a reader can evaluate should not be conflated.

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 most consequential limitation in the registry record is that it does not report numerical efficacy results. The primary endpoint is defined, but its group means, adjusted contrasts, uncertainty measures, and p-values are absent. Secondary outcome results are also not reported. This prevents assessment of the direction, size, or precision of treatment-group differences.

The described primary-outcome population is narrower than all randomized participants: it includes those who took at least one dose, and the change analysis requires baseline and at least one post-baseline measurement. Analysis counts by group and reasons for exclusion are not reported in the registry record. The outcome description mentions LOCF, but missingness patterns, the number of carried-forward values, and sensitivity analyses are not reported.

The rescue-medication clause in the population description is incomplete. The rules for including, excluding, or otherwise handling measurements after rescue medication are not reported in the registry record. This limits interpretation because the stated endpoint may depend on how post-rescue observations were treated.

The statistical model, planned treatment contrasts, multiplicity procedure, and sample-size justification are not reported in the registry record. Although the record describes assumptions for planning illustrations, it does not provide the observed effect needed to interpret them, and those illustrations do not establish the trial’s original design rationale.

There is also a mismatch between the brief purpose, which refers to increasing sulfonylurea, and the listed arms, which use glyburide 7.5 mg in every group. No higher-dose glyburide comparator is recorded. Finally, adverse-event counts are available, but death counts and a statistical safety comparison are not reported. These omissions limit what can be concluded about both efficacy and safety from the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Saxagliptin in Subjects With Type 2 Diabetes Who Have Inadequate Blood Sugar Control With Sulfonylureas — NCT00313313.
• Registry record last updated 2015-05-04; status last verified 2015-03.
• Results first posted 2009-09-25.
• 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.

18. Record Summary

NCT00313313 is recorded as a completed, multicenter, randomized, double-blind, placebo-controlled, parallel Phase 3 trial sponsored by AstraZeneca. Enrollment is listed as 768; the recorded start is 2006-04 and primary completion is 2007-09. The listed groups received saxagliptin 2.5 mg or 5 mg with glyburide 7.5 mg, or placebo with glyburide 7.5 mg. Metformin 500-2500 mg was listed as needed in each arm description.

The primary outcome was mean change from baseline in A1C at Week 24, adjusted for baseline. Secondary outcomes covered fasting plasma glucose, achievement of A1C < 7%, and postprandial glucose response. The registry record does not report numerical efficacy results, the statistical analysis plan, or the trial’s sample-size justification. It describes LOCF but does not fully specify rescue-medication handling. Adverse-event counts are recorded by group; death counts are not reported in the registry record.