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 18-week add-on to Metformin Comparison of Saxagliptin and Sitagliptin in Adult Patients With Type 2 Diabetes (T2D) (NCT00666458), a randomized phase 3 study. The primary outcome is Hemoglobin A1c (HbA1c) Change From Baseline to Week 18.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00666458 |
| Phase | Phase 3 |
| Condition | Type 2 Diabetes |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 822 |
| Arms in the primary outcome | n/a |
| Lead sponsor | AstraZeneca |
| Results posted | Yes |
2. Clinical Question
Population
Adults with type 2 diabetes whose glycaemic control was inadequate on metformin alone. The primary endpoint description focuses on a per-protocol analysis set: randomized participants who completed the 18 weeks of treatment, had both baseline and Week 18 HbA1c measurements, and had no significant protocol deviations.
Intervention
Saxagliptin added to metformin. The registry lists saxagliptin as a drug intervention. The dose and detailed administration schedule are not reported in the registry record.
Comparator
Sitagliptin added to metformin, also listed as a drug intervention. This is an active comparison: both groups received an add-on medicine alongside metformin. The registry record does not describe a placebo-only group.
Primary question
What was the adjusted mean change from baseline in HbA1c at Week 18 in the per-protocol comparison of saxagliptin plus metformin with sitagliptin plus metformin? HbA1c change is defined as the Week 18 value minus the baseline value. A numerical estimate and its precision are not reported in the registry record.
3. Trial Design
The registry describes a completed Phase IIIb, randomized, parallel-group trial of saxagliptin added to metformin compared with sitagliptin added to metformin in adults with type 2 diabetes. The official title describes the study as international, multicentre, double-blind, and active-controlled. The separate masking entry records quadruple masking. These descriptions are both part of the registry record; further detail about who was masked and how masking was maintained is not reported in the registry record.
The recorded enrollment is 822. The study start is listed as 2008-04 and primary completion as 2009-03. AstraZeneca is listed as the sponsor, and the status is completed. The parallel-group structure means participants were assigned to one of the two add-on strategies, rather than crossing between them during the comparison. The allocation entry identifies the trial as randomized, but the sequence-generation and allocation-concealment procedures are not reported in the registry record.
The primary outcome is HbA1c change from baseline to Week 18, measured in percent. Its description specifies an adjusted mean change and the per-protocol analysis set. The registry lists three secondary outcomes: the proportion achieving HbA1c of <= 6.5% at Week 18, and fasting plasma glucose change from baseline to Week 18 in mg/dL and mmol/L. The record reports four outcome measures and four analyses, but does not provide further detail about their statistical models or how the four analyses map to individual outcomes.
The registry does not report treatment doses, the precise randomization ratio, site locations, or a detailed schedule of assessments. Those details should not be inferred from the trial title or design labels.
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
| Stage | Saxa + Met | Sita + Met |
|---|---|---|
| STARTED | 403 | 398 |
| COMPLETED | 365 | 374 |
| NOT COMPLETED | 38 | 24 |
Reasons for not completing this period:
| Reason | Saxa + Met | Sita + Met |
|---|---|---|
| Adverse Event | 8 | 7 |
| Withdrawal by Subject | 5 | 4 |
| Lost to Follow-up | 0 | 3 |
| Administrative reason by sponsor | 1 | 1 |
| Incorrect enrollment | 7 | 1 |
| Study specific discontinuation criteria | 14 | 7 |
| Severe non-compliance to the protocol | 1 | 1 |
| Safety reasons | 1 | 0 |
| Medical history of anemia | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: Randomized participants who completed the 18 weeks of treatment had both baseline and week 18 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 | Saxa + Met | Sita + Met | Total |
|---|---|---|---|---|
| Participants analysed | 403 | 398 | 801 | |
| Age Continuous | years | 58.77 (SD 10.14) | 58.07 (SD 10.51) | 58.42 (SD 10.32) |
| Sex: Female, Male: Female | Participants | 213 | 196 | 409 |
| Sex: Female, Male: Male | Participants | 190 | 202 | 392 |
6. Primary Endpoint
The registry lists the primary outcome as Hemoglobin A1c (HbA1c) Change From Baseline to Week 18, measured over Baseline, Week 18.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Error |
| Unit | Percent |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is change in hemoglobin A1c, or HbA1c, from baseline to Week 18. The registry defines each participant’s change as the Week 18 value minus the baseline value and describes the comparison as an adjusted mean change. The unit is percent. The endpoint population is the per-protocol analysis set, consisting of randomized participants who completed 18 weeks of treatment, had both required HbA1c measurements, and had no significant protocol deviations.
General explanation: HbA1c is a continuous measure, so each participant contributes a numerical value and a change can be calculated between two assessment times. A group mean summarizes the average change, while an adjusted mean can account for variables included in a statistical model. The registry record does not identify which variables, if any, were used for adjustment. It also does not report the endpoint’s numerical group estimates, the between-group contrast, a confidence interval, or a p-value.
HbA1c is used as a measure of glycaemic status over time. In this trial, measuring change from baseline makes the endpoint responsive to differences between the start of treatment and Week 18. The secondary outcomes include a threshold-based HbA1c response and fasting plasma glucose change, but those do not replace the stated primary endpoint.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central statistical question is whether the adjusted mean change in HbA1c from baseline to Week 18 differed between the saxagliptin-plus-metformin group and the sitagliptin-plus-metformin group in the stated per-protocol population. The registry record describes that question and defines how individual change was calculated, but the numerical endpoint results are not reported in the registry record. There is therefore no posted mean change, between-group estimate, confidence interval, or p-value on which to base a quantitative interpretation.
The trial enrolled 822 participants according to the registry. That is the recorded enrollment, not necessarily the number included in the primary analysis. The endpoint description restricts the analysis population to participants who completed 18 weeks, had measurements at both specified times, and had no significant protocol deviations. The number meeting those conditions is not reported in the registry record. The primary estimate, if reported elsewhere, would need to be read in light of this population definition.
General explanation: A treatment contrast estimates the difference between groups on the endpoint; a confidence interval describes the range of values compatible with the data and the analysis assumptions. Precision depends on the amount and variability of the data, among other factors. Without the contrast and its interval, the size and precision of the comparison cannot be assessed. A p-value alone, even if available, would not describe the magnitude of the difference; in this record, a p-value is also not reported.
The registry identifies secondary outcomes involving a threshold of HbA1c <= 6.5% at Week 18 and fasting plasma glucose change in two units. Their numerical results are not reported in the registry record. The record also indicates that the primary outcome has several categories or time points and was not reduced to one comparison. This means the available information does not support selecting a single primary result or reconstructing a full result profile.
The serious adverse event entries give counts for each group, but they do not supply the primary efficacy result. Safety counts and an efficacy estimate address different questions and should be interpreted separately. Overall, the record establishes the trial’s design and intended endpoint, but does not provide enough outcome statistics to characterize the observed HbA1c contrast or its uncertainty.
9. Statistical Methods Explained
The registry record identifies a randomized, parallel-group comparison and an adjusted mean change in HbA1c as the primary endpoint. It does not report the statistical model, adjustment variables, hypothesis test, confidence interval method, or missing-data procedure. The explanations below describe general statistical concepts, not methods confirmed as having been used in this trial.
What does an adjusted mean represent?
General explanation: A mean is an average of measured values. An adjusted mean is an estimate derived from a model that accounts for specified characteristics or other terms. Adjustment can help compare groups while accounting for factors included in the model, but interpretation depends on what was included, how it was coded, and whether the model assumptions are reasonable. Since those details are not reported in the registry record, the specific meaning of adjustment in this analysis cannot be established from the record.
The endpoint description defines individual change as the Week 18 HbA1c value minus the baseline value. A negative change would indicate a lower Week 18 value than baseline, and a positive change would indicate a higher Week 18 value. This is a general interpretation of the stated subtraction, not a claim about the observed trial results. No group means or changes are reported in the registry record.
Why does the analysis population matter?
General explanation: An analysis population specifies which randomized participants contribute to an estimate. The stated per-protocol population excludes participants who did not complete treatment, lacked either required HbA1c measurement, or had significant protocol deviations. Restricting an analysis in this way can make the result describe a selected subset rather than all randomized participants. If completion or protocol deviations are related to participant characteristics or outcomes, the subset may differ systematically from the original randomized groups. The numbers excluded and the reasons for exclusion are not reported in the registry record.
Randomization is intended to support comparability between groups at the start of a trial. An analysis limited after randomization can affect how that protection applies to the analyzed sample. The record does not provide enough information to assess the consequences of the per-protocol restriction, nor does it report an intention-to-treat result for comparison.
How should uncertainty be assessed?
General explanation: An estimate gives a central value for a contrast; a confidence interval communicates its statistical uncertainty. A narrower interval indicates greater precision under the assumptions of the analysis, while a wider interval indicates less precision. The interval’s location and width are needed to assess both the plausible size of a contrast and how well it is distinguished from no difference. Neither an interval nor a numerical contrast is reported for the primary outcome in the registry record.
A p-value describes how unusual the observed data would be under a specified null hypothesis and statistical model. It is not the probability that a treatment effect is true, and it does not measure clinical importance. The registry record does not report a p-value or the testing procedure for the primary endpoint.
What do the planning illustrations mean?
The assumptions recorded for the planning illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They use fractions of the observed effect as planning targets. These are illustrations based on the observed effect, not the trial’s own design calculations. The registry record does not report its sample-size justification, and the registry entries do not state the numerical observed effect used for those illustrations.
General explanation: A sample-size planning illustration asks how many participants per arm might be needed under stated assumptions to detect a target effect with a chosen power. It is not an observed result, and it is not observed power. Planning values depend on the target effect, variability assumptions, allocation, significance level, and analysis method. In the absence of the trial’s original assumptions and justification, such an illustration cannot establish whether the actual trial was designed adequately for its primary endpoint.
Why can over-dispersion matter?
General explanation: Over-dispersion occurs when observed variability is greater than a model assumes. In models for counts or proportions, ignoring over-dispersion can make uncertainty appear too small and can lead to confidence intervals or tests that are too optimistic. That concern is general; the primary endpoint is continuous, and the registry record does not state that an over-dispersion model was used for any analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary HbA1c change outcome and secondary outcomes covering a Week 18 HbA1c threshold and fasting plasma glucose change in mg/dL and mmol/L. It reports four outcome measures and four analyses. However, the record does not describe a multiplicity strategy, such as an ordering of hypotheses, adjustment of significance levels, or a distinction between confirmatory and exploratory analyses. It is therefore not possible to determine from the registry record how the chance of false-positive findings across multiple outcomes or analyses was addressed.
The registry also notes that the primary outcome has several categories or time points and was not reduced to one comparison. Details identifying the relevant categories or time points and the analyses associated with them are not reported in the registry record. No numerical primary or secondary results are reported in the registry record for this account.
The endpoint population requires treatment completion, both baseline and Week 18 HbA1c measurements, and no significant protocol deviations. This implies that participants without the required measurements or with qualifying deviations do not meet the stated population definition. The registry record does not report the amount or pattern of missing data, reasons for missing measurements, or how missing data were handled in any analysis. It also does not report sensitivity analyses that might assess the effect of different missing-data assumptions.
The record does not identify an intention-to-treat analysis or state whether other analysis populations were prespecified. The primary endpoint’s per-protocol definition is available, but the number of participants included in that analysis is not reported in the registry record. These gaps limit assessment of how the randomized groups were represented in the efficacy analysis and how robust the results might be to missing data, exclusions, and multiple comparisons.
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 |
|---|---|---|---|
| Saxa + Met | 7 | 403 | 1.7% |
| Sita + Met | 5 | 398 | 1.3% |
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 |
|---|---|
| Saxa + Met | 54 of 403 |
| Sita + Met | 50 of 398 |
Most frequent serious adverse event terms
| Event term | Organ system | Saxa + Met | Sita + Met |
|---|---|---|---|
| Supraventricular Tachycardia | Cardiac disorders | 1 of 403 | 1 of 398 |
| Chills | General disorders | 1 of 403 | 0 of 398 |
| Ulcer Haemorrhage | General disorders | 1 of 403 | 0 of 398 |
| Faecaloma | Gastrointestinal disorders | 1 of 403 | 0 of 398 |
| ROAD TRAFFIC ACCIDENT | Injury, poisoning and procedural complications | 1 of 403 | 0 of 398 |
| Fall | Injury, poisoning and procedural complications | 0 of 403 | 1 of 398 |
| Hyperglycaemia | Metabolism and nutrition disorders | 1 of 403 | 0 of 398 |
| Hypoglycaemia | Metabolism and nutrition disorders | 0 of 403 | 1 of 398 |
| Hypoglycaemic Unconsciousness | Metabolism and nutrition disorders | 0 of 403 | 1 of 398 |
| Endometrial Cancer | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 1 of 403 | 0 of 398 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Saxa + Met | Sita + Met |
|---|---|---|---|
| Urinary Tract Infection | Infections and infestations | 28 of 403 | 26 of 398 |
| Influenza | Infections and infestations | 27 of 403 | 27 of 398 |
Serious events with the highest arm-level rate: supraventricular tachycardia (0.3%), fall (0.3%), hypoglycaemia (0.3%), hypoglycaemic unconsciousness (0.3%), aortic aneurysm (0.3%).
The registry reports serious adverse events for two groups. In the Saxa + Met group, 403 participants are listed as at risk and 7 as affected. In the Sita + Met group, 398 participants are listed as at risk and 5 as affected. These are the posted counts; they should be read as separate group summaries, not as a comparison adjusted for follow-up time or other factors. A formal between-group safety estimate and its uncertainty are not reported in the registry record.
The record separately lists 54 participants affected by other adverse events among 403 at risk in the Saxa + Met group, and 50 among 398 at risk in the Sita + Met group. The registry record does not provide further definition of the “other” category or a statistical comparison for these entries. Death counts are not reported in the registry record.
Several serious-event terms are listed with percentages: Supraventricular Tachycardia, Fall, Hypoglycaemia, Hypoglycaemic Unconsciousness, and Aortic Aneurysm are each listed as 0.3%; Chills is listed as 0.2%. The record does not identify the group attribution for each term in the registry entries. These term-level percentages should not be treated as comparative risks between groups. The event counts alone also do not establish causation or the relative safety of the add-on strategies.
12. What the Result Does, and Does Not, Mean
The record defines a randomized comparison of saxagliptin plus metformin and sitagliptin plus metformin in adults with type 2 diabetes whose control was inadequate on metformin alone. It identifies Week 18 HbA1c change as the primary endpoint and specifies a per-protocol population. Those details describe the question the trial was intended to address and the population for the stated primary analysis.
They do not establish the observed size or direction of the HbA1c contrast. The primary endpoint’s numerical estimates, uncertainty interval, and p-value are not reported in the registry record. The listed secondary outcomes likewise have no numerical results in the registry record for this account. As a result, the record does not support a conclusion about which group had a larger average change, whether any difference was statistically distinguishable, or how precise the estimate was.
General explanation: Statistical evidence and clinical importance are related but distinct. A difference can be estimated precisely yet be small in practical terms, or estimated imprecisely even when its point estimate appears substantial. Interpreting importance requires the effect estimate, its uncertainty, and relevant clinical context. Those outcome details are not reported in the registry record, so the clinical importance of the comparison cannot be assessed from these registry entries.
The reported serious adverse event counts provide limited descriptive safety information, but not a complete safety comparison. The record does not provide group-specific estimates for individual serious-event terms, an adjusted comparison, or uncertainty around the counts. The absence of such information in this record should not be interpreted as evidence that an event did not occur.
In short, the registry record documents the study framework, endpoint definition, selected analysis population, and some event counts. It does not provide the efficacy results required to characterize the primary comparison, or enough detail to assess the robustness of the analysis. The trial’s completion status and enrollment do not substitute for outcome estimates.
13. Why This Trial Matters Statistically
This trial is statistically notable for posing an active-comparator question rather than comparing an add-on medicine with no active comparator. Both groups are described as receiving a drug in addition to metformin. The intended primary comparison is therefore between two add-on strategies in a population described as having inadequate glycaemic control on metformin alone. The registry record does not report the numerical contrast, so the comparative result itself cannot be interpreted.
The endpoint is a continuous measure of change from baseline to a specified follow-up time. That structure makes the definition of the analysis population and the handling of incomplete measurements important. The primary population is explicitly per protocol and requires completion, paired HbA1c measurements, and no significant protocol deviations. This allows readers to identify the intended population but does not show how many participants qualified or how exclusions affected the comparison.
The record also highlights the distinction between a study’s design and its results. Randomized allocation, a parallel-group structure, and masking describe aspects of how a trial is organized; they do not reveal the size, precision, or importance of an outcome contrast. Similarly, an enrollment figure does not tell readers how much information was available for the primary analysis.
Planning illustrations can be useful for understanding how assumptions relate to sample size, but the registry’s stated illustrations are based on the observed effect and are not the trial’s own design justification. Keeping that distinction clear prevents planning quantities from being mistaken for observed evidence or proof of adequate design.
The statistical value of the registry entry is therefore primarily in documenting the question and analytic population. A fuller assessment would require the numerical outcome estimates, uncertainty, model specification, missing-data information, multiplicity approach, and analysis-population counts, none of which are reported in the registry record.
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 trial-specific limitation for interpretation is that the registry entries do not include numerical results for the primary HbA1c endpoint. The adjusted mean changes, the between-group contrast, its uncertainty, and a p-value are not reported in the registry record. The secondary outcome results are also not reported in the registry entries. Consequently, the magnitude, precision, and direction of the efficacy comparison cannot be described.
The primary endpoint is defined for a per-protocol population that includes participants who completed 18 weeks, had baseline and Week 18 measurements, and had no significant protocol deviations. The number included in that population, the number excluded, and reasons for exclusion are not reported in the registry record. No intention-to-treat result is reported in the registry entries, so the primary analysis cannot be compared with an analysis retaining all randomized participants.
The record does not report the statistical model, adjustment variables, methods for missing data, sensitivity analyses, multiplicity handling, or sample-size justification. The listed planning assumptions are not a substitute for the original design rationale. These omissions prevent assessment of whether the analysis choices were prespecified, how robust the conclusions are to assumptions, and how the number of outcomes or analyses was handled.
The record contains serious adverse event counts but does not give an inferential comparison or group attribution for the listed serious-event terms. Death counts are not reported in the registry record. Further details on adverse event definitions, follow-up duration for safety, and treatment exposure are also not reported in the registry record. The trial title describes double blinding, whereas the separate masking entry records quadruple masking; the record does not explain the terminology or identify masked roles. These limitations constrain interpretation without establishing that the underlying trial lacked the unreported information.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: 18-week add-on to Metformin Comparison of Saxagliptin and Sitagliptin in Adult Patients With Type 2 Diabetes (T2D) — NCT00666458.
- Registry record last updated 2010-04-15; status last verified 2010-03.
- Results first posted 2010-04-15.
- 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 a completed, randomized, parallel-group Phase IIIb trial sponsored by AstraZeneca. It compares saxagliptin added to metformin with sitagliptin added to metformin in adults with type 2 diabetes whose control was inadequate on metformin alone. Recorded enrollment is 822; the study start is 2008-04 and primary completion is 2009-03. The official title describes an international, multicentre, double-blind, active-controlled study, while the masking entry records quadruple masking.
The primary endpoint is adjusted mean HbA1c change from baseline to Week 18 in a per-protocol population. The record defines individual change as the Week 18 value minus baseline and requires completion, both measurements, and no significant protocol deviations. It also lists HbA1c threshold response and fasting plasma glucose change as secondary outcomes.
Numerical primary and secondary efficacy results, statistical uncertainty, model details, missing-data methods, and multiplicity handling are not reported in the registry record. Serious adverse event counts are listed by group, but the record does not provide a formal safety comparison.