Saxagliptin Compared to Glimepiride in Elderly Type 2 Diabetes Patients, With Inadequate Glycemic Control on Metformin: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Saxagliptin Compared to Glimepiride in Elderly Type 2 Diabetes Patients, With Inadequate Glycemic Control on Metformin (NCT01006603), a randomized phase 4 study. The primary outcome is Proportion of Patients Reaching HbA1c <7% After 52 Weeks of Treatment Without Confirmed or Severe Hypoglycaemia..

2. Clinical Question

The question is not simply whether patients reached an HbA1c threshold. It joins a measure of glycaemic response to a safety condition: a patient must reach the specified target and avoid the defined hypoglycaemic events to meet the primary endpoint. Consequently, the endpoint represents a combined treatment outcome rather than either glycaemic control or hypoglycaemia considered alone.

The study also lists secondary outcomes addressing hypoglycaemia, HbA1c, fasting plasma glucose, insulin, and beta-cell function. These outcomes could describe different aspects of the comparison, but their numerical results are not reported in the registry record. The primary endpoint is specified for the period from week 0 to week 52, with the outcome determined using central-laboratory results obtained on or before the eighth day after the last dosing day.

The question is framed for an elderly patient group receiving metformin, not for all people with type 2 diabetes or for patients using other background regimens. The registry record does not give the age range, the distribution of baseline glycaemic values, or other characteristics needed to describe how closely the enrolled participants match any particular clinical population.

3. Trial Design

This is a randomized, double-blind, active-controlled, parallel-group Phase 4 study. The official title describes a multicentre trial with a 52-week treatment period. The registry record lists enrollment of 957 and identifies AstraZeneca as the sponsor. The study is marked completed, with a start date of 2009-10 and primary completion in 2012-06.

The two listed treatment groups were saxagliptin 5 mg and glimepiride 1 - 6 mg. An active-controlled design compares two treatments rather than using an inactive control. Random allocation is intended, in general, to reduce systematic differences between groups at the start of a trial. Double masking means the trial was designed to conceal treatment assignment from participants and relevant study personnel, although the record does not specify who was masked or how masking was maintained.

A parallel-group design assigns participants to a treatment group and follows them in that group over the study period, rather than having each participant receive both treatments in sequence. The primary outcome is defined over 52 weeks. The registry record does not provide arm-specific randomized counts, treatment adherence, discontinuation counts, or the number included in each outcome analysis. It therefore does not allow a detailed account of participant flow or exposure to be reconstructed.

The primary endpoint population is called the safety analysis set and is described as a subset of the randomized analysis set including patients who took at least one investigational product dose. The record does not give the size of that set or describe other analysis populations in detail. It also does not report the trial's sample-size justification, planned interim analyses, allocation ratio, or a complete statistical analysis plan. These design details should not be inferred from the general study description or from planning illustrations.

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: A 2-week single-blind (to patient only) placebo lead-in period occurred from Week -2 to Week 0. Patients were from this period on,counselled on dietary and lifestyle modifications according to usual clinical routine. They were given a glucometer to check their plasma glucose at home at least every second day.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety analysis set (a subset of the randomised analysis set including patients who took at least one investigational product dose).

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 Proportion of Patients Reaching HbA1c <7% After 52 Weeks of Treatment Without Confirmed or Severe Hypoglycaemia., measured over From week 0 to week 52..

The primary endpoint is the proportion of participants who reached HbA1c <7% after 52 weeks of treatment without confirmed or severe hypoglycaemia. It is a binary participant-level outcome: each person either meets the combined criteria or does not. The result is intended to reflect whether a specified glycaemic target was achieved while avoiding the adverse events included in the endpoint definition.

HbA1c is used as a measure of glycaemic control over time. A threshold endpoint makes the outcome easy to interpret as the share of participants who crossed a stated target, but it does not show how far above or below that threshold individual values were. The registry separately lists change in HbA1c as a secondary outcome, which would address change rather than simply whether the threshold was met. Its numerical results are not reported in the registry record.

For this endpoint, confirmed hypoglycaemia includes a symptomatic event with blood glucose <3 mmol/L (<54 mg/dL) without need for external assistance, or an asymptomatic blood glucose measurement <3 mmol/L (<54 mg/dL). Severe hypoglycaemia is defined as a symptomatic event requiring external assistance because of severe impairment in consciousness or behaviour, with prompt recovery after glucose or glucagon administration. The definition allows that a plasma glucose measurement may not be available during such an event.

The endpoint is specified for week 0 through week 52 and uses central-laboratory determination obtained on or before the eighth day after the last dosing day. The registry record does not report the number or proportion of participants meeting this combined endpoint in either group, so the comparative result and its precision cannot be described.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods help distinguish the treatment question, the outcome definition, the analysis population, and the uncertainty around an estimate. The registry record specifies some of these elements, but it does not report a complete statistical analysis plan or numerical primary outcome findings. The explanations below describe general statistical concepts and identify what is and is not documented for this study.

What does a proportion measure?

General: a proportion for a binary outcome is the number of participants meeting the outcome divided by the number evaluated for that outcome. For this trial's primary endpoint, meeting the outcome requires both reaching HbA1c <7% and avoiding confirmed or severe hypoglycaemia. A proportion would therefore summarize the share of participants meeting both conditions in a treatment group. The registry record does not report those proportions or the corresponding participant counts.

A proportion is not the same as a change in a continuous measurement. The endpoint classifies participants according to whether they satisfy a threshold and safety condition; it does not express the amount by which HbA1c changed. The listed secondary outcome of change from baseline to week 52 in HbA1c addresses a different question. Results for that secondary outcome are not reported in the registry record.

How should a treatment comparison be read?

General: a treatment comparison can be expressed using a difference in outcome proportions, with an interval estimate describing the uncertainty around that comparison. A confidence interval represents a range of values compatible with the data and the assumptions of the statistical method; it is not a guarantee that the true value lies in that range. A p-value, when reported, evaluates the compatibility of the observed data with a specified null hypothesis under the chosen model. Neither an interval nor a p-value conveys the full clinical importance of an effect by itself.

No primary-outcome comparison, confidence interval, or p-value is reported in the registry record. No recalculated comparison is available from the posted summary information, because the outcome counts for the combined endpoint are absent. The record therefore does not support a numerical assessment of effect size or precision.

Why does the endpoint combine efficacy and safety?

General: a composite endpoint combines multiple conditions into one outcome, which can make a clinically framed question easier to summarize. Its interpretation depends on which components are included and how often each component occurs. Components may differ in frequency or importance, so a single combined result can be difficult to interpret unless component outcomes are also reported. Here, the primary endpoint combines the HbA1c target with freedom from confirmed or severe hypoglycaemia. The registry lists related secondary outcomes, but their numerical results are not reported in the registry record.

The endpoint definition also sets the event rules and assessment timing. Confirmed hypoglycaemia includes symptomatic or asymptomatic glucose measurements below the specified threshold, with the symptomatic definition distinguishing events that require external assistance from those that do not. Severe events are defined by symptoms, need for assistance, and recovery after glucose or glucagon administration. These operational definitions support consistent classification, but the record does not report how many events met each component definition.

What does randomization contribute?

General: randomization assigns participants to groups using a chance process. If implemented as intended, it helps balance measured and unmeasured baseline factors on average, reducing some sources of confounding in a group comparison. It does not guarantee that groups will be identical, and it does not remove uncertainty arising from the number of participants or incomplete outcome data.

The study is identified as randomized, double-blind, and parallel-group. These are design descriptors, not evidence of a particular result. The registry record does not state the allocation sequence method, concealment procedures, or the exact roles of people masked to treatment. It also does not report baseline characteristics by group, so the balance of the enrolled groups cannot be assessed from the record.

What is an analysis population?

General: results depend in part on which participants are included in an analysis. A safety population commonly includes participants exposed to at least one dose, while other analyses may use broader or differently defined populations. The record defines the primary endpoint population as a safety analysis set, a subset of the randomized analysis set that includes patients who took at least one investigational product dose. It does not report the size of this set, the number analyzed in each group, or a separate definition for the randomized analysis set.

The registry record does not describe how missing primary outcome measurements were treated, whether participants were analyzed according to assigned treatment, or whether sensitivity analyses were performed. Those details can affect interpretation when some participants do not have a measured endpoint. No missing-data method should be assumed from the design description alone.

What do the planning assumptions mean?

General: a sample-size calculation uses assumptions about the target effect, statistical error rate, allocation, and analysis method to illustrate how much information may be needed for a planned comparison. The registry record's planning illustration is based on a fraction of the observed effect, uses a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. It is explicitly a planning illustration based on the observed effect, not a description of the trial's own design justification.

The record does not provide numerical planning estimates in the information available for this interpretation, nor does it report the trial's original sample-size justification. A planning illustration is not observed power and does not determine whether a result is statistically or clinically persuasive. Observed power should not be substituted for an effect estimate and its uncertainty.

There are no analyses listed in the registry record for the endpoint results. It also does not report interim analyses, multiplicity procedures, or an adjustment strategy for multiple outcomes. In the absence of these details and the outcome estimates themselves, the methods can be described only at the level of the recorded design and general statistical principles.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcomes, including the combined primary endpoint and secondary measures of hypoglycaemia, HbA1c, fasting plasma glucose, insulin, and beta-cell function. It reports seven outcome measures, while the available list names several secondary outcomes. Numerical results for these outcomes are not reported in the registry record. The record also lists no analyses, so there is no reported set of outcome estimates from which to evaluate the statistical findings.

When a study examines multiple outcomes or multiple comparisons, the chance of at least one apparently unusual result can increase. General methods for addressing this include defining a primary outcome in advance, specifying a testing sequence, or adjusting interpretation for multiple testing. The registry record does not report whether any multiplicity procedure was planned or used. It would be inappropriate to assume that no adjustment was required or that a particular method was applied.

The record does not report how missing data were handled for the primary or secondary outcomes. It gives no counts of participants with missing measurements, no reasons for missingness, and no imputation or sensitivity-analysis methods. Missingness can matter when the probability of having an observed outcome is related to treatment or to the outcome itself. Without the relevant counts and methods, the possible effect of incomplete data cannot be assessed.

The primary endpoint is assigned to a safety analysis set consisting of randomized patients who took at least one investigational product dose. The set's size and its group-specific counts are not reported. The registry record does not clarify whether any efficacy analyses used another population or how treatment discontinuation was handled. These omissions limit assessment of how the analyzed population relates to all randomized participants.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cardiac failure (0.8%), myocardial infarction (0.8%), pneumonia (0.8%), acute myocardial infarction (0.6%), atrial fibrillation (0.6%).

The serious and other adverse-event entries are listed separately for saxagliptin 5 mg and glimepiride 1 - 6 mg. For serious adverse events, the record gives 41 affected participants among 359 at risk in the saxagliptin group and 32 among 359 at risk in the glimepiride group. For other adverse events, it gives 130 among 359 at risk with saxagliptin and 129 among 359 at risk with glimepiride. These are the registry's reported counts; they should not be read as a causal comparison without the study's full safety analyses and context.

The listed serious adverse-event terms include CARDIAC FAILURE, MYOCARDIAL INFARCTION, and PNEUMONIA at 0.8%, and ACUTE MYOCARDIAL INFARCTION, ATRIAL FIBRILLATION, and LUMBAR VERTEBRAL FRACTURE at 0.6%. The registry record does not identify those terms by treatment group in this listing, so they should not be assigned to one arm or interpreted as comparative rates.

Death counts are not reported in the registry record. The summary also does not provide event timing, recurrent-event counts, exposure-adjusted rates, or a detailed account of severity and relatedness beyond the serious-event classification. The figures describe recorded participants affected, not a complete assessment of the balance of safety between treatments.

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

The study addresses a clearly defined comparison: saxagliptin 5 mg versus glimepiride 1 - 6 mg over 52 weeks in elderly patients with type 2 diabetes whose control on metformin was inadequate. Its primary endpoint is a combined measure requiring both an HbA1c value below the stated threshold and no confirmed or severe hypoglycaemia. The record also documents randomization, double masking, and an active comparator.

Those details explain what the trial set out to assess, but they do not supply its primary finding. The proportion meeting the combined endpoint in either group, the treatment comparison, and uncertainty estimates are not reported in the registry record. Accordingly, the record does not establish the direction, magnitude, or precision of a difference in the primary outcome. It also cannot show whether any apparent difference would be driven more by HbA1c target attainment or by hypoglycaemia.

The same limitation applies to the listed secondary outcomes. Although the record names outcomes such as change in HbA1c and the proportion experiencing hypoglycaemia, it does not give numerical results for them. The serious and other adverse-event counts offer some safety information, but they are not a substitute for the missing efficacy results or for a complete comparative safety analysis.

General: a trial result is most informative when the estimated effect is considered alongside its uncertainty, the endpoint's clinical meaning, the analysis population, missing data, and the study's design. No single design feature guarantees a particular treatment effect. Nor does a lack of a reported result mean that the treatments were equivalent; equivalence requires an appropriate design and analysis with a prespecified margin, neither of which is reported.

The findings, if available from another source, would apply most directly to the enrolled population and treatment context described in the registry record. The record does not provide enough demographic or clinical detail to judge generalizability beyond that description. In the absence of comparative outcome results, the sound conclusion is limited: the study was designed to evaluate the stated combined endpoint, but its comparative result is not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable for using a composite endpoint that links a glycaemic target with avoidance of specified hypoglycaemic events. Such an endpoint poses an interpretation challenge: a single success proportion combines efficacy and safety dimensions, and the component outcomes may contribute differently. Separate reporting of component results would help explain what drives a combined estimate, but numerical component results are not reported in the registry record.

The design also illustrates the importance of distinguishing a trial's question from its reported evidence. Randomization, double masking, and an active comparator are relevant features for a treatment comparison, yet they do not provide an estimate of the treatment effect. The registry record describes the planned comparison and gives an overall enrollment count of 957, but it does not report the primary outcome proportions, comparative estimate, or confidence interval.

Statistical interpretation also depends on analysis populations and missing-data procedures. The primary endpoint is assigned to a safety analysis set, but the set size, arm-specific analysis counts, and missing-data handling are not reported. These details are needed to understand who contributed to the result and how incomplete observations might affect it. The record does not report a statistical analysis plan, multiplicity procedures, or interim analyses.

Finally, the registry's planning assumptions should not be mistaken for the original design rationale or a measure of observed evidential strength. The available description characterizes a planning illustration, while the trial's own sample-size justification is not reported in the registry record. With no comparative endpoint results, the statistical importance of this study lies chiefly in its clearly framed question and in the limits imposed by incomplete result reporting.

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.

Several trial-specific limitations of interpretation arise from the registry record. Most importantly, it does not report numerical results for the primary endpoint. There are no group-specific proportions, outcome counts, treatment-effect estimates, confidence intervals, or p-values. The comparison therefore cannot be quantified, and no conclusion about the relative primary outcome can be drawn from this record.

The record does not report participant flow by treatment group, the number randomized to each arm, or the number included in the primary endpoint analysis. Although the endpoint population is defined as a safety analysis set of randomized participants who took at least one investigational product dose, its size is not stated. Discontinuations, adherence, and the completeness of endpoint ascertainment are also not described.

The record does not report the age threshold defining elderly, baseline characteristics by group, or detailed eligibility criteria. That limits assessment of the population represented and of how comparable the groups were at baseline. It also does not report the statistical analysis plan, the sample-size justification, methods for missing data, multiplicity handling, or interim analyses.

The primary outcome is a composite, but the registry does not report its component results. The listed secondary outcomes are not accompanied by numerical findings, so they cannot clarify whether the components or related measures followed similar patterns. Safety information is also limited: deaths are not reported, and the serious-event term listing does not identify treatment groups for each term.

These are limitations of what can be concluded from the registry record, not proof that the study did not collect or analyze these data. The registry record does not report whether further results are available elsewhere, and no additional results should be inferred.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Saxagliptin Compared to Glimepiride in Elderly Type 2 Diabetes Patients, With Inadequate Glycemic Control on Metformin — NCT01006603.
• Registry record last updated 2013-11-28; status last verified 2013-09.
• Results first posted 2013-11-28.
• 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

NCT01006603 is a completed Phase 4, randomized, double-blind, active-controlled, parallel-group study sponsored by AstraZeneca. It enrolled 957 participants and compared saxagliptin 5 mg with glimepiride 1 - 6 mg over 52 weeks in elderly patients with type 2 diabetes and inadequate glycaemic control on metformin monotherapy. The record lists a start date of 2009-10 and primary completion in 2012-06.

The primary endpoint was the proportion reaching HbA1c <7% after 52 weeks without confirmed or severe hypoglycaemia, assessed in a safety analysis set. Comparative results for this endpoint, results for the listed secondary outcomes, and the trial's statistical analysis plan are not reported in the registry record. Serious and other adverse-event counts are reported by group, while death counts are not reported.