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 Insulin Glargine "All to Target" Trial (NCT00384085), a randomized phase 4 study. The primary outcome is Percentage of Patients Achieving Glycosylated Hemoglobin A1c (HbA1c) < 7.0% at Week 60 (Lantus/Apidra-3 Versus Novolog Mix 70/30 - Intent To Treat (ITT) Population Without Good Clinical Practices (GCP) Noncompliant Sites).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00384085 |
| Phase | Phase 4 |
| Condition | Diabetes Mellitus, Type 2 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 588 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Sanofi |
| Results posted | Yes |
2. Clinical Question
Population
Adults with type 2 diabetes whose condition was not adequately controlled with dual or triple therapy with oral agents. The primary analysis population was described as randomized participants with post-baseline follow-up information, excluding participants from non-GCP-compliant sites.
Intervention
Insulin glargine (Lantus) with up to 3 injections of mealtime insulin glulisine (Apidra), added to oral agents. The registry describes this as the Lantus/Apidra-3 regimen.
Comparator
Twice-daily premixed insulin (Novolog Mix 70/30), added to oral agents. A second insulin glargine regimen, with up to 1 mealtime insulin glulisine injection, was also compared with premixed insulin for a separate objective.
Primary question
At Week 60, what proportion of participants in the Lantus/Apidra-3 group achieved HbA1c below 7.0%, compared with the Novolog Mix 70/30 group? Participants with a missing Week 60 HbA1c measurement were classified as nonresponders. The registry record states an objective of demonstrating superiority for this comparison, but does not report the result or a statistical estimate for it.
The trial also described a noninferiority objective comparing Lantus/Apidra-1 with Novolog Mix 70/30, based on change in HbA1c from baseline to Week 60. The noninferiority margin and the result for that comparison are not reported in the registry record. The clinical questions therefore involve both a responder outcome and a continuous HbA1c change outcome, with different insulin regimens and comparison groups. These questions should not be treated as interchangeable: achieving the specified HbA1c threshold is a binary classification, while change in HbA1c retains the size and direction of measured change.
3. Trial Design
The Insulin Glargine “All to Target” Trial was a Phase 4, randomized, parallel-group, open-label clinical study. The registry lists an enrollment of 588 and describes the study as lasting 64 weeks. The recorded start was 2006-05, primary completion was 2010-03, and status is completed. The sponsor was Sanofi.
Participants were adults with type 2 diabetes described as failing dual or triple therapy with oral agents. The three regimens were insulin glargine plus up to 3 mealtime insulin glulisine injections, insulin glargine plus up to 1 mealtime insulin glulisine injection, and twice-daily premixed insulin. Each regimen was added to oral agents. The trial was randomized, but the allocation method beyond that designation is not reported in the registry record. Masking is listed as none, so the study was open-label.
The principal responder endpoint compared the Lantus/Apidra-3 regimen with Novolog Mix 70/30 at Week 60. The registry also describes a separate objective comparing Lantus/Apidra-1 with Novolog Mix 70/30 using baseline-to-Week 60 HbA1c reduction. These objectives differ in both treatment contrast and outcome definition. The registry record does not report the allocation ratio, details of treatment adjustment, adherence, or the statistical plan used to test either objective.
The main responder analysis was described as using an Intent-To-Treat population of randomized participants with any post-baseline follow-up information, with participants from non-GCP-compliant sites excluded. The record says an additional analysis including those participants was completed, but its result is not reported in the registry record. The exclusion of those sites is relevant when interpreting the stated primary analysis population: it is not identical to all randomized participants.
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: Patient were considered enrolled after informed consent were obtained, and randomized after completion of the 4-week run-in phase and assessment of the randomization criteria.
Overall Study
| Stage | Lantus/Apidra-3 | Lantus/Apidra-1 | Novolog Mix 70/30 |
|---|---|---|---|
| STARTED | 195 | 196 | 197 |
| Intent-To-Treat (ITT) | 195 | 196 | 197 |
| Treated (Safety Population) | 194 | 194 | 194 |
| Intent-To-Treat (Excluding nonGCP Sites) | 187 | 186 | 189 |
| COMPLETED | 156 | 150 | 141 |
| NOT COMPLETED | 39 | 46 | 56 |
Reasons for not completing this period:
| Reason | Lantus/Apidra-3 | Lantus/Apidra-1 | Novolog Mix 70/30 |
|---|---|---|---|
| Adverse Event | 5 | 5 | 4 |
| Lack of Efficacy | 0 | 3 | 2 |
| No longer Requires Study Treatment | 1 | 0 | 1 |
| Protocol Violation | 2 | 2 | 2 |
| Lost to Follow-up | 9 | 9 | 10 |
| Progressive Disease | 0 | 1 | 0 |
| Withdrawal by Subject | 14 | 19 | 29 |
| Patient non-compliant | 4 | 4 | 5 |
| Physician Decision | 1 | 0 | 0 |
| Loss of site personnel | 1 | 0 | 1 |
| Sponsor decision | 1 | 2 | 1 |
| Patient moved | 1 | 0 | 1 |
Analysis population for the primary outcome, as defined in the registry: Analysis was performed on Intent-To-Treat population which consisted of all patients who were randomized and for whom there was any post-baseline follow-up information. Patients from nonGCP compliant sites were excluded from this analysis. Additional analysis including those patients were completed to ensure that study results were not compromised.
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 | Lantus/Apidra-3 | Lantus/Apidra-1 | Novolog Mix 70/30 | Total |
|---|---|---|---|---|---|
| Participants analysed | 195 | 196 | 197 | 588 | |
| Age Continuous | years | 54.9 (SD 10.54) | 53.7 (SD 9.11) | 53.7 (SD 10.65) | 54.1 (SD 10.12) |
| Sex: Female, Male: Female | Participants | 78 | 88 | 86 | 252 |
| Sex: Female, Male: Male | Participants | 117 | 108 | 111 | 336 |
| Region of Enrollment: United States | participants | 195 | 196 | 197 | 588 |
| Weight | kilograms (kg) | 95.2 (SD 19.21) | 97.2 (SD 20.88) | 97.9 (SD 20.49) | 96.8 (SD 20.21) |
| Body Mass Index (BMI) | Kilogram/m^2 | 32.7 (SD 5.79) | 33.4 (SD 6.00) | 33.4 (SD 5.62) | 33.1 (SD 5.80) |
| Duration of diabetes at study entry | years | 9.4 (SD 6.80) | 9.1 (SD 5.70) | 9.5 (SD 5.87) | 9.3 (SD 6.13) |
| Age at onset of diabetes | years | 46.0 (SD 10.74) | 45.1 (SD 9.36) | 44.8 (SD 10.01) | 45.3 (SD 10.05) |
| Oral antidiabetic treatment combination at study entry: sulfonylurea (SU) or meglitinide plus metformin | participants | 97 | 99 | 97 | 293 |
| Oral antidiabetic treatment combination at study entry: SU or meglitinide plus thiazolinedione (TZD) | participants | 9 | 9 | 10 | 28 |
| Oral antidiabetic treatment combination at study entry: TZD plus metformin | participants | 26 | 27 | 27 | 80 |
| Oral antidiabetic treatment combination at study entry: SU or meglitinide plus metformin plus TZD | participants | 63 | 61 | 63 | 187 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Patients Achieving Glycosylated Hemoglobin A1c (HbA1c) < 7.0% at Week 60 (Lantus/Apidra-3 Versus Novolog Mix 70/30 - Intent To Treat (ITT) Population Without Good Clinical Practices (GCP) Noncompliant Sites), measured over At week 60.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | percentage of participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is the percentage of participants in the Lantus/Apidra-3 group and the Novolog Mix 70/30 group who achieved HbA1c below 7.0% at Week 60. HbA1c is a measure of glycemic control; the endpoint turns each participant’s Week 60 value into a responder or nonresponder classification according to the stated threshold. Participants who did not reach the threshold, and those with a missing Week 60 HbA1c value, were classified as nonresponders.
In general, a binary responder endpoint can make a clinical target easy to interpret: it reports the fraction of participants meeting a prespecified criterion rather than the average amount of change. That simplicity has a trade-off. Participants just on either side of the cutoff are placed in different categories, while variation in HbA1c within each category is not represented by the responder proportion. The registry also lists a separate HbA1c change objective, which addresses a different aspect of outcome.
The registry describes this endpoint as the basis for a superiority objective. However, the proportion achieving the threshold, the between-group effect estimate, its uncertainty, and the test result are not reported in the registry record. Consequently, the endpoint definition can be explained, but the extent or precision of any difference between the regimens cannot be assessed from the recorded information.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record gives the definition and analysis population for the primary responder endpoint, but it does not give the Week 60 response proportions or a between-group estimate. It therefore does not establish from the posted results whether the proportion achieving HbA1c below 7.0% differed between Lantus/Apidra-3 and Novolog Mix 70/30, nor how precisely any difference was estimated. A stated objective to demonstrate superiority is a study aim, not evidence that superiority was demonstrated.
The second objective concerns Lantus/Apidra-1 versus Novolog Mix 70/30 and baseline-to-Week 60 HbA1c reduction. Its result and the noninferiority margin are not reported in the registry record. Without the margin and the relevant estimate and uncertainty, the noninferiority question cannot be evaluated from this record. A noninferiority claim depends on comparing the uncertainty around the treatment contrast with a prespecified margin; merely describing the objective does not supply that evidence.
In general, an effect estimate describes the observed contrast, while a confidence interval conveys the range of values reasonably compatible with the data under the method used. A p-value addresses compatibility with a specified null hypothesis and model; it is not the probability that a treatment is effective or that the null hypothesis is true. None of these inferential results for the primary comparison is reported in the registry record. The posted information therefore supports interpretation of the endpoint and study design, but not a quantitative claim about the treatment contrast.
The analysis population also matters. The primary endpoint description says participants from non-GCP-compliant sites were excluded, despite describing an Intent-To-Treat population of randomized participants with post-baseline follow-up information. The record notes that an analysis including those participants was completed, but does not report its findings. Readers cannot judge whether the choice of analysis population affected the conclusion or how the additional analysis compares with the primary analysis.
The registry lists serious adverse event counts and other adverse event counts by regimen, as well as selected serious-event terms and percentages. These are descriptive safety figures, not substitutes for the missing efficacy estimates. The record does not give deaths, and does not state enough about event definitions, follow-up exposure, or comparative safety analyses to support a full assessment of relative risk. Counts from separate categories should not be combined or treated as participant totals without more information.
Overall, the posted results do not provide the numerical primary outcome needed to assess the stated superiority objective. They also do not provide the numerical result for the separate noninferiority objective. Interpretations should remain limited to what the record describes: the planned comparisons, endpoint definitions, analysis population, and recorded adverse-event figures.
9. Statistical Methods Explained
In general, randomized allocation is intended to make treatment groups comparable, on average, with respect to factors that could affect an outcome. Randomization can reduce confounding in a treatment comparison, but it does not guarantee identical groups, prevent missing outcome data, or eliminate bias from other design and analysis choices. The registry identifies the study as randomized and parallel-group; further details of the allocation process are not reported in the registry record.
What does the responder endpoint represent?
The primary endpoint classifies participants according to whether their Week 60 HbA1c was below 7.0%. A binary outcome records whether the criterion was met, rather than the exact measured value for each participant. In general, group response proportions can be compared using an absolute difference, a ratio, or another prespecified contrast. Each expresses the comparison differently, and the analysis method should match the endpoint and design. The registry record does not report the primary response proportions, effect estimate, confidence interval, or test result, so no such contrast can be interpreted for this trial from the recorded information.
The responder definition explicitly treats missing Week 60 HbA1c values as nonresponse. This is a rule for classifying the endpoint, not a recovery of the unobserved measurement. In general, this approach can make missingness part of the outcome definition and may affect the estimated response proportion if missingness differs between groups or is related to participants’ outcomes. The record does not report the number of missing Week 60 measurements, their distribution by group, or analyses using other assumptions for missing outcomes.
Why distinguish superiority from noninferiority?
Superiority asks whether the data support a specified direction of difference between treatments under a chosen statistical model and significance criterion. Noninferiority asks whether the data rule out an unacceptable loss relative to a comparator, using a prespecified margin. In general, noninferiority requires careful attention to the margin, analysis populations, adherence, and confidence interval because these features can affect the conclusion. The registry describes a superiority objective for the responder comparison and a noninferiority objective for HbA1c reduction in a different regimen comparison. The noninferiority margin, analysis details, and results are not reported in the registry record.
A confidence interval is a method-based expression of uncertainty around an estimate. Narrower intervals generally indicate greater precision, while wider intervals indicate less precision. A confidence interval must be considered alongside the endpoint and the clinically relevant comparison; an interval does not by itself show whether a difference is important. Because the registry does not report the primary estimate or its interval, readers cannot assess the precision of that comparison. The same limitation applies to the separate HbA1c-change objective.
What is the role of the planning illustrations?
In general, sample-size planning uses assumptions about the outcome, effect to detect, significance level, allocation, and statistical model. The planning assumptions described for the illustrations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is expressed as a fraction of the observed effect. These are planning-illustration assumptions based on an observed effect, not evidence about the trial’s original design. The registry record does not report the trial’s own sample-size justification or the assumptions used when the study was planned. Any such illustrations should be understood as planning scenarios, not as observed power or proof that a result is precise.
In general, power is a design property calculated under specified assumptions about the effect, variability or event frequency, significance threshold, and analysis. It is not a measure of the probability that a particular finding is true. Power calculated after observing an effect can be misleading when used to characterize the evidence; the estimate and its uncertainty are more informative. The record does not report a trial-specific power calculation or sample-size justification.
How should multiple outcomes be considered?
The record lists two primary objectives and several secondary outcome topics, including HbA1c measures and hypoglycemia outcomes. When multiple outcomes or comparisons are tested, the chance of at least one apparently notable finding can increase if no multiplicity strategy is used. In general, prespecifying a hierarchy or adjusting inference can address this issue, but those choices affect interpretation. The registry record does not report whether multiplicity was controlled, how the two primary objectives were ordered, or how secondary analyses were handled. It would therefore be inappropriate to infer a statistical plan from the list of objectives alone.
Over-dispersion is a general concern for event-count outcomes when observed variability exceeds what a simple count model assumes. It can make standard errors too small if not handled appropriately. The registry lists adjusted incidence and hypoglycemic event-rate outcomes, but does not report model specifications or dispersion assessments. This general issue is not a claim that over-dispersion occurred in this trial; it illustrates why analysis details matter for event-rate endpoints.
10. Multiplicity, Missing Data and Analysis Populations
The registry describes more than one primary objective: superiority for the Week 60 HbA1c responder proportion comparing Lantus/Apidra-3 with Novolog Mix 70/30, and noninferiority for baseline-to-Week 60 HbA1c reduction comparing Lantus/Apidra-1 with Novolog Mix 70/30. It also lists secondary outcomes. The hierarchy of the objectives, any multiplicity adjustment, and the planned approach to interpreting multiple tests are not reported in the registry record. In general, when several comparisons are examined, the method for controlling or interpreting multiplicity affects how strongly isolated findings should be treated as confirmatory.
For the responder endpoint, participants with missing Week 60 HbA1c values were classified as nonresponders. This rule is explicit, but the amount and pattern of missingness are not reported in the registry record. Nor are sensitivity analyses under alternative assumptions about missing outcomes reported. The rule does not reveal the HbA1c values that would have been observed had measurements been available.
The main analysis population is described as randomized participants with any post-baseline follow-up information, excluding participants from non-GCP-compliant sites. An additional analysis including those participants was completed, but its result is not reported in the registry record. The record does not report how many participants were excluded from the primary analysis, or whether conclusions differed between analyses. Thus, the stated population definition can be considered, but the robustness of findings to the site exclusion cannot be assessed.
The registry record does not report the statistical analysis plan, trial-specific sample-size justification, interim analyses, or the noninferiority margin. Those details should not be inferred from the study objectives or the planning assumptions described for sample-size illustrations.
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 |
|---|---|---|---|
| Lantus/Apidra-3 | 22 | 194 | 11.3% |
| Lantus/Apidra-1 | 23 | 194 | 11.9% |
| Novolog Mix 70/30 | 21 | 194 | 10.8% |
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 |
|---|---|
| Lantus/Apidra-3 | 140 of 194 |
| Lantus/Apidra-1 | 150 of 194 |
| Novolog Mix 70/30 | 152 of 194 |
Most frequent serious adverse event terms
| Event term | Organ system | Lantus/Apidra-3 | Lantus/Apidra-1 | Novolog Mix 70/30 |
|---|---|---|---|---|
| Angina pectoris | Cardiac disorders | 1 of 194 | 3 of 194 | 2 of 194 |
| Hypoglycaemia | Metabolism and nutrition disorders | 2 of 194 | 1 of 194 | 1 of 194 |
| Hypoglycaemic unconsciousness | Metabolism and nutrition disorders | 1 of 194 | 0 of 194 | 3 of 194 |
| Acute myocardial infarction | Cardiac disorders | 2 of 194 | 1 of 194 | 0 of 194 |
| Angina unstable | Cardiac disorders | 1 of 194 | 2 of 194 | 0 of 194 |
| Coronary artery occlusion | Cardiac disorders | 1 of 194 | 1 of 194 | 0 of 194 |
| Myocardial infarction | Cardiac disorders | 1 of 194 | 1 of 194 | 0 of 194 |
| Cellulitis | Infections and infestations | 1 of 194 | 1 of 194 | 0 of 194 |
| Urinary tract infection | Infections and infestations | 0 of 194 | 2 of 194 | 0 of 194 |
| Cerebrovascular accident | Nervous system disorders | 1 of 194 | 1 of 194 | 0 of 194 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Lantus/Apidra-3 | Lantus/Apidra-1 | Novolog Mix 70/30 |
|---|---|---|---|---|
| Any infections and infestations | Infections and infestations | 66 of 194 | 82 of 194 | 90 of 194 |
| Any musculoskeletal and connective tissue disorders | Musculoskeletal and connective tissue disorders | 45 of 194 | 44 of 194 | 40 of 194 |
| Any gastrointestinal disorders | Gastrointestinal disorders | 33 of 194 | 32 of 194 | 30 of 194 |
| Any nervous system disorders | Nervous system disorders | 39 of 194 | 29 of 194 | 27 of 194 |
| Any general disorders and administration site conditions | General disorders | 21 of 194 | 28 of 194 | 25 of 194 |
| Any respiratory, thoracic and mediastinal disorders | Respiratory, thoracic and mediastinal disorders | 19 of 194 | 27 of 194 | 23 of 194 |
| Upper respiratory tract infection | Infections and infestations | 18 of 194 | 20 of 194 | 25 of 194 |
| Sinusitis | Infections and infestations | 6 of 194 | 19 of 194 | 16 of 194 |
| Nasopharyngitis | Infections and infestations | 15 of 194 | 17 of 194 | 8 of 194 |
| Oedema peripheral | General disorders | 14 of 194 | 11 of 194 | 10 of 194 |
Serious events with the highest arm-level rate: angina pectoris (1.5%), hypoglycaemic unconsciousness (1.5%), acute myocardial infarction (1.0%), angina unstable (1.0%), urinary tract infection (1.0%).
The registry reports serious adverse events for Lantus/Apidra-3 as 22 among 194 at risk, for Lantus/Apidra-1 as 23 among 194 at risk, and for Novolog Mix 70/30 as 21 among 194 at risk. It also reports other adverse events as 140 among 194 at risk, 150 among 194 at risk, and 152 among 194 at risk, respectively. These are the recorded counts and denominators; the registry record does not provide a comparative estimate or confidence interval for these figures.
The listed serious-event terms and reported percentages are angina pectoris (1.5%), hypoglycaemic unconsciousness (1.5%), acute myocardial infarction (1.0%), angina unstable (1.0%), urinary tract infection (1.0%), and hypoglycaemia (1.0%). The registry record does not report deaths. It also does not explain whether the term-specific percentages represent participants, events, or another calculation, or whether participants could contribute to more than one term. These terms should therefore be read as the registry’s reported categories, not added together or interpreted as a total.
In general, adverse-event counts depend on how events are defined, collected, and followed. Counts alone do not show event timing, duration, severity beyond the recorded category, or comparative risk adjusted for exposure. The record does not report those details or a formal statistical comparison of safety between regimens. These figures describe recorded events but do not establish that the regimens have equivalent or different safety profiles.
12. What the Result Does, and Does Not, Mean
The registry record establishes the broad design and questions of the trial: a randomized, parallel, open-label comparison of insulin regimens in adults with type 2 diabetes whose control was inadequate on oral agents. It defines a Week 60 responder endpoint for the comparison of Lantus/Apidra-3 with Novolog Mix 70/30 and describes a separate HbA1c-change noninferiority objective for Lantus/Apidra-1 versus Novolog Mix 70/30.
What the record does not establish is the result of either primary objective. The primary response proportions, comparative estimate, confidence interval, and test result are not reported in the registry record. The noninferiority margin and numerical result are also not reported. The stated aims should not be mistaken for findings, and no conclusion about the size, direction, or precision of the treatment contrasts can be drawn from the available outcome details.
In general, a result for a threshold-based endpoint applies to the defined response criterion and analysis population. It does not by itself describe the full distribution of HbA1c values, the change for every participant, or outcomes beyond the assessed timeframe. Likewise, a noninferiority result would require the estimate and uncertainty to be assessed against the prespecified margin; the objective alone cannot establish noninferiority.
The recorded adverse-event counts and listed serious-event terms provide descriptive safety information. Without details about event definitions, exposure, missing safety data, and comparative analyses, they do not establish whether one regimen is safer or less safe than another. No medical advice or treatment recommendation follows from the registry record alone.
Accordingly, the most defensible interpretation is limited: the registry documents the trial’s design, endpoint definitions, analysis-population description, and selected adverse-event figures, but does not report the efficacy results needed to answer the central comparative questions. The information does not establish benefit, lack of benefit, equivalence, or a safety difference.
13. Why This Trial Matters Statistically
Statistically, this trial is notable for placing two kinds of treatment question in the same study record. One asks whether a regimen produces a higher proportion of participants meeting a defined HbA1c threshold; another asks whether a different regimen is noninferior to the comparator in HbA1c reduction. Those questions require different estimands and different standards for interpretation. A responder proportion and a continuous change measure should not be treated as interchangeable summaries of glycemic outcome.
The endpoint definition also illustrates how missing-data rules affect a binary outcome. Classifying participants with missing Week 60 HbA1c as nonresponders is explicit and reproducible as a rule, yet it does not eliminate uncertainty about unobserved outcomes. The amount and distribution of missingness, plus sensitivity analyses, would help readers assess how dependent the result is on that classification. Those details are not reported in the registry record.
The analysis population raises another useful statistical consideration. Although the registry describes an Intent-To-Treat population, it excludes participants at non-GCP-compliant sites from the primary analysis and notes a separate analysis including them. The result of that additional analysis is not reported. Comparing analyses across populations can help assess robustness, but the record does not allow that assessment.
Finally, the distinction between stated objectives and reported results is central to evidence interpretation. A superiority aim or noninferiority aim does not show that the corresponding criterion was met. Without effect estimates and uncertainty, statistical significance, precision, and clinical relevance cannot be evaluated. This record is therefore useful for understanding the trial’s questions and design, while its missing comparative outcomes sharply limit what can be concluded about those questions.
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.
A central limitation of the registry record is that it does not report the numerical result for the primary Week 60 responder comparison. It also does not report the result of the separate HbA1c-change noninferiority objective or the noninferiority margin. These omissions prevent evaluation of the treatment contrasts, their precision, or whether either objective was met.
The primary analysis excluded participants from non-GCP-compliant sites, while the registry says an additional analysis including those participants was completed. The result of that additional analysis is not reported in the registry record, so the effect of this exclusion on the findings cannot be assessed. The number of participants excluded and the reasons for site noncompliance are also not reported in the registry record.
The responder definition treats missing Week 60 HbA1c as nonresponse, but the record does not report missing-measurement counts by group, patterns of follow-up, or sensitivity analyses. The allocation ratio, allocation procedures, treatment adherence, treatment adjustment details, and statistical analysis plan are not reported in the registry record. These omissions make it difficult to evaluate implementation and analytic choices beyond the stated design features.
The record describes a 64-week study and an endpoint at Week 60, but does not report outcome estimates or uncertainty. It also does not report the trial’s own sample-size justification, interim analyses, multiplicity strategy, or the hierarchy for its two primary objectives. The recorded safety figures lack information on event definitions, exposure time, and formal comparative analyses; deaths are not reported in the registry record. These limitations are specific to the information available for this trial and should constrain interpretation rather than be filled by assumptions.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Insulin Glargine "All to Target" Trial — NCT00384085.
- Registry record last updated 2011-05-06; status last verified 2011-05.
- Results first posted 2011-04-26.
- 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 Insulin Glargine “All to Target” Trial was a Phase 4, randomized, parallel, open-label study sponsored by Sanofi. The registry lists enrollment of 588, a 64-week study, status as completed, a start of 2006-05, and primary completion in 2010-03. It studied insulin glargine with up to 3 mealtime insulin glulisine injections, insulin glargine with up to 1 such injection, and twice-daily premixed insulin, each added to oral agents.
The stated primary questions concerned Week 60 HbA1c response for Lantus/Apidra-3 versus Novolog Mix 70/30 and HbA1c reduction for Lantus/Apidra-1 versus Novolog Mix 70/30. The registry record defines the responder rule and analysis population, but does not report the primary efficacy result, the noninferiority margin, or the numerical result for the second objective. It lists selected adverse-event counts and terms, while deaths are not reported in the registry record. Interpretation is therefore limited to the documented design, endpoint definitions, and descriptive safety information.