Efficacy and Safety of Insulin Glargine/Lixisenatide Fixed Combination Versus Insulin Glargine Alone on Top of Metformin in Type 2 Diabetic Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Insulin Glargine/Lixisenatide Fixed Combination Versus Insulin Glargine Alone on Top of Metformin in Type 2 Diabetic Patients (NCT01476475), a randomized phase 2 study. The primary outcome is Change in HbA1c From Baseline to Week 24.

2. Clinical Question

The study addresses a comparison between two treatment strategies in participants already receiving metformin, not a comparison with no active treatment or with metformin alone. Its open-label design means participants and study staff were not masked to treatment assignment; the registry record states that masking was none. The stated primary focus was the between-group comparison of HbA1c change over 24 weeks. The result therefore pertains to that outcome, population, and follow-up period. It does not, on its own, establish what would happen with other background treatments, other populations, or longer follow-up.

3. Trial Design

This was a randomized, Phase 2, open-label, two-arm parallel-group study described as multicenter. The registry record lists 323 enrolled participants and a completed study. The study compared once-daily insulin glargine/lixisenatide fixed-ratio combination with once-daily insulin glargine; both groups received metformin as background treatment. Study treatment lasted 24 weeks, and doses were individually adjusted.

Random allocation supports a comparison of assigned treatment groups by reducing systematic differences in expectation, although randomization does not guarantee that every measured or unmeasured characteristic will be balanced. The parallel-group design means participants were assigned to one treatment strategy rather than crossing between the two. The record does not report the allocation ratio or the details of the randomization sequence, so these design features should not be inferred from the group counts.

The primary endpoint was change in HbA1c from baseline to Week 24. The posted arm summaries give least-squares mean changes with standard errors for 160 participants in the combination group and 161 in the insulin glargine group. Those analysis counts are endpoint-specific; the registry defines the modified intent-to-treat population as randomized participants who received at least one dose and had baseline and post-baseline efficacy assessment. The record's wording for the exact number analyzed is incomplete, so no more specific population count is asserted.

The registry describes efficacy analyses as on-treatment for this variable: the period began with the first dose and ended before rescue medication was introduced, with follow-up extending up to 14 days after the last injection. The record also states that missing primary endpoint data were imputed by last observation carried forward. The protocol's rationale for these design choices, the original sample-size justification, and details of any interim analyses 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.

Before assignment: A total of 323 participants were randomized in 1:1 ratio to insulin glargine/lixisenatide fixed ratio combination (FRC) and insulin glargine arms, stratified by screening HbA1c values (\<8% or ≥8%) and screening body mass index (BMI) values (\<30 kg/m\^2, ≥30 kg/m\^2).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified intent-to-treat (mITT) population consisted of all randomized participants received at least 1 dose of IMP and had both baseline and at least 1 post-baseline efficacy assessment. Number of participants analyzed = participants with both baseline and at least one post-baseline HbA1c assessment during on-treatment period.

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 in HbA1c From Baseline to Week 24, measured over Baseline, Week 24.

The primary endpoint was change in HbA1c from baseline to Week 24, expressed as a percentage of hemoglobin. The registry defines change by subtracting the baseline value from the Week 24 value. A negative change therefore represents a decrease from baseline in the measure. HbA1c is used in diabetes trials as a measure of longer-term glycemic control, making it a clinically relevant outcome for comparing glucose-lowering strategies. This is a general explanation of the measure, not a claim about the clinical importance of a particular difference in this study.

The posted arm estimates are least-squares means rather than simple raw averages. The registry reports a standard error for each estimate, which describes the estimated sampling uncertainty of that mean under the analysis. These summaries should be interpreted alongside the sponsor's reported ANCOVA comparison, which adjusts the group comparison according to the analysis model. The specific covariates included in that model are not reported in the registry record.

For the primary efficacy variable, the analysis used the modified intent-to-treat population and last observation carried forward for missing data. The on-treatment window excluded assessments from after rescue medication was introduced, while allowing the stated period after the last injection. Thus the endpoint is not simply a comparison of every participant's observed Week 24 value regardless of subsequent treatment or missingness. Interpretation depends on the specified analysis population, treatment window, and imputation rule as well as on the outcome itself.

7. Results Posted in the Registry

• Insulin Glargine: mean -1.64 versus -1.82 on control; difference 0.18 (95% CI 0.02 to 0.34), standardized effect d = 0.25; z-test p = 0.027.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change in 2-hour Postprandial Plasma Glucose (PPG) From Baseline to Week 24

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Insulin Glargine: mean -4.33 versus -7.49 on control; difference 3.16 (95% CI 2.39 to 3.93), standardized effect d = 0.92; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Insulin Glargine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in 2-hour Plasma Glucose Excursion From Baseline to Week 24

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Insulin Glargine: mean -0.67 versus -3.91 on control; difference 3.24 (95% CI 2.48 to 4.00), standardized effect d = 0.96; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Insulin Glargine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in Average 7-Point Self-Monitored Plasma Glucose (SMPG) Profiles From Baseline to Week 24

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Insulin Glargine: mean -2.93 versus -3.23 on control; difference 0.30 (95% CI 0.02 to 0.58), standardized effect d = 0.24; z-test p = 0.039.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Insulin Glargine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Across the posted secondary continuous outcomes, the arm summaries generally show decreases from baseline in both groups. The reported comparisons for meal-related glucose, glucose excursion, and average seven-point self-monitored glucose favor a larger decrease in the combination group, based on the outcome change direction and the listed contrast order. The record also lists other secondary outcomes, including body weight, daily insulin dose, fasting plasma glucose, rescue therapy, and HbA1c threshold outcomes, but their results are not included among the posted secondary comparisons summarized in the analysis data.

This overall pattern is directionally consistent with the primary HbA1c result: both the primary measure and the posted meal-related glucose outcomes indicate greater reductions in the combination group during the 24-week period. The secondary measures provide related but distinct information. Meal-test outcomes reflect glucose around a standardized meal, while the seven-point self-monitored profile summarizes measurements across the day. They do not substitute for the primary endpoint, and concordant directions do not prove that every outcome is independently established.

Secondary comparisons generally carry less confirmatory weight when the registry does not report a prespecified testing hierarchy or multiplicity adjustment. When many outcomes are tested, some apparently notable results may arise by chance. The registry reports no multiplicity procedure, so the secondary p-values should be interpreted as supportive or exploratory rather than as definitive evidence across all secondary objectives. This caution is separate from the size or direction of any one estimate.

The number of secondary outcomes listed is broader than the number with posted comparison analyses. The reasons for this difference are not reported in the registry record. Without a complete analysis plan and results for all outcomes, it is not possible to determine how outcome selection or testing order affected the pattern. The secondary findings can add context to the primary result, but they cannot resolve missing information about the analysis plan, clinical importance, or longer-term outcomes.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change in 2-hour Postprandial Plasma Glucose (PPG) From Baseline to Week 24
• Change in 2-hour Plasma Glucose Excursion From Baseline to Week 24
• Change in Average 7-Point Self-Monitored Plasma Glucose (SMPG) Profiles From Baseline to Week 24
• Change in Body Weight From Baseline to Week 24
• Average Daily Insulin Glargine Dose at Week 24
• Change in FPG From Baseline to Week 24
• Percentage of Participants Requiring Rescue Therapy During 24-week Treatment Period
• Percentage of Participants With HbA1c ≤6.5 % or <7.0 % at Week 24

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it records a measured quantity on a numeric scale. The registry reports least-squares mean changes for each arm and an ANCOVA estimate of the between-group difference. The methods explain these statistical terms in general, then distinguish what is and is not reported for this study.

What does an ANCOVA comparison do?

In general, analysis of covariance (ANCOVA) compares group outcomes while accounting for selected covariates in a statistical model. Adjustment can improve precision when the covariates are related to the outcome, and it can account for baseline differences included in the model. The adjusted estimate is conditional on the model and covariates actually used. The registry identifies ANCOVA as the method for the primary reported comparison, but the specific covariates, model formula, and handling of model assumptions are not reported in the registry record. It is therefore not possible to reproduce that adjusted estimate from the arm means alone.

The reported primary least-squares mean difference is -0.17, with a 95% confidence interval from -0.312 to -0.037. The direction depends on the order in which groups are contrasted. With the combination group listed first, a negative difference is consistent with a larger reduction in that group because more negative change denotes a larger decrease from baseline. The unadjusted recalculation uses the arm summaries and expresses the contrast as insulin glargine minus combination, so its positive value points in the same substantive direction. Opposite signs can describe the same ordering of group changes when the subtraction order is reversed.

What does the standard error describe?

In general, a standard error describes the estimated uncertainty of a statistic due to sampling variation, under the analysis and its assumptions. It is not the spread of individual participant outcomes. A standard error can be used to construct a confidence interval when the relevant distributional method is specified. The registry reports standard errors for the arm-level least-squares means, but it does not provide confidence limits for those individual arm estimates. The adjusted comparison does have a reported confidence interval.

The posted recalculation estimates a pooled standard deviation from the arm-level standard errors using an approximation. It is a simple unadjusted analysis of the posted summaries, not the trial's ANCOVA. Such a recalculation is useful for a transparent comparison of the displayed arm estimates, but it cannot recover information about covariate adjustment, participant-level data, or the exact analysis population beyond what the record reports.

What do the p-value and confidence interval contribute?

In general, a p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions are correct. It is not the probability that the null hypothesis is true, and it does not measure the size or practical importance of an effect. The reported primary p-value is 0.013. The corresponding 95% confidence interval gives a range of values compatible with the data and model at the stated confidence level; it also communicates uncertainty in the estimate. The interval excludes zero, which is consistent with the reported p-value for a test of no difference, though the registry does not provide the full testing specification.

The primary analysis includes a missing-data rule. In general, last observation carried forward assigns a participant's last observed value to a later missing assessment. This can retain participants in an analysis but relies on an assumption that the carried-forward value is a reasonable substitute for the unobserved value. When outcomes may change after the last observation, the imputation can misrepresent that later trajectory. The registry states that LOCF was used for the primary HbA1c endpoint; it does not report sensitivity analyses under alternative missing-data assumptions.

How should the analysis population be understood?

The registry defines modified intent-to-treat as randomized participants who received at least one dose and had baseline and at least one post-baseline efficacy assessment. In general, restricting an analysis to participants with observed efficacy information can make the analyzed group differ from everyone randomized, particularly if missingness or discontinuation is related to treatment or outcome. The exact consequences depend on reasons for missing data and the analysis method. Those reasons and a detailed accounting of missing primary measurements are not reported in the registry record.

The primary efficacy variable also had an on-treatment window. As defined in the record, this began at first dose and excluded data after rescue medication was introduced, while including follow-up up to 14 days after the last injection. This definition makes the endpoint analysis different from an analysis that includes all observations through the scheduled endpoint regardless of rescue treatment. The record does not describe additional treatment-policy analyses or sensitivity analyses.

What limits interpretation of multiple outcomes?

The registry lists a primary outcome and multiple secondary outcomes. In general, testing many outcomes can increase the chance that at least one result appears statistically notable by chance if no adjustment or prespecified hierarchy is used. The registry lists selected secondary comparisons and p-values, but a multiplicity adjustment or testing hierarchy is not reported in the registry record. Secondary p-values should therefore be treated as exploratory evidence rather than as equivalent in confirmatory weight to the primary result.

Finally, the randomized open-label parallel design informs how the comparison should be read. Randomization helps support a causal comparison of assigned groups, subject to adherence, missingness, and analysis choices. Open-label assignment can influence participant or investigator behavior and reporting, particularly for outcomes that require subjective assessment. The primary outcome is a laboratory measure, but no further details about outcome measurement safeguards are reported. These general considerations do not establish that bias occurred; they clarify which design features and missing information shape confidence in interpretation.

11. Confidence Intervals and Precision

The primary adjusted estimate is -0.17 percentage points, with a 95% confidence interval from -0.312 to -0.037. The interval is entirely below zero in the registry's listed contrast order, supporting a difference in the direction of a larger HbA1c reduction for the combination group. The confidence interval also shows that the estimated contrast is not known with exact precision: values across a range remain compatible with the data and model. It is not a prediction interval for individual participant outcomes.

In general, a confidence interval combines an estimate with its sampling uncertainty under a specified statistical model. A narrower interval indicates greater precision for that estimate, while a wider interval indicates more uncertainty. The interval's interpretation depends on the model, analysis population, outcome definition, and assumptions. It does not account automatically for bias from missing data, outcome measurement, or selective reporting.

The posted arm-level standard errors are 0.058 for the combination group and 0.057 for insulin glargine. These describe uncertainty in the reported least-squares mean changes, not the range of individual HbA1c changes. The sponsor's adjusted confidence interval and the simple unadjusted recalculation answer related but not identical statistical questions. The recalculated contrast is based on summary statistics and does not include the adjustment in ANCOVA. The registry record reports no discrepancy flag, but the analyses should still not be treated as interchangeable.

Precision should also be distinguished from clinical importance. The record does not report a threshold defining a clinically meaningful difference in HbA1c for this comparison. Consequently, the interval can be used to assess statistical uncertainty around the group contrast, but it cannot by itself determine the practical importance of the estimated difference. The time horizon is 24 weeks, so the interval likewise provides no direct information about precision of effects beyond the observed follow-up.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm for a two-arm continuous-outcome comparison; they are not total event counts. They are illustrations based on the observed effect, not the original study's sample-size justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. Each row considers detecting the stated fraction of the observed effect. For the primary outcome, the figures range from 258 participants per arm for 80% power and 345 participants per arm for 90% power when planning for 100% of the observed effect, to larger requirements when planning for smaller fractions.

In general, power is the probability that a planned statistical test will reject its null hypothesis under a specified true effect and analysis assumptions. The 80% power column gives the estimated number of participants per arm under which the design would have an 80% chance of detecting the assumed effect at the stated significance level. The 90% column represents a higher chance of detecting that same assumed effect, and therefore requires more participants under the same assumptions.

The rows for 75% and 50% of the observed effect show how the planning requirement changes when the assumed effect is smaller. Smaller effects are harder to distinguish from random variation, so more participants are needed to reach the same target power. These are hypothetical planning scenarios and are not observed power. The original trial's own design assumptions and sample-size justification are not reported in the registry record. The planning values should not be read as evidence about the probability that this completed trial detected an effect, nor as a recommendation for a future study without considering its design, outcome, population, and assumptions.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.180; pooled standard deviation σ = 0.728.
2. n = 2 × (1.96 + 0.842)² × 0.728² / 0.180² = 257.1, rounded up to 258 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 0.18                      # observed difference in means
sigma <- 0.728450926522231                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1476475)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 258 (80% power), 345 (90% power)
 75% of observed effect: n per arm = 458 (80% power), 612 (90% power)
 50% of observed effect: n per arm = 1,029 (80% power), 1,377 (90% power)
power.t.test: 258.1 
Simulated power at n = 258 per arm: 0.799 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry's sponsor-reported primary comparison is an ANCOVA estimate of -0.17, with a 95% confidence interval from -0.312 to -0.037 and p = 0.013. The simple recalculation from the posted arm summaries gives an unadjusted difference of 0.18 when expressed as insulin glargine minus the combination group, with an unadjusted p-value of 0.027. The signs differ because the contrasts are written in opposite group orders; both estimates point toward a larger reduction in the combination arm.

The analyses are not identical. The recalculation uses the group-level means and standard errors and does not reproduce the sponsor's covariate-adjusted ANCOVA model. The sponsor's analysis is the registry-reported analysis for the study and should take precedence when describing the trial's primary result. The recalculation is a check based on posted summaries, not a replacement for the participant-level analysis.

The registry's discrepancy flag is false. The difference between the unadjusted and adjusted estimates is not, by itself, a contradiction: an adjusted analysis can differ from an unadjusted summary comparison. The record does not report enough model detail to reproduce the ANCOVA or explain the exact contribution of adjustment.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome and a range of secondary outcomes. Testing multiple outcomes creates a multiplicity issue: in general, the more statistical tests conducted, the greater the chance of obtaining at least one apparently significant result by chance when all null hypotheses are true. A prespecified hierarchy or statistical adjustment can address this concern, but the registry record does not report a multiplicity adjustment, a testing hierarchy, or how secondary hypotheses were handled. The primary result is therefore distinct in status from the many secondary comparisons, whose p-values should be read with greater caution.

For the primary HbA1c outcome, the record states that missing data were imputed using last observation carried forward. In general, this method carries the last observed measurement forward to replace a later missing value. Its validity depends on assumptions about what would have happened after the last measurement; it may be inappropriate if the outcome would have changed. The record does not report sensitivity analyses using alternative imputation methods, the number or reasons for missing measurements, or how missingness differed by treatment group.

The primary efficacy population is described as modified intent-to-treat: randomized participants who received at least one dose and had baseline and at least one post-baseline efficacy assessment. The number analyzed is described as participants meeting those measurement criteria, but the registry wording is incomplete. The endpoint analysis also uses an on-treatment definition that excludes measurements after rescue medication was introduced and includes a stated period after the last injection. How these choices affect the estimate cannot be established from the available registry detail.

Other unreported elements include the original sample-size justification, any interim analyses, the full statistical analysis plan, model covariates, and sensitivity analyses. Each can matter when assessing whether a result is robust and how confirmatory it should be considered. Their absence does not show that a particular analysis was or was not performed; it means that these details are not reported in the registry record.

16. 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: cellulitis (0.6%), ovarian cancer (0.6%), diabetic neuropathy (0.6%), sciatica (0.6%), angina pectoris (0.6%).

The registry lists serious adverse events affecting 9 participants in the insulin glargine/lixisenatide combination group and 6 participants in the insulin glargine group. The corresponding groups at risk are listed as 161 and 162. These are participant counts, not event rates calculated for a person-time period, and the registry does not provide a comparative statistical analysis of these figures. The counts should be read descriptively rather than as evidence that one treatment caused more or fewer serious events.

The listed serious-event terms include cellulitis, ovarian cancer, diabetic neuropathy, sciatica, angina pectoris, and unstable angina, each reported as 0.6%. The registry record does not indicate from these summary terms whether the events were judged related to treatment. Death counts are not reported in the registry record.

Other adverse events are reported for 26 participants in the combination group and 21 participants in the insulin glargine group, among 161 and 162 participants at risk, respectively. These counts alone do not describe severity, timing, duration, or exposure-adjusted risk. The study lasted 24 weeks, and the registry record does not provide enough detail in these summaries to draw a broader comparative safety conclusion.

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

The primary registry analysis reports a difference in HbA1c change over 24 weeks between the two treatment groups on metformin. The adjusted estimate and its confidence interval favor a larger reduction in the combination group, and the reported p-value is 0.013. Both groups' posted mean changes are decreases from baseline. This is a description of the measured outcome and the statistical comparison, not a determination of whether the size of the difference is important to participants.

The result does not establish effects beyond the study's 24-week follow-up or in populations and treatment settings not described in the registry record. It does not show how the comparison would differ under another background therapy, and it does not answer outcomes that were not reported. The open-label design, the specified modified intent-to-treat population, the on-treatment window, and the LOCF rule are relevant to interpretation.

Statistical evidence is not the same as clinical importance. The registry does not report a threshold for a clinically meaningful HbA1c difference, and the confidence interval cannot supply that threshold. Nor does a statistically significant primary result mean that every participant experienced the estimated group-average change; group means do not describe individual responses.

The adverse-event counts provide a limited descriptive account of serious and other events recorded in the groups. They are not a complete evaluation of comparative safety, and the registry summaries do not give a causal assessment for the listed serious-event terms. The evidence should therefore be read as a bounded comparison of posted efficacy and safety information, with unanswered questions about longer-term outcomes, analysis robustness, and clinical relevance.

18. Why This Trial Matters Statistically

This trial illustrates why the estimated treatment contrast, rather than the within-group changes alone, is central to a randomized comparison. Both groups had negative mean HbA1c changes, but the key statistical question is whether their changes differed. The registry's ANCOVA estimate addresses that between-group question, while the unadjusted recalculation from the arm summaries provides a related but not equivalent check.

It also shows why analysis direction must be explicit. A negative contrast can indicate a greater reduction when one group is subtracted from another, while reversing the subtraction reverses the sign. Correct interpretation requires reading the group order together with the fact that a more negative change means a larger decrease from baseline.

The example also highlights the difference between a point estimate and its precision. A confidence interval describes uncertainty under the reported model; a p-value alone does not communicate the range of plausible effect sizes or whether those sizes matter clinically. Sample-size planning illustrations further show, in general, that planning for a smaller effect requires more participants to achieve a chosen power target. Those illustrations are separate from the original design justification, which is not reported in the registry record.

Finally, endpoint multiplicity, missing-data choices, analysis populations, and open-label design all shape how strongly results can be interpreted. These are not merely technical details: they define which question the statistical result answers and what uncertainty remains. The registry record offers a useful account of the primary comparison but does not include all information needed to fully assess its analysis plan or the broader evidence base.

19. 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 registry record does not report the full statistical analysis plan, the covariates used in the primary ANCOVA, the randomization sequence details, or the original sample-size justification. The allocation ratio is also not stated. These omissions limit independent reconstruction of the primary analysis and assessment of whether analysis decisions were prespecified.

The primary analysis used LOCF for missing HbA1c data and an on-treatment efficacy window that excluded measurements after rescue therapy began. The record does not describe reasons for missingness, the extent of missing data by group, or sensitivity analyses under alternative assumptions. The modified intent-to-treat definition requires at least one post-baseline assessment, so it does not include every randomized participant by definition. The exact primary analyzed population count is not clearly stated in the registry wording.

The study was open-label, and the record does not report details of safeguards for outcome assessment. HbA1c is a laboratory measure, but the available information does not establish how measurement and analysis were operationalized. The registry also does not report a multiplicity plan despite listing multiple secondary outcomes, and only selected secondary comparisons have results in the posted analysis data.

Follow-up was 24 weeks. The registry record therefore does not establish longer-term efficacy or safety. The adverse-event summary does not provide comparative inferential analyses or causal assessments of the listed serious events. The record also does not report a threshold for clinical importance of the primary contrast. These are limits on interpretation of the posted record, not evidence that any specific bias or harm occurred.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of Insulin Glargine/Lixisenatide Fixed Combination Versus Insulin Glargine Alone on Top of Metformin in Type 2 Diabetic Patients — NCT01476475.
• Registry record last updated 2017-02-10; status last verified 2016-12.
• Results first posted 2017-02-10.
• 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.

23. Record Summary

The registry describes a completed, randomized, open-label, parallel-group Phase 2 study in participants with type 2 diabetes treated with metformin. It compared once-daily insulin glargine/lixisenatide fixed-ratio combination with individually adjusted once-daily insulin glargine over 24 weeks. The listed enrollment is 323.

The primary endpoint was change in HbA1c from baseline to Week 24. The reported ANCOVA estimate was -0.17, with a 95% confidence interval from -0.312 to -0.037 and p = 0.013, favoring a larger reduction in the combination group under the listed contrast order. Missing primary endpoint data were imputed using LOCF, and the analysis used a modified intent-to-treat population with an on-treatment efficacy window.

The record includes selected secondary efficacy comparisons and descriptive adverse-event counts. Multiplicity procedures, sensitivity analyses, the full statistical plan, and a clinical-importance threshold are not reported in the registry record. Results should be interpreted within the stated population, analysis choices, and 24-week follow-up.