Efficacy and Safety of Alogliptin Used in Combination With Sulfonylurea in Participants With Type 2 Diabetes in Japan: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Alogliptin Used in Combination With Sulfonylurea in Participants With Type 2 Diabetes in Japan (NCT01318083), a randomized phase 2, phase 3 study. The primary outcome is Change From Baseline in Glycosylated Hemoglobin (Week 12)..

2. Clinical Question

The central statistical distinction is between describing the change within each group and estimating a difference between groups. The group means describe average baseline-to-week 12 change; the mean differences and confidence intervals address the uncertainty in contrasts. The comparison of the two alogliptin doses is relatively small in the posted summaries and its confidence interval includes no difference. Comparisons with the glimepiride group are larger in the recalculated summaries, but those estimates should be read with attention to the difference between an unadjusted re-calculation and the sponsor’s reported analysis. The registry record does not report enough detail to reconstruct the sponsor’s full analysis procedure.

3. Trial Design

The registry describes a phase 2/3, randomized, double-blind, placebo-controlled, parallel-group, multicenter study in participants with type 2 diabetes in Japan. It was sponsored by Takeda and is recorded as completed. Enrollment was 312. The listed treatment groups were alogliptin 12.5 mg once daily with glimepiride, alogliptin 25 mg once daily with glimepiride, and glimepiride alone. Glimepiride was taken once or twice daily, with the listed dose range of 1, 2, 3 or 4 mg in the comparator group.

Randomization means assignment to groups was determined by an allocation process rather than by participant or clinician choice. A parallel-group design assigns participants to a group and compares outcomes across groups. Double-blind describes masking, while the registry’s masking field says quadruple; the roles included in that masking designation are not reported in the registry record. The registry labels the study placebo-controlled, but the arm descriptions available for the primary outcome identify glimepiride alone as the comparator. More detail about any placebo procedures is not reported in the registry record.

The primary endpoint was change from baseline in glycosylated hemoglobin at week 12. Values were drawn from the Full Analysis Set, with last observation carried forward for missing scheduled postbaseline values. The registry lists secondary assessments of glycosylated hemoglobin at earlier visits, fasting plasma glucose at several visits, and blood glucose during a meal tolerance test at week 12. There are 9 outcome measures and 18 analyses recorded. The original sample-size rationale, randomization ratio, allocation-concealment procedure, and detailed statistical plan are not reported in the registry record. The posted sample-size illustrations use assumptions for planning and should not be mistaken for the assumptions used to design this trial.

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: Participants with a historical diagnosis of type 2 diabetes mellitus with uncontrolled blood glucose despite treatment with sulfonylurea as well as diet and exercise therapies were enrolled in one of 3, once-daily (QD) or twice daily (BID) treatment groups.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis based on last observation carried forward, where the last postbaseline double-blind observed value is carried forward and used for all subsequent scheduled time points where data is missing. Values are from the Full Analysis Set.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Glycosylated Hemoglobin (Week 12)., measured over Baseline and Week 12..

Glycosylated hemoglobin is the concentration of glucose bound to hemoglobin, expressed as a percentage of the absolute maximum that can be bound. Because it reflects glucose exposure over time rather than a single measurement, it is commonly used as a continuous outcome in studies of glucose-lowering interventions. That is general context; the registry identifies change in glycosylated hemoglobin as this study’s primary outcome.

The endpoint compares each participant’s value at baseline with the value at week 12 or the final visit. The analysis reports group means for the change, along with standard deviations. A negative mean change indicates that the recorded value at follow-up was lower than baseline on average; a positive mean indicates an increase. These signs describe the direction of the measured change, not a clinical judgment by themselves.

The registry states that the primary analysis used the Full Analysis Set and applied last observation carried forward: the last observed postbaseline double-blind value was carried forward to scheduled time points with missing data. This approach makes an assumption that the last available measurement can stand in for a later missing measurement. General statistical guidance is that this assumption may not represent the true outcome trajectory for every participant, particularly if missingness is related to health or response. The registry does not report the amount or pattern of missing primary outcome data, so the impact of this procedure cannot be determined from the posted record alone.

7. Results Posted in the Registry

• Alogliptin 25 mg QD and Glimepiride QD or BID: mean -0.66 versus -0.60 on control; difference -0.06 (95% CI -0.21 to 0.09), standardized effect d = -0.11; z-test p = 0.443.
• Glimepiride 1, 2, 3 or 4 mg QD or BID: mean 0.34 versus -0.60 on control; difference 0.94 (95% CI 0.78 to 1.10), standardized effect d = 1.60; z-test p = <0.001.

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 From Baseline in Glycosylated Hemoglobin (Week 2).

• 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 deviations.
• Result:
  • Alogliptin 25 mg QD and Glimepiride QD or BID: mean -0.13 versus -0.12 on control; difference -0.01 (95% CI -0.07 to 0.05), standardized effect d = -0.04; z-test p = 0.749.
  • Glimepiride 1, 2, 3 or 4 mg QD or BID: mean 0.08 versus -0.12 on control; difference 0.20 (95% CI 0.14 to 0.26), standardized effect d = 0.85; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Alogliptin 25 mg QD and Glimepiride QD or BID and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Glimepiride 1, 2, 3 or 4 mg QD or BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Glycosylated Hemoglobin (Week 4).

• 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 deviations.
• Result:
  • Alogliptin 25 mg QD and Glimepiride QD or BID: mean -0.36 versus -0.32 on control; difference -0.04 (95% CI -0.13 to 0.05), standardized effect d = -0.11; z-test p = 0.407.
  • Glimepiride 1, 2, 3 or 4 mg QD or BID: mean 0.09 versus -0.32 on control; difference 0.41 (95% CI 0.31 to 0.51), standardized effect d = 1.16; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Alogliptin 25 mg QD and Glimepiride QD or BID and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Glimepiride 1, 2, 3 or 4 mg QD or BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Glycosylated Hemoglobin (Week 8).

• 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 deviations.
• Result:
  • Alogliptin 25 mg QD and Glimepiride QD or BID: mean -0.59 versus -0.52 on control; difference -0.07 (95% CI -0.20 to 0.06), standardized effect d = -0.14; z-test p = 0.309.
  • Glimepiride 1, 2, 3 or 4 mg QD or BID: mean 0.18 versus -0.52 on control; difference 0.70 (95% CI 0.56 to 0.84), standardized effect d = 1.38; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Alogliptin 25 mg QD and Glimepiride QD or BID and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Glimepiride 1, 2, 3 or 4 mg QD or BID than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes extend the glycosylated hemoglobin assessment to earlier visits and include fasting plasma glucose at multiple visits and blood glucose measured during a meal tolerance test at week 12. In broad terms, the posted glycosylated hemoglobin summaries show changes developing across the measured visits, and the group pattern is directionally consistent with the week 12 primary summaries. Fasting plasma glucose and meal tolerance test outcomes address related but distinct aspects of glucose measurement; their inclusion broadens the outcome set rather than replacing the primary endpoint.

The sponsor-reported secondary glycosylated hemoglobin contrasts against glimepiride are also posted for the earlier assessments. These should be considered alongside the primary result but not as independent confirmation without qualification. Repeated measurements across time and multiple outcomes are statistically related, and the number of analyses increases the opportunity for chance findings if each is interpreted separately.

The registry records 8 secondary outcomes and 18 analyses overall. It does not report a multiplicity procedure, a hierarchy for testing secondary outcomes, or whether particular secondary analyses were confirmatory. Without that information, unadjusted secondary p-values or intervals carry less weight than a prespecified, multiplicity-controlled primary analysis. The registry also does not provide enough information to assess whether all secondary outcomes were measured and reported as planned.

The secondary findings can help describe whether the primary pattern appears across related measures and assessment times. They cannot, by themselves, resolve uncertainty in the primary analysis, establish a mechanism, or demonstrate that an observed difference is clinically important. The registry’s missing-data and analysis details for the secondary endpoints are not reported in the registry record, so those results should be interpreted with the same caution about analytic assumptions and selective interpretation.

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

• Change From Baseline in Glycosylated Hemoglobin (Week 2).
• Change From Baseline in Glycosylated Hemoglobin (Week 4).
• Change From Baseline in Glycosylated Hemoglobin (Week 8).
• Change From Baseline in Fasting Plasma Glucose (Week 2).
• Change From Baseline in Fasting Plasma Glucose (Week 4).
• Change From Baseline in Fasting Plasma Glucose (Week 8).
• Change From Baseline in Fasting Plasma Glucose (Week 12).
• Change From Baseline in Blood Glucose Measured by the Meal Tolerance Test (Week 12).

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous, so the registry reports a mean change and standard deviation for each group. A mean is the arithmetic average of the measured changes. The standard deviation describes how dispersed individual measurements are around that average; it is not the uncertainty interval for the mean. General explanation: a group can have a precisely estimated mean even when individual outcomes vary, because the precision of the mean also depends on how many observations contribute and how variable they are.

The re-calculated contrasts use posted group summaries to estimate unadjusted mean differences and confidence intervals. They are simple comparisons and are not a reconstruction of an adjusted model. For the alogliptin dose comparison, the estimated mean difference is -0.06, with a 95% confidence interval from -0.213 to 0.093 and p = 0.443. The interval crosses zero; the data therefore do not distinguish this contrast from zero at conventional levels using this unadjusted calculation. That is not the same as proving that the dose groups are identical. The registry does not report an equivalence margin or a non-inferiority margin.

A confidence interval gives a range of values reasonably compatible with the data under the calculation’s assumptions. General explanation: a 95% confidence interval is generated by a procedure that, over repeated studies under the same assumptions, would cover the true parameter in 95% of repetitions. It is not a probability statement that the specific fixed parameter has a 95% chance of lying in this one interval. A narrow interval indicates greater precision than a wide interval, but neither the interval nor the p-value resolves design or measurement limitations.

The p-value quantifies how unusual data at least as discrepant as those observed would be under a specified null hypothesis and model assumptions. It is not the probability that the null hypothesis is true, nor a measure of clinical importance. A p-value below a threshold does not establish that every modelling assumption is appropriate. The recalculated p-values are unadjusted; the registry record does not report whether or how the sponsor adjusted for multiple comparisons.

Over-dispersion is a general concern when the variability in data is greater than assumed by a statistical model. If a model underestimates variability, standard errors and confidence intervals may be too small and statistical evidence may appear stronger than it is. The primary measure is continuous, with group standard deviations posted; no over-dispersion adjustment or model details are reported in the registry record. Accordingly, over-dispersion is a general methodological consideration, not a reported finding about this study.

The primary population is the Full Analysis Set. The registry’s stated missing-data procedure carries the last observed postbaseline value forward to later scheduled visits when data are missing. This is a single imputation: it fills a missing value with an earlier observed value rather than representing uncertainty about the missing outcome. General statistical guidance recognizes that this can understate uncertainty and can bias an estimate if the carried-forward value is not a reasonable substitute. The registry does not report the amount of missing data, the reasons for missingness, or sensitivity analyses using other assumptions.

The sponsor’s reported primary comparisons are mean differences in final values, with 95% confidence intervals, while the posted recalculations use group summary statistics for change from baseline. The statistical method field is not stated for the sponsor analyses. The numerical comparisons should therefore be treated as related summaries, not assumed to be identical estimands or analyses. The sponsor’s own analysis takes precedence when describing its reported result; the re-calculations are useful as transparent, unadjusted checks of the posted group summaries.

The planning figures apply a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The assumed effect is the stated fraction of the observed effect. These are planning illustrations derived from observed results, not the original study design assumptions. General explanation: power is the probability that a specified analysis would reject a null hypothesis under a particular assumed effect and model. The planning figures do not describe the probability that this completed study detected an effect, and do not replace interpretation of the observed confidence intervals.

Multiplicity arises when many outcomes, time points, or pairwise contrasts are assessed. General explanation: if each test is interpreted as though it were the only test, the chance of at least one false-positive result across a family of tests can increase. Methods such as familywise error control or false-discovery-rate procedures can address this issue, but the registry does not state whether these were used. The secondary findings should therefore be read as supportive descriptive evidence unless their multiplicity plan is known.

11. Confidence Intervals and Precision

The confidence intervals express the range of values compatible with the posted comparison under the re-calculation assumptions. For the alogliptin 25 mg versus 12.5 mg contrast, the mean difference is -0.06 and the 95% confidence interval is -0.213 to 0.093. Because the interval includes zero, it leaves uncertainty about whether the underlying average difference is slightly negative, slightly positive, or near zero. The p-value is 0.443. The result is not evidence that the groups are exactly alike; it indicates limited evidence of a difference in this unadjusted comparison.

The sponsor-reported primary contrasts against glimepiride have 95% confidence intervals of -1.097 to -0.775 for the 12.5 mg combination group and -1.160 to -0.837 for the 25 mg combination group. Both intervals exclude zero. Their negative signs reflect the sponsor’s stated comparison ordering, which differs from the ordering used in the simple re-calculation. The sponsor’s analysis method is not reported in the registry record, so precision should be interpreted in the context of that missing analytical detail.

General explanation: confidence-interval width depends on the number of observations, outcome variability, and the analysis method. A confidence interval describes statistical precision, not whether a difference is important in practice. The registry does not report a threshold for a clinically important change, so the intervals cannot be classified against such a threshold. The primary outcome’s last-observation-carried-forward procedure may also affect uncertainty, and the registry does not report sensitivity analyses that would show how robust the estimates are to other missing-data assumptions.

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 sample-size illustrations are planning calculations based on observed effects, not the original trial’s design rationale. They use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They show participant counts per arm, not total event counts. The assumed effect is set at 100%, 75%, or 50% of the observed effect for the contrast being considered. The original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the approximate participants per arm required for an analysis to have an 80% chance of detecting the assumed effect under the stated planning assumptions. The 90% power column gives the corresponding count for a 90% chance. These are hypothetical design calculations for future planning, not a measure of the probability that the completed study found a true effect. This is not observed power.

Smaller assumed effects require more participants because they are harder to distinguish from random variation. That general pattern appears across the illustrations: planning for 75% or 50% of an observed effect requires more participants per arm than planning for 100% of that effect. The figures depend strongly on the effect estimate used as their starting point. An observed effect may be imprecise, and using it as a planning target can yield unstable sample-size estimates, particularly when the original estimate is small or uncertain.

The illustrations do not account for details not included in their stated assumptions, such as any multiplicity adjustment, loss to follow-up, or alternative analysis model. The registry’s own design assumptions are not reported, so these figures should not be used to infer why the study enrolled 312 participants or whether that enrollment met its original statistical objective.

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 = -0.060; pooled standard deviation σ = 0.564.
2. n = 2 × (1.96 + 0.842)² × 0.564² / -0.060² = 1,384.6, rounded up to 1,385 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 <- -0.0600000000000001                      # observed difference in means
sigma <- 0.563500221827818                     # 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(1318083)
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 = 1,385 (80% power), 1,854 (90% power)
 75% of observed effect: n per arm = 2,462 (80% power), 3,296 (90% power)
 50% of observed effect: n per arm = 5,539 (80% power), 7,415 (90% power)
power.t.test: 1385.6 
Simulated power at n = 1,385 per arm: 0.800 (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 sponsor’s registry-reported primary analyses compare each alogliptin-plus-glimepiride group with the glimepiride group. For alogliptin 12.5 mg, the reported mean difference in final values is -0.936, with a 95% confidence interval from -1.097 to -0.775. For alogliptin 25 mg, the reported mean difference is -0.998, with a 95% confidence interval from -1.160 to -0.837. The registry does not give p-values for these analyses or specify the statistical method.

The re-calculation uses the posted group means and standard deviations to form simple unadjusted comparisons. Its direction depends on which arm is subtracted from which: a contrast expressed as combination minus glimepiride has the opposite sign from glimepiride minus combination. The sponsor’s negative estimates and the positive re-calculated contrast for the 12.5 mg group can therefore reflect ordering rather than contradictory results. The registry reports no discrepancy flag, but the re-calculated estimate is not a substitute for the sponsor’s own analysis.

The two analyses are not necessarily identical in estimand or method. The sponsor’s records describe final-value mean differences, while the re-calculation is based on change-from-baseline summaries. Since the sponsor method is not reported, it is not possible to establish how covariate adjustment, missing data, or other analysis choices were handled. The sponsor’s own analysis takes precedence as the registry’s reported trial analysis; the re-calculation is best treated as a simple check and should be interpreted with its assumptions visible.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and multiple secondary outcomes, including measurements at several follow-up times. It also records 18 analyses. Testing multiple outcomes, time points, and pairwise contrasts raises a multiplicity question: when each test is considered separately, the overall chance of a false-positive result can be greater than for a single test. General statistical practice may address this through a prespecified testing sequence or adjustment, but whether such a procedure was used is not reported in the registry record.

The record does not state which comparisons were designated as confirmatory, whether the two alogliptin doses were compared under a formal hypothesis, or whether secondary outcomes were exploratory. It also does not report interim analyses or stopping rules. These aspects of the statistical plan cannot be inferred from the number of results posted.

For the primary analysis, the registry specifies the Full Analysis Set and last observation carried forward for missing scheduled postbaseline values. It does not give the number of participants with missing week 12 data, the reasons for missingness, or the timing of last observations. It also does not report sensitivity analyses under other missing-data assumptions. General explanation: carrying a previous value forward treats that value as if it represented a later scheduled measurement. If outcomes would have changed after the last observation, this can affect the estimated mean change and its uncertainty.

The analysis population and missing-data handling for each secondary outcome are not reported in the registry record. Detailed model specifications, covariate adjustments, assumptions about variance, and the sponsor’s method for the reported comparisons are also not reported. These omissions limit assessment of whether the posted estimates account for repeated measures or other features of the data. They do not establish that a particular method was absent; they mean the registry record does not document it.

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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: angina pectoris (1.0%), cerebral infarction (1.0%), lung neoplasm malignant (1.0%), breast cancer (1.0%).

The serious adverse event counts are reported by treatment group alongside the number at risk. The registry lists 3 of 105 participants in the alogliptin 12.5 mg group, 1 of 104 in the alogliptin 25 mg group, and 0 of 103 in the glimepiride group as affected by serious adverse events. These are counts of participants affected, not a comparison adjusted for follow-up time or other factors. No deaths values are posted; death information is not reported in the registry record.

The listed serious adverse event terms include angina pectoris, cerebral infarction, lung neoplasm malignant, and breast cancer, each reported as 1.0%. The registry also reports other adverse events affecting 27 of 105, 40 of 104, and 36 of 103 participants in the respective groups. These figures are descriptive. They do not establish that treatment caused an event, and the registry summary does not provide enough detail to compare event duration, severity beyond the serious classification, timing, or exposure-adjusted rates.

Small differences in counts should not be interpreted as evidence of a safety advantage or disadvantage. The record’s safety information is a summary and does not report the full clinical context, adjudication procedures, or a detailed statistical comparison. It supports describing what was recorded, not drawing a treatment recommendation.

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

The primary result describes average change in glycosylated hemoglobin over the study’s baseline-to-week 12 assessment. Group means and confidence intervals provide different kinds of information: the means describe the observed average changes, while intervals express uncertainty in estimated contrasts. The recalculated alogliptin dose comparison has a confidence interval that includes zero, so the posted summary does not provide clear evidence of a difference between those dose groups under the simple unadjusted calculation.

The registry’s sponsor-reported comparisons against glimepiride have confidence intervals that exclude zero. Those analyses are the sponsor’s reported results, but their method is not specified in the registry. The simple re-calculation is not an equivalent substitute, and comparison signs depend on the order in which groups are contrasted. These findings should therefore be understood as statistical descriptions of the recorded outcomes rather than as a complete assessment of treatment effects.

A statistically distinguishable mean difference is not automatically clinically important. The registry does not state a threshold for a clinically meaningful change in glycosylated hemoglobin, so the results cannot be judged against such a threshold from this record alone. The posted values also do not establish how effects might vary across participants, how durable changes are after week 12, or how results would generalize beyond the study population and setting.

The primary analysis used last observation carried forward for missing values. The amount and pattern of missing data are not reported, and no sensitivity analyses are described. The multiplicity approach is also not reported. Those gaps matter because both missing-data assumptions and multiple testing can affect confidence in the apparent pattern of results.

The safety summaries list serious adverse events and other adverse events by group, but counts alone do not establish causation or provide a complete comparison of harms. Taken together, the registry record gives a focused account of a randomized study and its posted results, but it does not contain enough methodological detail to settle all questions about robustness, clinical importance, or safety.

18. Why This Trial Matters Statistically

This trial is statistically instructive because its primary outcome is continuous and reported with group means, standard deviations, and between-group comparisons. It illustrates why the estimated effect, its confidence interval, and the p-value should be read together. The contrast between the two alogliptin dose groups is close to zero relative to its interval, while the sponsor-reported contrasts against glimepiride are more separated from zero. Statistical interpretation depends on which comparison is being asked, not merely on whether a result is labelled significant.

The record also highlights the importance of identifying the estimand and comparison direction. A mean difference changes sign when the subtraction order changes. In addition, final-value comparisons and change-from-baseline comparisons may not be identical, especially if the analysis includes adjustments that are not visible in summary statistics. Keeping the sponsor analysis distinct from an unadjusted re-calculation prevents a summary calculation from being mistaken for the trial’s formal analysis.

Missing-data handling and multiplicity are central to assessing reliability. Last observation carried forward is explicitly stated for the primary analysis, but the amount of missing information and sensitivity analyses are not reported. Multiple outcomes and times are recorded, while multiplicity handling is not stated. These are not minor details: they affect how confidently readers can interpret a pattern across outcomes.

Finally, the planning illustrations show how sample-size requirements depend on the effect size assumed. Smaller effects require more participants per arm under the stated assumptions. These calculations are useful for understanding planning logic, but they cannot explain the original enrollment or substitute for the design rationale, which is not reported in the registry record.

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.

Several trial-specific limitations affect interpretation of the posted summaries. First, the sponsor’s analysis method is not reported, although mean differences and 95% confidence intervals are posted. Without the method, it is difficult to know whether the comparisons were adjusted, how the analysis handled repeated measurements, or whether its assumptions align with the simple re-calculation.

Second, the primary analysis used last observation carried forward, but the registry does not report how many primary endpoint values were missing, why data were missing, or how results change under alternative assumptions. That makes the effect of the imputation procedure difficult to assess. The analysis population and missing-data procedures for secondary outcomes are not reported in the registry record.

Third, the record lists many outcome measures and analyses but does not state a multiplicity strategy, a testing hierarchy, or which secondary results were confirmatory. The absence of that information limits the weight that can be assigned to isolated secondary findings. The original sample-size justification and design assumptions are also not reported; the posted planning illustrations are based on observed effects rather than the original design.

Fourth, the registry’s study description calls the study placebo-controlled, while the listed primary outcome arms identify two combination groups and a glimepiride group. Further detail about placebo procedures is not reported in the registry record. The masking field says quadruple, but the roles masked are not specified. These reporting gaps make it difficult to reconstruct operational details from the registry alone.

Finally, the record presents safety counts without detailed exposure time, event timing, or a full comparative safety analysis. The sample is limited to the enrolled study population and the reported follow-up period. The posted results therefore provide a constrained statistical account, not a complete assessment of generalizability or long-term outcomes.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of Alogliptin Used in Combination With Sulfonylurea in Participants With Type 2 Diabetes in Japan — NCT01318083.
• Registry record last updated 2012-02-03; status last verified 2012-02.
• Results first posted 2011-07-06.
• 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, double-blind, parallel-group phase 2/3 study sponsored by Takeda, enrolling 312 participants with type 2 diabetes in Japan. The listed groups received alogliptin 12.5 mg or 25 mg once daily alongside glimepiride, or glimepiride alone. The primary outcome was change from baseline in glycosylated hemoglobin at week 12, analyzed in the Full Analysis Set with last observation carried forward for missing scheduled values.

The posted group summaries show mean changes of -0.6, -0.66, and 0.34 across the 12.5 mg combination, 25 mg combination, and glimepiride groups. The unadjusted recalculated contrast between the alogliptin dose groups has a 95% confidence interval that includes zero. Sponsor-reported comparisons against glimepiride have confidence intervals excluding zero, but the analysis method is not specified. Multiplicity procedures, detailed missing-data information, and the original sample-size justification are not reported in the registry record.