Comparative Study to Evaluate Efficacy and Safety When Metformin Hydrochloride 500 mg Once Daily is Added on to SYR-322 25 mg in Type 2 Diabetic Patients With Inadequate Glycemic Control Despite Treatment With SYR-322 25 mg in Addition to Diet and Exercise Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparative Study to Evaluate Efficacy and Safety When Metformin Hydrochloride 500 mg Once Daily is Added on to SYR-322 25 mg in Type 2 Diabetic Patients With Inadequate Glycemic Control Despite Treatment With SYR-322 25 mg in Addition to Diet and Exercise Therapy (NCT02068443), a randomized phase 3 study. The primary outcome is Change From Baseline in Glycosylated Hemoglobin A1c (HbA1c) National Glycohemoglobin Standardization Program (NGSP) at the End of Treatment (EOT) Period.

2. Clinical Question

The question is about comparative change in a continuous measure, not simply whether each group’s end-of-treatment value was lower than its baseline value. The primary outcome is expressed as change from baseline, and the registry states that a negative change indicates improvement. The central comparisons are therefore the estimated mean changes in the combination groups against the mean change in the alogliptin-alone group.

The study was randomized, quadruple-masked, and parallel-group. The registry describes a 24-week treatment period in participants whose glycemic control remained inadequate despite alogliptin together with diet and exercise. It does not report detailed eligibility criteria beyond that description, nor does the record establish that the findings apply to people with other treatment histories or clinical characteristics.

3. Trial Design

This completed Phase 3 study used randomized allocation, quadruple masking, and a parallel-group design. The registry reports enrollment of 374. Participants were assigned to alogliptin alone, alogliptin plus metformin hydrochloride 500 mg once daily, or alogliptin plus metformin hydrochloride 250 mg twice daily. Each regimen was described as lasting 24 weeks, with tablets and matching placebo used to support masking.

The treatment schedules differed in whether metformin was active and, for active metformin, how it was scheduled across the day. Alogliptin was 25 mg once daily in all three groups. The control group received metformin-matching placebo. The once-daily combination group received 500 mg metformin once daily and placebo at the other dosing time; the twice-daily group received 250 mg metformin twice daily and placebo at the additional dosing time.

The primary outcome was change in HbA1c from baseline to the end-of-treatment period, up to Week 24. The primary analysis population was the Full Analysis Set, defined as all randomized participants who received at least 1 dose of study drug. Group sizes reported for the primary outcome were 71 for alogliptin alone, 152 for the once-daily combination, and 151 for the twice-daily combination. These outcome-analysis counts describe the posted primary result, not a separate estimate of the number randomized to each arm.

The record gives the study start as 2014-02 and primary completion as 2015-02, and lists the status as completed. The sponsor was Takeda. Further operational details, including the randomization procedure, site locations, and trial-specific rationale for the enrollment target, 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: Participants with a diagnosis of Type 2 Diabetes Mellitus were enrolled equally in a 2:2:1 ratio to 1 of 3 treatment groups: alogliptin in combination with 500 mg total dose metformin hydrochloride administered either once a day (QD) or twice a day (BID) and alogliptin alone.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS) included all randomized participants who received at least 1 dose of study drug.

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.

Population: All randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Glycosylated Hemoglobin A1c (HbA1c) National Glycohemoglobin Standardization Program (NGSP) at the End of Treatment (EOT) Period, measured over Baseline and End of Treatment (EOT) (Up to Week 24).

The primary endpoint is change in glycosylated hemoglobin A1c, measured using the National Glycohemoglobin Standardization Program method, from baseline to the end-of-treatment period, up to Week 24. HbA1c reflects the concentration of glucose bound to hemoglobin as a percentage of the maximum that can be bound. The registry specifies that a negative change from baseline indicates improvement.

Change is useful because it compares each participant’s end-of-treatment measurement with their own baseline measurement, then estimates the group-level change. This focuses the analysis on movement over the treatment period rather than only on the final measurement. The posted group estimates are least-squares means, adjusted using the stated analysis model, rather than simple arithmetic means of the observed values.

The endpoint is continuous, so its estimated group change and between-group difference retain information about the size and direction of the result. The registry describes an analysis of covariance model with change from baseline as the dependent variable and baseline HbA1c and treatment as independent variables. Baseline adjustment can account statistically for differences in starting measurements while estimating treatment-group means.

The endpoint alone does not describe every aspect of glycemic control, safety, or longer-term outcomes. The registry lists secondary measures that include HbA1c values, achievement of target HbA1c levels, fasting blood glucose, and treatment-emergent adverse events. The primary result should therefore be read as evidence about this specified measure and treatment period, not as a complete assessment of every possible outcome.

7. Results Posted in the Registry

• Alogliptin + Metformin Hydrochloride QD: mean -0.49 versus 0.16 on control; difference -0.65 (95% CI -0.82 to -0.48), standardized effect d = -1.07; z-test p = <0.001.
• Alogliptin + Metformin Hydrochloride BID: mean -0.60 versus 0.16 on control; difference -0.76 (95% CI -0.93 to -0.59), standardized effect d = -1.26; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary analysis estimate?

The primary outcome is a continuous change score: each participant’s HbA1c at the end of treatment is compared with baseline. The analysis then estimates the mean change for each treatment group and contrasts those group estimates. A mean gives an overall location for a group’s measurements; it does not imply that every participant changed by that amount. In this study the posted values are least-squares means, which are model-based estimates rather than unadjusted averages.

The registry states that an analysis of covariance model was used, with change from baseline as the dependent variable and baseline HbA1c and treatment as independent variables. In general, analysis of covariance estimates group differences while accounting for a covariate, here the baseline value. This can improve the comparison’s precision when baseline measurements are related to later measurements. The registry does not give further model specifications, such as additional covariates, assumptions checks, or handling of departures from model assumptions.

What is a least-squares mean difference?

A least-squares mean is a model-adjusted estimate of the group mean. A least-squares mean difference contrasts two such estimates using the model. The sponsor’s reported comparison of alogliptin alone and the once-daily combination is -0.65. A negative contrast, read with the groups in their listed order and the endpoint’s direction, corresponds to a more negative estimated change for the combination group. The sponsor also reports a contrast of 0.11 between the once-daily and twice-daily metformin groups. The registry does not report the method details for this second analysis, so the contrast should be read as the posted estimate rather than as a fully specified inferential procedure.

How should the recalculations be understood?

The recalculated comparisons are simple, unadjusted re-calculations from posted summary statistics. They contrast each combination group with alogliptin alone and produce mean differences of -0.65 and -0.76. Their p-values are unadjusted and are reported as <0.001. These calculations do not reproduce the sponsor’s full ANCOVA model, because the summary statistics do not contain all participant-level information or model details. The standard deviations used in the recalculation are approximate, derived from the posted standard errors, so the re-calculated uncertainty is also approximate.

A p-value describes how compatible the observed data are with a specified null model under the calculation’s assumptions. It is not the probability that a treatment effect is absent, nor does it measure the size or importance of an effect. Here, the reported recalculated p-values are unadjusted; they do not account for the fact that the record includes more than one group comparison or multiple outcomes.

Why use a confidence interval?

A confidence interval gives a range of estimates compatible with the data under the analysis method and its assumptions. It communicates uncertainty around a point estimate, rather than only indicating whether a test crossed a chosen threshold. The sponsor reports a 95% interval for the comparison of alogliptin alone with the once-daily combination and for the comparison between the metformin schedules. The former is entirely negative; the latter extends from a negative value to a positive value. These intervals support different degrees of precision in the two reported comparisons, but neither interval addresses unreported sources of bias or model uncertainty.

Why can variability affect the analysis?

Participants’ measurements vary, so a mean difference is estimated with uncertainty. In general, greater variability makes it harder to distinguish a group contrast from the noise in individual measurements, while more information can make an estimate more precise. The registry reports standard errors for the least-squares means, but does not give standard deviations for the primary groups. The recalculation therefore estimates pooled variability from the standard errors and group sizes; it is not the same as using participant-level measurements.

Over-dispersion is a general concern when observed variability exceeds what a model assumes. For a continuous outcome analyzed with a normal approximation, an inaccurate variability assumption can affect standard errors, confidence intervals, and p-values. No over-dispersion assessment is reported in the registry record, and the posted summary statistics do not permit an independent check of the model’s fit.

What is not specified?

The primary outcome description gives the endpoint, analysis population, and core ANCOVA covariates. The registry record does not report the complete statistical analysis plan, missing-data method, multiplicity procedure, interim analyses, or the original sample-size justification. It also does not state the detailed procedure used to generate the sponsor’s confidence intervals or the method and criteria for the legacy comparison between metformin schedules. These omissions limit how fully the posted analyses can be reproduced or evaluated.

10. Confidence Intervals and Precision

The sponsor reports a least-squares mean difference of -0.65 for alogliptin alone compared with alogliptin plus metformin 500 mg once daily. Its 95% confidence interval runs from -0.821 to -0.480. The interval is entirely negative, indicating that the reported model-based estimates are compatible with a range of negative contrasts under the sponsor’s method. Its width conveys uncertainty around the point estimate; it does not account automatically for missing data, model misspecification, or unreported analysis choices.

For the comparison between once-daily and twice-daily metformin, the sponsor reports a least-squares mean difference of 0.11 and a 95% confidence interval from -0.026 to 0.247. This interval includes negative and positive contrasts. It therefore leaves uncertainty about the direction of the difference between schedules. The registry labels this a non-inferiority-or-equivalence legacy analysis, but the margin and decision rules are not reported in the registry record. The interval should not be treated as establishing equivalence or non-inferiority without those details.

The recalculated unadjusted comparisons have approximate intervals based on summary statistics, but the sponsor’s reported interval is the direct registry analysis for the alogliptin-alone versus once-daily comparison. The recalculated p-values are unadjusted and should not be mistaken for sponsor-reported p-values. The registry does not provide a sponsor p-value for the reported least-squares mean comparisons.

Precision is not the same as clinical importance. A narrow interval can still surround an effect whose relevance depends on clinical context, while a wider interval can leave several plausible values. The registry does not provide a threshold for clinically important HbA1c change. The interval estimates also pertain to the stated analysis population and treatment period, not necessarily to other populations or longer follow-up.

11. 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 entries are participants per arm for 80% power and participants per arm for 90% power. They are planning illustrations based on the observed effect, not the trial’s original design assumptions. The stated calculation assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The original trial’s sample-size justification and design assumptions are not reported in the registry record.

For the once-daily combination comparison, planning for 100% of the observed effect gives 14 participants per arm for 80% power and 19 for 90% power. Planning for 75% of the observed effect gives 25 per arm for 80% power and 33 for 90% power. Planning for 50% of the observed effect gives 55 per arm for 80% power and 73 for 90% power.

For the twice-daily combination comparison, planning for 100% of the observed effect gives 10 participants per arm for 80% power and 14 for 90% power. Planning for 75% of the observed effect gives 18 per arm for 80% power and 24 for 90% power. Planning for 50% of the observed effect gives 40 per arm for 80% power and 54 for 90% power.

In general, the 80% column describes a planning target with an 80% chance of detecting the specified effect under the stated assumptions; the 90% column uses a higher chance and therefore requires more participants for the same effect. Smaller effects need more participants to detect under otherwise similar assumptions. These figures are not observed power.

12. 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 = -0.650; pooled standard deviation σ = 0.605.
2. n = 2 × (1.96 + 0.842)² × 0.605² / -0.650² = 13.6, rounded up to 14 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 <- -0.65                      # observed difference in means
sigma <- 0.604927848059143                     # 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(2068443)
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 = 14 (80% power), 19 (90% power)
 75% of observed effect: n per arm = 25 (80% power), 33 (90% power)
 50% of observed effect: n per arm = 55 (80% power), 73 (90% power)
power.t.test: 14.6 
Simulated power at n = 14 per arm: 0.783 (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.

13. Analyses Reported by the Sponsor

The sponsor’s reported primary comparison is between alogliptin alone and alogliptin plus metformin 500 mg once daily. It gives a least-squares mean difference of -0.65 and a 95% confidence interval from -0.821 to -0.480. The recalculated unadjusted comparison for the same groups is -0.65, with an unadjusted p-value of <0.001. The estimates are similar at the displayed precision, but the recalculation is not a reproduction of the sponsor’s ANCOVA analysis.

The recalculation uses posted group summary statistics and approximate variability. It does not have access to participant-level data or the full model specification. For that reason, it can serve as a transparent check on the direction and approximate size of the comparison, but it should not displace the sponsor’s model-based result. The registry does not indicate a discrepancy between the recalculation and sponsor result.

The sponsor also reports a comparison between the metformin schedules, with a least-squares mean difference of 0.11 and a 95% confidence interval from -0.026 to 0.247. No method or p-value is listed for that analysis. The registry’s legacy label refers to non-inferiority or equivalence, but the necessary margin and decision criteria are not reported in the registry record. Thus, the label cannot be independently evaluated from the posted estimate alone.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures and 9 posted secondary outcomes, alongside the primary HbA1c change outcome. The primary comparison also involves two metformin schedules, and the reported analyses include a comparison between those schedules. When several outcomes or group comparisons are examined, the chance of at least one apparently notable result can increase. In general, multiplicity procedures address this issue by adjusting how evidence is assessed across a family of tests.

No multiplicity adjustment or prespecified testing hierarchy is reported in the registry record. The recalculated p-values for the comparisons with alogliptin alone are unadjusted. They should therefore be understood as tests for those individual contrasts, not as error-controlled results across all outcomes and comparisons. The reported sponsor intervals likewise cannot be assumed to have multiplicity adjustment, because the record does not say whether one was used.

The primary population is described as the Full Analysis Set, including all randomized participants who received at least 1 dose of study drug. The group counts for the primary outcome are reported, but the registry does not give a detailed accounting of participants excluded from the analysis or the reasons for any exclusions. It also does not report a missing-data strategy, such as how absent end-of-treatment measurements were handled or whether any sensitivity analyses were performed.

Those omissions matter because missing outcomes can affect estimated group means if missingness is related to treatment, outcome, or participant characteristics. Statistical methods can address missing data under stated assumptions, but those assumptions must be known to assess their plausibility. The registry record does not report interim analyses, stopping rules, or the complete statistical analysis plan. It is therefore not possible to determine from the record whether interim looks or other planned analytic safeguards affected interpretation.

The primary endpoint description does identify baseline HbA1c and treatment as model variables. Other covariates, model diagnostics, and analysis details are not reported in the registry record. The sponsor’s comparison between the metformin schedules is presented with a legacy non-inferiority-or-equivalence label, but its method, margin, and decision criteria are not specified. These gaps limit a full assessment of multiplicity, missing-data handling, and the confirmatory status of each analysis.

15. 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: diverticulitis (1.4%), brain contusion (0.7%), contusion (0.7%), fall (0.7%), fibula fracture (0.7%).

The registry reports serious adverse event counts and the numbers at risk for each group. Alogliptin alone had 1 serious adverse event among 71 at risk; the once-daily metformin combination had 3 among 152 at risk; and the twice-daily combination had 0 among 151 at risk. These are counts, not percentages calculated from the group sizes. No deaths value is reported for these groups; deaths are not reported in the registry record.

The record lists serious-event terms including diverticulitis at 1.4%, and brain contusion, contusion, fall, fibula fracture, and limb crushing injury at 0.7% each. These terms and percentages describe the listed serious-event information; they do not provide a complete account of event timing, severity, relatedness, or recurrence. The registry also reports other adverse-event counts of 15, 25, and 27 in the three groups, respectively, but those counts are separate from the serious-event counts.

Group counts alone do not establish whether an event was caused by a study drug, and small counts can be imprecise. The listed safety observations apply to the registry’s treatment period and enrolled study population. Detailed adverse-event definitions, exposure-adjusted analyses, causal assessments, and follow-up beyond the treatment period are not reported in the registry record.

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

The primary estimates show a more negative mean HbA1c change in each alogliptin-plus-metformin group than in the alogliptin-alone group over the stated end-of-treatment period. The reported group changes are 0.16 for alogliptin alone, -0.49 for the once-daily combination, and -0.6 for the twice-daily combination. The registry defines a negative change as improvement. The recalculated unadjusted differences versus alogliptin alone are -0.65 and -0.76, respectively, and both have recalculated unadjusted p-values of <0.001.

These results describe estimated group averages, not the outcome for every participant. They are specific to HbA1c change through up to Week 24 among participants in the stated Full Analysis Set. The registry does not provide enough detail to establish the relevance of the observed change for individual participants or to define a threshold for clinical importance. Statistical evidence about a group contrast is not, by itself, a complete judgment about clinical value.

The sponsor’s reported comparison between metformin schedules is 0.11, with a 95% confidence interval from -0.026 to 0.247. Since the interval covers contrasts in both directions, it leaves uncertainty about the relative difference between schedules. Although the registry attaches a legacy non-inferiority-or-equivalence label, it does not report the margin or method needed to assess that claim. The estimate should not be interpreted as proof that the schedules are equivalent.

The study also cannot answer questions that its posted primary endpoint does not address. It does not establish long-term outcomes beyond the stated treatment period, and the primary result does not summarize every secondary endpoint or safety outcome. Missing-data methods, multiplicity adjustments, and detailed analysis-plan information are not reported in the registry record. These limits do not erase the posted estimates, but they define the scope of what can be concluded from them.

17. Why This Trial Matters Statistically

This trial illustrates why a clinical comparison should be read through its endpoint, estimand, and analysis population rather than through a p-value alone. HbA1c change is continuous, so the group estimates and mean differences convey direction and magnitude. The primary outcome is a change from baseline, and the registry explicitly defines negative change as improvement. This makes it possible to interpret the sign of the contrast while keeping its scale tied to the stated measure.

The trial also illustrates the distinction between a model-based sponsor estimate and a simple recalculation from summary statistics. The sponsor reports a least-squares mean comparison from an ANCOVA description, while the recalculations use summary statistics and approximate variability. Similar displayed estimates for the once-daily comparison support a basic numerical check, but they are not analytically interchangeable. The model, population, and uncertainty procedure remain important.

Comparison of the two metformin schedules demonstrates why a legacy label cannot substitute for the details of a non-inferiority or equivalence analysis. A margin, analysis method, and decision criteria are necessary to interpret such a claim. The reported interval spans negative and positive values, while those essential specifications are not reported in the registry record.

Finally, the record shows how interpretation depends on transparency beyond the primary result. Multiplicity, missing-data handling, and the original sample-size rationale are not reported. Planning illustrations can show how hypothetical effect sizes relate to sample requirements, but they do not reveal the trial’s original design logic or observed power. Recognizing these distinctions helps keep conclusions proportional to the available statistical evidence.

18. 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 primary result covers a treatment period of up to Week 24. Longer-term HbA1c outcomes and outcomes beyond the stated period are not reported in the registry record. The endpoint is limited to HbA1c change and cannot by itself characterize all dimensions of glycemic control, safety, or clinical outcomes.

The registry’s population description is restricted to type 2 diabetic patients with inadequate glycemic control despite alogliptin 25 mg plus diet and exercise therapy. Detailed eligibility criteria and participant characteristics are not reported in the registry record. Consequently, the extent to which the estimates apply to people with different disease histories, treatment regimens, or clinical characteristics cannot be assessed from these registry entries.

Although the primary analysis population and core ANCOVA variables are described, the full analysis plan, handling of missing measurements, model diagnostics, and multiplicity procedure are not reported. The exact sponsor method for the reported treatment comparisons is not fully detailed. The recalculations use approximate variability derived from standard errors rather than participant-level data, limiting their ability to reproduce the sponsor’s model.

The sponsor’s comparison between the metformin schedules has a legacy non-inferiority-or-equivalence label, but the margin, method, and decision criteria are not reported in the registry record. The available estimate and interval are insufficient to verify that label. The safety information includes event counts and selected serious-event terms, but detailed causal assessment, timing, and longer follow-up are not reported. The trial’s original sample-size justification is also not reported, so the posted planning illustrations should not be treated as evidence about the trial’s design rationale.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Comparative Study to Evaluate Efficacy and Safety When Metformin Hydrochloride 500 mg Once Daily is Added on to SYR-322 25 mg in Type 2 Diabetic Patients With Inadequate Glycemic Control Despite Treatment With SYR-322 25 mg in Addition to Diet and Exercise Therapy — NCT02068443.
• Registry record last updated 2023-09-28; status last verified 2023-08.
• Results first posted 2016-06-06.
• 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.

22. Record Summary

The registry describes a completed, randomized, quadruple-masked, parallel-group Phase 3 study sponsored by Takeda. It enrolled 374 participants with type 2 diabetes and evaluated alogliptin alone against alogliptin plus metformin hydrochloride given as 500 mg once daily or 250 mg twice daily. Treatment lasted 24 weeks.

The primary outcome was change in HbA1c from baseline to the end-of-treatment period, up to Week 24. The stated primary analysis used an ANCOVA model with baseline HbA1c and treatment as independent variables, and the Full Analysis Set included randomized participants who received at least 1 dose of study drug. Posted group estimates were 0.16, -0.49, and -0.6, respectively.

The sponsor reported a least-squares mean difference of -0.65 for alogliptin alone versus the once-daily combination, with a 95% confidence interval from -0.821 to -0.480. A comparison between metformin schedules was also reported, but its method and non-inferiority or equivalence criteria are not reported in the registry record. Multiplicity procedures, missing-data methods, and the original sample-size rationale are not reported.