A Study to Evaluate the Efficacy and Safety of the Addition of Canagliflozin in Participants With Type 2 Diabetes Mellitus With Inadequate Glycemic Control on Metformin and Sitagliptin: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Efficacy and Safety of the Addition of Canagliflozin in Participants With Type 2 Diabetes Mellitus With Inadequate Glycemic Control on Metformin and Sitagliptin (NCT02025907), a randomized phase 4 study. The primary outcome is Change From Baseline in Glycosylated Hemoglobin (HbA1c) at Week 26.

2. Clinical Question

3. Trial Design

The registry identifies this as a Phase 4, randomized, quadruple-masked, parallel-group study with placebo control. The official title describes a 26-week, multicenter study with two arms. Enrollment is listed as 218, and the study status is completed. The record lists a start in 2014-02 and primary completion in 2015-09.

The canagliflozin intervention began at 100 mg once daily for the first 6 weeks, after which the dose could be increased to 300 mg once daily. The placebo description says participants received placebo once daily for 28 weeks. Because the official title gives a 26-week duration while the placebo description gives 28 weeks, the record contains a duration difference that it does not explain. The treatment allocation process beyond the randomized designation, including the method used to generate or conceal assignments, is not reported in the registry record.

Quadruple masking indicates that the registry labels the study as masked at four levels, but the record does not identify which participant, care, investigator, or assessment roles correspond to those levels. The study is described as parallel-group; participants were compared between the assigned canagliflozin and placebo groups rather than in a crossover description.

The primary endpoint is the change from baseline in HbA1c at Week 26. The registry also lists fasting plasma glucose, percent change in body weight, the proportion with HbA1c below 7.0 percent, and systolic blood pressure as secondary outcomes. Numerical summaries are posted for three of those four secondary outcomes; the systolic blood pressure result is not reported in the registry record. The registry lists 5 outcome measures and 0 reported analyses, while also posting numerical outcome summaries and comparison results. It does not describe the full statistical analysis plan or explain this distinction.

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: One participant was randomized at 2 different sites (once to placebo and once to canagliflozin) and was therefore counted twice in the total number of randomized participants. The participant was withdrawn from the study and not included in any efficacy or safety analyses.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: mITT population included all randomized participants who received at least 1 dose of double-blind study drug. A total of 3 participants were excluded from the mITT population due to potential misconduct and GCP compliance issues. Here, 'N' (number of participants analyzed) signifies those participants who were evaluable for this outcome measure.

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: mITT (modified intent-to-treat) population included all randomized participants who received at least 1 dose of double-blind study drug. A total of 3 participants were excluded from the mITT population due to potential misconduct and GCP compliance issues.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Glycosylated Hemoglobin (HbA1c) at Week 26, measured over Baseline and Week 26.

The primary endpoint measures change in glycosylated hemoglobin (HbA1c) from baseline to Week 26. HbA1c is a continuous measure of glycemia, and the registry reports it as a percentage of glycosylated hemoglobin. A change score expresses how the measured value at follow-up differs from the participant’s baseline value; a negative change in this outcome represents a decrease from baseline.

The registry reports least-squares mean changes, with standard errors, for the placebo and canagliflozin groups. The posted values are -0.01 for placebo and -0.91 for canagliflozin. The reported mean difference for canagliflozin relative to placebo is -0.9. These figures describe the group-level change and contrast. They are not individual participant trajectories, and the group means do not show how much participant-level variation occurred.

In general, a continuous endpoint can retain information about the magnitude of change across the measured scale rather than reducing the outcome to whether a threshold was reached. HbA1c is also relevant to the trial’s stated focus on glycemic control. That general rationale does not establish the clinical importance of a particular difference. A minimally important difference or threshold for interpreting this result is not reported in the registry record.

The primary outcome description identifies the mITT population as randomized participants who received at least 1 dose of double-blind study drug. It also says 3 participants were excluded because of potential misconduct and GCP compliance issues. The record does not provide further details about the exclusions or explain how missing Week 26 measurements were handled.

7. Results Posted in the Registry

• Canagliflozin: mean -0.91 versus -0.01 on control; difference -0.90 (95% CI -1.22 to -0.58), standardized effect d = -0.79; 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 Fasting Plasma Glucose (FPG) at Week 26

• 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:
  • Canagliflozin: mean -1.65 versus -0.14 on control; difference -1.51 (95% CI -2.27 to -0.75), standardized effect d = -0.54; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Canagliflozin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change From Baseline in Body Weight at Week 26

• 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:
  • Canagliflozin: mean -3.35 versus -1.60 on control; difference -1.75 (95% CI -2.67 to -0.83), standardized effect d = -0.52; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Canagliflozin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With HbA1c Less Than (<) 7.0 Percent at Week 26

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Canagliflozin: 32.3% (31/96) versus 12.2% (10/82) on control; risk difference 20.1 percentage points (95% CI 8.4 to 31.8), risk ratio 2.65 (1.38–5.07); unadjusted two-proportion z-test p = 0.002.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Canagliflozin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes give a broader picture of the measured changes at Week 26. The continuous fasting plasma glucose and body-weight comparisons have negative mean differences, and the binary HbA1c threshold comparison has a positive risk difference. Those directions are consistent with the direction of the primary HbA1c comparison in the limited sense that each posted contrast points toward a more favorable measured change or threshold proportion in the canagliflozin group. This is a description of the data pattern, not a conclusion about treatment effectiveness.

The outcomes are related but not interchangeable. HbA1c and fasting plasma glucose are distinct measures of glycemic status, while body weight and systolic blood pressure assess different dimensions. The threshold outcome recasts HbA1c as whether a participant is below a specified value, rather than reporting continuous change. Agreement in direction across related outcomes can be informative, but it does not make the endpoints independent or establish that each has been confirmed by a separate prespecified test.

The registry lists systolic blood pressure as a secondary outcome, but its numerical result is not reported in the registry record. It lists four secondary outcomes, while numerical summaries are present for three. The registry reports no secondary analyses, although it posts outcome summaries and unadjusted comparisons. The available record does not explain the distinction or provide a complete analysis plan.

Unadjusted secondary results carry less inferential weight when many outcomes or analyses may be considered, because the chance of at least one small p-value can rise across repeated testing. In general, multiplicity procedures address that issue by controlling a chosen error rate or by specifying a testing hierarchy. Whether such procedures were planned or applied is not reported in the registry record. The secondary p-values should therefore be read as unadjusted evidence for their individual comparisons, not as results known to have been protected against multiple testing.

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

• Change From Baseline in Fasting Plasma Glucose (FPG) at Week 26
• Percent Change From Baseline in Body Weight at Week 26
• Percentage of Participants With HbA1c Less Than (<) 7.0 Percent at Week 26
• Change From Baseline in Systolic Blood Pressure (SBP) at Week 26

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous, so the registry reports a least-squares mean change for each group and a mean difference between groups. Least-squares means are model-based adjusted means in general: they can account for terms included in a fitted statistical model. The registry labels the summaries as least-squares means but does not identify the model terms, covariates, or estimation details. It is therefore not possible to infer from the label alone what adjustment was made.

What does the standard error describe?

The reported dispersion is a standard error. In general, a standard error describes the sampling uncertainty of an estimated mean: it reflects how much an estimate would vary across repeated samples under the assumptions of the analysis. It is not the same as the spread of individual participant values. For the primary endpoint, the reported standard errors are 0.119 for placebo and 0.113 for canagliflozin. The registry does not report standard deviations for these arm summaries.

The comparison reports a mean difference of -0.9 and an unadjusted p-value of <0.001. The confidence interval describes a range of values compatible with the observed data under the interval procedure and its assumptions. The interval for the primary comparison is entirely negative, so zero is not within that posted interval. This does not establish that all assumptions are appropriate, and a confidence interval is not a guarantee that the true value lies within its limits.

How should the p-value be read?

In general, a p-value is calculated under a null hypothesis and a statistical model. It measures the probability, assuming that null model and its assumptions, of obtaining data at least as incompatible with the null as those observed. A small p-value is not the probability that the null hypothesis is true. It also does not measure effect size, practical importance, or the likelihood that an individual participant will have a particular outcome.

The primary p-value is explicitly unadjusted. The registry does not report whether the planned primary analysis used a multiplicity adjustment, nor whether other analyses were part of a confirmatory testing sequence. The unadjusted label should be retained when interpreting the p-value rather than treating it as evidence that multiple testing was accounted for.

What does the binary comparison measure?

The secondary endpoint for HbA1c below 7.0 percent is binary at the participant level: each measured participant either met the threshold or did not. The registry posts the percentages and a risk difference. The comparison is a difference in proportions, rather than a difference in mean change on a continuous scale. In general, a risk difference expresses the absolute contrast between two group proportions. The posted risk difference is 0.201, with a confidence interval above zero, and the unadjusted p-value is 0.002.

The registry also includes relative measures for this binary outcome, but those are not needed to understand the main absolute comparison. The analyzed counts are 82.0 in placebo and 96.0 in canagliflozin. The record does not describe whether missing threshold measurements were excluded, imputed, or otherwise addressed.

What do the recalculated planning estimates mean?

In general, planning estimates based on an observed effect are illustrative because the observed effect itself is uncertain and can be larger or smaller than the effect that would recur in another study. The planning assumptions listed for these estimates are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is a stated fraction of the observed effect. For continuous endpoints, the estimates depend on the posted summary information; the registry’s planning description does not establish that these were the assumptions used to design the original trial.

For the reported comparisons, the planning calculations are simple unadjusted recalculations from posted summary statistics. The registry discrepancy indicator is false, so no discrepancy between a sponsor analysis and these calculations is flagged. These calculations do not replace a prespecified sponsor analysis, and they do not supply missing details about the model used for the reported least-squares means.

What is not described?

The registry record does not report the trial’s original sample-size justification, the detailed primary analysis model, the method used to handle missing outcome data, or the timing and rules for any interim analyses. It also does not report a multiplicity strategy. Those details matter because different choices can affect standard errors, confidence intervals, p-values, and the interpretation of a collection of outcomes. General statistical explanations do not determine which methods were actually used in this study.

11. Confidence Intervals and Precision

The primary estimate is a mean difference of -0.9 in change from baseline in HbA1c. Its confidence interval lies entirely below zero, indicating that the full interval under the posted calculation points toward a negative group difference. The registry reports the interval, but its upper endpoint is given to more than three decimal places; the numerical endpoint is therefore not repeated. The key precision feature is that the interval does not include zero.

A confidence interval communicates both the estimated direction and the range of values reasonably compatible with the data and the interval’s statistical assumptions. A narrower interval generally indicates more precision than a wider interval, though the interval alone does not show whether the estimate is important in clinical practice. The registry does not give a clinical importance threshold for HbA1c change, so the interval cannot be compared with such a threshold from the posted record.

The group standard errors for the primary least-squares means are 0.119 for placebo and 0.113 for canagliflozin. These quantify uncertainty in the estimated group means, not variability among individual participants. The primary comparison interval is more directly relevant to uncertainty in the between-group contrast. The record does not fully describe the statistical model behind the least-squares means or confidence interval, which limits independent assessment of the interval’s assumptions.

For the posted secondary comparisons, the continuous fasting plasma glucose and body-weight intervals are below zero, while the interval for the HbA1c threshold risk difference is above zero. The registry gives interval endpoints for those results, but some are reported to more than three decimal places and are not repeated. The secondary interval directions describe the posted contrasts; they do not resolve the absence of reported multiplicity handling or the lack of numerical results for systolic blood pressure.

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 the continuous and binary outcomes in this study; they are not total event counts. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses a stated fraction of the observed effect as the effect to detect. The registry record does not report the original design’s sample-size assumptions or justification, so these figures should not be read as the trial’s original enrollment target calculations.

For the primary HbA1c comparison, planning for 100% of the observed effect gives 26 participants per arm for 80% power and 34 participants per arm for 90% power. Planning for 75% of the observed effect gives 45 participants per arm for 80% power and 60 participants per arm for 90% power. Planning for 50% of the observed effect gives 101 participants per arm for 80% power and 135 participants per arm for 90% power.

In plain language, the 80% power column gives the estimated per-arm sample size for a design with an 80% chance of detecting the stated effect under the listed assumptions if that effect is real. The 90% column uses a higher detection probability under those same assumptions. Smaller effects require more participants, which is why the planning counts rise as the fraction of the observed effect decreases. These are planning illustrations, not observed power; observed power is not a measure of how convincing the completed trial result is.

Each outcome has its own planning estimates because the observed effect and variability differ across outcomes. The figures are not evidence that the original study was designed around those effects, and they do not account for assumptions that are not listed, such as loss to follow-up or multiplicity adjustments. The registry record does not report whether such considerations were included in the original design.

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 = -0.900; pooled standard deviation σ = 1.139.
2. n = 2 × (1.96 + 0.842)² × 1.139² / -0.900² = 25.1, rounded up to 26 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 <- -0.9                      # observed difference in means
sigma <- 1.13875172771864                     # 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(2025907)
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 = 26 (80% power), 34 (90% power)
 75% of observed effect: n per arm = 45 (80% power), 60 (90% power)
 50% of observed effect: n per arm = 101 (80% power), 135 (90% power)
power.t.test: 26.1 
Simulated power at n = 26 per arm: 0.801 (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. Multiplicity, Missing Data and Analysis Populations

The registry posts an unadjusted p-value for the primary HbA1c comparison and unadjusted p-values for the secondary comparisons with numerical results. It does not report a multiplicity adjustment, a testing hierarchy, or a family of outcomes designated for a particular confirmatory procedure. The record therefore does not establish how the probability of false-positive findings across multiple outcomes or analyses was controlled.

In general, testing several outcomes can make it more likely that at least one result appears statistically notable by chance, even if each test is considered separately. Multiplicity methods can address this concern, but their consequences depend on which outcomes and analyses are included and how the procedure is specified. Without those details, the posted secondary p-values should not be treated as if they necessarily account for the full set of comparisons. The primary result is also reported with an unadjusted p-value; the registry does not describe whether any broader strategy applies to it.

The record does not report how missing baseline or Week 26 measurements were handled, whether any imputation was used, or what assumptions underlie the analysis of participants without an observed endpoint. This information is important because missing outcome data can affect the estimated group contrast, particularly if the amount or reasons for missingness differ between groups. The registry does not provide enough information to assess that possibility.

The primary outcome description defines the mITT population as all randomized participants who received at least 1 dose of double-blind study drug. It says 3 participants were excluded from that population due to potential misconduct and GCP compliance issues. It does not explain the specific circumstances, timing, or analysis consequences of those exclusions. The primary arm counts are 94.0 for placebo and 99.0 for canagliflozin, while other outcomes use different analysis counts. The record does not reconcile those denominators with the total enrollment of 218.

The registry says 0 analyses were reported, despite listing numerical outcome summaries and comparison results. It does not report a detailed statistical analysis plan, interim-analysis procedures, or whether outcome definitions or analysis populations changed during the study. Those omissions make it difficult to judge prespecification and to distinguish planned analyses from analyses reported after data collection.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: angina unstable (0.9%), diverticulitis (0.9%), cerebral infarction (0.9%), transient ischaemic attack (0.9%).

The registry lists 108 participants at risk in each group for both serious and other adverse events. Serious adverse events affected 2 participants in the placebo group and 2 in the canagliflozin group. Other adverse events affected 16 participants in the placebo group and 18 in the canagliflozin group. These are counts of participants affected, as presented in the registry, and should not be interpreted as counts of all event episodes.

The listed serious-event terms are angina unstable, diverticulitis, cerebral infarction, and transient ischaemic attack, each shown as 0.9%. The record does not indicate which treatment group each listed term belongs to in the displayed serious-event summary. Death counts are not reported in the registry record.

These figures describe the posted event summaries and their stated at-risk denominators. They do not, on their own, establish comparative safety: the record does not provide event timing, severity details beyond the serious classification, exposure-adjusted rates, or a full account of how events were collected and coded. The safety counts should be read separately from the efficacy outcomes and should not be used to infer causation from the group totals alone.

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

The primary result describes a difference in mean HbA1c change between randomized groups at Week 26. The posted estimate is -0.9, and its confidence interval lies below zero. In general terms, the observed group means show a larger decrease from baseline in the canagliflozin group than in the placebo group. That statement is limited to the measured endpoint, analysis population, follow-up point, and registry results.

The result does not establish what happened to every participant. Group means can conceal variation, including participants whose measurements changed differently from the group average. Nor does the result identify the clinical importance of the difference: a threshold for a meaningful HbA1c change is not reported in the registry record. A small p-value does not answer that question, since statistical evidence and practical importance are distinct.

The study examined canagliflozin in participants with type 2 diabetes who had inadequate glycemic control on metformin and sitagliptin. The posted result cannot automatically be extended to populations, treatment contexts, or follow-up periods not described by the registry. The record describes a 26-week study in its official title, while the placebo arm description gives a 28-week duration; it does not explain this difference. Longer-term outcomes are not established by the Week 26 endpoint.

Interpretation is also limited by information not reported in the registry record, including the original sample-size justification, missing-data handling, multiplicity procedures, detailed model specification, and any interim analysis plan. The mITT definition and exclusions are reported, but the record does not explain the specific exclusion circumstances or reconcile the analysis denominators across outcomes. These gaps do not erase the posted estimate, but they constrain how completely its design and analysis can be evaluated.

The secondary outcomes show a pattern across glycemic, weight, and threshold measures, but their comparisons are unadjusted and the systolic blood pressure result is not reported. The registry does not state whether a multiplicity plan was applied. Accordingly, the secondary results add descriptive context but do not independently settle the interpretation of the primary comparison. No medical decision should be based on these registry summaries alone.

17. Why This Trial Matters Statistically

This trial illustrates why a statistical reading should separate an effect estimate from its uncertainty and from its interpretation. The primary mean difference is reported alongside a confidence interval and an unadjusted p-value. The estimate conveys the observed magnitude and direction; the interval conveys uncertainty under the calculation’s assumptions; and the p-value addresses compatibility with a null model. None of these quantities alone describes clinical importance or all sources of uncertainty.

The record also shows why analysis populations and denominators matter. The primary endpoint uses an mITT description with 3 exclusions, and its group counts differ from those posted for secondary outcomes. Enrollment is listed as 218, but the registry does not reconcile that total with each outcome’s analyzed participants. Readers need to distinguish the enrolled population from the population contributing to a particular estimate rather than treating every outcome as having the same denominator.

Secondary outcomes create a further interpretive issue. The posted comparisons are unadjusted, the registry does not report a multiplicity strategy, and one listed secondary outcome lacks numerical results. A coherent direction across some endpoints can be noted, but repeated testing and incomplete analysis details mean that each secondary result should be considered in context rather than as a standalone confirmation.

Finally, planning calculations based on the observed effect can help illustrate how sample-size requirements change when the target effect changes. They are not a substitute for the original design rationale, which is not reported in the registry record. Keeping that distinction clear prevents a retrospective planning illustration from being mistaken for evidence about the study’s original power or design choices.

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 registry gives limited detail about the study’s methods beyond its randomized, quadruple-masked, parallel-group designation and placebo control. It does not report how the random sequence was generated, how allocation was concealed, or which roles were covered by the four masking levels. These details would help assess potential sources of bias but cannot be inferred from the design labels.

The official title describes a 26-week study, whereas the placebo description says placebo was taken for 28 weeks. The registry does not explain the duration difference. The canagliflozin dose could be increased from 100 mg once daily after the first 6 weeks to 300 mg once daily, but the number of participants receiving each dose and the dose-escalation rules beyond this description are not reported.

The primary endpoint summary identifies an mITT population and 3 exclusions due to potential misconduct and GCP compliance issues. The record does not provide more detail about those exclusions. Analysis counts vary across outcomes, and the registry does not reconcile those counts with enrollment. Missing-data handling and the statistical model underlying the least-squares means are not reported.

The record does not report the original sample-size justification, multiplicity procedures, a detailed analysis plan, or interim-analysis details. The secondary outcomes are listed, but numerical summaries are not present for systolic blood pressure. The adverse-event summary includes affected-participant counts and selected serious-event terms, but does not provide a complete safety analysis or report death counts. These limitations apply to interpretation of the registry record and do not establish that the underlying study lacked these procedures or information.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study to Evaluate the Efficacy and Safety of the Addition of Canagliflozin in Participants With Type 2 Diabetes Mellitus With Inadequate Glycemic Control on Metformin and Sitagliptin — NCT02025907.
• Registry record last updated 2016-10-31; status last verified 2016-10.
• Results first posted 2016-09-02.
• 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 Phase 4, randomized, quadruple-masked, parallel-group, placebo-controlled study of canagliflozin in participants with type 2 diabetes and inadequate glycemic control on metformin and sitagliptin. Enrollment is listed as 218. The official title describes a 26-week study, and the primary endpoint is change from baseline in HbA1c at Week 26.

The posted primary comparison reports a mean difference of -0.9, a confidence interval wholly below zero, and an unadjusted p-value of <0.001. Numerical results are also posted for fasting plasma glucose, body weight, and the proportion below the HbA1c threshold. Systolic blood pressure is listed as a secondary outcome but its result is not reported in the registry record.

The record does not report the original sample-size justification, missing-data methods, multiplicity plan, detailed analysis model, or interim-analysis procedures. It reports serious adverse events affecting 2 participants in each group and does not report death counts.