Efficacy and Safety of Sotagliflozin Versus Placebo in Patients With Type 2 Diabetes Mellitus Not Currently Treated With Antidiabetic Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Sotagliflozin Versus Placebo in Patients With Type 2 Diabetes Mellitus Not Currently Treated With Antidiabetic Therapy (NCT02926937), a randomized phase 3 study. The primary outcome is Change From Baseline in Hemoglobin A1c (HbA1c) at Week 26 (Sotagliflozin 400 mg Versus Placebo).

2. Clinical Question

HbA1c is a measure of average blood glucose over time and is used in this trial as the primary glycemic endpoint. The registry also describes secondary questions involving postprandial and fasting glucose, systolic blood pressure, body weight, and the proportions of participants meeting specified HbA1c thresholds. Those additional outcomes broaden the study’s assessment, but the primary comparison is the 400 mg group versus placebo at week 26.

3. Trial Design

This was a phase 3, randomized, double-blind, placebo-controlled, parallel-group, multicenter study sponsored by Lexicon Pharmaceuticals. The registry lists enrollment of 399 participants and a completed study. Participants had type 2 diabetes and inadequate glycemic control; the title specifies that they were not currently treated with antidiabetic therapy. The brief summary describes inadequate control on diet and exercise.

After a 2-week run-in period, participants were randomized to sotagliflozin 400 mg, sotagliflozin 200 mg, or matching placebo. The assigned tablets were taken once daily before the first meal of the day during the double-blind treatment period, for up to 26 weeks. The primary endpoint compared the 400 mg group with placebo. The 200 mg group was included in selected secondary comparisons, not in the stated primary comparison.

The primary endpoint was change from baseline in HbA1c at week 26. The registry describes the primary analysis population as intent-to-treat participants randomized to the 400 mg and placebo groups under the original protocol and Protocol Amendment 1. The arm-level primary results list 142 participants in the 400 mg group and 150 in the placebo group. These analysis counts should not be confused with total enrollment or assumed to represent every randomized participant.

Randomization and masking help limit predictable group differences and awareness-related influences on outcome assessment. A parallel-group design means the assigned groups were followed as separate groups rather than participants crossing between treatments. The registry record does not report the allocation ratio, the details of the randomization sequence, the identities of the people masked, or the original sample-size justification.

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 in 1 of 3 treatment groups: Placebo, Sotagliflozin 400 milligrams (mg) or Sotagliflozin 200 mg.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants in the Intent-to-treat (ITT) population who were randomized to Sotagliflozin 400 mg and Placebo under original Protocol and Protocol Amendment 1. Missing data are imputed using washout imputation method under the missing not at random framework.

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: The randomized population included any participant who had been allocated to a randomized treatment regardless of whether the treatment kit was used.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Hemoglobin A1c (HbA1c) at Week 26 (Sotagliflozin 400 mg Versus Placebo), measured over Baseline to Week 26.

The primary endpoint is the change from baseline in hemoglobin A1c (HbA1c) at week 26, expressed as a percentage of HbA1c. It measures how much the average HbA1c level changed during the study period. The reported arm values are least-squares means, which are model-based estimates rather than simple raw averages. For this endpoint, lower change values indicate a larger reduction from baseline.

HbA1c is used in diabetes research because it summarizes blood glucose over time, rather than capturing glucose at just one moment. As a continuous endpoint, it preserves information about the size of a change across participants. That can make it useful for comparing average changes between randomized groups. General statistical context: a continuous measure can show both the direction and magnitude of a group difference, while the uncertainty around that estimated difference is conveyed by its confidence interval.

The registered primary objective was to test whether sotagliflozin 400 mg was superior to placebo for HbA1c reduction. The registry describes an analysis of covariance (ANCOVA) model and a primary intent-to-treat population. It also states that missing values were imputed using a washout imputation method under a missing-not-at-random framework. The covariates included in the model and further details of the missing-data implementation are not reported in the registry record.

The endpoint does not, on its own, establish how participants felt or functioned, nor does it determine the clinical importance of a given difference. A threshold for what would count as a clinically important HbA1c change is not reported in the registry record.

7. Results Posted in the Registry

• Sotagliflozin 400 mg: mean -1.03 versus -0.34 on control; difference -0.69 (95% CI -1.03 to -0.35), standardized effect d = -0.47; 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 2-hour Postprandial Plasma Glucose (PPG) Following a Mixed Meal at Week 26 (Sotagliflozin 400 mg Versus Placebo)

• 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:
  • Sotagliflozin 400 mg: mean -1.16 versus -0.59 on control; difference -0.56 (95% CI -1.11 to -0.02), standardized effect d = -0.24; z-test p = 0.041.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Sotagliflozin 400 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in 2-hour Postprandial Plasma Glucose (PPG) Following a Mixed Meal at Week 26 (Sotagliflozin 200 mg Versus Placebo)

• 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:
  • Sotagliflozin 200 mg: mean -0.85 versus -0.50 on control; difference -0.35 (95% CI -1.03 to 0.33), standardized effect d = -0.14; z-test p = 0.319.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sotagliflozin 200 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Fasting Plasma Glucose (FPG) at Week 26 (Sotagliflozin 400 mg Versus Placebo)

• 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:
  • Sotagliflozin 400 mg: mean -1.76 versus -0.20 on control; difference -1.55 (95% CI -2.31 to -0.80), standardized effect d = -0.47; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Sotagliflozin 400 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes extend the primary HbA1c question to other measures of glucose regulation, blood pressure, and body weight. The registry lists postprandial glucose and fasting glucose outcomes, systolic blood pressure outcomes, and body-weight outcomes, with comparisons involving the 400 mg and, for selected endpoints, the 200 mg group. The numeric secondary analyses posted cover only some of those registered outcomes.

Across the posted numeric glucose comparisons, the estimated changes generally point in a direction of greater reduction for a sotagliflozin group than for its placebo comparison. The degree of statistical uncertainty varies, and the 200 mg postprandial comparison does not show the same clear separation in its sponsor-reported inferential result as the primary endpoint. This broad pattern is consistent with the primary result being the strongest registered evidence in the posted analyses, but it does not make the secondary outcomes interchangeable with the primary endpoint.

Secondary results should carry less weight because they are not the primary question and multiple outcomes and comparisons create more opportunities for chance findings. The registry record does not report a multiplicity adjustment or a hierarchy governing interpretation of the secondary tests. Their p-values should therefore be understood as posted results, rather than assumed to retain a particular overall error rate.

Not all secondary outcomes listed in the record have numerical results in the analysis summaries on this page. Where estimates, intervals, or tests are not posted, the direction and size of those outcomes cannot be assessed from this record. The secondary evidence is supplementary and should not be used to fill in missing outcome information or to make claims about endpoints whose numerical results are absent.

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

• Change From Baseline in 2-hour Postprandial Plasma Glucose (PPG) Following a Mixed Meal at Week 26 (Sotagliflozin 400 mg Versus Placebo)
• Change From Baseline in 2-hour Postprandial Plasma Glucose (PPG) Following a Mixed Meal at Week 26 (Sotagliflozin 200 mg Versus Placebo)
• Change From Baseline in Fasting Plasma Glucose (FPG) at Week 26 (Sotagliflozin 400 mg Versus Placebo)
• Change From Baseline in Systolic Blood Pressure (SBP) at Week 12 in Participants With Baseline SBP ≥130 Millimeter of Mercury (mmHg) (Sotagliflozin 400 mg Versus Placebo)
• Change From Baseline in SBP at Week 12 for All Participants (Sotagliflozin 400 mg Versus Placebo)
• Change From Baseline in SBP at Week 12 for All Participants (Sotagliflozin 200 mg Versus Placebo)
• Change From Baseline in Body Weight at Week 26 (Sotagliflozin 400 mg Versus Placebo)
• Change From Baseline in Body Weight at Week 26 (Sotagliflozin 200 mg Versus Placebo)

9. Statistical Interpretation

10. Statistical Methods Explained

What does ANCOVA contribute?

The primary analysis used analysis of covariance (ANCOVA). In general, ANCOVA compares groups while accounting for selected covariates in a statistical model. This can improve precision when the covariates are informative and can adjust the group comparison for measured baseline differences. The registry identifies the method but does not report the covariates, the model specification, or diagnostic checks. Those details should not be inferred.

The reported quantity is a difference in least-squares means. In general, least-squares means are model-based group estimates; their difference is therefore not necessarily identical to subtracting two unadjusted raw averages. The posted arm values and the reported contrast are consistent in direction and point estimate, but the contrast’s uncertainty comes from the ANCOVA analysis.

How should the confidence interval be read?

A confidence interval describes a range of estimates compatible with the data and the statistical procedure, under the model’s assumptions. In repeated use, a 95% confidence-interval procedure would cover the underlying quantity in 95% of comparable repetitions. It is not a probability statement that this particular interval contains the true value with a 95% chance. The sponsor’s interval for the primary difference extends from -0.975 to -0.415. Its endpoints are both negative, so the interval does not include zero.

The unadjusted recalculation uses a different route: it is based on posted arm-level estimates and standard errors, and its pooled standard deviation is approximate. It is useful as a simple cross-check, but it does not incorporate the sponsor’s ANCOVA adjustment. The recalculation’s interval therefore need not match the sponsor’s interval even when the reported point estimates are the same. The registry does not describe a discrepancy between the recalculated and reported primary point estimate.

What do the p-values tell us?

In general, a p-value summarizes how inconsistent the observed data are with a specified null hypothesis, assuming that hypothesis and the statistical model. Smaller p-values indicate greater incompatibility with the null under those assumptions. A p-value is not an effect size, a measure of clinical importance, or the probability that a finding is due to chance. The reported primary p-value is from the sponsor’s ANCOVA analysis; the recalculated p-value is unadjusted and should not be treated as the sponsor’s planned inferential result.

The registry lists a superiority analysis for the primary contrast. A superiority test evaluates whether the data support a difference in the prespecified direction or, depending on the test definition, a difference from no effect. The record identifies the result type but does not report enough detail to reconstruct every testing convention, such as whether the formal test was one-sided or two-sided.

What does the intent-to-treat population mean?

The primary population is described as participants in the intent-to-treat population randomized to sotagliflozin 400 mg or placebo under the original protocol and Protocol Amendment 1. In general, an intent-to-treat approach analyzes participants according to randomized assignment, helping preserve the comparability created by randomization. The registry’s description also narrows the stated primary population by protocol version. The precise participant disposition and reasons for any exclusions are not reported in the registry record.

How does missing-data handling matter?

The registry says missing values were imputed using washout imputation under a missing-not-at-random framework. General context: imputation replaces missing measurements with modeled or rule-based values so that an analysis can include participants who lack an observed endpoint. A missing-not-at-random framework allows the probability of missingness to depend on unobserved outcomes or other factors, rather than assuming that missing values are interchangeable with observed ones. Different assumptions can produce different estimates, so the method is part of the interpretation of the primary result.

The registry does not describe the exact imputation algorithm, the amount or pattern of missing primary endpoint data, sensitivity analyses, or how robustness to alternative assumptions was assessed. The named method should therefore be reported as written without inventing implementation details.

Why do the secondary results need care?

Several outcomes and dose comparisons are listed. General statistical context: as the number of statistical tests increases, the chance of at least one apparently unusual result can increase if each test is judged separately. Multiplicity procedures can address this, but a procedure and its results are not reported in the registry record. The secondary p-values should therefore be read as posted, without assuming they were adjusted or that they share the primary analysis’s status.

General context: over-dispersion refers to variability greater than a model expects. It matters when the analysis assumes less variability than the data contain, because uncertainty can then be understated. The primary endpoint is continuous and is analyzed with ANCOVA; over-dispersion is not identified as a feature of this analysis in the registry record. It is not appropriate to claim that it affected this trial.

Finally, power is a design concept, not a retrospective probability that a particular result is true. The planning illustrations shown with the trial use the observed effect as a starting point and specified assumptions. The registry’s original design assumptions and sample-size justification are not reported, so those illustrations cannot be presented as the trial’s original design plan.

11. Confidence Intervals and Precision

The sponsor’s primary 95% confidence interval runs from -0.975 to -0.415 percentage points for the difference in least-squares mean change in HbA1c. This interval gives the reader more information than the point estimate alone: it shows a range of values compatible with the ANCOVA estimate under its model and assumptions. Because both endpoints are below zero, the interval supports a difference in the direction of a larger HbA1c reduction in the 400 mg group, as defined by the comparison.

The interval is not a range containing most individual participants’ responses. It describes uncertainty about the group contrast, not the spread of individual outcomes. The arm-level standard errors, 0.12 for placebo and 0.122 for sotagliflozin 400 mg, describe uncertainty around the posted least-squares means. They are not standard deviations of participant-level HbA1c values.

The simple unadjusted recalculation has a 95% confidence interval from -1.025 to -0.355. It is wider than the sponsor’s reported interval at one end and narrower at the other, reflecting that it is a different calculation using summary statistics rather than the sponsor’s ANCOVA model. The recalculation is explicitly unadjusted and its pooled standard deviation is approximate. For inference about the trial’s primary analysis, the sponsor’s reported interval is the relevant one.

Precision is not the same as clinical importance. The interval indicates statistical uncertainty around an estimated average contrast; it does not specify what magnitude would matter to participants or clinicians. A threshold for clinical importance and longer-term outcome information are not reported in the registry record. The confidence interval also depends on the analysis model and missing-data assumptions, whose full details are not available.

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 table and curve are planning illustrations based on the observed effect, not the trial’s original design calculation. For these continuous outcomes, the figures are participants per arm, not total participants or total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each scenario uses the stated fraction of the observed effect as the effect to detect.

The 80% power column shows the illustrated number of participants per arm for a design that would have an 80% chance of detecting the specified effect under those assumptions, if the assumptions hold. The 90% power column shows the corresponding planning target for a 90% chance. These are prospective design concepts: they describe performance under a hypothetical repeated-study framework, not the probability that this completed trial’s finding is correct.

Rows that use a smaller fraction of the observed effect require more participants per arm in the illustration. That is because smaller differences are harder to distinguish from random variation when other assumptions are held fixed. For the primary endpoint, the table shows planning for 100%, 75%, and 50% of the observed effect; the listed per-arm requirements rise as the target effect becomes smaller. The curve visualizes the same planning relationship and should be interpreted alongside the stated assumptions.

This is not observed power. The registry record does not report the original trial sample-size justification or the assumptions used when the trial was designed. The illustration therefore cannot establish why enrollment was set at its recorded level, nor can it be used to evaluate the trial’s original planning process.

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 = -0.690; pooled standard deviation σ = 1.462.
2. n = 2 × (1.96 + 0.842)² × 1.462² / -0.690² = 70.5, rounded up to 71 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 <- -0.69                      # observed difference in means
sigma <- 1.46198665968724                     # 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(2926937)
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 = 71 (80% power), 95 (90% power)
 75% of observed effect: n per arm = 126 (80% power), 168 (90% power)
 50% of observed effect: n per arm = 282 (80% power), 378 (90% power)
power.t.test: 71.4 
Simulated power at n = 71 per arm: 0.796 (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 reports the primary comparison using ANCOVA, with a difference in least-squares means of -0.69, a 95% confidence interval from -0.975 to -0.415, and p < 0.001. This is the trial’s stated primary analysis and should be distinguished from the simple unadjusted recalculation shown alongside it.

The recalculation uses the posted arm-level estimates and standard errors to construct an unadjusted mean difference and interval. It gives the same point estimate, -0.69, but a 95% confidence interval from -1.025 to -0.355 and an unadjusted p-value of <0.001. The difference in uncertainty and p-value is expected to be possible because the calculations do not use the same model. In particular, the recalculation does not reproduce the sponsor’s ANCOVA adjustment, and its pooled standard deviation is approximate.

The registry record does not flag a discrepancy in the primary result. The recalculation is best treated as a transparent descriptive check, not as an alternative sponsor analysis or a correction. The ANCOVA result is the appropriate reference for what the sponsor reported; the recalculation illustrates how summary-statistic methods can produce a related but not identical inferential result.

15. Multiplicity, Missing Data and Analysis Populations

The primary analysis is identified as an ANCOVA comparison of sotagliflozin 400 mg with placebo for change in HbA1c at week 26. The registry gives the analysis population and missing-data method, but it does not report a full statistical analysis plan, model covariates, or a detailed hierarchy of hypotheses. The formal treatment of multiplicity across the primary and secondary tests is not reported in the registry record.

This matters because the record includes more than one dose, multiple endpoint types, and multiple time points for some outcomes. General statistical context: testing many hypotheses without adjustment can make an isolated small p-value less persuasive than a p-value from a single prespecified primary test. The registry does not establish whether the secondary comparisons were adjusted, ordered in a testing hierarchy, or considered exploratory. The posted secondary p-values should not be treated as multiplicity-controlled unless such control is documented.

The primary population is described as intent-to-treat participants randomized to 400 mg or placebo under the original protocol and Protocol Amendment 1. The exact handling of participants outside that definition, exclusions from analysis, and participant flow are not detailed in the posted primary description. Those details are not reported in the registry record.

For missing primary endpoint data, the registry specifies washout imputation under a missing-not-at-random framework. The amount of missingness, the practical imputation rules, and sensitivity analyses under alternative assumptions are not reported. Missing-data assumptions can influence estimated group differences, so the stated method is important, but the record does not permit a full assessment of its implementation or robustness.

Interim analyses, stopping rules, covariate selection, and the original sample-size justification are also not reported in the registry record. Their absence should not be taken as evidence that a particular procedure was or was not used.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

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: multiple fractures (0.9%), road traffic accident (0.9%), traumatic haemothorax (0.9%), traumatic lung injury (0.9%), chronic kidney disease (0.9%).

The safety summary lists serious adverse events for the placebo, sotagliflozin 400 mg, and sotagliflozin 200 mg groups. The posted counts are 3 of 150 participants for placebo, 3 of 142 for sotagliflozin 400 mg, and 2 of 107 for sotagliflozin 200 mg. The same groups list other adverse events affecting 20, 17, and 14 participants, respectively. The registry reports 0 deaths in each group.

These are descriptive counts from the registry record. They should be read with the group denominators and the different treatment groups in mind; they do not by themselves establish that a treatment caused or prevented an event. The record also lists serious-event terms with percentages, including multiple fractures, road traffic accident, traumatic haemothorax, traumatic lung injury, chronic kidney disease, and retinal detachment. No causal relationship should be inferred from the term list alone.

The registry summary does not provide the full event-level context, timing, exposure-adjusted rates, or a formal safety comparison. These figures therefore describe what is posted, not a complete assessment of comparative safety.

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

The primary result describes an average difference in HbA1c change between randomized groups at week 26. The sponsor’s ANCOVA estimate is -0.69 percentage points, with a 95% confidence interval from -0.975 to -0.415. The interval does not include zero, and the reported p-value is < 0.001. In statistical terms, the result supports a difference between the groups under the stated analysis and its assumptions.

That conclusion is narrower than saying that every participant experienced a change of the same size or direction. The analysis estimates an average group contrast; individual responses can vary. It is also narrower than a conclusion about the 200 mg dose, other endpoints, or outcomes after the treatment period. The 200 mg group is not part of the primary comparison.

Statistical evidence and clinical importance are different questions. The registry record does not state a threshold for a clinically important HbA1c difference, and the confidence interval does not answer that question on its own. Nor does the primary endpoint establish effects on symptoms, complications, quality of life, or longer-term outcomes; those conclusions cannot be drawn from the posted primary result.

The result is conditional on the primary population and methods described in the registry. In particular, it comes from an ANCOVA analysis and uses washout imputation under a missing-not-at-random framework for missing values. Further details about the model and sensitivity analyses are not reported in the registry record. The statistical result should be interpreted within those limits, alongside the secondary and safety information rather than as a complete account of every potential outcome.

18. Why This Trial Matters Statistically

This trial illustrates why the reported estimand and method matter when reading a clinical result. The primary question is not simply whether the two arm-level values differ; it is the week 26 difference in least-squares mean change analyzed with ANCOVA in a defined intent-to-treat population. The registry’s adjusted estimate and the simple unadjusted recalculation have the same point estimate but different confidence intervals and p-values. That comparison shows how analysis choices affect the uncertainty reported around an effect.

It also demonstrates the distinction between a point estimate and its precision. The reported difference summarizes the estimated average contrast, while the confidence interval shows the uncertainty associated with the sponsor’s model. A p-value alone would leave out both the range of plausible estimates and the practical size of the effect.

The record includes multiple secondary outcomes, dose comparisons, and safety summaries. These widen the statistical questions beyond the primary endpoint and make transparent reporting of analysis plans, missing-data methods, and multiplicity procedures important. Some such details are not reported in the registry record, limiting how fully readers can reconstruct the analysis.

Finally, the planning illustrations show how assumptions about the effect to detect relate to participant requirements, but they are not the trial’s original design justification. Keeping planning calculations separate from observed results prevents retrospective power figures from being mistaken for evidence about the credibility of the trial’s findings.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not provide the full statistical analysis plan, including ANCOVA covariates, model diagnostics, or the details of the testing hierarchy. It also does not report whether or how multiplicity was controlled across the several listed outcomes and dose comparisons. This makes it difficult to assess the inferential status of secondary results beyond their stated analyses.

Missing values for the primary analysis were handled with washout imputation under a missing-not-at-random framework, but the amount and pattern of missingness, implementation details, and sensitivity analyses are not reported. The record therefore identifies an important assumption without allowing a full assessment of how strongly the result depends on it.

The primary analysis population is limited to participants in the intent-to-treat population randomized to the 400 mg and placebo groups under the original protocol and Protocol Amendment 1. Detailed participant flow and the handling of other protocol versions are not described. The reported primary analysis counts should not be treated as a complete account of enrollment or follow-up.

The trial’s original sample-size justification, interim analyses, and stopping rules are not reported in the registry record. The planning figures accompanying the results are illustrations based on observed effects, not evidence of the original design assumptions. The safety summary is also limited: it gives counts but not full event narratives, timing, exposure-adjusted rates, or a formal comparative safety analysis.

Finally, the primary comparison concerns HbA1c change at week 26 in the specified groups. The record does not establish longer-term effects or provide a threshold for clinical importance. Those limits constrain interpretation beyond the posted endpoint and follow-up period.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of Sotagliflozin Versus Placebo in Patients With Type 2 Diabetes Mellitus Not Currently Treated With Antidiabetic Therapy — NCT02926937.
• Registry record last updated 2021-06-21; status last verified 2021-06.
• Results first posted 2021-06-21.
• 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

This completed phase 3 study randomized participants with type 2 diabetes to sotagliflozin 400 mg, sotagliflozin 200 mg, or placebo after a 2-week run-in. Double-blind treatment continued for up to 26 weeks. The primary question compared sotagliflozin 400 mg with placebo on change from baseline in HbA1c at week 26.

The sponsor’s ANCOVA analysis reports a difference in least-squares mean change of -0.69 percentage points, with a 95% confidence interval from -0.975 to -0.415 and p < 0.001. The primary population and missing-data method are described in the registry, while fuller model details, multiplicity procedures, and original sample-size assumptions are not reported in the registry record.

Secondary outcomes and descriptive safety counts provide additional context, but their interpretation is limited by the availability of posted analyses and the absence of reported multiplicity methods. The accompanying sample-size figures are planning illustrations, not the trial’s original design calculation.