Study to Investigate the Efficacy and Pharmacokinetic Profile of K-877-ER Compared to K-877-IR: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Investigate the Efficacy and Pharmacokinetic Profile of K-877-ER Compared to K-877-IR (NCT04447820), a randomized phase 2 study. The primary outcome is Estimated Percentage Change in Fasting Triglyceride(s) (TG).

2. Clinical Question

3. Trial Design

This was a Phase 2 randomized study with a parallel-group design and no masking. It compared K-877-ER Dose A once daily and K-877-ER Dose B once daily against K-877-IR twice daily. The registry lists planned enrollment of 96 and the condition as dyslipidemia. The study status is completed; its start date was 2020-06-10 and its primary completion date was 2020-12-03. The study duration is described as 4 weeks, with the primary outcome assessed from baseline to Day 28.

The listed primary outcome is estimated percentage change in fasting triglycerides. Its analysis population was the PPS, defined as patients in the Full Analysis Set without major protocol deviations and with both baseline and Day 28 fasting serum triglyceride measurements. The posted primary summaries contain 31.0 participants for K-877-IR, 30.0 for Dose A, and 27.0 for Dose B. These are analysis-group counts, not a statement about the number enrolled or all participants randomized.

Randomization can help balance groups at the outset, although the registry record does not report the allocation sequence method or whether allocation was concealed. The open-label design means treatment assignment was not masked. The registry identifies the study as multicenter, but the number and locations of centers are not reported in the registry record. Specific dose amounts, eligibility details beyond the description of adult patients with fasting triglyceride levels, and the trial's original sample-size justification are also not reported in the registry record.

The sponsor reported non-inferiority analyses using a mixed model for repeated measures (MMRM). The model specification, analysis time points, non-inferiority margin, handling of missing observations, and other details of the statistical plan are not reported in the registry record. Because the registry's primary endpoint summary and the sponsor's reported tests do not agree with the simple unadjusted recalculations, interpretation should give precedence to the sponsor's own analysis.

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: The Screening Period occurred no more than 70 days and no less than 7 days prior to the first dose of study drug. The up to 10-week Screening Period allowed for a 6-week washout period, if needed, followed by an up to 4-week Qualification Period that was used to assess patient eligibility. Prior to treatment on Day 1, eligible patients were randomized in a 1:1:1 ratio to 1 of 3 treatment groups. Randomization was stratified by sex, statin use, and PK assessment.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The Per-Protocol Set (PPS) was defined as all patients in the Full Analysis Set without any major protocol deviations and who had baseline and Day 28 fasting serum TG measurements.

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 Full Analysis Set (FAS) was defined as all randomized patients who took at least 1 dose of study drug and had a baseline fasting TG assessment.

6. Primary Endpoint

The registry lists the primary outcome as Estimated Percentage Change in Fasting Triglyceride(s) (TG), measured over Baseline to Day 28.

The primary endpoint is estimated percentage change in fasting triglycerides (TG) from baseline to Day 28. A percentage-change endpoint expresses how a measurement at follow-up relates to its baseline value, rather than reporting only the follow-up concentration. The registry labels the primary summary as a least-squares mean and gives dispersion as a standard deviation. The reported analysis population is the PPS.

Fasting triglycerides are the lipid measure directly named in the study question, so tracking their change addresses the short-term biochemical outcome the study set out to compare across formulations and dosing schedules. A continuous endpoint retains information about the size and direction of changes across participants, rather than reducing the result to a category such as improved or not improved. The negative estimated percentage changes in the posted group summaries indicate decreases relative to baseline within each group; the summaries do not by themselves establish the comparative result of the sponsor's non-inferiority tests.

The endpoint is a laboratory measure over the study's 4-week period, not a clinical event or a measure of long-term health outcomes. The registry record does not report a threshold for a clinically meaningful difference in triglycerides, nor does it explain how such a threshold was selected. It also does not report the detailed model specification underlying the least-squares means. These omissions limit interpretation of what magnitude of change would be important and how the sponsor's adjusted estimates relate to the descriptive summaries.

7. Results Posted in the Registry

• K-877-ER Dose A: mean -36.79 versus -41.57 on control; difference 4.78 (95% CI 3.12 to 6.44), standardized effect d = 1.45; z-test p = <0.001.
• K-877-ER Dose B: mean -38.94 versus -41.57 on control; difference 2.63 (95% CI 0.86 to 4.40), standardized effect d = 0.77; z-test p = 0.004.

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.

Percentage Change From Baseline to Day 28 in Total Cholesterol (TC)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • K-877-ER Dose A: mean -8.68 versus -6.07 on control; difference -2.61 (95% CI -9.82 to 4.60), standardized effect d = -0.18; z-test p = 0.478.
  • K-877-ER Dose B: mean 0.28 versus -6.07 on control; difference 6.35 (95% CI -2.37 to 15.07), standardized effect d = 0.38; z-test p = 0.154.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between K-877-ER Dose A and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between K-877-ER Dose B and the control. The interval gives the range of effects that remain compatible with the data.

Percentage Change From Baseline to Day 28 in Low-density Lipoprotein Cholesterol (LDL-C)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • K-877-ER Dose A: mean -2.16 versus 3.56 on control; difference -5.72 (95% CI -18.36 to 6.92), standardized effect d = -0.23; z-test p = 0.375.
  • K-877-ER Dose B: mean 22.71 versus 3.56 on control; difference 19.15 (95% CI -7.15 to 45.45), standardized effect d = 0.39; z-test p = 0.154.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between K-877-ER Dose A and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between K-877-ER Dose B and the control. The interval gives the range of effects that remain compatible with the data.

Percentage Change From Baseline to Day 28 in High-density Lipoprotein Cholesterol (HDL-C)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • K-877-ER Dose A: mean 15.39 versus 12.22 on control; difference 3.17 (95% CI -8.45 to 14.79), standardized effect d = 0.14; z-test p = 0.593.
  • K-877-ER Dose B: mean 20.04 versus 12.22 on control; difference 7.82 (95% CI -5.49 to 21.13), standardized effect d = 0.30; z-test p = 0.249.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between K-877-ER Dose A and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between K-877-ER Dose B and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes include changes in total cholesterol, LDL-C, HDL-C, remnant cholesterol, non-HDL-C, and free fatty acids, as well as K-877 pharmacokinetic parameters. The registry lists Cmax and AUC (Tau) among the pharmacokinetic outcomes. Posted summaries include eight secondary outcomes, but the registry record does not report numerical pharmacokinetic results in the secondary summaries. The lipid outcomes are additional laboratory measures assessed over the same baseline-to-Day 28 period.

Across the lipid summaries that are posted, estimates vary in direction and size across dose groups and measures. The uncertainty intervals for the recalculated comparisons generally allow a range of possible contrasts rather than establishing a consistent pattern across all lipid outcomes. Those secondary results provide context about other measured lipid changes, but they do not replace the primary fasting triglyceride endpoint or the sponsor's non-inferiority analysis. No secondary outcome should be treated as confirmation of the primary result merely because its estimate points in a particular direction.

Unadjusted secondary results carry less weight because multiple outcomes and comparisons create more opportunities for apparently notable findings by chance. Unless an adjustment or a prespecified hierarchy controls this issue, a nominal p-value is not equivalent to confirmatory evidence across the full set of analyses. The multiplicity strategy and the status of these outcomes as confirmatory or exploratory are not reported in the registry record. The secondary outcomes are therefore best read as descriptive and supportive evidence, with attention to their uncertainty and the absence of a reported multiplicity plan.

The registry record also does not provide a single integrated interpretation connecting the primary and secondary findings, nor does it state how pharmacokinetic findings were intended to inform the efficacy comparison. The secondary measurements broaden the set of short-term biological outcomes observed, but they do not establish effects beyond the reported study period or on clinical outcomes.

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

• Percentage Change From Baseline to Day 28 in Total Cholesterol (TC)
• Percentage Change From Baseline to Day 28 in Low-density Lipoprotein Cholesterol (LDL-C)
• Percentage Change From Baseline to Day 28 in High-density Lipoprotein Cholesterol (HDL-C)
• Percentage Change From Baseline to Day 28 in Remnant Cholesterol
• Percentage Change From Baseline to Day 28 in Non-HDL-C
• Percentage Change From Baseline to Day 28 in Free Fatty Acid (FFAs)
• K-877 PK Parameters Cmax
• K-877 PK Parameters AUC (Tau)

9. Statistical Interpretation

10. Statistical Methods Explained

For a continuous endpoint, a group mean summarizes the average observed or modeled outcome within that group. In this trial the primary summaries are least-squares means of percentage change in fasting triglycerides from baseline to Day 28. A least-squares mean is generally a model-based group estimate; it can differ from a raw arithmetic average when the model adjusts for features of the data. The registry does not describe the covariates or detailed model specification used to obtain these primary summaries.

What is the sponsor's MMRM analysis?

MMRM means mixed model for repeated measures. In general, this family of models can analyze repeated observations from participants while accounting for the fact that observations from the same person are related. It can estimate differences over time and use a specified covariance structure for repeated measurements. The registry record identifies MMRM as the sponsor's method for the non-inferiority analyses, but does not report its covariance structure, fixed effects, covariates, time points included, missing-data assumptions, or exact contrast estimates. The method name alone is not enough to reproduce the sponsor's analysis.

A non-inferiority analysis asks whether the evidence is consistent with the new regimen being no worse than a comparator by more than a prespecified margin. That margin is essential to interpretation: it defines what degree of disadvantage is considered acceptable for the trial's statistical question. The margin is not reported in the registry record. Nor are the sponsor's contrast estimates and confidence intervals. Consequently, the posted p-values for the tests cannot be translated into a direct statement about how close either formulation was to the margin.

What do the recalculated comparisons represent?

The simple recalculations contrast the posted primary group means for each extended-release regimen with the posted K-877-IR mean. They are unadjusted comparisons based on summary statistics, not participant-level data and not a reproduction of MMRM. The pooled standard deviations and effect-size summaries used for these recalculations are derived from the posted group summaries. The primary calculations use approximate standard deviations, as identified in the comparison details. Such approximations can be useful for a transparent descriptive check, but they need not match estimates from an adjusted repeated-measures model.

The recalculated p-values are unadjusted. In general, an unadjusted p-value describes how compatible the observed comparison is with a specified null model under the calculation's assumptions. It does not account for the fact that the study reports multiple comparisons and outcomes unless an adjustment has been applied. These values should not be substituted for the sponsor's MMRM non-inferiority tests. The registry marks the discrepancy as present, so the sponsor's own analysis takes precedence for the reported trial analysis.

Why does dispersion matter?

Standard deviation describes variation among observations around a group mean. Greater variability makes a mean difference harder to estimate precisely, all else being equal. The standard errors in a comparison reflect both the dispersion and the amount of information in the groups. For the primary comparisons, the listed group standard errors are 0.592 and 0.606 for K-877-IR and Dose A, and 0.592 and 0.684 for K-877-IR and Dose B. These calculations use the summary-statistic approach and approximate standard deviations; they are not standard errors from the sponsor's reported MMRM.

In general, over-dispersion means that variability is larger than a model assumes. When a model understates variability, its standard errors can be too small and its intervals too narrow. The registry does not report that over-dispersion was assessed or found in this study, so this is a general caution rather than a finding about these data. More broadly, statistical conclusions depend on whether the model and its assumptions are suitable for the endpoint and design.

How should a p-value be understood?

In general, a p-value measures how unusual data at least as incompatible with a null hypothesis would be under that null and the model assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. A small p-value from an unadjusted mean comparison does not answer a non-inferiority question unless the test, margin, and analysis are appropriate to that question. Likewise, a non-inferiority p-value without the margin and corresponding estimate cannot show the size or direction of the treatment contrast.

Randomization supports comparative interpretation by reducing systematic differences at allocation on average, but it does not eliminate uncertainty, missing data, deviations, or limitations from the analysis population. The primary PPS definition excludes patients with major protocol deviations and requires baseline and Day 28 measurements. The registry record does not report results for other analysis populations, the number excluded from the PPS, or the reasons for missing endpoint measurements. Those details would help readers understand how the analyzed group relates to all randomized participants.

11. Confidence Intervals and Precision

A confidence interval describes a range of values compatible with the data and the calculation's assumptions, at its stated confidence level. It is useful for assessing precision because a narrower interval indicates less statistical uncertainty about the estimated contrast, while a wider interval allows a broader range of values. It does not guarantee that the true value lies within the particular interval, and it does not capture biases or model misspecification.

For the simple recalculated primary comparison of Dose A with K-877-IR, the mean difference is 4.78 and the 95% confidence interval is 3.12 to 6.44. For Dose B compared with K-877-IR, the mean difference is 2.63 and the 95% confidence interval is 0.858 to 4.402. These intervals are based on the recalculation from posted summary statistics, not on the sponsor's MMRM non-inferiority model. The registry does not report confidence intervals for the sponsor's non-inferiority analyses, so their precision cannot be assessed from those entries.

Precision depends on the observed variability, the analyzed group sizes, and the calculation method. The primary group summaries have PPS sizes of 31.0, 30.0, and 27.0, while the reported recalculations use approximate standard deviations for these primary comparisons. The displayed intervals therefore should not be read as adjusted or as a substitute for the sponsor's inferential results. They also do not incorporate any multiplicity adjustment; whether the sponsor applied adjustments is not reported in the registry record.

Statistical precision is distinct from clinical importance. The registry record does not identify a clinically meaningful difference or non-inferiority margin against which these intervals could be judged. The results give estimates of short-term change, but the posted sponsor analysis lacks the interval information needed to assess its uncertainty directly.

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 figures are planning illustrations based on the observed effect, not a description of the trial's original design calculation. For these continuous-outcome comparisons, the entries 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 row uses the stated fraction of the observed effect as the effect to detect.

The 80% power column gives the illustrative number of participants per arm that would yield an 80% chance of detecting the chosen effect under those assumptions, if the assumed effect and other planning conditions held. The 90% power column applies the same idea with a higher target probability of detection. These are prospective planning quantities; they are not evidence about the probability that the completed study detected an effect, and this is not observed power.

The rows for 100%, 75%, and 50% of the observed effect show how the planning result changes when the assumed effect is reduced. Smaller effects need more participants to distinguish them from random variation at the same significance and power targets. For the primary comparison with Dose A, the illustration for 75% of the observed effect is 14 participants per arm for 80% power and 18 for 90% power. For Dose B, the corresponding values are 48 and 64 participants per arm. These figures illustrate sensitivity to the assumed effect; they do not establish the sample size the trial should have used or that it was adequately designed.

The sample-size illustrations are based on simple assumptions and observed effects from the posted summaries. The registry record does not report the trial's original sample-size justification or original design assumptions. The calculations do not automatically incorporate the sponsor's MMRM model, the PPS definition, multiplicity, or any difference between the observed and future effect. The sample-size curve should be read as a visual representation of the same planning relationship, not as an additional observed result.

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 = 4.780; pooled standard deviation σ = 3.307.
2. n = 2 × (1.96 + 0.842)² × 3.307² / 4.780² = 7.5, rounded up to 8 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 <- 4.78                      # observed difference in means
sigma <- 3.30681837718085                     # 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(4447820)
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 = 8 (80% power), 11 (90% power)
 75% of observed effect: n per arm = 14 (80% power), 18 (90% power)
 50% of observed effect: n per arm = 31 (80% power), 41 (90% power)
power.t.test: 8.6 
Simulated power at n = 8 per arm: 0.765 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The registry reports sponsor analyses comparing K-877-IR with each K-877-ER dose using MMRM, identified as non-inferiority analyses. The reported p-values are 0.308 for the comparison with Dose A and 0.593 for the comparison with Dose B. The registry does not report the sponsor's estimated contrasts, confidence intervals, or non-inferiority margin. Without those details, the p-values alone do not establish whether either regimen satisfied the prespecified non-inferiority criterion.

The simple recalculations compare the posted primary group means without the sponsor's repeated-measures model. They produce different unadjusted results from the sponsor-reported analyses, and the registry record marks a discrepancy. These recalculations are useful as a transparent check of the summary statistics, but they do not reproduce the sponsor's analysis or resolve why the analyses differ. The difference could not be attributed to a specific modeling choice from the registry entries alone.

Accordingly, the sponsor's own analysis takes precedence when describing the trial's reported inferential result. This does not make the sponsor p-values sufficient to evaluate non-inferiority independently; the missing margin and model estimates matter. The recalculated mean differences and intervals can be described as unadjusted statistical summaries, with their assumptions made clear, but should not be used to override the sponsor's MMRM results or to claim a non-inferiority conclusion.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several hypotheses are tested across outcomes, doses, or comparisons. In general, conducting more tests without an appropriate strategy increases the chance that at least one result appears statistically notable by chance. Possible approaches include adjusting significance thresholds, controlling a family-wise error rate, or defining a prespecified testing order. Which approach, if any, was used in this study is not reported in the registry record. The simple recalculated p-values are unadjusted, so they should not be treated as multiplicity-controlled evidence.

The registry includes a primary endpoint, multiple secondary outcomes, and comparisons of two extended-release doses with the immediate-release regimen. It does not report whether the hypotheses were organized into a hierarchy, whether the secondary analyses were exploratory, or how multiplicity was handled in the sponsor's tests. This limits how strongly nominal results across the outcome set can be interpreted together.

The primary analysis population is the PPS. It includes patients in the Full Analysis Set who had no major protocol deviations and had baseline and Day 28 fasting serum triglyceride measurements. This definition makes clear that participants without the required measurements or with major deviations were not included in that population. The registry record does not report the number or reasons for exclusions, outcome completion by randomized group, results from the Full Analysis Set, or whether the sponsor used additional analysis populations.

Missing-data handling is not reported in the registry record. For an MMRM analysis, assumptions about how observed measurements relate to missing measurements can affect interpretation; the model name alone does not specify those assumptions. The record also does not describe sensitivity analyses for missingness or protocol deviations. These omissions matter because the PPS definition conditions inclusion on observed endpoint data and protocol adherence. The available results can be interpreted for the defined PPS summaries, but the robustness of the findings to missing observations and alternative populations cannot be assessed from the registry entries.

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 non-serious adverse event terms

The posted safety figures list 32 participants at risk in each group for serious adverse events. The number reported as affected by a serious adverse event is 0 in K-877-IR, 0 in K-877-ER Dose A, and 0 in K-877-ER Dose B. Deaths are reported as 0 in each group. These figures describe the recorded serious adverse event and death summaries for the groups and should not be read as proof that no risks exist.

For other adverse events, the registry reports 0 affected in K-877-IR, 1 in Dose A, and 5 in Dose B, with 32 at risk in each group. These entries concern participants affected, not necessarily the total count of event episodes. The registry does not provide serious adverse event terms, and no top serious terms are listed. It also does not give further detail about event severity, timing, relatedness, or withdrawals due to adverse events.

Safety summaries from a short study describe events recorded during that study and do not establish comparative safety over longer periods. The group counts should be read alongside the study duration and the limits of the registry's reported safety detail.

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

The results address a short-term comparison of once-daily extended-release K-877 at two doses with twice-daily immediate-release K-877, using percentage change in fasting triglycerides from baseline to Day 28. The posted group estimates are negative in all arms, indicating decreases from baseline in the PPS summaries. The simple unadjusted comparisons estimate contrasts between those group means, while the sponsor's reported analysis uses MMRM for non-inferiority.

These are different statistical questions and analyses. The recalculated mean differences and their intervals describe unadjusted contrasts from posted summary statistics. The sponsor's p-values are from non-inferiority tests, but the registry does not report the non-inferiority margin or the model estimates and confidence intervals. Because the record flags a discrepancy, the sponsor's own analysis takes precedence. The information posted is not sufficient to independently verify the non-inferiority conclusion from the sponsor p-values alone.

The results do not establish that either extended-release dose is equivalent to immediate-release treatment, nor do the simple recalculations establish that one regimen is better. A non-inferiority conclusion depends on a prespecified margin and an appropriate analysis, not just on a comparison of means or a p-value. The margin is not reported in the registry record. Statistical differences, even when estimated precisely, also do not automatically indicate clinical importance; a meaningful difference threshold is not reported.

The primary measure is a short-term laboratory endpoint. The findings do not describe long-term lipid changes, cardiovascular outcomes, symptoms, or other patient-important outcomes. Nor can the results be generalized beyond the study population and conditions on the basis of this registry record alone. The trial's open-label design, PPS analysis, limited reported model detail, and missing-data information further constrain interpretation. These statistical results are not medical advice and do not determine what treatment an individual should receive.

18. Why This Trial Matters Statistically

This trial illustrates why the name of an analysis and the result of a simpler calculation should not be treated as interchangeable. The registry reports sponsor MMRM non-inferiority tests, while the posted summaries allow a straightforward unadjusted comparison of group means. The resulting analyses differ, and the record marks that discrepancy. Giving precedence to the sponsor's analysis is important, but the absence of its contrast estimates, confidence intervals, and non-inferiority margin means readers cannot fully assess its interpretation from the registry entries alone.

It also shows why a confidence interval and a p-value answer different questions. A p-value summarizes compatibility with a null under a model; an interval conveys a range of effect sizes consistent with the data and assumptions. In a non-inferiority setting, the margin must be considered alongside the estimate and interval. The sponsor p-values without those supporting details do not communicate the distance from the margin or the precision of the treatment contrast.

The analysis population is another important part of the statistical story. The PPS excludes participants with major protocol deviations and requires endpoint measurements at both baseline and Day 28. The registry does not report the exclusions or results from other analysis populations. This makes it difficult to assess how sensitive the result is to the population definition or missing endpoint data.

Finally, the sample-size illustrations show how planning depends on an assumed effect, but they are not the trial's design justification or evidence of observed power. A planning estimate based on the observed effect can vary substantially when the assumed effect is reduced. Keeping these distinctions clear helps prevent descriptive summaries, planning calculations, and sponsor inferential tests from being conflated.

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 provides limited detail about the methods behind the primary sponsor analysis. It identifies MMRM but does not report the model specification, covariates, covariance structure, or analysis time points. The sponsor's non-inferiority margin, treatment contrast estimates, and confidence intervals are not reported. These omissions prevent an independent assessment of the sponsor's non-inferiority results using the posted entries.

The simple primary comparisons are unadjusted recalculations from summary statistics and use approximate standard deviations. They do not reproduce the sponsor's model, and the record marks a discrepancy between them and the sponsor analysis. They should therefore be treated as descriptive checks rather than replacement inferential results. The recalculated confidence intervals and p-values do not account for multiplicity.

The primary analysis uses the PPS, which requires baseline and Day 28 measurements and excludes major protocol deviations. The record does not report how many participants were excluded, why measurements were missing, or how missing data were handled in the sponsor model. Results for alternative populations and sensitivity analyses are also not reported in the registry record.

The trial was open-label, and the endpoint reflects change over the 4-week study period. The registry does not report the specific dose amounts, detailed eligibility criteria, a clinically meaningful difference threshold, or the original sample-size justification. It also does not provide a multiplicity strategy for the primary and secondary outcomes. These limits constrain assessment of masking-related influences, generalizability, the clinical importance of estimated changes, and the strength of conclusions across the full outcome set.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study to Investigate the Efficacy and Pharmacokinetic Profile of K-877-ER Compared to K-877-IR — NCT04447820.
• Registry record last updated 2023-12-22; status last verified 2023-11.
• Results first posted 2023-12-22.
• 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

NCT04447820 is a completed Phase 2 randomized, open-label, parallel-group study of K-877-ER once daily compared with K-877-IR twice daily in adults with dyslipidemia and fasting triglyceride levels. It assessed estimated percentage change in fasting triglycerides from baseline to Day 28. The primary outcome summaries use the PPS, defined by protocol adherence and availability of baseline and Day 28 fasting serum triglyceride measurements.

The sponsor reported MMRM non-inferiority tests for both extended-release dose groups. Their p-values are posted, but the non-inferiority margin and model estimates and confidence intervals are not reported in the registry record. Simple recalculations from the group summaries differ from the sponsor results; the sponsor's own analysis takes precedence. Secondary lipid outcomes are posted, while numerical pharmacokinetic results and several statistical-plan details are not reported in the registry record.