MK0524B Bioequivalence Study (0524B-070): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for MK0524B Bioequivalence Study (0524B-070) (NCT00943124), a randomized phase 1 study. The primary outcome is Plasma Area Under the Curve (AUC(0 to 48hr)) for Simvastatin Acid.

2. Clinical Question

The study asks whether the pharmacokinetic exposure to simvastatin acid differs when simvastatin is taken as part of the triple-combination tablet MK0524B versus as a separate simvastatin tablet taken with MK0524A. It also belongs to a broader study evaluating bioequivalence of simvastatin, laropiprant, and extended-release niacin across the combination formulations. The primary endpoint in the registry record is plasma area under the curve for simvastatin acid through 48 hours after dosing.

This is a pharmacokinetic comparison, not a trial of clinical outcomes. The endpoint describes drug exposure over a specified period after dosing. The registry record does not report the equivalence margin needed to assess the ratio against a prespecified bioequivalence boundary.

3. Trial Design

This phase 1 study was randomized, open-label, and used a crossover model. Participants were assigned to one of two treatment sequences. In one sequence, participants received MK0524B in period 1 and separate simvastatin plus MK0524A in period 2. In the other, the order was reversed. Each described period involved a single tablet of MK0524B or the separate tablets of simvastatin and MK0524A.

A crossover design compares treatments within participants across periods. In general, this can reduce the influence of stable differences between participants, because each participant receives both regimens. Interpretation also depends on the period structure and on how the analysis addresses period or sequence effects. Those details are not reported in the registry record. The washout schedule is not reported in the registry record.

The study was open-label, meaning masking was not used. The registry lists 220 participants enrolled. For the primary simvastatin acid AUC endpoint, data were available for 200 participants in the MK0524B analysis and 202 in the simvastatin plus MK0524A analysis. The registry attributes unavailable data to early dropout (N=4), mishandling of plasma samples (N=10), and assay limitations (N=10). These reasons are reported in the endpoint description; the record does not explain how each affected the analysis sets.

The study is marked completed, with a start in 2007-07 and primary completion in 2007-08. The original design’s sample-size assumptions, including its planned statistical power and target effect, are not reported in the registry record. The available planning illustrations based on the observed effect are not the study’s original design 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.

Period 1

Reasons for not completing this period:

Period 2

Analysis population for the primary outcome, as defined in the registry: Two hundred-twenty (220) subjects were enrolled in this study. Due to early dropout (N=4), mis-handling of plasma samples (N=10) and the limitation of the assay (N=10), data from a total of 200 and 202 subjects were available for simvastatin acid AUC(0 to 48 hour) analysis for MK0524B and Simvastatin + MK0524A, respectively.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Plasma Area Under the Curve (AUC(0 to 48hr)) for Simvastatin Acid, measured over Through 48 Hours Post Dose.

The primary endpoint is plasma area under the concentration-time curve for simvastatin acid, the active metabolite of simvastatin, from zero to 48 hours after dosing. AUC integrates drug concentration over time, so it summarizes exposure across the stated post-dose interval rather than describing a single concentration measurement. The unit is ng/mL * Hour.

Pharmacokinetic exposure measures are used in bioequivalence studies to compare how much of a drug or active metabolite becomes available over time under different formulations. For this study, the question is whether simvastatin acid exposure following MK0524B is comparable with exposure following simvastatin taken with MK0524A. The endpoint is therefore suited to a formulation comparison; it does not directly measure symptom change, disease progression, or another clinical outcome.

The registry reports least-squares mean values and standard deviations for the two endpoint analyses. The posted estimates are 9.19 for MK0524B and 8.03 for simvastatin plus MK0524A. These values describe the endpoint as recorded for the analyzed participants, but the registry’s available entry does not give a detailed explanation of the least-squares model used. The separate sponsor-reported analysis expresses the comparison as a least-squares mean ratio.

The registry record does not report the bioequivalence acceptance limits or explain how the 48-hour collection interval was selected. Those details are important when interpreting whether a ratio falls within a prespecified range.

7. Results Posted in the Registry

• Simvastatin + MK0524A: mean 8.03 versus 9.19 on control; difference -1.16 (95% CI -2.74 to 0.42), standardized effect d = -0.14; z-test p = 0.150.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry includes two kinds of comparison for the primary continuous endpoint: a sponsor-reported least-squares mean ratio and a simple unadjusted recalculation based on posted summary statistics. These should not be treated as interchangeable estimates. The crossover design, endpoint scale, and statistical model all matter when interpreting the comparison.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based estimate of a group’s mean after accounting for terms included in the statistical model. It can differ from a raw arithmetic mean when the model adjusts for design features or other terms. The registry reports a least-squares mean ratio of 1.14, with a 90% confidence interval from 1.09 to 1.20, but the specific model terms and estimation method are not reported in the registry record. The ratio summarizes relative exposure between the listed treatment groups; the record does not describe its numerator and denominator in prose.

What does the unadjusted recalculation measure?

The unadjusted comparison uses the posted endpoint estimates of 9.19 and 8.03 and the posted standard deviations of 8.38 and 7.78. It reports a mean difference of -1.16, an approximate pooled standard deviation of 8.084, and an unadjusted p-value of 0.15. The confidence interval for the difference extends from -2.741 to a positive upper endpoint. The endpoint is continuous, so the mean difference is expressed on the same scale as the AUC measure, ng/mL * Hour.

This recalculation is not adjusted for the crossover structure, and it is not the least-squares ratio analysis reported by the sponsor. Its standard error uses an approximation from the available group summaries. A p-value is a measure of how compatible the observed data are with a specified null model under the calculation used; it is not the probability that the null hypothesis is true, and it does not by itself determine bioequivalence.

Why does the ratio scale matter?

For bioequivalence, relative exposure is commonly evaluated on a ratio scale. In general, a ratio can be useful when the scientific question concerns proportional rather than absolute differences. The confidence interval then needs to be considered against prespecified lower and upper equivalence bounds. The registry record does not report those bounds, so the reported ratio interval cannot be assessed against the study’s acceptance criterion from the record alone.

A difference scale is also informative, but it asks about the absolute separation between the reported endpoint estimates. A difference near zero and a ratio near one are corresponding concepts only when the estimates and model are appropriately defined; the two analyses use different methods and may not align exactly. The unadjusted comparison should therefore be treated as a descriptive recalculation rather than a reconstruction of the sponsor’s analysis.

How does crossover affect interpretation?

In general, crossover analyses can use within-participant comparisons and may account for period, sequence, and participant effects. The registry identifies a randomized crossover design and describes two treatment sequences, but the sponsor’s statistical model, period handling, sequence handling, and participant-level covariance structure are not reported in the registry record. The group summaries do not provide enough information to reproduce those features.

The open-label status describes masking, not the endpoint’s objectivity. Plasma AUC is a laboratory-based pharmacokinetic measure, but the registry notes sample mishandling and assay limitations as reasons that some endpoint data were unavailable. The record does not report assay validation details, sample handling procedures, or how measurements below assay limits were treated.

Overall, the sponsor-reported least-squares ratio is the analysis presented for the equivalence or non-inferiority assessment. The summary-based difference provides a supplementary, unadjusted perspective. Neither should be read as a clinical outcome, and a definitive equivalence interpretation requires the criterion and analysis details not reported in the registry record.

10. Confidence Intervals and Precision

The sponsor-reported ratio is 1.14, and its 90% confidence interval extends from 1.09 to 1.20. A confidence interval describes the uncertainty around an estimated effect under the analysis that produced it. In general, a narrower interval indicates greater precision, while a wider interval indicates greater uncertainty. The interval is expressed on the ratio scale and should be interpreted as uncertainty around relative AUC exposure, not as a range of individual participants’ values.

For a bioequivalence decision, the interval’s location must be compared with the prespecified acceptable limits. Those limits are not reported in the registry record. Consequently, the confidence interval alone does not show whether the sponsor’s criterion was met. The registry’s classification of the analysis as non-inferiority or equivalence does not replace the missing limits or state the conclusion.

The recalculated unadjusted mean difference is -1.16. Its confidence interval spans negative and positive values, so the summary-based estimate is not highly precise about the direction of the unadjusted difference. The unadjusted p-value is 0.15. This interval and p-value arise from a different calculation than the sponsor’s ratio analysis and should not be used to judge the sponsor’s equivalence criterion.

The available endpoint analysis counts are 200 for MK0524B and 202 for simvastatin plus MK0524A. Unavailable measurements due to dropout, sample mishandling, and assay limitations may affect precision and could affect interpretation if missingness is related to exposure. The registry record does not report an analysis of missingness or sensitivity analyses. The reported estimates therefore describe the analyzed data under methods whose full details are not available in the registry record.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total event counts. They illustrate the sample size that would be needed under the stated planning assumptions when targeting 80% or 90% power. The 80% column means an 80% chance of detecting the specified effect under the assumed model if that effect is present; the 90% column represents a higher target chance under those same assumptions.

The assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The effect used for each illustration is the stated fraction of the observed effect: 100%, 75%, or 50%. The figures are participants per arm for 80% power or participants per arm for 90% power, respectively. For example, planning for 100% of the observed effect gives 763 participants per arm for 80% power and 1,021 participants per arm for 90% power. Planning for 75% of the observed effect gives 1,356 participants per arm for 80% power and 1,815 participants per arm for 90% power. Planning for 50% of the observed effect gives 3,050 participants per arm for 80% power and 4,083 participants per arm for 90% power.

Smaller effects require more participants to distinguish them from random variation under the same assumptions. These are planning illustrations based on the observed effect, not the original trial’s documented design assumptions; those original assumptions are not reported in the registry record. They are not observed power and do not describe the probability that the completed study reached a particular conclusion.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -1.160; pooled standard deviation σ = 8.084.
2. n = 2 × (1.96 + 0.842)² × 8.084² / -1.160² = 762.4, rounded up to 763 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 <- -1.16                      # observed difference in means
sigma <- 8.08406828273982                     # 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(943124)
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 = 763 (80% power), 1,021 (90% power)
 75% of observed effect: n per arm = 1,356 (80% power), 1,815 (90% power)
 50% of observed effect: n per arm = 3,050 (80% power), 4,083 (90% power)
power.t.test: 763.4 
Simulated power at n = 763 per arm: 0.803 (20000 replicates)

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

13. Analyses Reported by the Sponsor

The registry reports the sponsor’s analysis as a least-squares mean ratio of 1.14, with a 90% confidence interval from 1.09 to 1.20. This is the result classified in the record as a non-inferiority or equivalence analysis. The registry does not report the model details or the equivalence limits used to interpret the interval.

The separate recalculation uses posted endpoint summaries to estimate an unadjusted mean difference of -1.16, with an unadjusted p-value of 0.15. It is a simple comparison on the difference scale, not a replication of the sponsor’s ratio analysis or a model-based crossover analysis. It is therefore useful as a supplementary check, but it does not supersede the sponsor’s analysis for interpreting the reported equivalence assessment.

The record does not flag a discrepancy between the posted results and recalculation. The recalculation’s interval spans values on both sides of zero, but that does not contradict the sponsor’s ratio interval: the estimates use different scales and methods. For the sponsor’s bioequivalence interpretation, the ratio interval and the study’s prespecified acceptance range would need to be considered together. The acceptance range is not reported in the registry record, so the sponsor’s conclusion cannot be independently assessed from the registry entry alone.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures and eight analyses, but no secondary outcome analyses are posted. The available results focus on the primary endpoint, plasma AUC for simvastatin acid through 48 hours. The registry record does not report a multiplicity adjustment or explain whether the primary endpoint was tested alongside other prespecified endpoints in a formal testing hierarchy.

In general, when several outcomes or comparisons are tested, the chance of at least one apparently notable result can increase unless the analysis plan addresses multiplicity. Whether that issue applies, and how it was handled, cannot be determined from the registry record. The record does not report an interim analysis plan, stopping rules, or a detailed statistical analysis plan.

The primary endpoint description reports 220 enrolled and states that data from 200 and 202 participants were available for the two treatment analyses. It attributes data loss to early dropout (N=4), mishandling of plasma samples (N=10), and assay limitations (N=10). The record does not report the exact participant-level flow by sequence and period, the reason for each unavailable measurement, or whether any participants contributed data to one treatment analysis but not the other.

In general, missing data can affect precision and can introduce bias if the chance of a measurement being unavailable is related to the outcome. The registry record does not report a missing-data method, imputation procedure, sensitivity analysis, or assessment of whether missingness was related to treatment or measured exposure. It also does not specify the analysis population definition, such as whether the primary result used all randomized participants, a pharmacokinetic population, or another set.

The crossover design makes participant-level treatment data and period information relevant to interpretation. Those details, as well as the model used for the sponsor’s least-squares ratio, are not reported in the registry record. The available record therefore supports a description of the posted endpoint estimates but not a full reconstruction of the analysis plan or handling of missing data.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for both treatment groups. For MK0524B, 220 participants were listed as at risk and none were recorded as affected by a serious adverse event. For simvastatin plus MK0524A, 220 were listed as at risk and none were recorded as affected by a serious adverse event.

The registry also lists other adverse events affecting 61 participants in the MK0524B group and 58 in the simvastatin plus MK0524A group. These are counts, not rates; no percentages are calculated. The record does not provide the event terms or further detail about severity, timing, or whether an event occurred in one or both crossover periods. No serious adverse event terms are listed.

Deaths are not reported in the registry record. The at-risk counts are reported separately by treatment group; because this was a crossover study, they should not be read as a count of distinct participants across the whole study. These adverse-event summaries describe what was recorded and do not establish a comparative safety conclusion.

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

The main statistical result concerns simvastatin acid exposure, measured as plasma AUC through 48 hours after dosing. The registry reports a least-squares mean ratio of 1.14 with a 90% confidence interval from 1.09 to 1.20. The endpoint estimates are 9.19 for MK0524B and 8.03 for simvastatin plus MK0524A. The sponsor-reported ratio is the relevant posted analysis for the equivalence or non-inferiority classification.

The registry record does not report the equivalence limits, so the ratio interval cannot be judged against the study’s acceptance criterion. The result therefore cannot, from the registry entry alone, be translated into a determination that the formulations met or failed a bioequivalence standard. The unadjusted difference of -1.16 and its p-value of 0.15 are a separate summary-based calculation, not a substitute for that standard.

This study addresses pharmacokinetics rather than clinical benefit or harm. AUC describes drug exposure over a defined interval; it does not directly show changes in symptoms, disease outcomes, or long-term effects. The registry lists dyslipidemia as the condition and identifies the products being compared, but the primary result is limited to the measured exposure endpoint.

The study’s randomized crossover design enables comparison across treatment periods, but the full statistical model and handling of period, sequence, and missing data are not reported in the registry record. Endpoint measurements were unavailable for some participants for reported reasons including dropout, sample mishandling, and assay limitations. These limitations affect how fully the posted estimate can be independently evaluated.

The posted serious adverse event counts were zero in both treatment groups, while other adverse events were recorded for 61 and 58 participants, respectively. The registry does not provide event terms for those other events. These summaries should be read as descriptive records, not as proof of equivalent safety.

17. Why This Trial Matters Statistically

This trial is statistically useful because it illustrates how a pharmacokinetic crossover comparison can be reported on more than one effect scale. The sponsor’s result is a least-squares mean ratio, which is aligned with a relative-exposure question. A separate recalculation expresses the contrast as an unadjusted mean difference. These estimates may help readers see the distinction between relative and absolute comparisons, but they are not interchangeable.

The registry also illustrates why an equivalence result requires more than a point estimate and confidence interval. The reported ratio and its interval convey an estimated relative exposure and uncertainty, but interpretation depends on the prespecified acceptance limits. Those limits are not reported in the registry record. Without them, readers cannot independently determine whether the interval meets the intended equivalence criterion.

The crossover structure is another important feature. In general, within-participant comparisons can account for stable differences between participants, but correct interpretation depends on the model and treatment of periods and sequences. The record describes randomized treatment order and two periods, while leaving the sponsor’s detailed model unreported. This makes the distinction between the sponsor’s model-based result and the summary-based recalculation especially important.

Finally, the endpoint analysis counts are smaller than the enrollment count, with reasons for unavailable data noted in the endpoint description. The record does not report missing-data sensitivity analyses or a detailed analysis-population definition. These omissions limit independent assessment of precision and robustness. The trial is therefore a useful example of both the value of reporting pharmacokinetic estimates and the importance of reporting the analysis criteria and methods needed to interpret them.

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 record does not report the equivalence margin or acceptance limits for the sponsor’s ratio analysis. This prevents an independent assessment of whether the reported 90% confidence interval met the study’s bioequivalence criterion. The record also does not explain the specific statistical model behind the least-squares mean ratio.

Although the study is identified as a randomized crossover trial with two treatment sequences, the handling of period and sequence effects is not reported. The washout schedule and the planned approach to within-participant correlation are not reported in the registry record. These details are relevant to reproducing or evaluating the model-based estimate.

For the primary endpoint, the registry reports 200 and 202 available analyses, alongside 220 enrolled and reasons for some unavailable data. It does not provide a participant flow by period and sequence, define the analysis population, explain the handling of partial endpoint data, or report sensitivity analyses for missing measurements. Sample mishandling and assay limitations are noted, but their specific impact on estimates is not described.

The study was open-label, and detailed assay methods, sample collection procedures, and handling of values limited by the assay are not reported. The open-label design does not by itself establish bias in a laboratory endpoint, but it remains a design feature relevant to the full assessment of study conduct.

There are no secondary outcome analyses posted, and the registry does not report multiplicity procedures, interim analyses, or the original sample-size justification. The safety record does not list serious event terms and does not report deaths; it also gives no terms for the other adverse events. These gaps constrain interpretation beyond the available pharmacokinetic summaries.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: MK0524B Bioequivalence Study (0524B-070) — NCT00943124.
• Registry record last updated 2015-06-19; status last verified 2015-05.
• Results first posted 2010-01-06.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT00943124 is a completed, randomized, open-label, phase 1 crossover study of MK0524B versus separate simvastatin and MK0524A tablets. It enrolled 220 participants and lists dyslipidemia as the condition. The primary endpoint is simvastatin acid plasma AUC through 48 hours after dosing.

The registry reports a least-squares mean ratio of 1.14 with a 90% confidence interval from 1.09 to 1.20. A separate unadjusted recalculation reports a mean difference of -1.16 and an unadjusted p-value of 0.15. The equivalence limits, detailed model, missing-data methods, and original sample-size assumptions are not reported in the registry record.

No serious adverse events were recorded in either treatment group; other adverse events affected 61 and 58 participants, respectively. Secondary outcome analyses are not posted.