Levetiracetam 750 mg Tablets Under Fasting Conditions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Levetiracetam 750 mg Tablets Under Fasting Conditions (NCT00849862), a randomized phase 1 study. The primary outcome is Cmax - Maximum Observed Concentration.

2. Clinical Question

This was a two-way crossover comparison rather than a parallel-group study in which each subject would receive only one treatment. The registry describes two treatment sequences: Levetiracetam followed by Keppra®, and Keppra® followed by Levetiracetam. That design makes the comparison of interest the relationship between the test and reference formulations across periods. The record does not describe the timing between periods or the statistical method used to account for the crossover.

The primary endpoint is Cmax, the maximum observed concentration in the blood samples collected over a 36 hour period. The registry also describes the broader objective as comparing the rate and extent of absorption. The posted sponsor analysis reports a geometric test/reference ratio for Cmax. The record does not report the sponsor’s decision rule or the criteria used to judge bioequivalence, so the reported estimate and interval should be read as results rather than as a complete account of how the study’s formal objective was evaluated.

3. Trial Design

The registry identifies this as a Phase 1, randomized, open-label, two-way crossover bioequivalence study in healthy subjects. Teva Pharmaceuticals USA is named as the sponsor. The listed study dates are 2005-10 for start and 2005-11 for primary completion, and the study status is completed. Enrollment is reported as 22.

Participants were assigned to one of two sequences. One sequence received the Levetiracetam 750 mg test tablet in the first period and Keppra® 750 mg reference tablet in the second. The other sequence received the reference tablet first and the test tablet second. These sequence descriptions are important when interpreting the arm summaries: the arm labels correspond to treatment order, not simply to a group that received one product throughout.

Randomization assigns treatment sequence by chance. Open-label means that the study is described as having no masking; the registry does not state that subjects or study personnel were unaware of treatment. In a crossover design, each sequence includes exposure to both products over different periods. General statistical consideration: such designs can use within-subject comparisons, but their analysis needs to account for the paired treatment observations and period or sequence structure. The method used to do so is not reported in the registry record.

The primary outcome is Cmax, with blood samples collected over 36 hour period. The record lists 3 outcome measures and 3 analyses, but it does not provide details of all of them in the posted results summarized for the primary endpoint. It identifies the primary analysis population as data from all subjects who completed the study. The number who completed, the crossover interval, and any additional design details are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

First Intervention

Washout: 7 Days

Second Intervention

Analysis population for the primary outcome, as defined in the registry: Data from all subjects who completed the study were included in the statistical analysis.

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 Cmax - Maximum Observed Concentration, measured over Blood samples collected over 36 hour period.

Cmax means the maximum observed concentration of a substance in the measured blood samples. In this study, it is reported as a continuous measure in mcg/mL, with the endpoint described as the mean and its spread described by the standard deviation. The measurement timeframe is blood samples collected over 36 hour period.

Cmax is used in absorption comparisons because it describes the highest measured concentration after dosing. As a general statistical and pharmacokinetic concept, peak concentration can inform comparisons of the rate and extent of absorption, but it is not a complete description of the concentration-time profile. The registry’s stated objective includes both rate and extent of absorption, while Cmax is the identified primary endpoint. The record does not identify the other endpoint measures by name in the posted information summarized for this analysis.

The arm summaries list a mean of 21.62 mcg/mL with a standard deviation of 5.699 for Levetiracetam, and a mean of 23.5 mcg/mL with a standard deviation of 6.247 for Keppra®. These are descriptive summaries; in this crossover study, the sequence descriptions mean that they should not be treated automatically as though they were independent parallel groups. The sponsor’s analysis reports a geometric test/reference ratio rather than a difference between arithmetic means. The registry record does not explain the transformation or model used to obtain that ratio.

Cmax does not by itself describe every aspect of absorption. The broader objective mentions rate and extent, but no other posted primary endpoint result is included in the available record. Interpretation should therefore stay focused on the reported Cmax summaries and ratio rather than imply that every feature of absorption was measured by this single endpoint.

7. Results Posted in the Registry

• Keppra®: mean 23.50 versus 21.62 on control; difference 1.88 (95% CI -1.66 to 5.41), standardized effect d = 0.31; z-test p = 0.298.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, open-label, two-way crossover design. The randomized sequences were Levetiracetam followed by Keppra®, and Keppra® followed by Levetiracetam. Randomization concerns which sequence a participant follows. Open-label indicates that masking is listed as none. The record does not explain how randomization was generated or concealed.

What does crossover mean for the comparison?

General statistical explanation: in a crossover study, participants receive multiple treatments in separate periods. A treatment comparison can use the fact that measurements from the same participant are related. That within-participant relationship may affect the uncertainty of the estimated treatment contrast. A suitable analysis also needs to account for relevant features such as period and sequence. The registry record does not report the sponsor’s model, whether the analysis used paired observations, or how period and sequence were handled. It also does not report the interval between periods.

The posted arm summaries give means and standard deviations for the Levetiracetam and Keppra® sequence labels. Because the arm descriptions specify treatment order, a direct contrast of those summaries does not necessarily recover the within-participant treatment contrast. The recalculated mean difference of 1.876 and interval from -1.657 to 5.409 are explicitly unadjusted recalculations from posted summary statistics. They should not be mistaken for an estimate from the sponsor’s crossover model.

What do the mean and standard deviation describe?

General statistical explanation: a mean describes the center of a set of continuous observations. A standard deviation describes how dispersed observations are around that mean. The endpoint is Cmax in mcg/mL, and the registry labels its parameter as a mean and its dispersion as a standard deviation. The Levetiracetam-labeled arm has a posted mean of 21.62 and standard deviation of 5.699; the Keppra®-labeled arm has a posted mean of 23.5 and standard deviation of 6.247. Those summaries do not reveal the distribution of individual observations or the within-subject correlation.

What does the recalculated difference represent?

The recalculated mean difference is an absolute contrast on the reported measurement scale. It is not the same quantity as the sponsor’s geometric test/reference ratio multiplied by 100. The recalculation also reports a standardized difference of 0.314, which expresses the difference relative to a pooled standard deviation under the recalculation’s assumptions. General statistical explanation: standardization can help describe the size of a contrast relative to observed variability, but it does not convert the result into a clinical or regulatory decision.

The recalculated p-value is 0.298 and is labeled unadjusted. A p-value is a measure of how incompatible the observed data are with a specified null model, assuming that model and the calculation’s assumptions are appropriate. It is not the probability that the null hypothesis is true, and it does not assess equivalence by itself. The registry does not report the method used to calculate this p-value. It should be understood as belonging to the simple unadjusted recalculation, not attributed to the sponsor’s ratio analysis.

How is the sponsor’s ratio different?

The sponsor reports a geometric test/reference ratio multiplied by 100 of 92.2, with a 90% confidence interval from 85.0 to 100. A geometric ratio is a relative rather than absolute expression of the comparison. General statistical explanation: ratio analyses of this kind are often based on transformed measurements or model-based estimates, but the precise transformation and model for this study are not reported in the registry record. The posted result should therefore be described using the sponsor’s stated metric without reverse-engineering an unreported method.

The registry classifies this report as a non-inferiority or equivalence analysis, but its method, null hypothesis, and acceptance limits are not given. A formal equivalence interpretation depends on those criteria as well as on the estimate and interval. The information available does not permit an independent reconstruction of that decision.

What do the planning calculations mean?

The planning illustrations use the observed effect and assume a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They show participants per arm for 80% power and participants per arm for 90% power, using 100% of the observed effect, 75% of the observed effect, or 50% of the observed effect. General statistical explanation: power is the probability that a specified statistical procedure would detect an effect of a stated size under its assumptions. These illustrations are not the trial’s original design calculation. The original design assumptions and the sponsor’s sample-size justification are not reported in the registry record.

10. Confidence Intervals and Precision

The sponsor’s ratio estimate is 92.2, and its reported 90% confidence interval runs from 85.0 to 100. This interval expresses uncertainty around the sponsor’s ratio estimate on the reported scale. The registry identifies the interval as 90%, but does not provide the model or method used to calculate it. It is therefore possible to describe the reported range, but not to verify its construction from the information in the record.

The unadjusted recalculation reports a mean difference of 1.876 and an interval from -1.657 to 5.409. The confidence level for this recalculated interval is not reported in the registry record. It is a separate summary from the sponsor’s ratio interval: the former is an absolute difference based on the posted arm-level summaries, while the latter is a geometric test/reference ratio. Their endpoints cannot be compared as if they were on the same measurement scale.

General statistical explanation: a confidence interval conveys both an estimate and its uncertainty under the calculation’s assumptions. A wide interval permits a broader range of underlying values than a narrow interval. Whether an interval is sufficiently precise for a particular decision depends on the decision criteria and the analysis design. The registry does not report the sponsor’s equivalence limits, so the sponsor’s interval cannot be judged against those limits from the record alone.

The crossover structure matters for precision because measurements within a participant may be related. The unadjusted recalculation does not incorporate reported within-participant covariance, and the record does not disclose enough analysis detail to assess how that dependence affected the sponsor’s interval. Thus the posted interval is the sponsor’s reported precision estimate, while the recalculated interval is a separate, assumption-dependent summary.

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. The record describes a continuous Cmax outcome, and its planning assumptions specify equal allocation between two arms. The original trial’s own design assumptions are not reported in the registry record; the listed figures are planning illustrations based on the observed effect, not a reconstruction of the trial’s original sample-size calculation.

The calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For each assumed effect, one column gives participants per arm for 80% power and another gives participants per arm for 90% power. In plain language, those columns describe the planned number per arm under the assumptions for an 80% or 90% chance of detecting the specified effect using the stated calculation.

For 100% of the observed effect, the illustrations give 160 participants per arm for 80% power and 214 participants per arm for 90% power. For 75% of the observed effect, they give 284 participants per arm for 80% power and 380 participants per arm for 90% power. For 50% of the observed effect, they give 638 participants per arm for 80% power and 854 participants per arm for 90% power. The pattern reflects a general statistical principle: smaller effects are harder to detect and require more participants under otherwise similar assumptions.

These calculations are planning illustrations, not observed power. They do not describe the probability that this completed study succeeded, and they do not establish the adequacy of the original enrollment. The planning assumptions for this illustration are reported; the rationale and assumptions used when the trial was designed are not reported in the registry record.

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 = 1.876; pooled standard deviation σ = 5.979.
2. n = 2 × (1.96 + 0.842)² × 5.979² / 1.876² = 159.5, rounded up to 160 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 <- 1.876                      # observed difference in means
sigma <- 5.97928131132831                     # 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(849862)
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 = 160 (80% power), 214 (90% power)
 75% of observed effect: n per arm = 284 (80% power), 380 (90% power)
 50% of observed effect: n per arm = 638 (80% power), 854 (90% power)
power.t.test: 160.4 
Simulated power at n = 160 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 sponsor’s posted result is a geometric test/reference ratio multiplied by 100 of 92.2, with a 90% confidence interval from 85.0 to 100. The registry classifies the analysis as non-inferiority or equivalence. The method and decision limits are not reported, so the estimate and interval can be described, but the formal conclusion cannot be independently reconstructed from the registry record.

The recalculation uses the posted arm means to estimate an unadjusted mean difference of 1.876, with an interval from -1.657 to 5.409 and an unadjusted p-value of 0.298. It is an absolute-scale summary and does not account for a reported within-subject crossover analysis. The sponsor’s ratio is a relative-scale result and is identified as the sponsor’s analysis. These estimates are not competing calculations of precisely the same quantity.

The recalculation is useful for examining what follows from the displayed arm-level means under a simple unadjusted approach. It cannot replace the sponsor’s analysis of the test/reference ratio or verify the sponsor’s model. The record does not report the sponsor’s model details, the analysis population beyond the completed-subject description, or the criteria applied to the interval. Consequently, the sponsor result should be read as the reported study analysis, while the recalculation should be read as a separate, limited summary rather than a correction to it.

14. Multiplicity, Missing Data and Analysis Populations

The registry identifies Cmax as the primary outcome and reports 3 outcome measures and 3 analyses. It also records 0 secondary outcomes posted in the summary of secondary results. The available record does not identify all outcome measures or provide the results and methods for every analysis. It therefore does not show how many hypotheses were formally tested or how the analyses were ordered.

Multiplicity arises when several outcomes, comparisons, or analyses are considered. General statistical explanation: as the number of statistical tests increases, the chance of at least one apparently notable result can increase if no adjustment or other prespecified strategy is used. Whether adjustment is needed and how it should be done depend on the study’s objectives and analysis plan. The registry record does not report a multiplicity strategy, so no claim can be made about whether the sponsor adjusted for multiple testing.

The primary analysis population is described as data from all subjects who completed the study. Enrollment is reported as 22, but the number who completed the study is not reported in the registry record. The record does not state how many subjects were excluded from the analysis, how missing Cmax observations were handled, or whether any values were imputed. It also does not describe additional analysis populations or sensitivity analyses. The phrase about completed subjects should not be expanded into an assumption that all enrolled subjects contributed to the primary analysis.

General statistical explanation: missing data can affect an estimate if the observations that are unavailable differ systematically from those that remain. In crossover studies, an absent measurement in one period can also affect whether a participant contributes a complete paired comparison. The registry record does not describe the amount, reason, or handling of missing data, so its possible influence cannot be assessed.

The analysis method for the primary comparison, the handling of period and sequence, any interim analyses, and the complete statistical plan are not reported in the registry record. These omissions limit the ability to evaluate whether the posted analysis followed a prespecified approach or whether all planned outcomes and analyses are represented.

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

The posted results describe Cmax after the test and reference products in a randomized, open-label crossover study in healthy subjects under fasting conditions. The sponsor reports a geometric test/reference ratio multiplied by 100 of 92.2, with a 90% confidence interval from 85.0 to 100. Separately, an unadjusted recalculation from arm summaries reports a mean difference of 1.876, with an interval from -1.657 to 5.409 and an unadjusted p-value of 0.298. Each result must be interpreted on its own scale and in light of its own method.

The record classifies the sponsor’s analysis as non-inferiority or equivalence, but does not report the analysis method or the criteria for equivalence. A ratio and interval, without the decision limits and model details, are not enough to establish a formal equivalence conclusion from the registry information alone. The unadjusted mean difference and p-value also do not establish equivalence: they come from a different, simplified calculation and are not identified as the sponsor’s crossover analysis.

General statistical explanation: failure to distinguish a difference from zero is not the same as demonstrating that two treatments are equivalent. Equivalence asks whether the uncertainty around a comparison is sufficiently contained within prespecified limits. Those limits and the sponsor’s detailed statistical procedure are not reported in the registry record.

The study concerns a specified 750 mg dose, healthy subjects, fasting conditions, and Cmax measured through blood samples collected over 36 hour period. The registry does not support extending these findings to other doses, populations, conditions, or outcomes. Nor does the concentration endpoint by itself answer every question about absorption. The appropriate reading is limited to what the posted measurements and analyses describe, with the noted uncertainty and reporting gaps.

These statistical results do not constitute medical advice. They describe a study comparison and its reported precision; they do not establish an individual response or a treatment decision.

16. Why This Trial Matters Statistically

This study illustrates why the design and analysis structure matter when interpreting clinical-trial statistics. It is a randomized crossover study: the two sequences receive the test and reference formulations in opposite orders. General statistical explanation: crossover designs can make within-participant comparisons informative, but the analysis must reflect the repeated measurements and their dependence. Arm-level means and standard deviations alone do not show that dependence or reproduce a paired analysis.

The record also illustrates why the scale of an estimate matters. The sponsor reports a geometric test/reference ratio, while the recalculation reports an unadjusted absolute mean difference. Both can describe a comparison, but they answer on different scales and rely on different analysis details. Treating them as interchangeable would obscure what each estimate means.

Finally, the record distinguishes a reported point estimate from a formal statistical conclusion. The sponsor’s 90% interval is available, but the equivalence method and limits are not. The separate recalculated interval and p-value do not fill in those missing details. Statistical interpretation depends not only on the numerical estimate, but also on the endpoint, design, analysis population, model, and decision criteria.

The trial is a Phase 1 study in healthy subjects, with enrollment of 22 and a single stated 750 mg dose under fasting conditions. Those characteristics define the setting of the evidence. The record does not establish how its results would apply beyond that setting. Its value for statistical learning is in seeing how crossover structure, ratio-based reporting, uncertainty intervals, and incomplete methodological reporting shape what can responsibly be concluded.

17. 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 reports enrollment of 22, but does not report the number of subjects who completed the study. Although the primary analysis population is described as all subjects who completed, the number included in that analysis and the handling of any incomplete observations are not stated. This limits assessment of the analysis population and missing-data impact.

The crossover is described through two treatment sequences, but the record does not report the interval between periods, the method used to account for period or sequence, or whether a within-subject model was used. These details matter for evaluating how the sponsor’s ratio and uncertainty were estimated. The posted arm summaries do not provide within-subject covariance, so the unadjusted recalculation cannot reproduce a paired comparison.

The trial is open-label, with masking listed as none. The registry does not describe whether this could have influenced any study procedures or measurements. The endpoint is an objectively measured concentration, but the record does not provide details of laboratory procedures or measurement quality. Those details are not reported in the registry record.

The primary endpoint is Cmax after a 750 mg dose under fasting conditions, with samples collected over 36 hour period. The results therefore pertain to this endpoint and setting. The record does not provide sufficient posted information to characterize all listed outcome measures or analyses. It also does not report the sponsor’s equivalence limits, full statistical method, multiplicity strategy, or original sample-size assumptions. These omissions prevent a full independent evaluation of the sponsor’s formal bioequivalence analysis.

The recalculated comparison is unadjusted and based on the posted arm-level summaries. It is not a substitute for the analysis suited to the crossover design. The sponsor’s reported ratio and interval remain distinct from that recalculation, and the registry record does not provide enough detail to reconstruct either the analysis model or its decision rule.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Levetiracetam 750 mg Tablets Under Fasting Conditions — NCT00849862.
• Registry record last updated 2009-09-11; status last verified 2009-09.
• Results first posted 2009-08-04.
• 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.

21. Record Summary

NCT00849862 is a completed Phase 1, randomized, open-label, two-way crossover study sponsored by Teva Pharmaceuticals USA. It enrolled 22 healthy subjects and compared a 750 mg Levetiracetam test tablet with Keppra® 750 mg as the reference under fasting conditions. The two sequences received the products in opposite orders.

The primary endpoint was Cmax in mcg/mL, measured through blood samples collected over 36 hour period. The sponsor reported a geometric test/reference ratio multiplied by 100 of 92.2 with a 90% confidence interval from 85.0 to 100. An unadjusted recalculation from arm summaries reports a mean difference of 1.876, an interval from -1.657 to 5.409, and an unadjusted p-value of 0.298. These are distinct analyses on different scales.

The registry does not report the sponsor’s detailed model, equivalence limits, number completing, or missing-data methods. Those gaps limit independent assessment of the formal bioequivalence analysis.