Leflunomide 20 mg Tablets, Non-Fasting: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Leflunomide 20 mg Tablets, Non-Fasting (NCT00834405), a randomized phase 1 study. The primary outcome is Cmax - Maximum Observed Concentration - Metabolite A77 1726 in Plasma.

2. Clinical Question

This is a pharmacokinetic comparison in healthy participants, not a trial of clinical outcomes in people with a diagnosed condition. Its primary question concerns the concentration profile represented by Cmax under the specified non-fasting conditions. The registry record does not report a clinical efficacy outcome, a safety comparison, or the sponsor’s criteria for deciding whether the products met a bioequivalence standard.

3. Trial Design

The registry identifies this as a Phase 1, randomized, parallel study with no masking. It compared leflunomide 20 mg tablets with ARAVA 20 mg tablets under non-fasting conditions. The recorded enrollment was 37. The arm-level outcome records list 20.0 participants for leflunomide and 17.0 for ARAVA; these are the sample sizes reported for the primary outcome.

The study population was described as healthy, adult, non-smoking subjects, specifically females unable to become pregnant and vasectomized males. The primary outcome analysis included data from all subjects who completed the study. The registry record does not report how many participants completed, whether any participants were excluded after randomization, or how any missing measurements were handled.

The primary completion date was 2002-05, and the study start date was 2002-04. The record lists the study as completed and names Teva Pharmaceuticals USA as sponsor. These details describe the registered study but do not establish the timing or conduct of any unreported analyses.

A parallel design assigns participants to one of the comparison groups rather than having each participant receive both products in sequence. Randomization can help balance measured and unmeasured participant characteristics between groups on average, although it does not guarantee exact balance in a study of this size. Because masking was absent, participants and study personnel were not described as masked in the registry. The registry record does not report the randomization method, allocation concealment, dosing schedule, washout considerations, or the procedures used to collect and assay blood samples.

The record lists two outcome measures and two analyses, but the available primary outcome description concerns Cmax of metabolite A77 1726. No secondary posted outcomes are listed. The registry does not report a fuller protocol, the study’s original sample-size justification, or its planned statistical methods.

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.

Overall Study

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 - Metabolite A77 1726 in Plasma, measured over Blood samples collected over 72 hour period.

The primary endpoint is Cmax, the maximum observed concentration of metabolite A77 1726 in plasma. Concentration is reported in ng/mL. Blood samples were collected over a 72 hour period, and the recorded summary measure is the mean with standard deviation for each group.

Cmax describes the highest measured plasma concentration in the sampled profile. In a relative bioavailability comparison, it is one way to compare how much of a drug-related substance reaches the circulation, as reflected by observed concentration, after administration of different products. The registry explicitly describes the primary outcome as bioequivalence based on Cmax. That makes the endpoint relevant to the study’s stated product-comparison objective, but it does not make Cmax a direct measure of clinical benefit or patient experience.

Observed Cmax depends on the sampling schedule and the measurements captured during the collection period. The registry record identifies the 72 hour sampling period but does not report the sampling times, assay details, how peak values were selected, or whether values below a detection threshold occurred. These details matter when interpreting how closely the recorded maximum represents the concentration profile.

The outcome summaries report a mean and standard deviation for each arm, while the sponsor’s posted analysis reports a geometric test/reference ratio. These are different forms of summary. The mean difference describes an absolute contrast in the recorded concentration scale; a ratio describes a relative contrast. Neither alone supplies the full protocol criteria for concluding bioequivalence, which are not reported in the registry record.

7. Results Posted in the Registry

• Arava®: mean 2,345 versus 2,488 on control; difference -142.71 (95% CI -353.67 to 68.25), standardized effect d = -0.42; z-test p = 0.185.

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical summaries appear in the registry record: group means and standard deviations for Cmax, an unadjusted recalculated mean difference with a confidence interval and p-value, and the sponsor’s geometric test/reference ratio with a 90% confidence interval. Each addresses the comparison in a different way. The summaries should be read as complementary descriptions rather than as interchangeable answers to one question.

What does a group mean describe?

A mean is the arithmetic average of the observed values in a group. For this continuous outcome, it gives a summary of the recorded Cmax values in ng/mL. The group means are 2488.0 for leflunomide and 2345.0 for ARAVA. A mean is useful for describing the center of a distribution, but it does not show the full distribution of individual measurements. Two groups can have similar means while differing in variability or in the shape of their distributions.

The standard deviation describes how spread out observations are around a group mean. The recorded standard deviations are 385.2 for leflunomide and 266.2 for ARAVA. It is not a confidence interval for the mean, and it does not show the uncertainty in the comparison by itself. To describe uncertainty about a group contrast, an analysis uses the group sizes and variability as well as the observed difference.

What does the recalculated mean difference express?

The recalculated comparison is an unadjusted difference in arithmetic means, reported for ARAVA relative to leflunomide. Its estimate is -142.7. A negative estimate means the ARAVA group mean was lower on the recorded scale than the leflunomide group mean. The accompanying interval, from -353.7 to 68.25, indicates uncertainty around that estimated difference. It includes both negative and positive values, so the posted interval does not pin down the direction of the population mean difference with precision.

“Unadjusted” means this recalculation is not reported as controlling for other variables or applying a multiplicity adjustment. It is a simple comparison based on the posted summary statistics. The registry record does not report covariate-adjusted analyses or enough detail to establish whether the sponsor’s analysis used adjustments not reflected in this recalculation.

How should the p-value be understood?

The recalculated unadjusted p-value is 0.185. A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be under the statistical model if that null hypothesis were true. It is not the probability that the null hypothesis is true, the probability that the findings occurred by chance, or a measure of clinical importance. A p-value alone cannot establish that two products are equivalent. The registry record does not report a p-value for the sponsor’s geometric ratio analysis.

General statistical principle: a non-significant result may arise because the true difference is small, because the estimate is imprecise, or because the study contains limited information for the question. Those possibilities cannot be distinguished by the p-value alone. Here, the confidence interval is important because it shows the range of mean differences compatible with the recalculated analysis.

Why is a ratio analysis used?

The sponsor’s analysis reports a geometric test/reference ratio multiplied by 100. Ratio summaries express a relative comparison rather than an absolute difference in ng/mL. Geometric summaries are often used for concentration outcomes when a relative scale is of interest. General statistical principle: taking a logarithmic scale for analysis and then expressing the result as a ratio can make multiplicative contrasts easier to interpret. The registry record does not state the sponsor’s exact calculation method, so that general explanation should not be mistaken for a description of its undocumented procedure.

The sponsor’s reported ratio is 106, and its 90% confidence interval is 98.0 to 114. This interval describes uncertainty for the sponsor’s ratio estimate on the reported scale. To decide whether a ratio meets a bioequivalence standard, the relevant acceptance bounds and prespecified decision rule are needed. Those are not reported in the registry record, so the estimate and interval cannot independently establish the sponsor’s formal conclusion.

Why can estimates on different scales differ?

An absolute mean difference and a geometric ratio answer related but distinct questions. The mean difference is measured in ng/mL and compares arithmetic means. The geometric ratio is relative and concerns a different summary of the measurements. Their numerical estimates and intervals should not be compared as though they were two versions of the same quantity. A discrepancy flag is not reported, but the distinction in scale and estimand remains relevant when interpreting the two analyses.

General statistical principle: the choice of scale affects how a contrast is expressed and which patterns in the data are emphasized. An absolute difference can be more directly tied to units, while a ratio describes proportional separation. Neither scale is universally preferable; the study objective and prespecified analysis determine which is appropriate. The registry record does not report the protocol’s rationale for selecting the sponsor’s analysis or the detailed model used.

What does a confidence interval add?

A confidence interval communicates uncertainty around an estimate. Under the assumptions of the method, intervals generated repeatedly in the same way would cover the target parameter at the stated confidence level at the corresponding rate. It is not a guarantee that the particular interval contains the true value. The recalculated mean-difference interval’s confidence level is not stated in the record, while the sponsor’s interval is explicitly identified as 90%.

General statistical principle: interval width reflects both the variability in the data and the amount of information available. An interval that includes values on both sides of a reference point may leave important possibilities unresolved. For a difference, zero is the no-difference reference; for a ratio, the corresponding no-relative-difference point is 100 on a ratio-multiplied-by-100 scale. These reference points do not themselves define a bioequivalence margin.

Why does variability matter?

Greater variability generally makes an estimate less precise when other conditions are comparable. For count outcomes, over-dispersion means that observed variability exceeds what a simple count model expects and can make standard errors too small if ignored. This trial’s primary outcome is continuous, not a count, so that count-specific issue does not apply to the reported Cmax summary. For this comparison, the group standard deviations and participant counts contribute to the uncertainty of the mean difference.

The assumptions of any statistical model also matter. The registry record does not report diagnostic checks, distributional assumptions, missing-data procedures, or the sponsor’s exact statistical method. Accordingly, the recalculated comparison should be understood as a simple unadjusted re-analysis of the posted summary statistics rather than a reproduction of an undocumented sponsor model.

10. Confidence Intervals and Precision

The recalculated mean difference is -142.7 ng/mL, with a confidence interval from -353.7 to 68.25. Because the interval spans zero, the estimate does not narrowly identify the direction of the mean difference. The interval also allows a range of differences that may be relevant to interpretation; the registry record does not define a clinically meaningful margin for the absolute Cmax difference.

The confidence level for this recalculated interval is not reported in the registry record. It should not be assigned a confidence level based on assumption. The sponsor’s interval is explicitly reported as a 90% confidence interval: the geometric test/reference ratio multiplied by 100 was 106, with limits of 98.0 and 114. This is a ratio-scale result, not an interval for the arithmetic mean difference.

Confidence intervals express statistical uncertainty under the method’s assumptions. They do not account automatically for unreported sources of uncertainty, such as missing measurements, selection after randomization, or deviations from the planned analysis. The record says that completed subjects were included in the primary analysis, but does not report completion counts or missing-data handling.

The sponsor’s interval cannot be judged against a bioequivalence boundary from the registry record, because no boundary or decision rule is stated. Likewise, the recalculated interval cannot establish equivalence merely because it contains zero. Precision must be considered alongside the scale of the analysis, the study population, and the criteria specified for the scientific question.

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 illustrations give participants per arm, not total events. They are based on the observed effect and are not the original trial’s design calculation. The assumptions listed are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect.

For planning at 100% of the observed effect, the illustration gives 88 participants per arm for 80% power and 117 participants per arm for 90% power. At 75% of the observed effect, it gives 155 participants per arm for 80% power and 208 participants per arm for 90% power. At 50% of the observed effect, it gives 349 participants per arm for 80% power and 467 participants per arm for 90% power.

The 80% power column represents a planning scenario in which the study would have an 80% chance of detecting the specified effect under the stated assumptions, if that effect were the true one. The 90% power column uses a higher target chance of detecting that specified effect under the same planning framework. These are design concepts, not probabilities that the observed study result is correct.

Smaller effects require more participants to detect with the same target power, which is why the planning counts increase as the target effect becomes a smaller fraction of the observed effect. These figures are not observed power, and they do not describe what the completed trial achieved. The trial’s own original sample-size assumptions and justification 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 = -142.706; pooled standard deviation σ = 336.042.
2. n = 2 × (1.96 + 0.842)² × 336.042² / -142.706² = 87.0, rounded up to 88 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 <- -142.706                      # observed difference in means
sigma <- 336.042000514179                     # 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(834405)
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 = 88 (80% power), 117 (90% power)
 75% of observed effect: n per arm = 155 (80% power), 208 (90% power)
 50% of observed effect: n per arm = 349 (80% power), 467 (90% power)
power.t.test: 88 
Simulated power at n = 88 per arm: 0.800 (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 a sponsor analysis of the geometric test/reference ratio multiplied by 100. The estimate is 106, and the reported 90% confidence interval is 98.0 to 114. No p-value is reported for this analysis. The registry record also does not report the sponsor’s exact statistical method, acceptance bounds, or decision rule for bioequivalence.

The recalculated analysis is an unadjusted difference in arithmetic means, with an estimate of -142.7 and a confidence interval from -353.7 to 68.25. It is expressed in ng/mL and has an unadjusted p-value of 0.185. Because the sponsor analysis is a geometric ratio and the recalculation is an absolute mean difference, they do not estimate the same contrast on the same scale. Their results should be interpreted separately, not treated as competing calculations of an identical quantity.

The registry record does not flag a discrepancy between the sponsor’s reported analysis and the recalculation. The sponsor’s analysis is the study’s reported analysis; the recalculation is a simple unadjusted comparison based on the posted summary statistics. The record does not provide enough information to reproduce the sponsor’s method or determine whether the ratio interval met the study’s intended bioequivalence criteria.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists two outcome measures and two analyses, but no secondary posted outcomes are identified. The primary outcome description specifies that data from all subjects who completed the study were included in the statistical analysis. This identifies the stated analysis population in broad terms, but it does not report the number completing or the number analyzed in each group beyond the arm-level sample sizes attached to the outcome.

Missing-data handling is not reported in the registry record. It is therefore not possible to determine whether any Cmax values were missing, how incomplete sampling profiles were treated, or whether participants who did not complete the study were excluded for particular reasons. The record also does not describe sensitivity analyses for missing data or deviations from the study procedures.

Multiplicity refers to the increased chance of a false-positive finding when multiple outcomes, comparisons, or analyses are tested. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for outcomes. Since no secondary posted outcomes are listed, the available record does not establish how many comparisons were planned or whether analyses beyond those posted were conducted. The recalculated p-value is unadjusted and should be interpreted as such.

The record identifies randomized allocation but does not report allocation concealment or the method used to generate the random sequence. It also states that masking was absent. These details affect how risk of bias is assessed, but the available information is insufficient to infer whether group assignment or outcome measurement was influenced by awareness of treatment.

The record does not report a full statistical analysis plan, protocol deviations, exclusions, or a participant flow description. These omissions limit assessment of whether the completed analysis followed the original plan and whether the analyzed population differs from the randomized population. No assumptions about those aspects can be recovered from the posted summary alone.

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

The results describe plasma Cmax for metabolite A77 1726 after administration of leflunomide 20 mg tablets or ARAVA 20 mg tablets under non-fasting conditions in the study’s healthy population. The group means, standard deviations, recalculated mean difference, and sponsor’s geometric ratio provide statistical descriptions of the observed pharmacokinetic comparison.

The recalculated mean difference and its confidence interval do not establish that the arithmetic mean concentrations are identical. The interval spans zero, and the unadjusted p-value is 0.185; neither result proves equality or rules out differences. Conversely, the sponsor’s ratio estimate and 90% confidence interval do not, without reported acceptance bounds and a decision rule, establish whether the formal bioequivalence criterion was met.

The endpoint is a concentration measure, not a clinical outcome. These results do not establish comparative symptom relief, disease control, safety, or other patient-centered outcomes. The study was conducted in healthy adults under non-fasting conditions, with the registry description restricting the eligible population to non-smoking females unable to become pregnant and vasectomized males. Findings cannot be assumed to apply to populations or conditions not described in the record.

The results also do not answer how the products compare under fasting conditions, in a different dosing context, or on outcomes not reported. The record does not provide detailed sampling times, assay procedures, a complete analysis plan, or the sponsor’s bioequivalence margins. Those omissions constrain interpretation of both the endpoint measurement and the sponsor’s decision framework.

Statistically, the central lesson is to separate the estimate from the decision. An estimate summarizes the observed contrast; an interval describes uncertainty under a method; a p-value addresses a specified null hypothesis; and a bioequivalence conclusion requires defined criteria. The registry record contains some of these elements but not the full criteria needed to connect them to a formal conclusion.

16. Why This Trial Matters Statistically

This trial illustrates why the statistical scale matters in product-comparison studies. The primary outcome is a continuous concentration measure, while the sponsor’s reported analysis is a geometric ratio and the simple recalculation is an absolute mean difference. Those summaries address related aspects of the data but have different interpretations. Readers need to distinguish the relative comparison used by the sponsor from the absolute difference expressed in ng/mL.

It also illustrates why a p-value should not be treated as a decision rule for equivalence. The recalculated p-value is 0.185, but a test of no mean difference is not the same as an equivalence assessment. Equivalence or bioequivalence judgments depend on a prespecified margin and a corresponding analysis. The registry record reports the sponsor’s ratio and 90% confidence interval, but not the acceptance bounds or decision rule.

The study is also a useful example of what registry summaries can and cannot convey. The record gives randomized parallel allocation, no masking, arm-level outcome summaries, and an analysis population defined by completion. It does not give the original sample-size rationale, detailed missing-data procedures, or a full statistical plan. Such absences do not invalidate the reported numbers, but they limit what can be concluded about design intent and analytic choices.

Finally, the context matters: this was a Phase 1 comparison in healthy participants under non-fasting conditions, with Cmax as the primary measure. Statistical interpretation should remain within that scope. The record supports a focused discussion of concentration summaries and their uncertainty, not broader claims about clinical effects or performance in other populations.

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 provides limited detail about the conduct and analysis of this specific study. Although it reports randomized allocation and no masking, it does not describe sequence generation, allocation concealment, sampling times, assay methods, or the exact dosing procedures. These omissions make it difficult to assess how the measurements were obtained and how potential sources of bias were managed.

The primary outcome population is described as all subjects who completed the study, but completion counts and reasons for non-completion are not reported. Missing measurements, exclusions, and the handling of incomplete concentration profiles are also not described. As a result, the degree to which the outcome analysis reflects all randomized participants cannot be assessed from the registry record.

The record does not report the original statistical analysis plan, a justification for the enrollment, interim analyses, or the sponsor’s exact method for the geometric ratio. It also omits the acceptance limits and decision rule needed to interpret the sponsor’s bioequivalence analysis. The recalculated mean difference is a simple unadjusted comparison and should not be mistaken for a complete reconstruction of the sponsor’s analysis.

The population and conditions further limit interpretation. Participants were healthy, adult, non-smoking subjects from the specified eligibility groups, and administration was under non-fasting conditions. The primary measure is Cmax of metabolite A77 1726, not a clinical outcome. The record does not report results for fasting administration, broader patient populations, or outcomes beyond the posted primary analysis.

These are limitations of what can be evaluated from the registry record. They do not establish that particular procedures were absent; they indicate that the record does not report them.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Leflunomide 20 mg Tablets, Non-Fasting — NCT00834405.
• Registry record last updated 2024-08-19; status last verified 2024-08.
• Results first posted 2009-08-18.
• 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

NCT00834405 is a completed Phase 1, randomized, unmasked, parallel study sponsored by Teva Pharmaceuticals USA. It compared leflunomide 20 mg tablets with ARAVA 20 mg tablets under non-fasting conditions in healthy adult, non-smoking subjects described as females unable to become pregnant and vasectomized males. The record lists enrollment of 37.

The primary outcome was Cmax of metabolite A77 1726 in plasma, measured in ng/mL from blood samples collected over a 72 hour period. The recalculated unadjusted mean difference for ARAVA relative to leflunomide was -142.7, with a confidence interval from -353.7 to 68.25 and an unadjusted p-value of 0.185. The sponsor reported a geometric test/reference ratio multiplied by 100 of 106, with a 90% confidence interval from 98.0 to 114.

The registry record does not report the sponsor’s bioequivalence acceptance bounds or full statistical method, nor does it provide detailed missing-data procedures or the original sample-size justification.