A Clinical Study in Subjects With Neuropathic Pain From PHN Who Have Had an Inadequate Response to Gabapentin Treatment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Clinical Study in Subjects With Neuropathic Pain From PHN Who Have Had an Inadequate Response to Gabapentin Treatment (NCT00617461), a randomized phase 2 study. The primary outcome is Change From Baseline in the Mean 24-hour Average Pain Intensity (API) Score at the Last Week of Each Treatment Period Using Last Observation Carried Forward (LOCF) Data.

2. Clinical Question

The clinical question is narrowly about the relative results of two doses in this study population. The outcome was a participant-rated pain score, recorded on an 11-point scale from 0, no pain, to 10, pain as bad as you can imagine. Change was defined as end-of-treatment score minus baseline score, so negative changes represent lower scores at the end of a treatment period than at baseline. The trial record does not establish how these results apply to people outside the enrolled population or to comparisons with treatments that were not studied.

3. Trial Design

This was a phase 2, randomized, triple-masked, crossover study sponsored by XenoPort, Inc. Its registry title describes participants with neuropathic pain from post-herpetic neuralgia who had an inadequate response to gabapentin. The study was completed. Its recorded start was 2008-03, and primary completion was 2009-07.

Participants received both gabapentin enacarbil dose regimens across treatment periods: 1200mg/day and 3600mg/day, each for 4 weeks. Either dose could be given in the first or second treatment period. Randomization and masking are recorded, but details of the randomization sequence and masking procedures are not reported in the registry record. The length and conduct of any interval between treatment periods are not reported in the registry record.

The planned enrollment was 96. The primary outcome description identifies 93 participants in the ITT population. It also notes that 3 participants did not take GEn 1200 mg in the second period and 8 did not take GEn 3600 mg in the second period. One participant did not provide post-baseline 24-hour API data during the GEn 3600mg treatment period. The dose-specific primary outcome summaries use 90 participants for 1200 mg and 84 for 3600 mg.

A crossover design allows dose periods to be compared within a study in which participants receive both regimens. The registry record reports the dose-specific summaries and a sponsor ANCOVA analysis, but it does not describe the full crossover model, treatment sequence analysis, or any period or carryover adjustments. Those details should not be inferred from the design label alone.

4. Participant Flow and Accounting

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

Before assignment: Participants (par.) were enrolled in a two-week Baseline Period, which included treatment with 1800 milligrams (mg)/day gabapentin. Participants who met entry criteria were then randomized. Inv., investigator.

First Treatment Intervention Period

Reasons for not completing this period:

4-Day Crossover Period

Reasons for not completing this period:

Second Treatment Intervention Period

Reasons for not completing this period:

6-Day Down-Titration Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Of the 93 participants in the ITT Population, there were 3 and 8 participants who did not take GEn 1200 and 3600 mg, respectively, in the second period. In addition, 1 participant did not provide post-baseline data for 24-hour API assessments during the GEn 3600mg treatment period, and was therefore not included in this 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 Change From Baseline in the Mean 24-hour Average Pain Intensity (API) Score at the Last Week of Each Treatment Period Using Last Observation Carried Forward (LOCF) Data, measured over Baseline and End of Treatment (Weeks 4 and 9, representing the last week of each treatment period).

The primary endpoint was change from baseline in mean 24-hour average pain intensity at the last week of each treatment period, using last observation carried forward (LOCF) data. Baseline was the mean of daily 24-hour API scores during the last 7 days before randomization. End of treatment was the mean of daily scores during the last 7 days on treatment within a period. The endpoint therefore represents a participant’s average pain rating over a day, rather than a single rating at one time.

Participants rated pain over the preceding 24 hours on an 11-point PI-Numerical Rating Scale, with 0 meaning no pain and 10 meaning pain as bad as you can imagine. The reported change is end-of-treatment score minus baseline. A negative value indicates a lower average rating at the end of treatment than at baseline; it is not itself a direct comparison between doses.

This kind of continuous endpoint retains information about the size of score changes rather than reducing participants to categories. The comparison between dose periods is relevant to the trial’s stated purpose of evaluating the difference between two doses. The registry record reports additional day-time, night-time, worst-pain, current-pain, and responder outcomes, but the primary endpoint remains the specified 24-hour average pain measure. The rationale for choosing this endpoint over other pain measures is not reported in the registry record.

7. Results Posted in the Registry

• GEn 3600 mg: mean -1.47 versus -1.18 on control; difference -0.29 (95% CI -0.77 to 0.19), standardized effect d = -0.18; z-test p = 0.233.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in the Mean Day-time Average Pain Intensity (API) Score at the Last Week of Each Treatment Period Using LOCF Data

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • GEn 3600 mg: mean -1.48 versus -1.17 on control; difference -0.31 (95% CI -0.79 to 0.17), standardized effect d = -0.19; z-test p = 0.205.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GEn 3600 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in the Mean Day-time Worst Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF Data

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • GEn 3600 mg: mean -1.50 versus -1.17 on control; difference -0.33 (95% CI -0.83 to 0.17), standardized effect d = -0.20; z-test p = 0.194.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GEn 3600 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in the Mean Current (Evening) Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF Data

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • GEn 3600 mg: mean -1.39 versus -1.10 on control; difference -0.29 (95% CI -0.79 to 0.21), standardized effect d = -0.17; z-test p = 0.259.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GEn 3600 mg and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists 18 outcome measures and 16 posted secondary outcomes. These include day-time and night-time average or worst pain scores, current pain ratings, and counts of participants reaching different levels of reduction in average pain. The outcomes address related aspects of pain, rather than independent questions wholly separate from the primary endpoint.

For the available continuous secondary results, the mean changes are negative for both dose periods, and the 3600 mg estimates are more negative than the 1200 mg estimates. The simple recalculations for day-time average pain, day-time worst pain, and current evening pain all have confidence intervals that include zero. Their unadjusted p-values are 0.205, 0.194, and 0.259, respectively. Taken together, these results show a broadly similar direction of dose-specific mean changes but do not provide a clear, consistent separation in the unadjusted comparisons.

These secondary findings should be read alongside, rather than substituted for, the primary outcome. They use related pain ratings and may be correlated with each other. When many outcomes are examined, the chance of at least one apparently notable result can increase. General statistical practice often addresses this through a prespecified testing hierarchy or multiplicity adjustment. Such a strategy is not reported in the registry record. The unadjusted secondary comparisons therefore carry less weight than the designated primary analysis, especially when interpreting isolated p-values. Detailed results for every listed secondary endpoint are not reported in the registry record.

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

• Change From Baseline in the Mean Day-time Average Pain Intensity (API) Score at the Last Week of Each Treatment Period Using LOCF Data
• Change From Baseline in the Mean Day-time Worst Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF Data
• Change From Baseline in the Mean Current (Evening) Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF Data
• Change From Baseline in the Mean Night-time Average Pain Intensity (API) Score at the Last Week of Each Treatment Period Using LOCF
• Change From Baseline in the Mean Night-time Worst Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF
• Change From Baseline in the Mean Current Morning Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF
• Number of Participants Achieving Various Levels of Percent Reduction From Baseline in the Mean 24-hour Average Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF Data
• Number of Participants Achieving Various Levels of Percent Reduction From Baseline in the Mean 24-hour Average Pain Intensity Score at the Last Week of Each Treatment Period Using LOCF Data by Period

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports least-squares means for the primary outcome, with standard errors as the dispersion measure. A least-squares mean is a model-based mean, often adjusted for terms included in the model; it is not necessarily the same as a simple arithmetic average. The registry reports the sponsor’s comparison as an ANCOVA adjusted mean difference. The covariates and other model terms are not reported in the registry record, so the specific adjustments cannot be reconstructed from the description.

What does the standard error describe?

A standard error describes the estimated uncertainty in a mean or other model estimate. It is not the spread of individual participants’ scores. In the primary outcome, the registry gives standard errors of 0.171 for the 1200 mg mean and 0.173 for the 3600 mg mean. Smaller standard errors generally indicate more precise estimates, while larger standard errors indicate greater uncertainty. The standard error alone does not show the range of individual responses.

How should the mean difference be read?

The sponsor’s adjusted mean difference is -0.29 for GEn 3600 mg relative to GEn 1200 mg. The negative direction means the adjusted change score for the higher dose was lower under the reported comparison. The difference is on the original points scale. It should be interpreted alongside the confidence interval and the model that produced it, not as a stand-alone measure of certainty or clinical importance.

What does the confidence interval contribute?

A confidence interval gives a range of estimates compatible with the data and the method used to calculate it. The sponsor reports a 90% interval from -0.48 to -0.10. That interval does not include zero. By contrast, the simple unadjusted recalculation’s interval includes zero. The different intervals arise from different analyses, not from a change in the displayed primary mean difference. Confidence intervals describe statistical uncertainty; they do not account for every possible source of bias or prove that an effect is clinically important.

Why can a crossover analysis differ from an independent-groups recalculation?

In a crossover study, the same participant contributes data under both treatments. General statistical principle: when outcomes from the same participant are correlated, a paired or appropriately modeled comparison can use that relationship. An independent-groups calculation ignores it. The posted summary statistics do not include the within-participant covariance needed to reproduce a paired calculation. The recalculation also uses an approximate standard deviation derived from the posted standard errors and sample sizes. It is consequently a simple, unadjusted check, not a reconstruction of the sponsor’s ANCOVA.

What is LOCF?

The registry states that LOCF was used if fewer than 4 days of diary data were provided. In general, last observation carried forward fills a missing later measurement with the participant’s most recent available observation. This can retain participants with incomplete diary records, but it relies on an assumption that the carried-forward value is a suitable stand-in for the missing outcome. That assumption may not reflect the participant’s unobserved pain ratings. The registry record does not report sensitivity analyses comparing LOCF with other missing-data methods.

How should the p-values be understood?

The sponsor reports p = 0.013 for its ANCOVA comparison. The unadjusted recalculation gives p = 0.233. A p-value describes how unusual data at least as incompatible with a null hypothesis would be under the specified statistical model, if that null hypothesis and model assumptions held. It is not the probability that the null hypothesis is true. The values should not be treated as interchangeable because they arise from different methods. The sponsor’s result is the trial’s reported analysis; the recalculation is an approximate, unadjusted analysis of summary statistics.

The registry reports one primary analysis and lists 18 outcome measures. The trial’s original sample-size rationale, model specification, testing hierarchy, and any planned interim analyses are not reported in the registry record. The methods available support a focused reading of the reported primary result, but not a complete independent audit of the analysis plan.

11. Confidence Intervals and Precision

The sponsor’s adjusted estimate is -0.29, with a 90% confidence interval from -0.48 to -0.10. The interval stays below zero, indicating that the reported adjusted comparison is not compatible with no difference at that interval level under the sponsor’s model. Its width also shows that the analysis does not identify a single exact value for the difference. The point estimate is the center of the reported result, not a guarantee of the true difference.

The unadjusted recalculation has a confidence interval that includes zero, and its p-value is 0.233. This is a less conclusive result from the simple summary-statistic calculation. The contrast with the sponsor interval reflects the different analytical approaches; it should not be read as though one interval is a second measurement of the same model-based estimate. The registry record does not provide the participant-level data or the covariance between dose periods needed to reconstruct the crossover comparison independently.

Precision is not the same as clinical importance. The outcome is measured in points, but a threshold for an important change is not reported in the registry record. Nor does a confidence interval include every uncertainty, such as uncertainty about modeling assumptions, missing-data handling, or how well results generalize beyond the trial participants. The sponsor interval is the interval from the trial’s reported analysis, while the recalculated interval illustrates how an unadjusted approach to the summaries yields greater uncertainty.

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 planning figures for the primary comparison are participants per arm, not total events. They are illustrations computed from the observed effect, not the trial’s original design calculation. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. The trial’s original sample-size assumptions and justification are not reported in the registry record.

Using 100% of the observed effect, the illustration gives 481 participants per arm for 80% power and 644 participants per arm for 90% power. Using 75% of the observed effect, it gives 855 participants per arm for 80% power and 1,144 participants per arm for 90% power. Using 50% of the observed effect, it gives 1,923 participants per arm for 80% power and 2,574 participants per arm for 90% power.

In plain language, the 80% column describes a design that would detect the specified effect in 80% of repeated studies under the stated assumptions. The 90% column uses a higher chance of detection and therefore calls for more participants for the same effect. Smaller effects are harder to distinguish from random variation, so the planning illustrations require more participants as the target effect gets smaller. These figures are not observed power and do not describe the probability that this completed trial detected an effect.

13. How the Sample Size Is Derived

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

Formula

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

Worked calculation for this trial

1. Observed difference Δ = -0.290; pooled standard deviation σ = 1.605.
2. n = 2 × (1.96 + 0.842)² × 1.605² / -0.290² = 480.6, rounded up to 481 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.29                      # observed difference in means
sigma <- 1.60465420095769                     # 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(617461)
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 = 481 (80% power), 644 (90% power)
 75% of observed effect: n per arm = 855 (80% power), 1,144 (90% power)
 50% of observed effect: n per arm = 1,923 (80% power), 2,574 (90% power)
power.t.test: 481.6 
Simulated power at n = 481 per arm: 0.799 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The sponsor’s reported analysis is an ANCOVA with an adjusted mean difference of -0.29 for GEn 3600 mg relative to GEn 1200 mg. It gives a 90% confidence interval from -0.48 to -0.10 and p = 0.013. This is the trial’s own reported comparison. The registry labels its parameter as an adjusted mean difference versus placebo, although the listed groups are the two gabapentin enacarbil doses; the primary outcome’s comparison is between those doses.

The simple recalculation from the posted summary statistics also gives a mean difference of -0.29, but its confidence interval includes zero and p = 0.233. It is unadjusted, uses an approximate standard deviation, and does not use the crossover pairing or reproduce ANCOVA. The matching point estimates do not make the inferential results equivalent: uncertainty depends on the method and information used.

The registry record flags a discrepancy between the reported analysis and recalculation. The sponsor’s own analysis takes precedence. The recalculation should be understood as a limited check on summary data, not a correction to the sponsor’s result. Details of the ANCOVA covariates and complete analysis specification are not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary outcome and lists secondary outcomes covering several pain ratings and responder definitions. It reports one analysis for the primary endpoint. The number and variety of outcomes make it important to know whether the trial specified a hierarchy of tests, an adjustment for multiple comparisons, or another approach to controlling false-positive findings. Those details are not reported in the registry record. The registry record also does not report whether interim analyses were planned or conducted.

Without a stated multiplicity strategy, the secondary results should be treated as exploratory unless other documentation establishes their planned status. Examining multiple outcomes can make isolated nominally notable results less persuasive, because the overall chance of a chance finding may rise as more comparisons are considered. No adjustment should be presumed from the presence of a primary endpoint alone.

The primary outcome description states that LOCF was used if fewer than 4 days of diary data were provided. This addresses incomplete diary records by carrying forward the last observed value, but it does not establish that missingness had no effect on the result. The registry record does not report a sensitivity analysis using a different missing-data approach or describe reasons for missing diary observations. It notes that one participant had no post-baseline 24-hour API data during the 3600mg treatment period.

The primary outcome description identifies an ITT population of 93 and gives the dose-period counts of 90 and 84. It also reports participants who did not take each dose in the second period. The record does not fully explain all exclusions or the precise population used in every analysis. The original sample-size justification, full statistical analysis plan, and detailed handling of period effects are not reported in the registry record. These omissions limit assessment of prespecification and robustness, but do not alter the numerical results that the registry does report.

16. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hallucination, auditory (1.2%).

The registry lists serious adverse events by treatment or study period. It reports 0 affected participants among 94 at risk for baseline gabapentin 1800 mg, 0 among 91 for GEn 1200 mg, 0 among 82 for crossover GEn 2400 mg, and 0 among 85 for GEn 3600 mg. In the down-titration period, 1 of 80 participants at risk was affected. For overall GEn, 1 of 94 was affected. The serious adverse event term listed is hallucination, auditory, at 1.2%.

Because this was a crossover study, period-specific groups may include participants who also appear in other treatment-period groups. The counts should not be added to estimate a number of unique participants. The overall GEn figure is reported separately. The registry record does not provide event descriptions beyond the listed term, or details to establish relatedness to treatment. Death counts are not reported in the registry record; a blank value should not be interpreted as zero. These counts describe reported serious adverse events, not every possible safety outcome.

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

The primary outcome records a reduction in mean pain scores from baseline during both dose periods. The sponsor’s adjusted comparison reports a difference between the doses, with an estimate of -0.29 and a 90% confidence interval from -0.48 to -0.10. The simple unadjusted recalculation does not exclude zero. The registry record flags this discrepancy, and the sponsor’s own analysis takes precedence for the trial’s reported inference.

The result does not establish the size of an effect that would be important to an individual participant. The record does not report a clinical-importance threshold. A statistically significant result in the sponsor analysis does not by itself establish that participants experienced a noticeable or worthwhile difference. Conversely, the recalculation’s interval including zero does not prove that the doses are identical. It indicates that this particular unadjusted method leaves no-difference values compatible with its estimate.

The trial compared two gabapentin enacarbil doses in a defined population with post-herpetic neuralgia and inadequate response to gabapentin. It did not compare either dose with placebo in the primary outcome, and the primary result cannot answer how either dose compares with treatments not studied. The design and population also do not establish the result for all people with post-herpetic neuralgia.

The registry record gives limited detail about the ANCOVA specification, missing-data sensitivity analyses, period effects, multiplicity, and the original design assumptions. These gaps make it difficult to assess the robustness of the result from the record alone. The responsible interpretation is therefore specific: the sponsor’s analysis reports a dose difference for the primary measure, while the approximate unadjusted recalculation is more uncertain; the magnitude’s clinical meaning and broader applicability are not determined by these statistics alone.

18. Why This Trial Matters Statistically

This trial is statistically instructive because its crossover design and the difference between its reported analysis and a summary-statistic recalculation show how analysis choices affect uncertainty. The dose-specific means and standard errors are not enough to reproduce a paired comparison: the relationship between each participant’s outcomes across periods also matters. A recalculation that treats the dose summaries as independent can yield the same point estimate while producing a different confidence interval and p-value.

The registry also makes clear why the trial’s own reported model should be distinguished from an illustrative recalculation. The sponsor used ANCOVA and reported an adjusted comparison; the unadjusted calculation is a limited approximation. When a discrepancy is flagged, the recalculation can help readers understand sensitivity to method, but it should not displace the sponsor analysis.

Finally, the trial illustrates the difference between a statistical result and a clinical conclusion. The sponsor’s interval excludes zero, but the record does not report a clinical-importance threshold. Multiple related pain outcomes and incomplete reporting of multiplicity procedures further caution against selecting isolated secondary results as decisive. Careful interpretation keeps the endpoint, design, model, missing-data approach, and uncertainty in view together.

19. Limitations

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

The registry record does not report the full ANCOVA specification, including the covariates and handling of crossover-period effects. It also does not provide the within-participant covariance needed to reconstruct the dose comparison from the posted summaries. The recalculation is therefore an approximate independent-groups analysis rather than a faithful reanalysis of the crossover data.

LOCF was used when fewer than 4 days of diary data were provided. The record does not report a sensitivity analysis with alternative missing-data methods or detail the reasons for missing diary data. The primary outcome description notes one participant without post-baseline 24-hour API data during the 3600mg period, as well as participants who did not take a dose in the second period. Full details of all analysis exclusions are not reported in the registry record.

The original sample-size rationale and trial analysis plan are not reported in the registry record. Neither are any multiplicity adjustments, interim analyses, or a clinical-importance threshold for the pain scale. The registry lists 18 outcome measures, but detailed results are not available for all of them in the record. Safety reporting includes serious adverse event counts, but death counts are not reported in the registry record and the listed adverse-event information does not establish causality.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Clinical Study in Subjects With Neuropathic Pain From PHN Who Have Had an Inadequate Response to Gabapentin Treatment — NCT00617461.
• Registry record last updated 2013-07-22; status last verified 2013-01.
• Results first posted 2011-05-20.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

This completed phase 2 randomized, triple-masked crossover trial enrolled 96 participants with neuropathic pain associated with post-herpetic neuralgia and inadequate response to gabapentin. Participants received gabapentin enacarbil at 1200mg/day and 3600mg/day in 4-week treatment periods.

The primary endpoint was change from baseline in mean 24-hour average pain intensity. The sponsor’s ANCOVA reported an adjusted mean difference of -0.29, with a 90% confidence interval from -0.48 to -0.10 and p = 0.013. An unadjusted recalculation gave the same point estimate but a confidence interval including zero and p = 0.233. The registry flags the discrepancy; the sponsor’s analysis takes precedence.

The record does not report the complete analysis plan, multiplicity strategy, original sample-size justification, or sensitivity analyses for missing data. The numerical comparison describes the trial’s primary pain outcome, not clinical importance or effects in populations and comparisons not studied.