A Study of E2020 in Patients With Dementia With Lewy Bodies (DLB), Followed by a Long-term Extension Phase: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of E2020 in Patients With Dementia With Lewy Bodies (DLB), Followed by a Long-term Extension Phase (NCT01278407), a randomized phase 3 study. The primary outcome is Change From Baseline in Mini-Mental State Examination (MMSE) Score.

2. Clinical Question

The central question is whether mean change in cognitive function over the confirmatory phase differed among participants assigned to donepezil 5 mg, donepezil 10 mg, or matched placebo. The primary measure was change from baseline in Mini-Mental State Examination (MMSE) score at Week 12. Positive change indicated improvement from baseline, and higher MMSE scores indicated better function.

The posted comparison statistics contrast each donepezil dose with placebo. They describe average change in the analyzed participants, not the experience of every participant or an individual-level treatment effect. The registry reports the number analyzed and the mean change with its standard deviation for each arm. It also gives unadjusted mean differences and confidence intervals for the dose-versus-placebo comparisons.

These results address the confirmatory phase, not the longer-term effect of treatment during the extension phase. Although the study included an extension, the primary endpoint’s timeframe is Week 12. The extension descriptions indicate changes in treatment for participants who began on placebo and dose changes for some participants who began on donepezil. The extension therefore should not be interpreted as a continued randomized comparison of the original treatment groups based on the information in the registry record.

The results are descriptive of the posted efficacy population and analyses. The record does not report an adjusted inferential analysis or a multiplicity procedure. Accordingly, the unadjusted comparisons should be read as statistical contrasts from the posted summary results, with uncertainty and analysis limitations considered alongside the point estimates.

3. Trial Design

This was a Phase 3, randomized, triple-masked, parallel-group, placebo-controlled study of E2020 in people with dementia with Lewy bodies. The registry lists enrollment of 142 participants. The confirmatory phase compared placebo with donepezil 5 mg and donepezil 10 mg. Treatment was taken orally once daily for 12 weeks.

In the 5 mg arm, treatment began at 3 mg for 2 weeks and then increased to 5 mg for 10 weeks. In the 10 mg arm, treatment began at 3 mg for 2 weeks, increased to 5 mg for 4 weeks, and then increased to 10 mg for 6 weeks. Participants in the placebo arm received matched placebo. These descriptions identify the assigned regimens; the registry record does not report individual adherence or dose exposure.

The primary endpoint was assessed at Week 12, the stated timeframe for the confirmatory phase. The efficacy arm counts were 44 for placebo, 43 for donepezil 5 mg, and 49 for donepezil 10 mg. These are the numbers used for the posted endpoint summaries and are distinct from the safety risk counts reported elsewhere in the record.

The study also included a long-term extension phase. The registry describes participants previously receiving placebo as continuing placebo until Week 16, then beginning donepezil with subsequent dose increases. Participants previously receiving donepezil maintained their assigned treatment and doses until Week 24, with an increase to 10 mg for those in the 5 mg group at Week 24. The extension descriptions are truncated in the registry record, so later treatment details are not reported in the registry record.

The registry lists a start in 2011-02, primary completion in 2013-03, and study status as completed. It does not report the original sample-size justification, the allocation ratio, or the detailed procedures used to generate the random sequence. The masking label is triple, but which specific groups were masked is 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.

Before assignment: In the 'Placebo to Donepezil (5+10 mg) Extension Phase' arm, 37 of 46 participants started active treatment at Week 16.

Confirmatory Phase

Reasons for not completing this period:

Extension Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set: Efficacy was analyzed in the full analysis set, including the randomized participants who received the study drug at least once and had valid efficacy assessment data at more than one point.

5. Baseline Characteristics

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

Population: The data is based on Full Analysis Set, defined as all the randomized patients who received the study drug at least once and had valid efficacy assessment data at more than one point.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Mini-Mental State Examination (MMSE) Score, measured over Week 12 for Confirmatory Phase.

The primary endpoint was change from baseline in Mini-Mental State Examination (MMSE) score at Week 12 of the confirmatory phase. The MMSE is a questionnaire with 11 items covering cognitive domains such as orientation, memory, attention, and construction. The score ranges from 0 to 30, with higher scores indicating better function. A positive change score represents improvement relative to baseline.

The endpoint is a continuous measure, and the registry reports the mean change and standard deviation in each arm. A mean gives a compact description of the group’s average change; the standard deviation describes how much individual results vary around that average. These are different kinds of information: a mean does not show how consistent the changes were among participants, while a standard deviation does not by itself indicate the direction of the average change.

The MMSE is used in this study to assess cognitive impairment and overall cognitive function in dementia. Reporting change from baseline focuses on how scores shifted during the confirmatory phase rather than only on the Week 12 score. The direction of the scale matters when reading the results: a more positive change indicates a larger average improvement from baseline on this measure. It does not, by itself, describe the clinical importance of a difference or establish how an individual participant’s function changed.

The endpoint population was the Full Analysis Set, defined in the registry as randomized participants who received study drug at least once and had valid efficacy assessment data at more than one point. The record does not report how many randomized participants were excluded from that set or the reasons for any exclusions.

7. Results Posted in the Registry

• Donepezil 5 mg - Confirmatory Phase: mean 1.40 versus 0.60 on control; difference 0.80 (95% CI -0.55 to 2.15), standardized effect d = 0.25; z-test p = 0.245.
• Donepezil 10 mg - Confirmatory Phase: mean 2.20 versus 0.60 on control; difference 1.60 (95% CI 0.40 to 2.80), standardized effect d = 0.54; z-test p = 0.009.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous, so it is summarized with a mean and standard deviation. The mean change describes the average movement in MMSE score from baseline for the participants analyzed in an arm. The standard deviation describes the spread of individual changes around the arm mean. In general, a larger standard deviation means observations are more dispersed; it does not mean the estimated arm mean is necessarily less accurate by the same amount, because precision also depends on how many participants contribute data.

What does a mean difference represent?

The posted mean difference compares an active-treatment arm’s average change with the placebo arm’s average change. A positive difference means the active-treatment arm’s mean change is numerically higher than the placebo mean change, given the scale direction stated in the registry. For the 5 mg comparison, the posted difference is 0.8; for the 10 mg comparison, it is 1.6. These are contrasts between group averages, not individual changes and not adjusted effects from a model.

The registry also reports pooled standard deviations and standardized effect-size values. A standardized effect expresses a mean contrast relative to the variability used in its calculation. Such a measure can help describe the size of a contrast on a scale that is less tied to the MMSE’s original units. It does not replace the mean difference in MMSE units, and it does not establish clinical importance. The record does not report a clinical threshold for interpreting the size of a change.

How should confidence intervals be read?

In general, a confidence interval gives a range of values compatible with the data and the calculation’s assumptions. It communicates uncertainty around an estimate: a narrower interval is more precise than a wider one, all else being equal. It is not a range that contains every participant’s response, and it is not a probability statement about a particular treatment effect after the study is complete. The intervals posted for these unadjusted mean differences reflect uncertainty in the comparisons, not the full uncertainty associated with unreported modeling choices or missing-data procedures.

The interval for the donepezil 5 mg contrast includes zero; the interval for the donepezil 10 mg contrast does not. Zero represents no difference between the two group means on the comparison scale. This interval-based distinction is consistent with the corresponding unadjusted p-values, but neither interval nor p-value accounts for multiplicity unless an adjustment is made. No multiplicity adjustment is reported in the registry record.

What does an unadjusted p-value tell the reader?

In general, a p-value measures how incompatible the observed data, or more extreme data, are with a specified null model under its assumptions. It is not the probability that the null hypothesis is true, the probability that the finding arose by chance, or a measure of the size or importance of a difference. The posted p-values are unadjusted: 0.245 for donepezil 5 mg versus placebo and 0.009 for donepezil 10 mg versus placebo. Because the record presents two dose-versus-placebo contrasts and does not report an adjustment, these should be treated as nominal statistical summaries rather than as multiplicity-controlled results.

The comparison statistics are unadjusted summaries calculated from the posted arm results. The registry record does not report an analysis model, covariate adjustments, standard error method, or a sponsor-generated inferential result for these comparisons. There is no discrepancy flagged between the posted summaries and the comparison calculations. That does not supply the missing details of the original analysis plan.

Why does variation matter?

MMSE changes vary across participants, as reflected in the arm standard deviations. In general, greater variability makes it harder to distinguish a group difference from random variation unless the data also provide enough information to estimate the group means precisely. This is one reason an average difference should be read together with its confidence interval and the number analyzed. The standard deviations do not indicate the shape of the data distribution, and the registry record does not report distributional checks or sensitivity analyses.

Over-dispersion is a general concern in count outcomes when observed variation exceeds the level assumed by a statistical model. It is not the main issue for this continuous MMSE-change endpoint, which is reported using means and standard deviations. The registry does not describe a model that would permit assessment of whether over-dispersion was considered. More generally, the appropriateness of a statistical method depends on the outcome and assumptions; the method used for the trial’s original analysis is not reported in the registry record.

Who was included in the efficacy summary?

The Full Analysis Set included randomized participants who received the study drug at least once and had valid efficacy assessment data at more than one point. This definition identifies the analysis population, but it does not tell the reader how many otherwise randomized participants were excluded or how missing endpoint measurements were treated. The efficacy counts are 44 for placebo, 43 for donepezil 5 mg, and 49 for donepezil 10 mg. The record does not report whether an imputation method or another missing-data strategy was used.

In general, missing-data methods can affect estimates when outcome availability is related to participants’ outcomes or treatment experience. The registry does not report the amount, timing, or reasons for missing data, so the potential impact cannot be assessed from these summaries. The available means should therefore be understood as summaries for the reported analysis population, not automatically as results for every randomized participant.

10. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimate. The mean-difference estimate for donepezil 5 mg versus placebo was 0.8, and its confidence interval extends across zero. The estimate for donepezil 10 mg versus placebo was 1.6, and its confidence interval lies above zero. The posted unadjusted p-values are 0.245 and 0.009, respectively. These intervals and p-values are unadjusted comparisons.

The uncertainty should be considered with the arm sizes: the efficacy summaries include 44 participants in the placebo arm, 43 in the donepezil 5 mg arm, and 49 in the donepezil 10 mg arm. The standard deviations were 3.0, 3.4, and 2.9, respectively. They show that individual changes varied around each arm’s mean. A difference between two means can be estimated with more or less precision depending on both the variation in the data and the number of observations.

In general, a confidence interval is a range produced by a statistical procedure that, under its assumptions, reflects uncertainty about the estimated contrast. It is not a range for individual participant outcomes, and it does not cover all uncertainty about the analysis. Here, the registry record does not state the original analysis method, covariate adjustments, missing-data method, or multiplicity approach. The intervals therefore should not be interpreted as resolving those unreported aspects of the trial analysis.

The interval for the 5 mg comparison leaves substantial uncertainty about the direction of the mean difference. The interval for the 10 mg comparison is more consistent with a positive mean difference under the unadjusted comparison, but it does not establish the size or importance of an effect for an individual or beyond the endpoint’s Week 12 timeframe. The registry does not report a clinically meaningful threshold for MMSE change.

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 sample-size display is a planning illustration based on the observed mean differences, not the trial’s original design calculation. The figures are participants per arm for 80% power and participants per arm for 90% power; they are not total event counts. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed to have an 80% chance of detecting the selected effect under those assumptions, if that effect and the assumptions were correct. The 90% column uses a higher target chance of detecting the selected effect and therefore generally requires more participants per arm. These are prospective planning quantities, not evidence about the probability that the completed trial’s result is true. This is not observed power.

Each comparison has planning rows for 100%, 75%, and 50% of its observed effect. For the 5 mg comparison, planning for 100% of the observed effect gives 252 participants per arm for 80% power and 338 for 90% power; planning for 50% gives 1,008 and 1,349, respectively. For the 10 mg comparison, the corresponding 100% row gives 54 and 72 participants per arm, while the 50% row gives 214 and 286. The 75% rows fall between those effect assumptions in the displayed calculations.

The smaller the effect targeted for detection, the more participants the planning illustration requires, because a smaller difference is harder to distinguish from variability. These values should not be read as the number recruited or analyzed in the trial. They use observed effects as inputs, so they are illustrations of how planning requirements change under the stated assumptions, not a retrospective assessment of whether the trial was adequately powered.

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 = 0.800; pooled standard deviation σ = 3.204.
2. n = 2 × (1.96 + 0.842)² × 3.204² / 0.800² = 251.8, rounded up to 252 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 <- 0.8                      # observed difference in means
sigma <- 3.20389468872973                     # 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(1278407)
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 = 252 (80% power), 338 (90% power)
 75% of observed effect: n per arm = 448 (80% power), 600 (90% power)
 50% of observed effect: n per arm = 1,008 (80% power), 1,349 (90% power)
power.t.test: 252.7 
Simulated power at n = 252 per arm: 0.802 (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. Multiplicity, Missing Data and Analysis Populations

The registry record presents two primary endpoint comparisons: donepezil 5 mg versus placebo and donepezil 10 mg versus placebo. It reports unadjusted p-values and confidence intervals for both. No multiplicity procedure, testing hierarchy, or adjusted result is reported in the registry record. When several comparisons are considered, the chance of obtaining at least one apparently unusual result can differ from the chance for a single planned comparison. In general, multiplicity methods address that issue; without a reported method, the posted p-values should be read as unadjusted rather than as error-controlled across the comparisons.

The record lists two outcome measures and no posted secondary outcomes. It reports no analysis entries. That absence means the registry record does not describe additional statistical analyses; it does not establish that no further analyses were conducted outside the registry. The original analysis plan, including any prespecified primary comparison or secondary analyses, is not reported in the registry record.

The efficacy population is the Full Analysis Set: randomized participants who received study drug at least once and had valid efficacy assessment data at more than one point. The registry does not report the number randomized but excluded from this set, the reasons for exclusion, or the number with missing Week 12 MMSE data. It also does not report whether missing outcomes were imputed, whether a model used all available measurements, or whether sensitivity analyses assessed alternative assumptions about missing data.

In general, missing-data handling matters because an analysis of available observations can differ from an analysis that accounts for unobserved outcomes, particularly if outcome availability is related to treatment or participants’ underlying status. The registry summaries do not permit evaluation of that possibility. The original statistical analysis plan, including assumptions and methods for missing data, is not reported in the registry record.

The Full Analysis Set definition should not be assumed to mean that every randomized participant contributed to the endpoint summary. The arm counts for efficacy are 44, 43, and 49, while safety risk counts are reported separately. Those differences in counts should not be reconciled by inference: the registry does not explain the reasons for the distinct populations or counts.

14. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: spinal compression fracture (5.4%), myocardial infarction (2.7%), cataract (2.7%), inguinal hernia (2.7%), femoral neck fracture (2.7%).

The serious adverse event summaries report both the number at risk and the number affected. In the confirmatory phase, the counts were 5 of 46 for placebo, 4 of 47 for donepezil 5 mg, and 1 of 49 for donepezil 10 mg. These figures describe participants with at least one serious adverse event as recorded in the registry; they are not the efficacy analysis counts.

The registry also lists other adverse events affecting 23 of 46 participants in the placebo arm, 18 of 47 in the donepezil 5 mg arm, and 17 of 49 in the donepezil 10 mg arm. In the extension phase, it reports 9 of 37 participants affected by serious adverse events in the placebo-to-donepezil group and 12 of 96 in the donepezil extension group. Other adverse events affected 31 of 37 and 62 of 96, respectively.

These are counts of affected participants and do not by themselves establish causation, severity beyond the serious-event classification, or comparative risk under a prespecified safety analysis. Extension groups reflect treatment histories and changes described in the record, so they are not equivalent to the original confirmatory randomized comparisons. Death counts are not reported in the registry record. The listed serious-event terms and percentages are registry summaries, not evidence that an event was caused by treatment.

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

The posted endpoint results show positive mean MMSE changes in all three confirmatory arms: 0.6 with placebo, 1.4 with donepezil 5 mg, and 2.2 with donepezil 10 mg. The unadjusted mean differences versus placebo were 0.8 for the 5 mg arm and 1.6 for the 10 mg arm. The first comparison’s confidence interval includes zero; the second comparison’s interval does not. Their unadjusted p-values were 0.245 and 0.009.

These statistics describe group averages at Week 12 for the Full Analysis Set. They do not say that every participant improved, quantify an individual participant’s expected change, or establish a clinically meaningful threshold. The registry record does not report such a threshold. Standard deviations of 3.0, 3.4, and 2.9 indicate variation in participant changes within the respective arms, so the means do not capture the full range of observed experiences.

The two dose comparisons are each against placebo. The registry does not report a direct statistical comparison of donepezil 5 mg with donepezil 10 mg, so the difference between the two posted means should not be presented as a tested dose contrast. Nor do the results establish the durability of the Week 12 pattern: the primary endpoint timeframe is the confirmatory phase, and later extension results for this endpoint are not reported in the registry record.

The comparisons are unadjusted, and the registry does not report a multiplicity procedure, analysis model, or missing-data method. These limits matter when interpreting the strength and scope of the statistical evidence. The unadjusted p-value for the 10 mg comparison is a statistical summary under its assumptions, not the probability that a treatment effect is true or a measure of clinical importance. The record’s safety summaries also cannot establish causation or substitute for a full benefit-risk assessment.

In general, a trial result is most informative when the estimate, its uncertainty, the analysis population, the endpoint, and the methods are considered together. Here, the mean differences and intervals supply useful information about the posted Week 12 summaries, while unreported analysis details constrain broader conclusions. The data should be understood as registry-reported statistical results, not as a complete reconstruction of the trial’s protocol or analysis.

16. Why This Trial Matters Statistically

This trial illustrates why a reported mean difference needs context. The MMSE endpoint is expressed in the scale’s units, while standard deviations describe the spread of participant changes. Reading the means alone would omit the variability in each arm; reading a p-value alone would omit both the estimated difference and its uncertainty. The posted summaries make those distinct aspects visible.

The two dose-versus-placebo comparisons also show how the interpretation of an estimate depends on its interval and on the analysis context. The 5 mg comparison has a positive point estimate but an interval that includes zero. The 10 mg comparison has a positive point estimate and an interval that excludes zero. Both p-values are unadjusted, and the registry does not report a method for handling multiple comparisons. A statistical reading therefore needs to distinguish nominal evidence from a result adjusted for a stated testing strategy.

The study’s efficacy counts and safety risk counts are not identical. That difference is a reminder that analysis populations are outcome-specific and should not be treated as interchangeable without explanation. The registry defines the Full Analysis Set for efficacy but does not report why the safety and efficacy denominators differ.

The sample-size illustrations demonstrate the dependence of planning requirements on the effect chosen for detection. A smaller target effect requires more participants per arm under the listed assumptions. Because the calculations use observed effects and do not represent the trial’s original design justification, they are best read as conditional planning examples, not as a verdict on the completed study’s power.

Overall, the record supports a focused statistical discussion of a continuous endpoint, unadjusted comparisons, uncertainty, analysis populations, and planning assumptions. It also shows why registry summaries need to be separated from an original statistical analysis plan, which is not reported in the registry record.

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 does not report the original sample-size justification, randomization ratio, sequence-generation procedures, or the detailed masking arrangements. Although the study is identified as randomized and triple-masked, those labels do not reveal the operational procedures used. The reported enrollment of 142 is not the same as the number in each efficacy analysis arm.

The main efficacy summaries are based on 44 placebo participants, 43 donepezil 5 mg participants, and 49 donepezil 10 mg participants. The registry does not report a participant flow accounting for differences between enrollment, randomization, treatment receipt, and inclusion in the Full Analysis Set. It also does not explain why the safety risk denominators are 46, 47, and 49 in the confirmatory groups rather than the efficacy denominators.

The registry record does not report the detailed statistical analysis plan, covariate adjustment, model diagnostics, handling of missing measurements, or sensitivity analyses. It gives unadjusted comparison statistics, but does not report a multiplicity approach for the two dose comparisons. Those omissions limit assessment of the robustness of the inferential results.

The endpoint’s timeframe is Week 12 in the confirmatory phase. Although an extension phase is described, the posted primary endpoint results do not establish whether the differences persisted over a longer period. Extension treatment changed over time, and the descriptions are truncated; later details are not reported in the registry record. No direct donepezil-dose comparison is reported.

The registry reports serious adverse event counts and selected serious-event terms but does not provide death counts. The listed terms and percentages do not establish treatment causation. Safety interpretation is further limited by the lack of a detailed analysis plan and by the different follow-up and treatment histories in the extension groups.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Study of E2020 in Patients With Dementia With Lewy Bodies (DLB), Followed by a Long-term Extension Phase — NCT01278407.
• Registry record last updated 2023-06-29; status last verified 2015-12.
• Results first posted 2015-11-30.
• 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

The registry record describes a completed Phase 3, randomized, triple-masked, parallel-group, placebo-controlled study of E2020 in people with dementia with Lewy bodies. Enrollment was 142. The confirmatory phase compared matched placebo with donepezil 5 mg and 10 mg over 12 weeks.

The primary endpoint was change from baseline in MMSE score at Week 12. Mean changes were 0.6 for placebo, 1.4 for donepezil 5 mg, and 2.2 for donepezil 10 mg. The unadjusted mean differences versus placebo were 0.8 and 1.6, with unadjusted p-values of 0.245 and 0.009, respectively.

The record also describes a long-term extension and posts adverse-event counts. The original sample-size rationale, detailed analysis plan, multiplicity approach, missing-data method, and death counts are not reported in the registry record.