A Study to Investigate the Safety and Effectiveness of Arbaclofen Extended-Release Tablets for Patients With MS: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Investigate the Safety and Effectiveness of Arbaclofen Extended-Release Tablets for Patients With MS (NCT03290131), a randomized phase 3 study. The primary outcome is Change From Baseline in Total Numeric-transformed Modified Ashworth Scale Score of the Most Affected Limb (TNmAS-MAL).

2. Clinical Question

The clinical question is whether change in muscle tone over 84 days differed between patients assigned to either dose of arbaclofen extended-release tablets and patients assigned to placebo. The primary outcome was change from baseline in the Total Numeric-transformed Modified Ashworth Scale score for the most affected limb.

3. Trial Design

This was a Phase 3, randomized, parallel-group study with placebo control. The official title describes it as double-blind, while the registry’s masking field records single masking. Which people were masked, and how the two descriptions relate, are not reported in the registry record.

The registry lists three groups: AERT 40 mg, AERT 80 mg, and placebo. The active groups are described as receiving 40 mg or 80 mg arbaclofen extended-release tablets. The placebo group received placebo. The recorded enrollment was 536. The registry status is completed, with a study start date of 2018-01-28 and a primary completion date of 2018-12-03.

The primary endpoint was a continuous measure: change from baseline in the Total Numeric-transformed Modified Ashworth Scale score of the most affected limb at 84 days. Results are reported as group means with standard deviations. The registry lists two outcome measures and no posted analyses. It does not report the original statistical analysis plan, the trial’s original sample-size justification, details of allocation procedures, or the original analysis population.

Randomization is intended to support a fair comparison by assigning participants to study groups by chance; it does not guarantee that all baseline characteristics are identical. A parallel-group design assigns each participant to a study group rather than comparing multiple treatments within the same participant. A placebo group provides a concurrent reference for interpreting change over the study period. These are general features of the design; the registry record does not provide enough detail to characterize implementation beyond the recorded design labels and group descriptions.

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

Reasons for not completing this period:

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 Total Numeric-transformed Modified Ashworth Scale Score of the Most Affected Limb (TNmAS-MAL), measured over 84 days.

The primary endpoint is change from baseline in the Total Numeric-transformed Modified Ashworth Scale score for the most affected limb, measured at 84 days. The scale assesses abnormality in muscle tone or resistance to passive movement. Its scores are based on three main joints in a limb; the joint values are summed to form a limb score. The most affected limb is the limb with the highest score. The stated limb range is 0 to 15, and a higher score is a worse outcome.

Using change from baseline focuses the comparison on how the recorded score shifted during the study, rather than only on its value at the end. A negative mean change indicates that the recorded score decreased from baseline; it does not by itself establish why the score changed or whether the difference is important to patients. The endpoint is used as a quantitative measure of muscle tone because it places the observed change on a defined scale and permits comparison of group means.

The reported means describe average change, while the standard deviations describe variation in individual measurements within each group. The endpoint captures the score for the most affected limb; it should not be read as a measure of every limb, every symptom, or overall health. The registry describes a total-limb score separately, but that is not the stated primary endpoint.

7. Results Posted in the Registry

• AERT 40 mg: mean -1.70 versus -1.40 on control; difference -0.30 (95% CI -0.69 to 0.09), standardized effect d = -0.16; z-test p = 0.135.
• AERT 80 mg: mean -2.00 versus -1.40 on control; difference -0.60 (95% CI -0.97 to -0.23), standardized effect d = -0.33; z-test p = 0.002.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean change summarize?

The endpoint is continuous, so each participant’s change can be represented numerically and the group result summarized by a mean. A mean is an average: it gives a compact description of the group’s changes, but it does not show the full distribution or indicate that every participant had a change close to that average. In the posted results, the means are negative in all three groups. Given the scale direction, that means the average recorded score decreased from baseline in each group.

The standard deviation describes the spread of individual observations around the group mean. The reported standard deviations are 1.97 for AERT 40 mg, 1.78 for AERT 80 mg, and 1.82 for placebo. Greater spread means that individual outcomes are more variable around the mean. A standard deviation is not the uncertainty of the mean itself, and it should not be interpreted as a range containing a specified share of participants.

How is a mean difference read?

The comparison estimate is the mean change in an active-treatment group minus the mean change in the placebo group. A negative estimate therefore means that the active-treatment group’s mean change was more negative than placebo’s mean change. The estimate for AERT 40 mg was -0.3, and the estimate for AERT 80 mg was -0.6. These differences are in units on the scale. The scale direction matters: negative change is a reduction in score, while higher scores correspond to worse outcomes.

A mean difference describes the size and direction of the estimated contrast; it does not on its own communicate how uncertain that estimate is. That uncertainty is described by the confidence interval. The p-value addresses a different question, assessing how compatible the observed result is with a specified null model under the calculation used. The unadjusted p-values in the registry are 0.135 for AERT 40 mg and 0.002 for AERT 80 mg. They should be read as unadjusted comparisons, not as a complete account of the analysis plan or as a measure of clinical importance.

Why do confidence intervals matter?

A confidence interval gives a range of values compatible with the estimate under the method used. General statistical interpretation: a narrower interval indicates greater precision, while a wider interval indicates less precision. An interval that crosses zero includes no mean difference among the values it covers; an interval entirely below zero supports a negative estimated contrast under that calculation. The AERT 40 mg interval runs from -0.693 to 0.093, and the AERT 80 mg interval runs from -0.974 to -0.227. The latter lies below zero; the former includes zero. The registry record does not report whether these intervals are from the sponsor’s prespecified analysis or describe their adjustment for multiple comparisons.

What does “unadjusted” mean?

Here, unadjusted means the posted p-values are not reported as corrected for the number of comparisons. When multiple hypotheses are tested, the chance of at least one apparently notable result can increase; multiplicity methods may be used to account for that. This is a general statistical point. The registry record does not report a multiplicity procedure, a hierarchy among outcomes, or a planned testing strategy. The posted p-values should therefore be understood as unadjusted rather than as evidence that multiplicity has been addressed.

The registry lists no posted analyses. The comparison calculations are based on the group summary statistics, rather than a model or covariate adjustment documented in the registry. The record does not report the handling of missing outcome data, analysis population definitions, or any imputation method. Those details can affect interpretation, especially if the availability of outcome measurements differs between groups. Without them, the summary statistics cannot establish whether all randomized participants contributed to the endpoint analysis or how incomplete observations were treated.

General statistical explanation: randomization supports causal comparison when allocation is preserved and analysis follows an appropriate plan, but it does not remove the need to understand missingness, adherence, outcome measurement, and analysis choices. The registry’s design labels alone do not answer those questions. The results should be read as group-level estimates and uncertainty based on the information recorded, rather than as a complete reconstruction of the trial’s statistical conduct.

10. Confidence Intervals and Precision

The confidence interval for the AERT 40 mg comparison extends from -0.693 to 0.093 units around a mean difference of -0.3. It includes zero and also covers values on either side of zero. The posted summary statistics therefore leave uncertainty about the direction of the underlying mean contrast. The interval does not establish that the groups had identical outcomes; it indicates that the estimate is not precise enough to exclude zero under the reported calculation.

For AERT 80 mg, the mean difference is -0.6 units, with an interval from -0.974 to -0.227. This interval is entirely below zero. It is compatible with a negative contrast of varying magnitude within that range under the calculation used. The interval’s location describes direction, while its width reflects uncertainty. Neither feature determines whether the difference is clinically important; a threshold for clinical importance is not reported in the registry record.

Confidence intervals are useful because they show a range of plausible values rather than only a point estimate. In general, they reflect both the observed data and the method’s assumptions. They are not guarantees that the true value falls inside the particular interval, and they do not account for analysis choices that are not represented in the calculation. The registry record does not report whether the intervals account for multiple comparisons or other adjustments.

The group standard deviations—1.97, 1.78, and 1.82—provide context about variation in observed change, but they are not the uncertainty intervals for the mean differences. Precision should be judged from the comparison intervals, alongside awareness that details of the original analysis and missing data are not reported. The posted estimates give more precision about the direction of the AERT 80 mg contrast than about the AERT 40 mg contrast, but they do not resolve questions about clinical relevance or the broader outcome profile.

11. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm, not total event counts. They illustrate the sample size that would be required under the stated assumptions for a future comparison, using a chosen fraction of the observed effect. They are not the trial’s original design calculation, and they are not observed power.

The planning assumptions 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 to the stated fraction of the observed effect. The figures are shown separately for planning around 80% power and 90% power. In plain language, those columns describe the number of participants per arm the calculation estimates would be needed to have the specified probability of detecting the target effect if that effect and the assumptions were correct. A higher power target generally requires a larger planned sample under otherwise comparable assumptions.

Rows for 100% of the observed effect use the full observed contrast as the planning target; rows for 75% or 50% of the observed effect represent smaller target differences. Smaller effects need more participants because they are harder to distinguish from random variation. The required figures differ between the AERT 40 mg and AERT 80 mg comparisons because the observed contrasts differ. These are planning illustrations, not evidence that the completed trial had a particular probability of finding a result.

The original design assumptions, including any sponsor sample-size justification, are not reported in the registry record. The planning display therefore should not be used to infer the trial’s planned power, nor should the observed results be described as having a particular observed power.

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 = -0.300; pooled standard deviation σ = 1.897.
2. n = 2 × (1.96 + 0.842)² × 1.897² / -0.300² = 627.5, rounded up to 628 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 <- -0.3                      # observed difference in means
sigma <- 1.89669467870434                     # 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(3290131)
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 = 628 (80% power), 840 (90% power)
 75% of observed effect: n per arm = 1,116 (80% power), 1,494 (90% power)
 50% of observed effect: n per arm = 2,510 (80% power), 3,360 (90% power)
power.t.test: 628.4 
Simulated power at n = 628 per arm: 0.795 (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 lists two outcome measures and no posted analyses. The recorded primary endpoint is the change in the most affected limb’s score at 84 days. The registry record does not report a full statistical analysis plan, any prespecified hierarchy of outcomes, or a procedure for controlling multiplicity across doses or outcomes.

Multiplicity arises when several hypotheses are tested. General statistical explanation: the more comparisons considered, the greater the opportunity for a result to appear notable by chance, so a trial may prespecify how to interpret or adjust related tests. The p-values posted for the dose comparisons are explicitly unadjusted. No multiplicity correction should be assumed, and the registry record does not state that one was applied. The comparison results are therefore best read as separate unadjusted estimates rather than as a family-wise testing conclusion.

Missing-data handling is also not reported in the registry record. The record does not describe how participants without an endpoint measurement were treated, whether observations were imputed, or whether any sensitivity analyses were conducted. The outcome group sizes are listed, but those figures alone do not establish how missing measurements were handled or whether the endpoint analysis followed an intention-to-treat approach.

Analysis populations are not defined in the registry record. It does not specify whether analyses were based on all randomized participants, participants who received treatment, or another population. No covariate adjustment or model details are reported. These omissions limit the ability to assess whether the posted summary comparisons align with the trial’s prespecified analysis or how robust the estimates are to alternate assumptions.

These are reporting limitations, not proof that the trial did or did not use particular methods. The registry record does not provide enough information to characterize the missing-data strategy, analysis populations, or multiplicity control. Interpret the posted estimates and p-values accordingly, without assuming unreported procedures.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

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: multiple sclerosis relapse (1.7%), withdrawal syndrome (0.6%), urosepsis (0.6%), schwannoma (0.6%), somnolence (0.6%).

The registry reports serious adverse events for each study group as counts among those listed at risk. The AERT 40 mg group had 7 participants affected among 179 at risk; the AERT 80 mg group had 6 among 179; and the placebo group had 6 among 178. Deaths were recorded as 0 in each group. These are descriptive counts and do not, by themselves, establish whether a treatment caused an event or whether groups differ reliably.

The registry also reports other adverse events affecting 148 participants in the AERT 40 mg group, 154 in the AERT 80 mg group, and 133 in the placebo group. Serious events and other adverse events are separate categories in the record and should not be combined. The listed serious-event terms include Multiple Sclerosis Relapse at 1.7%, and Withdrawal Syndrome, Urosepsis, Schwannoma, Somnolence, and Trigeminal Neuralgia at 0.6% each.

The record does not provide event-level timing, exposure-adjusted rates, severity details beyond the serious-event category, or a statistical comparison for these safety counts. Counts alone do not measure the full safety profile. They should be read as the adverse-event information recorded, not as proof of causation or as medical advice.

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

The posted results show that the mean score decreased from baseline in all three groups at the 84-day assessment. The mean difference versus placebo was -0.3 units for AERT 40 mg and -0.6 units for AERT 80 mg. The confidence interval for the AERT 40 mg comparison includes zero, while the interval for AERT 80 mg lies below zero. These statements describe the numerical results and their reported precision; they do not, by themselves, establish clinical importance.

The endpoint is a scale-based measure of muscle tone in the most affected limb. It is not a direct measure of every aspect of multiple sclerosis, overall function, quality of life, or a participant’s own assessment. A change on the score should therefore be interpreted within the limits of what the scale measures. The registry record does not report a clinically important difference threshold or additional posted secondary outcome results.

The result also does not explain why scores changed. Changes in the placebo group show that decreases were recorded in that group as well, but the group averages alone cannot distinguish the contribution of study participation, natural fluctuation, measurement, or other influences. Randomization is intended to support a comparative interpretation, but the limited registry reporting does not describe the full analysis plan, missing-data handling, or analysis populations.

Safety counts should be considered alongside the outcome estimates, but the registry’s descriptive counts do not establish causal relationships or provide a complete comparison of risks. Nor do these results address outcomes beyond the recorded timeframe. The posted evidence should not be generalized beyond the study population, endpoint, groups, and timeframe described. No treatment recommendation follows from these statistics; decisions about care require clinical judgment and information beyond this record.

16. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison should be read through both its point estimate and its uncertainty. The mean differences offer a concise measure of how the average score change compared with placebo, while the confidence intervals show how much uncertainty remains around each estimate. The AERT 40 mg interval includes zero; the AERT 80 mg interval does not. That distinction is clearer from the intervals than from comparing the group means alone.

It also demonstrates that statistical evidence and clinical importance are different questions. The registry reports scale-unit estimates and uncertainty, but it does not report a threshold for a clinically meaningful change. A result can be described statistically without claiming that it is important to patients or that it affects broader outcomes.

The trial’s parallel-group, randomized, placebo-controlled design provides a framework for comparison, yet design labels do not supply missing analytic details. The registry lists no analyses and does not report how missing data, analysis populations, or multiple comparisons were handled. These details matter when deciding how directly to interpret unadjusted summary comparisons.

Finally, the planning illustrations show how sample-size needs depend on the target effect and the assumptions used. They are not evidence about the trial’s original planning or its observed power. Distinguishing planning calculations from trial results helps prevent an apparent calculation from being mistaken for a finding about the completed study. Together, these points show why estimates, intervals, analysis transparency, and scope all matter in reading a clinical trial.

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 a full statistical analysis plan, the original sample-size justification, or whether the posted confidence intervals and p-values reflect the sponsor’s prespecified analysis. It lists no analyses. The available group summaries therefore do not permit reconstruction of the full inferential approach.

Details about allocation concealment, masking implementation, adherence, and treatment discontinuation are not reported in the registry record. The official title says double-blind, while the registry masking field says single; the record does not identify who was masked or explain the difference. These details are relevant to assessing how the study was conducted, but should not be inferred from the labels.

Missing-data methods and analysis populations are not reported. The outcome group sizes do not reveal whether all randomized participants contributed usable endpoint data or how any absent measurements were addressed. The registry also does not report baseline score distributions, adjustment for baseline values, or other covariates in the posted comparisons.

The primary endpoint is limited to the most affected limb’s score at 84 days. The registry record does not report a clinical-importance threshold, and the listed secondary outcome measures are not described in the posted results. Safety reporting is descriptive: the record does not provide exposure-adjusted rates, event timing, or inferential comparisons for adverse events. These reporting limits constrain conclusions about the broader clinical picture and should not be interpreted as evidence that unreported methods or outcomes were absent from the trial.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Study to Investigate the Safety and Effectiveness of Arbaclofen Extended-Release Tablets for Patients With MS — NCT03290131.
• Registry record last updated 2022-07-15; status last verified 2022-05.
• Results first posted 2022-07-15.
• 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

This completed Phase 3 study evaluated arbaclofen extended-release tablets against placebo for spasticity in patients with multiple sclerosis. It used randomized, parallel-group allocation and had recorded enrollment of 536. The study dates listed in the registry are 2018-01-28 to 2018-12-03.

The primary outcome was change from baseline in the most affected limb’s Total Numeric-transformed Modified Ashworth Scale score at 84 days. The posted mean differences versus placebo were -0.3 for AERT 40 mg and -0.6 for AERT 80 mg. The AERT 40 mg confidence interval included zero; the AERT 80 mg interval was below zero. The comparisons are unadjusted, and details of the original analysis plan, missing-data handling, and analysis populations are not reported in the registry record.

Serious adverse-event counts and other adverse-event counts are recorded by group, with 0 deaths reported in each group. These are descriptive results and do not establish causation.