Efficacy and Safety of BIA 9-1067 in Idiopathic Parkinson's Disease Patients.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of BIA 9-1067 in Idiopathic Parkinson's Disease Patients. (NCT01227655), a randomized phase 3 study. The primary outcome is Efficacy of 2 BIA 9-1067 (25 mg, and 50 mg) Compared With Placebo, When Administered With the Existing Treatment of L-DOPA Plus a DDCI (DOPA Decarboxylase Inhibitor).

2. Clinical Question

The primary measure is a continuous outcome expressed in minutes. The posted arm results are mean changes from baseline, so a negative value indicates that the measured OFF-time at the endpoint was lower than at baseline on average. The comparisons describe the difference in mean change between each active-treatment arm and placebo; they do not by themselves establish why the groups differed.

The question concerns an add-on treatment in people already receiving levodopa plus a dopa decarboxylase inhibitor, rather than a comparison with no background treatment. The registry record does not report eligibility thresholds, the precise definition of an OFF period used for measurement, or further details of the assessment procedure. Those omissions limit how specifically the population and measurement can be characterized.

3. Trial Design

The registry record describes a completed, randomized, parallel-group Phase 3 clinical study. Its official title identifies it as multicentre, double-blind, and placebo-controlled, while the masking field lists triple masking. The record does not explain which roles were included in the masking designation, so the terms cannot be reconciled further from the available information.

The trial enrolled 427 participants according to the registry record. The posted outcome groups are BIA 9-1067 25 mg, BIA 9-1067 50 mg, and placebo. Both active regimens are described as once daily and were administered with existing levodopa plus a dopa decarboxylase inhibitor treatment. The placebo group served as the comparator while participants remained on existing treatment.

The record gives a study start of 2011-03 and a primary completion of 2012-07, and lists the status as completed. These dates describe the registry record and do not provide the dates of individual recruitment or follow-up. The primary outcome timeframe is 14-15 weeks.

The posted arm sizes for the primary outcome are 125 for BIA 9-1067 25 mg, 147 for BIA 9-1067 50 mg, and 135 for placebo. The safety record gives different at-risk counts for some groups, so outcome-specific denominators should not be assumed to be interchangeable. Reasons for differences in the reported counts, including any exclusions or missing measurements, are not reported in the registry record.

The registry record does not report the sequence-generation procedure, allocation concealment, details of sites, or the trial’s original sample-size justification. It also does not describe protocol deviations or the analysis population used for the primary comparison. Randomization and masking are important design features, but the record alone does not establish how successfully they were implemented.

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

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: Four subjects from safety set who had no post-baseline OFF-time efficacy assessment were excluded from the full analysis set: three (1.9%) subjects in the OPC 50 mg group and one (0.7%) subject in the placebo group.

6. Primary Endpoint

The registry lists the primary outcome as Efficacy of 2 BIA 9-1067 (25 mg, and 50 mg) Compared With Placebo, When Administered With the Existing Treatment of L-DOPA Plus a DDCI (DOPA Decarboxylase Inhibitor), measured over 14-15 weeks.

The primary endpoint is the change from baseline in absolute OFF-time, measured in minutes, at the end of the double-blind period. OFF-time refers to periods of motor symptoms when the usual benefit of levodopa has worn off; the registry record frames the study population as having end-of-dose motor fluctuations. The measure is continuous, and the posted results describe its mean and standard deviation in each group.

Change from baseline focuses the comparison on how the measured OFF-time shifted during the study, rather than on endpoint values alone. A negative mean change represents a reduction from baseline on average; it should not be read as a percentage or as a result for every participant. The spread values are standard deviations, which describe variation among participants and are not confidence intervals for the group means.

Using a continuous endpoint can retain information about the extent of change rather than reducing outcomes to a simple event/no-event classification. The primary description does not state how daily OFF-time was recorded, whether assessments were based on diaries, or how multiple observations were combined. Those measurement details are not reported in the registry record, so interpretation should remain tied to the endpoint as defined: mean change in absolute OFF-time at 14-15 weeks.

The registry record also lists secondary measures including UPDRS sections, the Parkinson’s Disease Sleep Scale, and the Non-motor Symptoms Scale. Their presence does not change the status of OFF-time change as the stated primary efficacy variable.

7. Results Posted in the Registry

• BIA 9-1067 25 mg: mean -102.80 versus -64.50 on control; difference -38.30 (95% CI -76.61 to 0.01), standardized effect d = -0.24; z-test p = 0.050.
• BIA 9-1067 50 mg: mean -124.00 versus -64.50 on control; difference -59.50 (95% CI -98.45 to -20.55), standardized effect d = -0.35; z-test p = 0.003.

8. Statistical Interpretation

9. Statistical Methods Explained

The following explanations are general statistical context for interpreting the posted results. The registry record does not report enough detail to identify the full prespecified statistical methods used by the trial.

What does a group mean describe?

A mean is the arithmetic average of the observed values in a group. Here, the continuous measure is change from baseline in OFF-time, in minutes. A mean of -102.8 minutes describes the average change among the participants contributing to that posted result; it does not mean that every participant changed by that amount. Individual changes can be more or less pronounced, and some may move in a different direction.

The standard deviation describes the spread of individual values around the group mean. The posted standard deviations are 159.4 for BIA 9-1067 25 mg, 178.2 for 50 mg, and 155.3 for placebo. A standard deviation is not the standard error of the mean and is not a confidence interval. A relatively broad spread indicates that individual outcomes vary substantially around the average, but it does not identify the shape of that variation.

What is a mean difference?

A mean difference compares the average change in one group with the average change in another. The recalculated comparisons use placebo as the reference. The estimate of -38.3 minutes for 25 mg and -59.5 minutes for 50 mg is negative because the active-treatment group’s mean change is more negative than the placebo group’s mean change. These are unadjusted comparisons based on posted summary statistics, rather than results from a documented adjusted model.

In general, a difference in means is expressed in the same units as the endpoint, so these estimates are in minutes. Its interpretation depends on the endpoint’s direction and definition. Since the endpoint is change from baseline in absolute OFF-time, a more negative change corresponds to a larger mean reduction from baseline. This description does not state that an individual participant will experience the group-average change.

What does a confidence interval communicate?

General explanation: a confidence interval describes statistical uncertainty around an estimate under the assumptions of the method used to construct it. It gives a range of values compatible with the observed data and those assumptions; it is not a guarantee that the true effect lies within the interval, nor does it describe the range of individual participant outcomes.

The 25 mg comparison has an interval that extends from a negative value to just above zero. That means the posted estimate is imprecise enough that the interval includes zero difference. The 50 mg interval runs from -98.45 to -20.55, so its endpoints are both negative. Neither interval removes uncertainty about the magnitude of the difference. The registry record does not report the exact confidence-interval procedure used in the trial’s own analysis.

How should a p-value be read?

General explanation: a p-value evaluates how unusual data at least as incompatible with a specified null model would be, if that model and the statistical assumptions were correct. It is not the probability that the null hypothesis is true, the probability that a result happened by chance, or a measure of clinical importance.

The recalculated unadjusted p-values are 0.05 for 25 mg and 0.003 for 50 mg. The first is at the conventional 0.05 reference threshold, while the second is smaller. These values should be considered alongside the effect estimates and intervals, not used alone. The record does not report whether the sponsor used these same tests or whether its primary analysis adjusted for other factors.

Why can variation affect precision?

General explanation: when observations vary widely within groups, a group mean is estimated less precisely than it would be with less variation, all else equal. The posted standard deviations are large relative to the magnitudes of the group mean changes, illustrating why group averages should not be mistaken for uniform participant responses. Precision also depends on the amount of information available for the comparison and on analysis assumptions.

Over-dispersion is a general issue when outcome variability is greater than a statistical model assumes. If present and ignored, uncertainty can be understated, making intervals too narrow and p-values too small. The registry record does not report an over-dispersion assessment, and it would not be appropriate to infer that a particular model or correction was used for this continuous endpoint.

What do the recalculations represent?

The posted comparison estimates are labeled as unadjusted recalculations from the group summary statistics. They are useful as a transparent descriptive comparison, but they do not reproduce an undocumented sponsor model. They do not account for baseline differences, missing measurements, covariates, or multiplicity unless those features are incorporated in the original analysis; whether they were incorporated is not reported in the registry record. The record does not flag a discrepancy, but the absence of a flagged discrepancy does not supply details of the original analysis.

The registry record lists three secondary outcome measures, but numerical secondary results and the associated analysis methods are not available in the registry record for these explanatory notes. There is also no reported account of how missing primary outcome values were handled. That means the posted group summaries can be interpreted as reported, but the consequences of missingness or analysis-population choices cannot be assessed from the registry record alone.

10. Confidence Intervals and Precision

The confidence intervals indicate that the estimated mean differences are not exact values. For the 25 mg comparison, the interval extends from -76.61 minutes to just above zero. Because it includes zero, the posted summaries do not rule out no difference in mean change under the unadjusted comparison. They are also compatible with a larger negative difference. The interval therefore conveys uncertainty in both the direction and size of the comparison near zero.

For the 50 mg comparison, the interval ranges from -98.45 to -20.55 minutes. Both endpoints are negative, while the width of the interval shows that a range of mean differences remains compatible with the data and calculation. The estimate of -59.5 minutes is the central point estimate, not a guarantee that the true difference is exactly that amount.

Confidence intervals concern uncertainty in a group-level estimate, not the spread of participant-level outcomes. The standard deviations of 159.4, 178.2, and 155.3 minutes describe variability within the respective outcome groups. They answer a different question from the intervals around the differences. The registry record does not report the confidence-interval method for the trial’s own analysis, and these intervals accompany unadjusted recalculations.

Precision should be considered alongside study design, analysis choices, and completeness of outcome data. The registry record does not report missing-data handling, the primary analysis population, or multiplicity adjustment. Those omissions make it difficult to determine whether the displayed uncertainty captures all relevant sources of uncertainty in the trial’s planned inference.

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 estimates are participants per arm for 80% power and participants per arm for 90% power. They are not total event counts. The calculation uses the observed effect as a reference, with scenarios targeting 100%, 75%, or 50% of that observed effect. For example, at 100% of the observed effect, the estimates for the 25 mg comparison are 265 participants per arm for 80% power and 355 for 90% power; for the 50 mg comparison, they are 125 and 167 participants per arm, respectively.

In plain language, the 80% power column gives the estimated participants per arm under the stated assumptions for an 80% chance of detecting the target difference if that difference is the true effect. The 90% column applies a higher probability of detection and therefore gives a larger planning estimate in the displayed scenarios. These are planning illustrations based on the observed effect, not observed power.

The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is the stated fraction of the observed effect. Planning for 75% or 50% of the observed effect requires more participants than planning for the full observed effect, because smaller differences are harder to distinguish from random variation. The trial’s original sample-size assumptions and justification are not reported in the registry record. The planning calculations should therefore be kept separate from the design decisions actually made for the trial.

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 = -38.300; pooled standard deviation σ = 157.319.
2. n = 2 × (1.96 + 0.842)² × 157.319² / -38.300² = 264.9, rounded up to 265 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 <- -38.3                      # observed difference in means
sigma <- 157.319266656186                     # 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(1227655)
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 = 265 (80% power), 355 (90% power)
 75% of observed effect: n per arm = 471 (80% power), 631 (90% power)
 50% of observed effect: n per arm = 1,060 (80% power), 1,419 (90% power)
power.t.test: 265.8 
Simulated power at n = 265 per arm: 0.800 (20000 replicates)

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

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists the primary endpoint as change from baseline in absolute OFF-time and lists three secondary outcome measures: UPDRS sections, the Parkinson’s Disease Sleep Scale, and the Non-motor Symptoms Scale. Testing multiple outcomes or multiple comparisons can increase the chance of finding an apparently unusual result when all null hypotheses are true. General statistical practice may address this through prespecified testing hierarchies or adjustment procedures, but whether either approach was used in this trial is not reported in the registry record.

The two posted recalculated comparisons each compare an active-treatment group with placebo. Their p-values are unadjusted. No multiplicity adjustment is reported for these recalculations, and the registry record does not state whether the sponsor’s analysis adjusted for testing both doses or for secondary outcomes. The unadjusted values should therefore not be treated as evidence that multiplicity was controlled in the trial’s formal analysis.

Missing-data handling is also not reported in the registry record. The record does not say how participants without an endpoint measurement were handled, whether repeated observations were used, or what assumptions were made about missing outcomes. The primary outcome group sizes are 125, 147, and 135, while the safety at-risk counts are 125, 150, and 136. These different denominators are reported for different outcomes; the reason for the differences is not given.

The primary analysis population is not reported in the registry record. It is therefore not possible to determine from these details whether the posted result represents all randomized participants, participants who completed treatment, or another defined group. The registry record also does not describe covariate adjustment, sensitivity analyses, or an interim analysis plan. These missing details limit assessment of how robust the original inferential analysis was to analysis choices and incomplete data.

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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: osteoarthritis (0.8%), dyskinesia (0.8%), renal failure acute (0.8%), urinary retention (0.8%), pneumonia (0.7%).

The registry safety record lists serious adverse events affecting 4 of the 125 participants at risk in the BIA 9-1067 25 mg group, 9 of the 150 at risk in the 50 mg group, and 5 of the 136 at risk in the placebo group. These are posted counts and denominators; they should be read as group-specific safety information, not as a direct estimate of comparative risk adjusted for differences between groups.

Serious-event terms listed among the most frequent include osteoarthritis, dyskinesia, acute renal failure, and urinary retention, each reported as 0.8%. Pneumonia and femoral neck fracture are each reported as 0.7%. The record does not provide further context about timing, recurrence, severity within each term, or how events were judged to relate to treatment. The percentages should not be added together because the record does not establish whether participants could contribute to more than one event term.

Deaths are not reported in the registry record. Serious adverse event counts describe a different outcome from the primary OFF-time measure and do not, on their own, establish a causal relationship with an intervention.

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

The posted primary results describe mean decreases in OFF-time from baseline in all three groups. The active-treatment groups had more negative mean changes than placebo, with unadjusted recalculated mean differences of -38.3 minutes for 25 mg and -59.5 minutes for 50 mg. The confidence interval for the 25 mg comparison includes zero, while the interval for the 50 mg comparison is entirely negative. These summaries describe group averages and uncertainty; they do not predict the response of a particular person.

The results do not by themselves establish a direct difference between the two active doses. Each recalculated comparison uses placebo as its reference, and no direct active-dose comparison is reported in the registry record. Nor can the summaries establish whether the differences were due to treatment rather than chance, bias, measurement issues, missingness, or other factors. Randomization is a design strength, but complete assessment also depends on trial conduct and analysis details not reported in the registry record.

The posted safety information gives serious adverse event counts and selected event terms. It does not include enough detail to determine causality or to characterize every possible safety outcome. Death counts are not reported in the registry record. Efficacy and safety should therefore be considered as separate parts of the posted information, without treating either as a complete account of participant experience.

Other outcome measures are listed, but their numerical results are not reported in the registry record for this analysis. The record also does not describe missing-data methods, the primary analysis population, or multiplicity procedures. Those omissions prevent a full reconstruction of the trial’s formal statistical inference. The meaning of the results is consequently limited to the reported endpoint summaries, their unadjusted recalculations, and the recorded safety counts.

Nothing in these statistical results is individual medical advice. They describe trial-level data and do not determine whether any particular treatment is appropriate for an individual.

16. Why This Trial Matters Statistically

This trial illustrates how the same continuous endpoint can be described at several levels: group mean change, variation within groups, a between-group mean difference, and uncertainty around that difference. These quantities answer different questions. The group means describe average changes, standard deviations describe variation among participants, and confidence intervals address uncertainty in the estimated comparisons.

The results also demonstrate why an estimate should not be reduced to a p-value. The 25 mg comparison has an unadjusted p-value of 0.05 and an interval that reaches across zero; the 50 mg comparison has a smaller unadjusted p-value and an interval wholly below zero. Looking at the estimates and intervals together makes the difference in precision and the range of compatible values more apparent than a threshold alone.

There is also a useful distinction between the registry’s trial enrollment and the outcome-specific denominators. Enrollment is listed as 427, while efficacy and safety groups have their own reported counts. Those figures describe different aspects of the record and should not be combined or assumed to represent the same analysis population. Reasons for the denominator differences are not reported in the registry record.

Finally, the registry record demonstrates the limits of interpreting abbreviated statistical reporting. A randomized, masked design and posted outcome summaries are informative, but the missing details on analysis populations, missing data, multiplicity, and the original sample-size rationale constrain assessment of the formal analysis. Transparent reporting of those elements is important for understanding how robust a trial’s conclusions are.

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 detailed eligibility criteria, the measurement procedure for OFF-time, or how the primary endpoint was constructed from observations. It also does not explain the masking designation, despite identifying the study as double-blind in its official title and listing triple masking in the design field. The extent to which participants, investigators, outcome assessors, or other roles were masked cannot be determined from the record.

The original sample-size justification, planned statistical analysis, and analysis population are not reported in the registry record. Neither are missing-data procedures, covariate adjustments, multiplicity handling, sensitivity analyses, or interim analyses. The comparisons discussed are unadjusted recalculations from posted summary statistics, not a reconstruction of an undocumented sponsor analysis.

Outcome-specific denominators differ between the primary efficacy and safety results, and the record does not explain why. The numerical results for the listed secondary measures are not reported in the registry record for this analysis. Safety information is limited to recorded serious adverse event counts and selected event terms; deaths are not reported in the registry record. These gaps limit interpretation of the completeness of the outcome reporting and prevent assessment of the full balance of measured benefits and harms.

The primary endpoint is a mean change, which does not show the spread of individual changes beyond the group standard deviations or the proportion of participants with particular responses. The registry record also does not provide a direct comparison between active doses. The results therefore cannot resolve whether one dose differs from the other or how the mean differences map to individual outcomes.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Efficacy and Safety of BIA 9-1067 in Idiopathic Parkinson's Disease Patients. — NCT01227655.
• Registry record last updated 2015-10-19; status last verified 2015-09.
• Results first posted 2015-01-13.
• 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, randomized, parallel-group Phase 3 study of BIA 9-1067 added to levodopa plus a dopa decarboxylase inhibitor in people with Parkinson’s disease and end-of-dose motor fluctuations. The active regimens were 25 mg once daily and 50 mg once daily, compared with placebo. Enrollment is listed as 427; the primary outcome timeframe is 14-15 weeks.

The primary endpoint was change from baseline in absolute OFF-time. Posted mean changes were -102.8 minutes for 25 mg, -124.0 minutes for 50 mg, and -64.5 minutes for placebo. Unadjusted recalculated mean differences versus placebo were -38.3 minutes and -59.5 minutes, respectively. The 25 mg interval includes zero; the 50 mg interval is entirely negative.

The record lists three secondary outcome measures and reports serious adverse event counts by group. Further details on analysis methods, missing data, multiplicity, and the primary analysis population are not reported in the registry record.