Study of the Safety and Efficacy of OPC-34712 as Adjunctive Therapy in the Treatment of Patients With Major Depressive Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of the Safety and Efficacy of OPC-34712 as Adjunctive Therapy in the Treatment of Patients With Major Depressive Disorder (NCT00797966), a randomized phase 2 study. The primary outcome is Change From the End of Phase A (Week 8 Visit) to the End of Phase B (Week 14 Visit) in Montgomery Asberg Depression Rating Scale (MADRS) Total Score..

2. Clinical Question

The question concerns change over the Phase B treatment period among participants meeting the stated analysis-population criteria. It is not a comparison of OPC-34712 alone with no treatment: both intervention and comparator groups received the assigned open-label antidepressant treatment. The registry record identifies the trial as Phase 2 and lists an enrollment of 850. It does not report the clinical or demographic characteristics of the participants in the results, nor does it report the criteria used to define an incomplete response beyond the prospective trial description.

3. Trial Design

The registry describes a Phase 2, randomized, triple-masked, parallel-group, placebo-controlled trial. Participants were assigned to OPC-34712 or placebo as adjunctive treatment to an assigned open-label marketed antidepressant treatment. The three OPC-34712 groups were a 0.15 mg fixed dose, a 0.5 ± 0.25 mg low dose, and a 1.5 ± 0.5 mg high dose. The comparator was placebo, also given with the assigned antidepressant treatment.

The treatment sequence described in the record begins with a prospective eight-week trial of the assigned open-label antidepressant. Participants demonstrating an incomplete response proceeded to the adjunctive-treatment comparison. The primary outcome measures change from the end of Phase A at Week 8 to the end of Phase B at Week 14. The record gives the trial start as 2009-05 and primary completion as 2010-06, and lists the study status as completed.

Enrollment is recorded as 850, but enrollment should not be read as the analysis count for every outcome. The primary outcome results list their own group counts, and the adverse-event results use separate at-risk counts. Details of the randomization procedure, allocation concealment, masking roles, dose adjustment rules, and original sample-size justification are not reported in the registry record. The record also does not state whether the open-label antidepressant assignment was chosen by participants, investigators, or another process. Those design details cannot be inferred from the broad randomized and triple-masked labels.

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: The study consisted of a 28-day Screening period, an 8-Week single-blind placebo + Antidepressant therapy (ADT) prospective Phase A. A 6-week double-blind Randomization Phase (Phase B) or single-blind Phase A+ for those participants who did not meet criteria for randomization and a Follow-up of 30 (+2) days after the last dose of study medication.

Phase A

Reasons for not completing this period:

Phase B

Reasons for not completing this period:

Phase A+

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) dataset consisted of all randomized participants who had an End of Phase A value and at least one post-randomization efficacy evaluation for MADRS Total Score in Phase B. The Last Observation Carried Forward (LOCF) method was used to impute missing data.

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 the End of Phase A (Week 8 Visit) to the End of Phase B (Week 14 Visit) in Montgomery Asberg Depression Rating Scale (MADRS) Total Score., measured over Week 8 to Week 14.

The primary endpoint is change in Montgomery-Åsberg Depression Rating Scale (MADRS) total score from Week 8, the end of Phase A, to Week 14, the end of Phase B. The registry describes MADRS as an efficacy assessment of a participant’s level of depression. It comprises 10 items, each rated from 0 to 6; 0 is the best rating and 6 the worst. The total is the sum of the item ratings, with possible totals from 0 to 60. Thus, a negative change indicates a lower score at the later assessment than at the Week 8 starting point.

The score was considered unevaluable if fewer than 8 of the 10 items were recorded. If 8 or 9 items were recorded, the total was calculated from the mean of recorded items, multiplied by 10, and rounded to the first decimal place. The reported endpoint is a least-squares mean, with standard error as its dispersion measure. Least-squares means are model-based group estimates; they are used to summarize the outcome while accounting for terms in the fitted analysis model. The registry identifies ANCOVA for the reported comparisons, but does not report the model’s covariates or full specification.

This endpoint is used to compare change in symptom-scale scores over the adjunctive-treatment period. It is a scale-based outcome, not by itself a complete account of functioning, quality of life, or adverse effects.

7. Results Posted in the Registry

• OPC-34712 0.15 mg Fixed Dose: mean -6.62 versus -6.09 on control; difference -0.53 (95% CI -2.93 to 1.87), standardized effect d = -0.07; z-test p = 0.665.
• OPC-34712 0.5 ± 0.25 mg Low Dose: mean -6.46 versus -6.09 on control; difference -0.37 (95% CI -2.38 to 1.64), standardized effect d = -0.05; z-test p = 0.718.
• OPC-34712 1.5 ± 0.5 mg High Dose: mean -8.23 versus -6.09 on control; difference -2.14 (95% CI -4.16 to -0.12), standardized effect d = -0.27; z-test p = 0.038.

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 End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Mean Clinical Global Impression - Severity of Illness Scale (CGI-S) Score.

• 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:
  • OPC-34712 0.15 mg Fixed Dose: mean -0.83 versus -0.71 on control; difference -0.12 (95% CI -0.43 to 0.19), standardized effect d = -0.12; z-test p = 0.448.
  • OPC-34712 0.5 ± 0.25 mg Low Dose: mean -0.81 versus -0.71 on control; difference -0.10 (95% CI -0.36 to 0.16), standardized effect d = -0.10; z-test p = 0.457.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.15 mg Fixed Dose and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.5 ± 0.25 mg Low Dose and the control. The interval gives the range of effects that remain compatible with the data.

Change From End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Mean Quality of Life, Enjoyment, and Satisfaction Questionnaire - Short Form (QLES-Q-SF) Subscale Score - the Overall General Subscore (Sum of First 14 Items).

• 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:
  • OPC-34712 0.15 mg Fixed Dose: mean 7.60 versus 5.92 on control; difference 1.68 (95% CI -2.75 to 6.11), standardized effect d = 0.12; z-test p = 0.457.
  • OPC-34712 0.5 ± 0.25 mg Low Dose: mean 6.53 versus 5.92 on control; difference 0.61 (95% CI -3.16 to 4.38), standardized effect d = 0.04; z-test p = 0.751.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.15 mg Fixed Dose and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.5 ± 0.25 mg Low Dose and the control. The interval gives the range of effects that remain compatible with the data.

Change From End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Sheehan Disability Scale (SDS) Mean Score (the Mean of 3 Individual Item Scores).

• 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:
  • OPC-34712 0.15 mg Fixed Dose: mean -0.84 versus -0.61 on control; difference -0.23 (95% CI -0.89 to 0.43), standardized effect d = -0.11; z-test p = 0.494.
  • OPC-34712 0.5 ± 0.25 mg Low Dose: mean -0.80 versus -0.61 on control; difference -0.19 (95% CI -0.76 to 0.38), standardized effect d = -0.09; z-test p = 0.512.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.15 mg Fixed Dose and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.5 ± 0.25 mg Low Dose and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several domains, including illness severity, quality of life and satisfaction, disability, and additional symptom assessments over Phase B. The pattern in the secondary results available for review is not uniform across domains or dose groups. These outcomes can add context to the primary symptom-scale result, but they do not all measure the same construct, and their scales should not be treated as interchangeable.

The registry record lists 13 posted secondary outcomes, but detailed results are not reported for every listed outcome in the information presented. The available secondary comparisons should therefore not be taken as a complete account of every planned or registered secondary measure. They also should not be treated as independent confirmation of the primary result: secondary endpoints may be correlated with one another and with the primary outcome.

Unadjusted secondary results carry less inferential weight when many outcomes and comparisons are examined. In general, the more comparisons considered, the greater the chance that at least one result will appear unusual by chance, even when there is no underlying difference for that outcome. A multiplicity procedure can address this issue, but the registry record does not report whether such a procedure was applied to the secondary analyses. The reported secondary findings are best read as supplementary evidence and interpreted alongside their estimates, uncertainty, analysis populations, and the primary endpoint, rather than as a separate set of definitive tests.

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

• Change From End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Mean Clinical Global Impression - Severity of Illness Scale (CGI-S) Score.
• Change From End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Mean Quality of Life, Enjoyment, and Satisfaction Questionnaire - Short Form (QLES-Q-SF) Subscale Score - the Overall General Subscore (Sum of First 14 Items).
• Change From End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Sheehan Disability Scale (SDS) Mean Score (the Mean of 3 Individual Item Scores).
• Change From End of Phase A (Week 8 Visit) in MADRS Total Score for Every Study Week Visit in Phase B Other Than the Week 14 Visit.
• Change From End of Phase A (Week 8 Visit) in Mean CGI-S Score for Every Study Week Visit in Phase B Other Than the Week 14 Visit.
• Change From End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Mean Q-LES-Q-SF Item 15 Score (Satisfaction With Medication).
• Change From End of Phase A (Week 8 Visit) to End of Phase B (Week 14 Visit) in Mean Q-LES-Q-SF Item 16 Score (Overall Life Satisfaction).
• Change From End of Phase A (Week 8 Visit) for Every Study Week Visit in Phase B in Inventory of Depressive Symptomatology (Self-Report) (IDS-SR) Total Score.

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: MADRS total score is numerical, and the record reports a group mean change rather than an event count. The group summaries are least-squares means with standard errors. A least-squares mean is an estimate from a statistical model, rather than simply the arithmetic mean of the observed scores. The registry names ANCOVA for the sponsor-reported comparisons. In general, analysis of covariance compares groups while accounting for selected covariates; which covariates were used, and how the model was otherwise specified, are not reported in the registry record.

What the mean difference represents

The mean difference compares the estimated change in an OPC-34712 group with the estimated change in the placebo group. Negative differences in this endpoint indicate a more negative change in MADRS score in the OPC-34712 group than in placebo. The difference should be read alongside its confidence interval: an estimate is a point summary, while the interval describes a range of values compatible with the data and analysis, subject to the model assumptions.

The results also include simple unadjusted recalculations based on posted summary statistics. Those calculations are not a substitute for ANCOVA. Their pooled standard deviations are marked as approximate, and the recalculated p-values are unadjusted. The registry record flags a discrepancy, so the sponsor’s own analysis takes precedence over those recalculations. This distinction is important: an unadjusted calculation from summary information may not reproduce an analysis that uses participant-level data and model adjustments.

What a p-value can and cannot say

In general, a p-value describes how unusual results at least as incompatible with a specified null hypothesis would be under that hypothesis and the analysis assumptions. It does not measure the probability that the null hypothesis is true, the size or importance of an effect, or the chance that the finding will repeat. The recalculated p-values are unadjusted, so they do not account for the existence of multiple dose comparisons or multiple outcomes. The trial’s own multiplicity procedure is not reported in the registry record.

For continuous outcomes, the mean difference and its confidence interval are usually more informative about direction and precision than a p-value alone. A confidence interval spanning zero is compatible with no difference as well as differences in either direction. An interval that excludes zero indicates that zero is outside the range of values covered by that interval under the reported analysis; it does not resolve issues such as multiplicity, model assumptions, or clinical importance.

Analysis population and missing outcomes

The primary analysis population is defined as all randomized participants who had an End of Phase A value and at least one post-randomization MADRS efficacy evaluation in Phase B. The record says that the last observation carried forward (LOCF) method was used to impute missing data. In general, LOCF fills a missing later measurement with the participant’s last observed measurement. This approach can be convenient, but it treats an earlier value as if it represents the later outcome and does not itself account for uncertainty in the imputation. The registry record does not report the amount or pattern of missingness, reasons for missing observations, sensitivity analyses, or alternative missing-data models.

Other outcomes and analyses may have different analysis counts. The displayed group counts for the primary endpoint, secondary endpoints, and adverse-event summaries are not interchangeable. The record does not report complete analysis-population definitions for every secondary endpoint. Consequently, interpretation should stay tied to the population and outcome specified for each result rather than treating every reported count as if it came from the same analysis set.

11. Confidence Intervals and Precision

For the primary endpoint, the sponsor-reported 95% confidence interval for the 0.15 mg comparison runs from -2.87 to 1.81, and the interval for the low-dose comparison runs from -2.30 to 1.55. Both include zero. Each therefore leaves room for no difference and for differences in either direction, within the range covered by that analysis. The high-dose comparison has a reported 95% confidence interval from -4.08 to -0.21. It lies entirely below zero, indicating a more negative estimated change in that group than in placebo under the reported ANCOVA.

In general, a confidence interval expresses statistical uncertainty around an estimate under the analysis assumptions. It is not a range containing every participant’s response, and it does not say that values outside it are impossible. Its width conveys precision: wider intervals allow a broader range of underlying effects, while narrower intervals allow a more limited range. The interval also does not establish whether an effect is important to participants; that requires context beyond the interval itself.

Precision should be considered with the analysis design and the endpoint’s measurement process. The primary outcome is a least-squares mean change, and its reported uncertainty comes from ANCOVA. The registry record does not give the model details needed to independently assess all assumptions behind those intervals. The recalculated intervals are based on summary statistics and are approximate, and the record flags a discrepancy. For that reason, the sponsor-reported intervals are the primary reference for interpretation. Neither interval width nor exclusion of zero addresses unreported multiplicity procedures, missing-data sensitivity, or the generalizability of results beyond the analysis population.

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 are participants per arm for this continuous outcome, not total events. They are illustrations based on the observed effect and do not describe the trial’s original design calculation. 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 original design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the assumed effect under the stated planning assumptions, if that effect and assumptions were correct. The 90% power column represents the corresponding planning target for a 90% chance. These are design-planning quantities, not guarantees about the results of a future study. This is not observed power.

Rows using a smaller fraction of the observed effect require more participants per arm than rows using a larger fraction. That follows from a general statistical principle: smaller differences are harder to distinguish from random variation, so more information is needed to estimate them precisely. These figures are especially sensitive to the observed effect used as their starting point. If the observed effect is unstable or differs from the effect expected in a future study, the planning figures may not provide a reliable design target. They also do not incorporate any trial-specific considerations beyond the stated assumptions, because additional original design assumptions are not reported in the registry record.

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 = -0.530; pooled standard deviation σ = 7.989.
2. n = 2 × (1.96 + 0.842)² × 7.989² / -0.530² = 3,566.8, rounded up to 3,567 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 <- -0.53                      # observed difference in means
sigma <- 7.98908469059807                     # 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(797966)
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 = 3,567 (80% power), 4,775 (90% power)
 75% of observed effect: n per arm = 6,341 (80% power), 8,489 (90% power)
 50% of observed effect: n per arm = 14,268 (80% power), 19,100 (90% power)
power.t.test: 3567.8 
Simulated power at n = 3,567 per arm: 0.801 (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 registry reports ANCOVA comparisons from the sponsor as well as simple unadjusted recalculations from posted summary statistics. For the primary endpoint, the sponsor reported mean differences of -0.53, -0.37, and -2.14 for the 0.15 mg, low-dose, and high-dose comparisons. The simple recalculations give the same displayed point estimates, but the confidence limits and p-values differ somewhat. For example, the reported sponsor confidence interval for the high-dose comparison is -4.08 to -0.21; the recalculation uses an approximate pooled standard deviation derived from summary information.

The discrepancy flag means the sponsor’s own analysis takes precedence. The ANCOVA uses the trial analysis, whereas the recalculation is unadjusted and cannot reproduce all details of a participant-level model from summary values. The recalculated p-values of 0.665, 0.718, and 0.038 should therefore be treated as checks based on the posted summaries, not as replacements for sponsor results. This distinction does not imply that either analysis is automatically free of limitations: the registry record does not report the full ANCOVA specification or the procedures for handling multiplicity.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists three OPC-34712 dose comparisons against placebo for the primary endpoint, alongside multiple secondary outcomes and 87 analyses overall. Considering several comparisons raises a general multiplicity concern: the more hypotheses examined, the greater the opportunity for a nominally unusual result to occur by chance. This does not make each reported estimate invalid, but it affects how a single result should be interpreted in relation to the full set of analyses.

The registry record does not report a prespecified testing hierarchy, family-wise error procedure, false-discovery procedure, or another adjustment for multiplicity. It also does not report whether analyses were designated as confirmatory or exploratory, or whether interim analyses were conducted. Those details should not be inferred from the presence of reported p-values. Without a stated multiplicity strategy, the unadjusted recalculations should be read cautiously, especially when comparing several doses and outcomes.

Missing-data handling is described for the primary endpoint: LOCF was used in the stated ITT dataset. In general, this method carries a participant’s last observed measurement forward when a later measurement is missing. Its use does not remove uncertainty about the unobserved outcome, and the record does not report missingness by group, reasons for missing data, or sensitivity analyses assessing other assumptions. The record also does not report whether the same imputation method was used for every secondary outcome.

The primary analysis population was not simply all randomized participants: it required an End of Phase A value and at least one post-randomization Phase B MADRS efficacy evaluation. This definition and LOCF procedure are important to interpreting the primary result. The registry record does not provide further detail on the number excluded by this criterion or the consequences of missing outcome data for the estimates.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: radius fracture (0.8%), ulna fracture (0.8%), chronic obstructive pulmonary disease (0.8%), renal mass (0.8%).

The safety results list serious adverse events for participants at risk in each group. The reported number affected was 0 in the 0.15 mg group among 62 at risk, 0 in the low-dose group among 120 at risk, 2 in the high-dose group among 121 at risk, and 1 in the placebo group among 126 at risk. These are group-specific counts and should be read with their respective at-risk counts. They are not measures of the primary efficacy endpoint.

The registry also lists Radius fracture, Ulna fracture, Chronic obstructive pulmonary disease, and Renal mass among the top serious adverse-event terms, each at 0.8%. These terms and percentages describe the entries identified in the record; they do not establish a causal relationship to study treatment. Death counts are not reported in the registry record.

Adverse-event counts are descriptive. The figures alone do not establish comparative risk, account for event timing or duration, or show whether an event was judged related to treatment. The registry record does not provide enough information to infer causation or to characterize the full safety profile. Safety interpretation should remain separate from conclusions about the primary score change.

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

The primary outcome describes change in MADRS total score between Week 8 and Week 14 among the stated analysis population, comparing three adjunctive OPC-34712 dose groups with placebo alongside the assigned antidepressant treatment. The sponsor-reported estimate and confidence interval for the high-dose comparison indicate a more negative change than with placebo within this outcome and period. The sponsor-reported intervals for the fixed and low-dose comparisons include zero, leaving uncertainty about the direction of those differences.

This finding does not establish the effect of OPC-34712 used alone, because the trial compared adjunctive treatment against placebo adjunctive treatment while participants were on an assigned open-label antidepressant. It does not establish effects beyond the Week 14 endpoint, or on outcomes not measured or reported. Nor does a difference on MADRS alone describe every aspect of illness, functioning, quality of life, or safety.

In general, statistical evidence about a group average does not predict an individual participant’s outcome. A mean difference does not show how responses vary among participants, and the registry record does not provide individual-level response distributions in the results described. The confidence interval reflects uncertainty under the sponsor’s analysis, not a guarantee about the size or direction of an effect in other populations or settings.

The interpretation is also bounded by what the registry record reports. It does not give a multiplicity adjustment, complete ANCOVA specification, or sensitivity analyses for missing data. It reports LOCF for the primary endpoint, and its analysis population requires an End of Phase A measurement and at least one post-randomization efficacy evaluation. The results should therefore be understood as estimates from this study’s stated endpoint, population, treatment context, and analysis, rather than as a universal conclusion.

18. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison should be interpreted through the between-group estimate and its uncertainty, not only through the direction of change within each group. All four groups had negative mean MADRS changes, but that fact alone does not distinguish adjunctive OPC-34712 from adjunctive placebo. The estimated difference against placebo is the relevant contrast for the primary question.

It also shows why multiple dose comparisons require care. The reported confidence intervals differ across dose groups, and the registry record does not report a multiplicity strategy. A result that stands out among several comparisons can be informative, but its evidential weight depends on the analysis plan and the complete set of tests, not only on a single estimate or p-value.

The record also makes visible a distinction between the sponsor’s ANCOVA and a simple recalculation from summary statistics. The point estimates align, while some uncertainty measures differ. Understanding what each analysis represents helps prevent an approximate unadjusted check from being mistaken for the trial’s own analysis. The discrepancy flag indicates that the sponsor’s analysis takes precedence.

Finally, the outcome population and missing-data method matter. The primary analysis uses a defined subset of randomized participants and LOCF imputation. Those choices shape what the estimate represents and what assumptions underlie it. This is a useful example of how design, analysis population, missingness, multiplicity, and precision jointly determine the strength and scope of a clinical-trial conclusion.

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.

Several limitations specific to the trial follow from information not reported in the registry record. The full ANCOVA model specification, including covariates and other modeling choices, is not reported, so the sponsor’s primary comparison cannot be independently reconstructed from the information available. The record also does not report a multiplicity procedure for the several dose comparisons and outcome analyses, or whether interim analyses took place.

The primary analysis used LOCF, but the registry record does not report the extent or pattern of missing outcome data, the reasons for missing measurements, or sensitivity analyses under alternative assumptions. The analysis population required both an End of Phase A value and at least one post-randomization efficacy evaluation. How many randomized participants did not meet that definition, and how that selection affected the estimates, are not reported in the registry record.

The posted secondary outcome list is broader than the detailed secondary results available in the record information. Results for every listed secondary measure are not reported. The record also does not provide participant characteristics in the results described, limiting assessment of how well the study population represents other populations.

The summary-statistic recalculations are approximate and unadjusted; they do not reproduce the sponsor’s ANCOVA. The discrepancy flag makes the sponsor’s own analysis the relevant reference. Finally, the primary outcome is a scale change over the specified Week 8 to Week 14 period. The registry record does not report longer-term results in the information presented, and the scale-based result alone cannot resolve questions about other outcomes or broader effects.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of the Safety and Efficacy of OPC-34712 as Adjunctive Therapy in the Treatment of Patients With Major Depressive Disorder — NCT00797966.
• Registry record last updated 2016-02-29; status last verified 2016-02.
• Results first posted 2015-12-03.
• 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

The registry record describes a completed, randomized, triple-masked, parallel-group Phase 2 study in major depressive disorder. OPC-34712 at three dose levels was compared with placebo as adjunctive treatment to an assigned open-label marketed antidepressant after an eight-week trial with incomplete response. Recorded enrollment was 850.

The primary endpoint was change in MADRS total score from Week 8 to Week 14. The sponsor-reported ANCOVA comparison for the high-dose group was -2.14, with a 95% confidence interval from -4.08 to -0.21; the reported intervals for the 0.15 mg and low-dose comparisons included zero. The primary analysis used the stated ITT definition and LOCF for missing data.

The registry record does not report a multiplicity procedure, full ANCOVA specification, or sensitivity analyses for missing data. The recalculations are approximate and unadjusted, and the sponsor’s analysis takes precedence where results differ. Secondary outcomes and adverse-event counts add context but do not change the scope of the primary endpoint comparison.