Study to Evaluate the Efficacy, Safety, and Tolerability of Oral OPC-34712 and Aripiprazole for Treatment of Acute Schizophrenia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Evaluate the Efficacy, Safety, and Tolerability of Oral OPC-34712 and Aripiprazole for Treatment of Acute Schizophrenia (NCT00905307), a randomized phase 2 study. The primary outcome is Change From Baseline to Week 6 in Positive and Negative Syndrome Scale (PANSS) Total Score (Double Blind Phase).

2. Clinical Question

The primary outcome was change from baseline to Week 6 during the double-blind phase. PANSS total score combines ratings of positive symptoms, negative symptoms, and general psychopathology. A lower score indicates less severe symptoms; therefore, a more negative change represents a larger reduction from baseline. The posted arm summaries and sponsor analyses describe group-level results, not the experience of every participant.

The trial was designed to evaluate efficacy, safety, and tolerability. The registry record describes the hospitalized adult population and the parallel comparison of four OPC-34712 dose arms, placebo, and aripiprazole. The results should be read as estimates for this study population and follow-up period, rather than as an answer to questions about other populations, longer treatment, or outcomes not measured in the record.

3. Trial Design

The registry identifies NCT00905307 as a Phase 2, randomized, double-blind, parallel-group, placebo-controlled study. Its official title describes a 6-week, multicenter trial in hospitalized adult patients with acute schizophrenia. The study is listed as completed. The registry records enrollment of 459; the brief summary had described an aim to obtain approximately 450 randomized subjects after screening approximately 563 at an estimated 75 sites worldwide. The record does not establish that the planned screening or randomization figures were the final numbers screened or randomized.

Participants were assigned to OPC-34712 0.25 mg, low-dose OPC-34712, mid-dose OPC-34712, high-dose OPC-34712, aripiprazole, or placebo. The summary describes OPC-34712 doses spanning 0.25 to 6.0 mg and aripiprazole doses spanning 10 to 20 mg. The specific doses corresponding to the low-, mid-, and high-dose OPC-34712 arms are not reported in the registry record. Aripiprazole was included as a positive control intended to confirm assay sensitivity, according to the brief summary; that stated purpose is not itself a result about the study’s sensitivity.

Double-blind masking means the registry describes a design in which treatment assignment was masked, but details about who was masked and how masking was maintained are not reported in the registry record. The primary endpoint was mean change in PANSS total score from baseline to Week 6. Other posted outcomes include PANSS subscales, personal and social functioning, clinical global ratings, response, and discontinuation-related measures. The record does not report the protocol’s sample-size justification, original power assumptions, or detailed statistical analysis plan.

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: Partipants were randomized in a 1:2:2:2:2:1 ratio to the following groups: OPC-34712 0.25 mg arm, OPC-34712 low-dose, OPC-34712 mid-dose, OPC-34712 high-dose, Placebo, Aripiprazole. All other prohibited medications were discontinued at least 24 hours before the first dose of double-blind study medication.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Consists of all participants who received at least one dose of study medication and have baseline and at least one post-baseline efficacy evaluation. 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.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline to Week 6 in Positive and Negative Syndrome Scale (PANSS) Total Score (Double Blind Phase), measured over Baseline to Week 6.

The primary endpoint was change from baseline to Week 6 in PANSS total score during the double-blind phase. PANSS is a clinician-rated symptom scale made up of three subscales and 30 symptom constructs. Each construct is rated on a 7-point scale, from absence of symptoms to extremely severe symptoms. The total is formed from positive-symptom, negative-symptom, and general-psychopathology items. The registry describes a total-score range of 30-210, with higher scores indicating more severe symptoms.

Change scores put each participant’s Week 6 assessment in relation to that participant’s baseline score. In these results, negative group means indicate that the average score decreased from baseline. A larger negative change is a larger mean reduction, but the group mean alone does not describe the spread of individual changes or how many participants improved. The posted standard deviations provide information about variability within each arm.

PANSS total score is used in this trial because it aggregates several symptom domains into one continuous measure for the primary comparison. That makes it possible to summarize average change between randomized groups while retaining the distinction between the total score and its separate positive and negative subscales. The endpoint is a measure of symptoms over the stated assessment period; it does not by itself represent functioning, quality of life, long-term outcomes, or safety. The record’s primary analysis population required at least one dose and baseline plus at least one post-baseline efficacy evaluation, with missing values imputed by last observation carried forward.

7. Results Posted in the Registry

• OPC-34712 0.25 mg: mean -9.76 versus -14.40 on control; difference 4.64 (95% CI -2.54 to 11.82), standardized effect d = 0.23; z-test p = 0.206.
• OPC-34712 Low-dose: mean -18.73 versus -14.40 on control; difference -4.33 (95% CI -10.22 to 1.56), standardized effect d = -0.21; z-test p = 0.150.
• OPC-34712 Mid-dose: mean -16.19 versus -14.40 on control; difference -1.79 (95% CI -7.40 to 3.82), standardized effect d = -0.09; z-test p = 0.531.
• OPC-34712 High-dose: mean -18.25 versus -14.40 on control; difference -3.85 (95% CI -9.70 to 2.00), standardized effect d = -0.19; z-test p = 0.197.

8. Secondary Efficacy Endpoints

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

Change From Baseline to Week 6 in PANSS Positive Subscale Score (Double Blind Phase)

• 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 deviations.
• Result:
  • OPC-34712 0.25 mg: mean -3.22 versus -4.82 on control; difference 1.60 (95% CI -0.46 to 3.66), standardized effect d = 0.27; z-test p = 0.128.
  • OPC-34712 Low-dose: mean -5.97 versus -4.82 on control; difference -1.15 (95% CI -3.12 to 0.82), standardized effect d = -0.17; z-test p = 0.252.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.25 mg 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 Low-dose and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 6 in PANSS Negative Subscale Score (Double Blind Phase)

• 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 deviations.
• Result:
  • OPC-34712 0.25 mg: mean -1.93 versus -3.17 on control; difference 1.24 (95% CI -0.62 to 3.10), standardized effect d = 0.25; z-test p = 0.192.
  • OPC-34712 Low-dose: mean -3.61 versus -3.17 on control; difference -0.44 (95% CI -1.89 to 1.01), standardized effect d = -0.09; z-test p = 0.553.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.25 mg 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 Low-dose and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 6 in Personal and Social Performance Scale (PSP) (Double Blind Phase)

• 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 deviations.
• Result:
  • OPC-34712 0.25 mg: mean 4.54 versus 7.56 on control; difference -3.02 (95% CI -8.19 to 2.15), standardized effect d = -0.21; z-test p = 0.253.
  • OPC-34712 Low-dose: mean 11.36 versus 7.56 on control; difference 3.80 (95% CI -0.59 to 8.19), standardized effect d = 0.26; z-test p = 0.090.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between OPC-34712 0.25 mg 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 Low-dose and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts secondary outcomes that cover positive and negative PANSS subscales, personal and social performance, global severity and improvement ratings, response, and discontinuation-related outcomes. The continuous secondary summaries show arm-level changes or scores, but the presence of multiple endpoints means each addresses a different aspect of the study rather than serving as an independent confirmation of the primary question.

Across the posted continuous secondary outcomes, the group summaries vary in direction and size by measure and arm. The PANSS subscale summaries include negative mean changes in the arms, while the personal and social performance summaries show positive mean changes. As with the primary endpoint, those are descriptions of group averages, not proof that differences between treatments are reliable or attributable to treatment. Sponsor-reported comparisons are available for only some of the secondary outcome contrasts represented in the detailed record.

The secondary findings should be considered alongside, not in place of, the primary PANSS total score result. They can provide context about symptom domains and functioning, but they do not remove the uncertainty in the primary treatment-placebo estimates. A pattern that appears favorable on one outcome may be less informative when several outcomes, arms, and comparisons are considered together.

Unadjusted secondary results carry less weight because many comparisons create more opportunities for apparently notable results to occur by chance. The registry does not report a multiplicity strategy for the secondary analyses, a hierarchy of confirmatory endpoints, or a rule controlling the overall error rate. The different endpoints also have their own analysis populations and missing-data patterns. These results are consequently best read as exploratory or descriptive unless the registry record identifies a prespecified confirmatory role; no such role is reported for these secondary comparisons.

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

• Change From Baseline to Week 6 in PANSS Positive Subscale Score (Double Blind Phase)
• Change From Baseline to Week 6 in PANSS Negative Subscale Score (Double Blind Phase)
• Change From Baseline to Week 6 in Personal and Social Performance Scale (PSP) (Double Blind Phase)
• Change From Baseline to Week 6 in Clinical Global Impression-Severity of Illness Scale (CGI-S) Score (Double Blind Phase)
• Mean Clinical Global Impression - Improvement (CGI-I) at Week 6
• Response Rate at Week 6
• Discontinuation Rate for Lack of Efficacy or Receipt of Open Label OPC-34712

9. Statistical Interpretation

10. Statistical Methods Explained

What does the mean change represent?

The endpoint is continuous: each participant’s PANSS total score is measured at baseline and Week 6, and the analysis summarizes the change. The registry reports the arm mean and standard deviation. A mean is the arithmetic center of the observed or imputed changes in an arm. A standard deviation describes how dispersed those changes are around the mean. When the standard deviation is large relative to the mean difference between arms, the comparison can be imprecise even when group means differ numerically.

Negative mean changes indicate a decrease in PANSS total score from baseline. Because higher PANSS scores indicate more severe symptoms, a decrease is a change toward a lower score. This scale interpretation does not establish why the score changed, and it does not mean every participant had the same change.

How should the between-group estimate be read?

The registry’s own primary comparisons are reported as ANCOVA treatment differences. ANCOVA is a regression-based method that can compare groups while accounting for specified covariates. The covariates and full model specification are not reported in the registry record, so the adjusted estimate cannot be reconstructed from the arm means alone. The sponsor’s estimate is the relevant reported analysis; an unadjusted calculation from summary means is a separate, simplified comparison.

In the recalculation, the mean difference is formed in the treatment-minus-placebo direction. Since the endpoint is a change score, a negative difference means the treatment-arm mean change was more negative than the placebo-arm mean change. A positive difference means it was less negative. The recalculated standard error uses the posted group standard deviations and sample sizes under an unadjusted comparison. The confidence interval describes uncertainty around that estimate under the calculation’s assumptions. This is not the same model as the sponsor’s ANCOVA.

What does a confidence interval add?

A confidence interval communicates estimation uncertainty and the range of values compatible with the observed data under the model used. Its endpoints should be considered with the point estimate: a broad interval signals limited precision, while an interval that includes zero does not rule out no difference on the scale of the contrast. An interval that also includes differences in either direction indicates that the data are not pinning down the direction with confidence. A confidence interval is not a guarantee that the true effect lies within its endpoints, nor does it show the range of individual participant outcomes.

Why are the recalculations labeled unadjusted?

The posted summary statistics do not contain the participant-level data or the adjustment details needed to reproduce ANCOVA. A simple comparison of means therefore leaves out any covariate adjustment in the reported model. It also cannot account for any analysis details that depend on participant-level information. These recalculations are useful as a transparent check on the scale and broad magnitude of the posted results, but they should not be treated as a replacement for the sponsor analysis.

The record’s discrepancy indicator is false, so no stated conflict between the recalculation and sponsor result is flagged. Even when the numerical estimates are close, their interpretation differs because one is based on a simple unadjusted calculation and the other is the reported ANCOVA. The sponsor’s own analysis remains the primary registry-reported comparison.

Why can variability affect precision?

General statistical explanation: greater variation within arms makes a mean difference harder to estimate precisely, all else being equal. The standard deviations in the primary outcome are around 18 to 21 units across arms, while the arm sample sizes vary. These features contribute to uncertainty in the pairwise comparisons. The recalculated standard errors and confidence intervals reflect the posted group sizes and variability, but they do not incorporate ANCOVA adjustments.

General statistical explanation: equal allocation between two arms is a common planning assumption for a simple comparison because, for a fixed total sample size, it often gives efficient precision. The registry’s planning illustration uses equal allocation, but the actual primary outcome group sizes are not equal across all arms. The planning figures are therefore illustrations under specified assumptions, not a description of the enrolled design or a demonstration of the trial’s power.

What role does the analysis population play?

The primary population included participants who had received at least one dose and had baseline and at least one post-baseline efficacy evaluation. This is a defined efficacy population, not necessarily everyone randomized. The record says missing values were imputed using last observation carried forward (LOCF), which carries the last available observation forward to fill a later missing value. LOCF creates a complete analysis value under that rule, but it relies on an assumption about the unobserved outcome and can understate or distort uncertainty if that assumption is not appropriate. The record does not report sensitivity analyses for alternative missing-data assumptions.

The sample size attached to each arm’s outcome summary can differ from the enrollment figure and from the group counts in the adverse-event record. Different outcomes and analysis populations can have different available measurements. The registry does not provide enough detail to reconcile every such difference, so counts should be read within the specific outcome or safety summary in which they appear.

11. Confidence Intervals and Precision

The sponsor’s primary ANCOVA intervals are 95% confidence intervals. They quantify uncertainty in each estimated treatment difference under the reported model. For low-dose OPC-34712, the estimate was -4.70 and the interval ran from -10.2 to 0.82. For mid-dose, the estimate was -1.44 and the interval ran from -6.96 to 4.07. For high-dose, the estimate was -3.86 and the interval ran from -9.32 to 1.59. For OPC-34712 0.25 mg, the estimate was 4.62 and the interval ran from -2.89 to 12.12. For aripiprazole, the estimate was -3.64 and the interval ran from -10.7 to 3.38.

Every interval includes zero, so the registry’s reported estimates do not exclude a zero treatment difference at the 95% confidence level. They also extend to values on both sides of zero, allowing for effects in either direction on the treatment-minus-placebo change-score scale. The ranges are broad enough that the point estimates alone should not be treated as precise descriptions of the underlying difference. In particular, an estimate near zero is not proof that the groups are equivalent; equivalence requires a specified margin and an analysis designed for that question, neither of which is reported in the registry record.

The recalculated unadjusted intervals also include zero, but they are not substitutes for the sponsor intervals. The sponsor used ANCOVA, while the recalculation uses posted arm-level summary statistics. Differences in estimates and interval endpoints can arise from those different methods. Confidence intervals address uncertainty in the estimated group contrast; they do not express uncertainty about individual outcomes, and they do not correct for multiple comparisons unless the method explicitly does so. A multiplicity adjustment is not reported in the registry record.

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 sample-size figures are planning illustrations based on the observed effect, not the trial’s original design assumptions. For these continuous outcomes, the entries are participants per arm, not total participants or total events. The calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers the stated fraction of the observed effect as the effect to detect. The original trial’s sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the illustrative participants per arm for a design that, under those assumptions, would have an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column gives the corresponding planning number for a 90% chance. These are design probabilities, not statements about the probability that the observed finding is true. They are not observed power.

The rows for 100%, 75%, and 50% of the observed effect illustrate how the planning requirement changes as the effect to detect is made smaller. A smaller target effect generally requires more participants to distinguish it from random variation under the same assumptions. For example, the primary low-dose comparison lists 342 participants per arm for 80% power at 100% of the observed effect and 608 participants per arm for 80% power at 75% of that effect. Those figures are hypothetical planning values based on the observed summary contrast, not prospective targets validated by the original protocol.

The curve should be read as the same kind of sensitivity illustration across candidate effect sizes and sample sizes, not as a plot of the trial’s realized power. Because the calculation uses a simplified unadjusted framework and the observed effect, it may not match a future study’s design, analysis, or outcome variability. The registry record does not report the assumptions used to plan this completed trial.

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 = 4.640; pooled standard deviation σ = 19.879.
2. n = 2 × (1.96 + 0.842)² × 19.879² / 4.640² = 288.1, rounded up to 289 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 <- 4.64                      # observed difference in means
sigma <- 19.8792031576172                     # 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(905307)
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 = 289 (80% power), 386 (90% power)
 75% of observed effect: n per arm = 513 (80% power), 686 (90% power)
 50% of observed effect: n per arm = 1,153 (80% power), 1,543 (90% power)
power.t.test: 289.1 
Simulated power at n = 289 per arm: 0.797 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The sponsor’s reported primary comparisons use ANCOVA and are the registry’s own analysis results. They should be distinguished from the recalculated comparisons, which use only the posted group means, standard deviations, and sample sizes for a simple unadjusted comparison. The two approaches answer closely related but not identical statistical questions because ANCOVA can adjust for specified covariates, while the summary-statistic recalculation cannot.

For the primary outcome, the sponsor reports treatment differences and 95% confidence intervals for each OPC-34712 dose arm versus placebo and for aripiprazole versus placebo. Each sponsor interval includes zero. The recalculated estimates are generally close in direction and scale to the sponsor results, but they are not numerically identical in every comparison. The record’s discrepancy indicator is false; no discrepancy is flagged. The sponsor analysis is not displaced by the simple recalculation.

The registry does not report the ANCOVA covariates, the full model specification, or enough detail to reproduce the sponsor’s estimates independently. The recalculation is therefore best treated as a transparent, unadjusted reference point. For conclusions about the trial’s reported primary comparisons, the sponsor’s own ANCOVA results take precedence over these recalculations.

15. Multiplicity, Missing Data and Analysis Populations

The primary outcome involves several treatment-versus-placebo comparisons: four OPC-34712 dose arms and an aripiprazole arm. The record also lists multiple secondary outcomes and numerous analyses. When many comparisons are considered, the chance of at least one result appearing more extreme by chance can increase. Methods such as a prespecified testing hierarchy or multiplicity adjustment can address that issue, but the registry record does not report whether such a method was planned or used. The reported p values should therefore be read as individual comparison results rather than as evidence that an overall error rate was controlled.

The record identifies the primary analysis population as participants who received at least one dose and had a baseline and at least one post-baseline efficacy evaluation. Missing values were handled using last observation carried forward. This method replaces a missing later value with the last observed value. It is a specific imputation rule, not a direct observation of the missing outcome. Its suitability depends on assumptions about how outcomes would have changed after the last measurement. Those assumptions, the extent and reasons for missing primary outcome data, and sensitivity analyses using other approaches are not reported in the registry record.

The record does not report the detailed covariates in the ANCOVA model, the analysis plan’s handling of protocol deviations, or interim analyses. It also does not report a full account of participants excluded from each outcome analysis. The posted analysis counts vary by outcome and arm, so a result should be interpreted using the population attached to that particular endpoint rather than assuming the same participants contributed to every summary.

These omissions do not show that a method was not used; they mean its use cannot be established from the registry record. They limit independent assessment of how missing data, multiple comparisons, and analysis choices affected 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: rhabdomyolysis (2.0%), complex partial seizures (2.0%), ankle fracture (1.1%), schizophrenia (1.1%), hypoglycaemia (1.1%).

The registry’s serious adverse event summary lists affected counts alongside the number at risk in each arm. It reports 0 affected among 42 at risk for OPC-34712 0.25 mg; 3 among 89 for low-dose; 5 among 90 for mid-dose; 4 among 93 for high-dose; 2 among 50 for aripiprazole; and 3 among 95 for placebo. These are counts in the registry summary, not a comparison adjusted for exposure time or other differences between groups. The record does not provide a statistical comparison of serious adverse events.

Listed serious-event terms include rhabdomyolysis and complex partial seizures, each reported as 2.0%, and ankle fracture, schizophrenia, hypoglycaemia, and dizziness, each reported as 1.1%. The registry summary does not assign those term percentages to treatment arms in the displayed information, so they should not be attributed to a particular group from this record alone.

Deaths are not reported in the registry record. These safety counts are descriptive and do not establish whether an event was caused by study treatment. They should be considered separately from the efficacy endpoint and in light of the trial’s 6-week follow-up.

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

The primary outcome measures average change in PANSS total score from baseline to Week 6. The registry reports a decrease in the group mean in every arm, including placebo. Its sponsor-reported treatment differences are estimates of how much mean change differed between each active group and placebo under the ANCOVA model. Every reported 95% confidence interval includes zero and spans values in both directions.

Those results do not establish a clear between-group difference on the primary measure in this study. They also do not establish that the groups are equivalent: the intervals allow for differences on either side of zero, and the registry does not report an equivalence or noninferiority design with a prespecified margin. Nor does a within-arm decrease establish that the treatment caused that decrease. A randomized placebo comparison, rather than a before-and-after change alone, is the relevant contrast for evaluating between-group evidence.

The interpretation applies to the listed hospitalized adult population, the specific regimens studied, the 6-week period, and the registry’s defined analysis population and LOCF approach. It does not answer questions about longer follow-up, other populations, other doses not studied, or outcomes not included in the registry results. Secondary outcomes offer additional descriptive context, but multiple outcomes and the absence of a reported multiplicity strategy mean that isolated secondary comparisons should not be treated as definitive confirmation.

The safety summary reports serious adverse event counts by arm, but the counts alone do not establish causality or characterize all possible safety risks. The record’s limited follow-up and the absence of a reported death count also constrain what can be concluded from the posted safety information. Overall, the statistics describe observed group summaries and their uncertainty; they are not medical advice or a standalone basis for treatment decisions.

18. Why This Trial Matters Statistically

This trial illustrates why group change and treatment-placebo contrast are different quantities. The posted mean PANSS total score changes are negative in all arms, including placebo. Looking only at a treatment group’s change from baseline would miss the central comparative question: whether its change differs from the change in the randomized placebo group. The sponsor’s ANCOVA estimates address that comparison, while the confidence intervals show the uncertainty around each estimate.

It also illustrates how several dose arms and an active control create a collection of comparisons around one primary outcome, followed by further outcomes in different domains. That structure can be useful for characterizing a study, but it makes prespecification and multiplicity handling important for interpreting individual p values. The registry record does not report a multiplicity strategy, so the results need to be read with that limitation in mind.

The distinction between sponsor analysis and simple recalculation is another statistical lesson. Posted means and standard deviations allow an unadjusted comparison that helps readers inspect the scale of the results. They cannot reproduce an ANCOVA analysis without its model details and participant-level information. A close numerical match is not evidence that the methods are interchangeable.

Finally, the planning illustrations show how sample-size requirements depend on the effect size a future study aims to detect and the probability of detection it targets. Because those figures use the observed effects and simplified assumptions, they are not a retrospective score for this trial. Keeping design planning, observed results, and uncertainty separate is central to sound interpretation.

19. Limitations

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

The primary outcome analysis used LOCF for missing data, but the registry does not report the amount or pattern of missingness, reasons for missing assessments, or sensitivity analyses under alternative assumptions. The analysis population also required a post-baseline efficacy assessment, so it is not necessarily identical to all randomized participants. These features matter when interpreting the group summaries and their precision.

The registry’s ANCOVA results are reported without the covariates or full model specification. This limits independent reproduction of the sponsor estimates and assessment of the role of adjustment. The unadjusted recalculations cannot resolve that gap, because they use arm-level summaries rather than participant-level data.

There are multiple dose arms, an active control, and secondary outcomes. A multiplicity procedure or confirmatory hierarchy is not reported in the registry record. This limits the weight that can be placed on individual p values, particularly for secondary comparisons. The primary intervals also include zero and are broad enough to allow differences in either direction, limiting precision about the size of any treatment-placebo contrast.

The study followed participants for 6 weeks, so the posted efficacy results do not address longer-term outcomes. The registry summary describes dose ranges but does not report the specific low-, mid-, and high-dose OPC-34712 levels. The safety summary gives serious adverse event counts but does not provide a reported death count or a statistical comparison of safety outcomes. These are limits of what the registry record establishes, not evidence that unreported procedures or events did not occur.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study to Evaluate the Efficacy, Safety, and Tolerability of Oral OPC-34712 and Aripiprazole for Treatment of Acute Schizophrenia — NCT00905307.
• Registry record last updated 2015-10-20; status last verified 2015-09.
• Results first posted 2015-09-03.
• 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

NCT00905307 is a completed Phase 2, randomized, double-blind, parallel-group, placebo-controlled study in hospitalized adults with acute schizophrenia. The official title describes a 6-week, multicenter trial evaluating oral OPC-34712 and aripiprazole. The registry records enrollment of 459 and lists four OPC-34712 dose arms, aripiprazole, and placebo.

The primary outcome was mean change from baseline to Week 6 in PANSS total score. The registry’s sponsor-reported ANCOVA treatment differences all have 95% confidence intervals that include zero. The primary population required at least one dose and baseline plus at least one post-baseline efficacy evaluation; LOCF was used for missing data.

The registry also reports secondary outcomes and descriptive serious adverse event counts. The statistical plan’s multiplicity approach, ANCOVA model details, and the trial’s original sample-size assumptions are not reported in the registry record. The sample-size illustrations are planning calculations based on observed effects, not the trial’s original design or observed power.