Study of Rapastinel as Monotherapy in Patients With Major Depressive Disorder (MDD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Rapastinel as Monotherapy in Patients With Major Depressive Disorder (MDD) (NCT03675776), a randomized phase 3 study. The primary outcome is Change From Baseline on Montgomery-Asberg Depression Rating Scale (MADRS) Total Score at End of Double-blind Treatment (End of Week 6)..

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, parallel-group, placebo-controlled phase 3 study of Rapastinel monotherapy in patients with major depressive disorder. The registry lists an enrollment of 50 and gives the study status as terminated. The study start date is 2018-10-30, and the primary completion date is 2019-07-11. The reason for termination is not reported in the registry record.

Participants were assigned to placebo, Rapastinel 225 mg, or Rapastinel 450 mg. All three interventions are described as weekly intravenous administration using a prefilled syringe. The registry does not report the allocation ratio, the method used to generate or conceal the randomization sequence, or whether masking was maintained successfully. It identifies the study as double-blind but does not provide further operational details about who was masked.

The primary outcome was change from baseline in MADRS total score at the end of Week 6. The posted arm summaries include 15 participants for placebo, 17 for Rapastinel 225 mg, and 18 for Rapastinel 450 mg. The reported endpoint population was the mITT Population, defined as patients in the Safety Population with at least one postbaseline MADRS assessment. How many randomized participants did not meet that endpoint-population definition, and why, are not reported in the registry record.

The registry also lists a secondary MADRS change outcome measured one day after the first dose. It does not report the trial’s original sample-size justification, planned allocation ratio, full statistical analysis plan, or the reason the study was terminated. No sponsor-reported inferential analyses are listed; comparisons discussed in the statistical interpretation are unadjusted recalculations from the posted arm summaries.

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: Total of 68 participants were screened for eligibility; 50 participants randomized to receive double-blind treatment.

Double Blind Treatment Period

Reasons for not completing this period:

Safety Follow-up Period

Analysis population for the primary outcome, as defined in the registry: The Modified Intent-to-Treat (mITT) Population consisted of all patients in the Safety Population who had at least 1 postbaseline assessment of the MADRS total score.

5. Baseline Characteristics

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

Population: The Modified Intent-to-Treat (mITT) Population consisted of all patients in the Safety Population who had at least 1 postbaseline assessment of the MADRS total score.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline on Montgomery-Asberg Depression Rating Scale (MADRS) Total Score at End of Double-blind Treatment (End of Week 6)., measured over Baseline to end of Week 6.

The primary endpoint was change from baseline in the Montgomery-Asberg Depression Rating Scale (MADRS) total score at the end of double-blind treatment, at the end of Week 6. The MADRS is a clinician-rated scale intended to assess depressive symptomatology during the preceding week. It covers ten items, including feelings of sadness, lassitude, pessimism, inner tension, suicidality, sleep or appetite reduction, difficulty concentrating, and lack of interest.

Each item is rated on a 7-point scale from 0, indicating no symptoms, to 6, indicating symptoms of maximum severity. The total score ranges from 0 to 60, with higher scores indicating more depression. A negative change from baseline therefore represents a reduction in the measured symptom score. The endpoint uses change over time rather than the final score alone, which describes how the measured score moved from the participant’s baseline assessment.

A clinician-rated scale provides a structured way to quantify symptoms for group comparisons. However, the group mean is not a description of every participant’s experience, and a change in score does not by itself establish the cause of that change. Interpretation also depends on the comparison with placebo, the uncertainty around the estimated difference, the participants included in the analysis, and the analysis procedures. Those procedures are not fully reported in the registry record.

7. Results Posted in the Registry

• Rapastinel 225mg: mean -21.30 versus -11.30 on control; difference -10.00 (95% CI -16.71 to -3.29), standardized effect d = -1.03; z-test p = 0.003.
• Rapastinel 450mg: mean -12.90 versus -11.30 on control; difference -1.60 (95% CI -8.57 to 5.37), standardized effect d = -0.15; z-test p = 0.653.

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 in MADRS Total Score at 1 Day After First Dose of Treatment

• 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:
  • Rapastinel 225mg: mean -7.40 versus -6.70 on control; difference -0.70 (95% CI -5.81 to 4.41), standardized effect d = -0.09; z-test p = 0.788.
  • Rapastinel 450mg: mean -6.40 versus -6.70 on control; difference 0.30 (95% CI -4.71 to 5.31), standardized effect d = 0.04; z-test p = 0.907.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Rapastinel 225mg 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 Rapastinel 450mg and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists a secondary outcome measuring change from baseline in MADRS total score one day after the first dose. Its overall pattern differs in emphasis from the Week 6 primary endpoint: mean changes were negative in all three groups, and the unadjusted recalculated comparisons with placebo were close to zero relative to their uncertainty. The confidence intervals for both dose comparisons include zero. These observations do not establish that the groups had identical responses; they show that the available secondary summaries do not precisely distinguish the mean changes between groups.

The one-day measure is an earlier assessment of the same rating scale, while the primary measure is at the end of Week 6. It can therefore add information about the timing of measured change, but it is not a substitute for the primary endpoint or a direct replication of it. Different assessment times answer different questions, and a pattern at one time point need not determine the pattern at another.

The recalculated secondary comparisons are unadjusted. When multiple outcomes, doses, time points, or comparisons are examined, the chance of obtaining an apparently notable result by chance can increase. Statistical methods may control that risk, but the registry record does not report a multiplicity procedure or a hierarchy of testing. For that reason, an unadjusted secondary result generally carries less confirmatory weight than a prespecified primary analysis with a documented testing strategy.

The registry does not list sponsor-reported inferential analyses for the secondary outcome. The comparisons are recalculated from posted arm summaries and should be treated as descriptive evidence, not as a complete account of the planned secondary analysis. The record also does not report whether any other secondary outcomes were planned but not posted, or how missing measurements at the one-day assessment were handled.

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

• Change From Baseline in MADRS Total Score at 1 Day After First Dose of Treatment

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant has a MADRS total score, and the registry reports the mean change from baseline and standard deviation by group. A mean describes the center of the group’s change scores. A standard deviation describes how much individual scores vary around that group mean. It is not the uncertainty of the mean difference itself.

What does the recalculated mean difference represent?

The posted comparison estimates the difference between the active-treatment group’s mean change and the placebo group’s mean change. For Rapastinel 225 mg, the estimate is -10.0; for Rapastinel 450 mg, it is -1.6. These are unadjusted recalculations from the posted group summaries, not estimates identified as sponsor-reported analyses. The registry record lists no reported analyses, and it does not give a full statistical analysis plan against which to check these calculations.

Because the endpoint is change from baseline, a negative mean difference indicates a more negative average change in the active-treatment group than in placebo. The sign should be interpreted using the scale’s stated direction: a negative change score indicates improvement. A mean difference describes the average contrast, not the proportion of participants who improved, the range of individual responses, or the clinical importance of the difference.

Why does variation matter?

Group means are estimated from participants whose scores vary. The standard deviation is one description of that variation. When the outcome is more variable, it is harder to estimate a group mean precisely from a given amount of information. The standard deviations reported for the primary outcome are 9.06 for placebo, 10.31 for Rapastinel 225 mg, and 11.36 for Rapastinel 450 mg.

Some statistical methods for continuous outcomes pool information about variation across comparison groups to estimate uncertainty. The recalculated comparison values include pooled standard deviations, but the registry record does not report the sponsor’s method for handling the primary outcome or its assumptions about variation. General statistical context: when observations are more dispersed than a model assumes, standard errors and confidence intervals may be too narrow if the model does not account for that extra dispersion. The registry does not report evidence of over-dispersion or a model that would require a special correction.

How should a confidence interval be read?

A confidence interval gives a range of values compatible with the data and the method used to calculate it. In repeated use of a method under its assumptions, intervals constructed in this way would cover the underlying parameter at the stated rate. A particular interval is not a probability statement that the true effect lies within its endpoints. It communicates both the estimate and its precision.

For the 225 mg comparison, the recalculated interval is entirely below zero. For the 450 mg comparison, the interval spans zero. This distinction describes the uncertainty in these recalculations; it does not by itself establish clinical importance, confirm that the analysis was prespecified, or settle whether results would replicate. The interval method and any adjustment for multiple comparisons are not reported in the registry record.

What does the p-value contribute?

A p-value describes how incompatible the observed data are with a specified null model, under the assumptions used for the calculation. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The p-values 0.003 and 0.653 are unadjusted recalculations from posted summaries. They do not account for any multiplicity adjustment, because no adjustment procedure is reported.

Results should not be divided into a simple proof/no-proof distinction based only on a p-value. The estimated difference, interval, group summaries, analysis population, endpoint definition, and missing-data handling all inform interpretation. Here, the registry defines the mITT population as members of the Safety Population with at least one postbaseline MADRS assessment, but it does not report imputation procedures or a sensitivity analysis for missing outcomes.

What cannot be inferred from the analysis?

The registry record does not provide the sponsor’s analysis code, a complete analysis plan, the original sample-size assumptions, or a reported inferential analysis. It therefore cannot establish whether the recalculations reproduce the intended primary analysis. They are simple, unadjusted comparisons of posted summaries. The record also does not report interim analyses, a multiplicity strategy, or the reason for termination. These gaps limit conclusions about whether the displayed comparisons reflect the complete prespecified testing strategy.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data locate a comparison estimate, not whether a result is important or valid. For Rapastinel 225 mg versus placebo, the recalculated mean difference is -10.0 and the 95% confidence interval runs from -16.71 to -3.289. The interval lies below zero and still spans a range of possible mean differences. It indicates that the estimate is more precise about direction than about the exact magnitude.

For Rapastinel 450 mg versus placebo, the estimate is -1.6 and the 95% confidence interval runs from -8.569 to 5.369. This wider range includes zero and values on either side of it. The observed estimate alone should not be used to conclude that the true difference is small or absent: the interval reflects substantial uncertainty about both direction and magnitude.

The observed endpoint groups include 15 placebo participants, 17 participants in the 225 mg group, and 18 participants in the 450 mg group. Group variability is reflected in the reported standard deviations. The amount of information in these groups, together with that variability and the analysis method, affects precision. These comparisons are unadjusted recalculations from summary statistics, and the registry does not report whether the sponsor used a different model or procedure.

Confidence intervals do not address every uncertainty. They do not account automatically for unreported missing-data methods, multiplicity, analysis choices, or possible selective reporting. They also do not describe participant-level variation in the treatment contrast. The registry record contains no sponsor-reported confidence intervals or inferential analyses, so the recalculated intervals should be understood as descriptive statistical estimates rather than confirmation of the trial’s planned analysis.

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 planning illustrations use participants per arm, not a total across groups and not total events. They are based on the observed effect and the stated planning assumptions: a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set at the stated fraction of the observed effect. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

For the comparison of Rapastinel 225 mg with placebo, planning for 100% of the observed effect gives 15 participants per arm for 80% power and 20 participants per arm for 90% power. Planning for 75% of the observed effect gives 27 participants per arm for 80% power and 36 participants per arm for 90% power. Planning for 50% of the observed effect gives 60 participants per arm for 80% power and 80 participants per arm for 90% power.

For Rapastinel 450 mg versus placebo, the corresponding figures are 662 and 886 participants per arm for 100% of the observed effect, 1,176 and 1,574 for 75% of the observed effect, and 2,646 and 3,541 for 50% of the observed effect, with each pair referring to 80% and 90% power respectively.

In plain terms, the 80% and 90% columns describe the planned chance of detecting the specified effect under the stated assumptions if that effect were present. Higher power requires more participants under otherwise unchanged assumptions. Smaller effects also need more participants to detect, which is why the figures rise as the planning effect moves from 100% to 75% to 50% of the observed effect. These are planning illustrations, not observed power.

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 = -10.000; pooled standard deviation σ = 9.747.
2. n = 2 × (1.96 + 0.842)² × 9.747² / -10.000² = 14.9, rounded up to 15 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 <- -10                      # observed difference in means
sigma <- 9.74663702685872                     # 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(3675776)
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 = 15 (80% power), 20 (90% power)
 75% of observed effect: n per arm = 27 (80% power), 36 (90% power)
 50% of observed effect: n per arm = 60 (80% power), 80 (90% power)
power.t.test: 15.9 
Simulated power at n = 15 per arm: 0.782 (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. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when more than one hypothesis is tested, such as comparisons of two doses with placebo or analyses at more than one time point. Without an appropriate strategy, the chance of at least one chance finding across a collection of tests can exceed the chance associated with a single test. The registry identifies a primary Week 6 outcome and a secondary one-day outcome, and the primary results include comparisons for both active doses against placebo. However, the registry record does not report a multiplicity adjustment, a gatekeeping procedure, or a hierarchy specifying how the tests were to be interpreted.

The reported p-values for the comparisons discussed are unadjusted recalculations. They should not be read as adjusted evidence across the set of outcomes and comparisons. The registry does not report whether the sponsor used another approach, because no sponsor-reported analyses or full statistical analysis plan are listed. The recalculated results therefore do not establish how the trial’s planned confirmatory testing would have treated multiplicity.

The primary analysis population is defined as the mITT Population: patients in the Safety Population who had at least one postbaseline MADRS total-score assessment. This definition makes the presence of a postbaseline assessment part of eligibility for that endpoint population. The registry does not report how many randomized participants lacked such an assessment, the reasons for missing measurements, or whether missing data were imputed. It also does not report sensitivity analyses examining alternative assumptions about missing outcomes.

Details about protocol deviations, treatment discontinuation, adherence, and the handling of measurements after discontinuation are not reported in the registry record. Those details can matter when interpreting a treatment comparison, particularly when analysis populations depend on postbaseline measurements. Interim analyses, stopping rules, and the original sample-size justification are also not reported. The trial’s terminated status is recorded, but the reason is not reported; the status alone does not show whether the termination was related to efficacy, safety, recruitment, or another reason.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry safety counts list no serious adverse events among those at risk in any group: 0 affected among 15 at risk for placebo, 0 among 17 for Rapastinel 225 mg, and 0 among 18 for Rapastinel 450 mg. The record also lists 0 deaths in each group. No serious adverse-event terms are listed.

For other adverse events, the registry lists 6 affected among 15 at risk for placebo, 1 among 17 for Rapastinel 225 mg, and 6 among 18 for Rapastinel 450 mg. These are counts, not a complete description of event severity, timing, duration, or relationship to treatment. The registry record does not provide event-level details in the registry record for this analysis.

Safety counts should be read in the context of the groups observed and the length of follow-up. The primary outcome was assessed at the end of Week 6, and the intervention was described as weekly IV administration. The counts do not establish whether rarer events could occur, and the absence of reported serious events is not proof that serious risks are impossible. The reason for trial termination is not reported in the registry record.

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

The posted results describe average MADRS change in the endpoint population over the period from baseline to the end of Week 6. The 225 mg group had a more negative mean change than placebo, and the unadjusted recalculated mean difference has a confidence interval below zero. The 450 mg comparison is less precise: its interval includes zero and values in either direction. These are descriptions of the posted data and recalculations, not medical advice or a definitive account of the trial’s prespecified analysis.

The findings do not show what happened to every participant. Group means can conceal substantial variation, and the registry does not provide individual-level outcomes or the distribution of changes. They also do not establish a direct difference between the two rapastinel doses, because the posted comparisons are each against placebo and no dose-to-dose analysis is reported.

The registry marks the study as terminated but does not report why. It does not provide the original sample-size justification, a complete statistical analysis plan, multiplicity procedures, or detailed missing-data methods. The reported mITT definition includes only patients in the Safety Population with at least one postbaseline MADRS assessment, and the record does not describe sensitivity analyses for missing outcomes. These omissions limit how confidently the recalculated comparisons can be connected to the trial’s original analysis strategy.

The one-day secondary outcome offers an earlier measurement, but its unadjusted results are not a replacement for the primary endpoint and carry less confirmatory weight when multiplicity handling is unknown. Safety counts are also limited summaries; they do not explain event details or the termination. Overall, the registry supports a cautious reading of the observed group summaries and their uncertainty, not a broad conclusion about treatment effects beyond these data.

17. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the difference between an estimated group contrast and the uncertainty around that contrast. The primary comparison for Rapastinel 225 mg has a recalculated interval below zero, whereas the comparison for Rapastinel 450 mg has a broader interval that includes zero. Reading estimates alongside their intervals avoids treating the point estimate or p-value as a complete description of the evidence.

The record also highlights why a registry result should be distinguished from a prespecified sponsor analysis. Arm means and standard deviations permit simple unadjusted recalculations, but the registry lists no reported inferential analyses and does not provide the full statistical plan. Without that plan, it is not possible to determine whether the recalculations match the intended model, missing-data procedure, or testing strategy.

The secondary endpoint raises a related issue: outcomes and time points can multiply the number of potential comparisons. The registry does not report how multiplicity was addressed, so unadjusted secondary results should be interpreted with restraint. The sample-size illustrations further show how planning estimates depend on the effect chosen: smaller effects require more participants under the stated assumptions. Those illustrations are not the original trial design calculation and are not observed power.

Finally, termination status, limited outcome follow-up, and incomplete reporting of analysis details matter for interpretation. The registry does not give a reason for termination, so no statistical explanation should be inferred. Together, these features make this record a useful example of separating what the posted data quantify from what remains uncertain or unreported.

18. 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 trial is recorded as terminated, but the reason is not reported in the registry record. It is therefore not possible to attribute termination to efficacy, safety, recruitment, or any other cause. The registry reports enrollment of 50, while the primary endpoint summaries include 15 participants for placebo, 17 for Rapastinel 225 mg, and 18 for Rapastinel 450 mg. The record does not explain the relationship between enrollment and the endpoint population or describe missing endpoint data in detail.

The defined mITT population required at least one postbaseline MADRS assessment among patients in the Safety Population. The registry does not report imputation methods, sensitivity analyses, or the effects of missingness on the estimates. It also does not provide detailed information on adherence, discontinuation, protocol deviations, baseline group characteristics, or reasons participants may have lacked assessments.

Comparisons in this analysis are simple unadjusted recalculations from posted summaries. No sponsor-reported inferential analyses are listed, and the registry does not give the original sample-size justification, full analysis plan, multiplicity method, or interim-analysis procedures. The recalculations may not reflect every feature of the planned analysis.

The primary endpoint is measured at the end of Week 6, and the secondary endpoint is measured one day after the first dose. The registry entries do not establish longer-term outcomes. Safety reporting is limited to the posted counts and does not describe event details or provide a reason for termination. These limitations constrain interpretation to the registered population, time points, and posted results.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study of Rapastinel as Monotherapy in Patients With Major Depressive Disorder (MDD) — NCT03675776.
• Registry record last updated 2020-07-28; status last verified 2020-07.
• Results first posted 2020-07-28.
• 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.

22. Record Summary

The registry record describes a randomized, double-blind, parallel-group, placebo-controlled phase 3 study of Rapastinel monotherapy in patients with major depressive disorder. It lists enrollment of 50, a start date of 2018-10-30, a primary completion date of 2019-07-11, and terminated status.

The primary endpoint was change from baseline in MADRS total score at the end of Week 6 in the defined mITT Population. The posted arm summaries show a more negative mean change for Rapastinel 225 mg than placebo, while the 450 mg comparison is less precise. Unadjusted recalculations from the posted summaries are not sponsor-reported analyses. The registry also lists a one-day secondary MADRS outcome and safety counts, but does not report a complete statistical analysis plan, multiplicity procedures, detailed missing-data methods, or the reason for termination.