Sequential Therapy for the Treatment of Severe Bipolar Depression.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sequential Therapy for the Treatment of Severe Bipolar Depression. (NCT02974010), a randomized phase 2 study. The primary outcome is BDM Score (BISS-derived MADRS) Change From Baseline at Day 42.

2. Clinical Question

3. Trial Design

This is a randomized Phase 2 study with a sequential design and quadruple masking. The registry lists an enrollment of 22 and a completed status. The recorded study start is 2018-01-15, and primary completion is recorded as 2018-08-31. The sponsor is NeuroRx, Inc.

The brief summary describes recruitment of patients with severe bipolar depression and acute suicidal ideation or behavior, followed by an intravenous ketamine infusion. Patients described as having a satisfactory clinical response to ketamine were to be randomly allocated to NRX-101 or lurasidone alone. The stated purpose of this phase was feasibility work for a pivotal Phase 2b/3 trial. The summary describes blood levels of NRX-101 as the primary outcomes for this phase, with remission from depression and relapse as secondary outcomes.

The registered primary endpoint, however, is BISS-derived MADRS score change from baseline at Day 42. The adverse-event record separates participants into four sequences: ketamine followed by NRX-101, ketamine followed by lurasidone, saline followed by NRX-101, and saline followed by lurasidone. The registry record does not explain how these saline sequences relate to the ketamine-response allocation described in the brief summary, or provide enough detail to reconcile the differing descriptions of the phase’s primary outcomes.

Randomization is intended to reduce systematic differences between assigned groups, while masking is intended to limit the influence of treatment knowledge on participants, clinicians, or outcome assessment. Those are general design principles; the registry record does not report the detailed allocation procedure, the roles covered by quadruple masking, or how masking was maintained. The actual number randomized to each sequence and the analysis populations are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Last Observation Carrier Forward

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 BDM Score (BISS-derived MADRS) Change From Baseline at Day 42, measured over Day 42.

The primary endpoint is change from baseline in BISS-derived MADRS score at Day 42. The record describes baseline as the end of stage 1, Day 1, and says that change data are presented using LOCF, expanded in the record as Last Observation Carrier Forward. The score is based on the MADRS, a clinician-rated scale with 10 items. Each item is rated on a 0-6 severity scale, responses are equally weighted and summed, and total scores range from 0 to 40. Higher scores indicate worse severity, so decreases in average BDM are described as a better outcome.

The BISS is a validated clinician-rated scale with 42 items: 21 for depression and 21 for mania. Each item is rated on a 0-4 severity scale, with higher values indicating worse severity. The registry identifies the endpoint as BISS-derived MADRS rather than reporting a standalone BISS total.

A change score expresses how a participant’s recorded score differs from the baseline assessment. Comparing average changes between assigned groups can describe the treatment-group contrast on the measured symptom scale. The registry’s stated endpoint therefore focuses on symptom severity at the specified follow-up, but a result estimate, its uncertainty, and the number of participants contributing to the endpoint are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Randomization assigns participants to treatment groups using a chance-based process. In general, this helps prevent deliberate selection from determining who receives each treatment and supports comparisons between groups. Randomization does not guarantee that groups will be identical, especially in a small study. The registry identifies the study as randomized but does not report the allocation sequence, concealment procedure, or numbers assigned to each group.

What does sequential design mean in this record?

A sequential design studies treatment sequences rather than only a single treatment assignment considered in isolation. Here, the registry’s adverse-event groups identify sequences involving ketamine or saline first, followed by NRX-101 or lurasidone. The brief summary describes ketamine infusion followed by random allocation to NRX-101 or lurasidone among patients with a satisfactory clinical response. The registry does not explain how the saline sequences fit with that description, whether the sequence stages had separate allocation processes, or how the primary comparison accounted for the initial infusion. Those design details are not reported in the registry record.

What does quadruple masking tell the reader?

Masking means that knowledge of assigned treatment is withheld from specified people involved in a study. In general, masking can reduce the risk that expectations affect symptom ratings, participant reports, clinical care, or interpretation. The label “quadruple” indicates the registry’s masking classification, but the record does not identify which four roles were masked or describe procedures used to maintain masking. It is therefore possible to state the registered classification, but not to assess masking success from the record.

How should a mean change be understood?

The primary measure is continuous: it records a score along a scale rather than assigning each participant to a simple event category. A group mean summarizes the central level of change among participants included in the analysis. A between-group contrast in mean change would describe the average difference in change between NRX-101 and lurasidone. A mean is useful for summarizing an overall shift, but it does not show the full spread of individual responses or how many participants improved, remained similar, or worsened.

The registry identifies the planned summary as a least-squares mean and its dispersion as a standard error. In general, a least-squares mean is a model-based adjusted mean; interpretation depends on the model and covariates used. The registry record does not report the model specification, adjustment variables, or a numerical least-squares mean. A standard error describes the estimated sampling uncertainty of a mean, rather than the spread of individual participant scores. No numerical standard error is reported for this endpoint.

What does a confidence interval contribute?

In general, a confidence interval gives a range of values compatible with the data and the statistical model at a stated confidence level. It helps readers assess precision and the range of effects that remain plausible under the analysis. It does not measure the probability that a particular treatment effect is true. The registry record does not report a confidence interval for the primary endpoint, so the precision of the treatment contrast cannot be evaluated.

Why can over-dispersion matter?

For count outcomes, over-dispersion means that observed variability is greater than a model assumes. If a model understates variability, uncertainty estimates can be too narrow and inferential tests can be misleading. This is a general issue for some count-data analyses. The primary endpoint in this record is a continuous score, and the registry does not report an over-dispersion analysis or a count model for that endpoint. It would not be appropriate to assume that a particular model or correction was used.

How should the planning assumptions be read?

The record describes planning illustrations using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says that the effect to detect is expressed as a stated fraction of the observed effect. These are assumptions for planning illustrations, not a report of the study’s own design justification. The registry does not report the original sample-size justification. Since the numerical observed effect and the resulting planning values are not reported in the registry record, no numerical sample-size conclusion can be drawn. Any such illustration is a hypothetical planning calculation, not observed power.

In general, a significance level defines a threshold for a statistical testing procedure under its assumptions; it does not describe the size or clinical importance of an effect. A p-value, if reported, would quantify how unusual the data are under a specified null model, not the probability that the null hypothesis is true. No primary endpoint p-value is reported in the registry record. The analysis model, handling of covariates, and inferential procedure are also not reported.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, time points, treatment contrasts, or statistical tests are considered. In general, testing many hypotheses can increase the chance of at least one apparently notable result if no adjustment or prespecified strategy is used. A statistical plan may address this through a hierarchy, adjustment procedure, or a clearly defined primary comparison. The registry record does not report a multiplicity-control strategy.

The record lists one primary outcome measure and one analysis, while the result structure is not reduced to a single comparison because it involves several categories or time points. The registry record does not explain the complete set of comparisons or how the several categories or time points relate to the stated Day 42 endpoint. It does not report whether any adjustment was used or which estimate was designated as the principal inferential result.

The outcome description states that data are presented using LOCF, expanded as Last Observation Carrier Forward. In general, last-observation-carried-forward methods fill a missing follow-up value with an earlier recorded value. This approach makes assumptions about what would have happened after the last observed measurement; those assumptions can be unrealistic when symptoms change over time. The registry does not report the amount or pattern of missing endpoint data, the specific implementation of LOCF, or sensitivity analyses using other missing-data assumptions.

The analysis population is not reported in the registry record. It is therefore unclear whether the endpoint analysis included all randomized participants, only participants who received assigned treatment, only those with an available assessment, or another defined group. The record also does not report exclusions, withdrawals, or the handling of participants who did not complete the sequence. These omissions limit assessment of attrition bias and of how closely the analysis preserved the benefits of randomization.

11. 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 reports serious adverse events for four treatment sequences. The number at risk was 12 for ketamine followed by NRX-101, 5 for ketamine followed by lurasidone, 4 for saline followed by NRX-101, and 1 for saline followed by lurasidone. Serious adverse events affected 0 participants in each group. The record lists no serious adverse-event terms. Deaths were 0 in each group.

For other adverse events, the record reports 8 affected participants among 12 at risk after ketamine followed by NRX-101, 3 among 5 after ketamine followed by lurasidone, 4 among 4 after saline followed by NRX-101, and 1 among 1 after saline followed by lurasidone. These are counts as recorded for the separate sequences; they do not establish that the treatment sequence caused the events or that the groups differ reliably in risk.

The group sizes are small, particularly in the saline-followed sequences. Counts can change substantially in meaning when the number of participants at risk is limited, and the record does not provide event descriptions, timing, duration of exposure, severity categories for other events, or statistical comparisons. The registry record also does not provide enough detail to determine whether a participant could contribute more than one event. These safety summaries should therefore be read as descriptive counts, not as a basis for comparative safety conclusions.

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

The record establishes that a randomized, quadruple-masked Phase 2 sequential study was registered for severe bipolar depression with acute suicidal ideation or behavior. It identifies the intended oral comparison as NRX-101 versus lurasidone and names change in BISS-derived MADRS score at Day 42 as the primary endpoint. It also describes a sequence involving ketamine infusion and lists saline-based sequences in the adverse-event record.

Those design details are not the same as outcome evidence. A numerical result for the primary endpoint is not reported in the registry record. Without a group-level estimate and a measure of its uncertainty, it is not possible to characterize the size, direction, or precision of the symptom-score comparison. It is also not possible to determine from the record whether any reported difference would be statistically compatible with no difference or whether it would be clinically important.

The registry describes the study as feasibility work intended to inform a later pivotal trial. That context can help explain why a small study might be conducted, but it does not establish that feasibility objectives were achieved. The record does not provide explicit feasibility thresholds or their results, and it does not report the sample-size justification.

The adverse-event figures provide counts of participants affected in each listed sequence. They do not determine the relative safety of NRX-101 and lurasidone, especially because group sizes differ and the registry does not provide event-level detail or comparative analyses. Likewise, these counts do not answer the primary symptom-score question.

The record therefore supports a description of the registered design, intended endpoint, and posted safety counts, but not a conclusion about the comparative efficacy of the treatments. It cannot establish that one treatment is better or worse, nor can it establish the absence of an effect. Those interpretations require endpoint results and an adequately described analysis.

13. Why This Trial Matters Statistically

This study illustrates why a registered design and an outcome result need to be distinguished. The registry identifies a randomized, masked comparison and a clinically relevant symptom measure, but the numerical endpoint estimate is not reported. Without the estimate and its uncertainty, readers cannot judge the observed contrast or its precision.

It also illustrates the importance of distinguishing a feasibility study from a confirmatory study. The brief summary describes this Phase 2 work as preparation for a pivotal Phase 2b/3 trial. Feasibility studies can inform whether procedures and measurements are workable, while confirmatory studies are generally designed to address a prespecified treatment-effect question. The record does not state formal feasibility criteria or their outcomes, so the extent to which the study informed later planning cannot be assessed from the registry record.

The sequence structure matters statistically. The brief summary describes response to ketamine before randomization to oral treatments, while the adverse-event record includes saline followed by both oral treatments. Understanding the estimand—the exact treatment contrast the analysis is intended to estimate—requires clarity about each stage, assignment, and analysis population. Those details are not reported sufficiently to reconstruct the intended comparison across all sequences.

Finally, the enrollment of 22 emphasizes why estimates and uncertainty are essential to interpretation. The enrollment alone does not reveal the number analyzed in each sequence or the precision of the primary comparison. The posted safety counts are informative as descriptive observations, but do not replace a defined endpoint analysis. Together, these features show how design transparency, complete outcome reporting, and explicit analysis methods are needed to assess what a small clinical study contributes.

14. Limitations

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

The registry record does not report numerical results for the primary endpoint, despite identifying the endpoint and its assessment time. It therefore does not permit evaluation of the treatment contrast, its uncertainty, or its statistical significance. It also does not provide confidence intervals, endpoint p-values, or group-specific mean changes.

The study summary and the recorded sequences are not fully aligned in their descriptions. The summary describes allocation after a satisfactory clinical response to ketamine, while adverse-event groups include sequences beginning with saline. The registry record does not explain the relationship between these pathways, the sequence-specific allocation procedures, or the analysis contrast intended for each pathway.

The record does not report the numbers randomized or analyzed in each group, participant flow, withdrawal reasons, baseline group characteristics, or the endpoint missingness pattern. Although it states that the outcome uses LOCF, it does not describe implementation details or sensitivity analyses. These limitations make it difficult to assess the risk of bias from missing data or exclusions.

The statistical analysis plan is not reported in enough detail to identify the model, covariate adjustments, multiplicity strategy, or analysis population. The record also does not provide the original sample-size justification. The listed planning assumptions do not supply a numerical planning result or substitute for a prespecified design rationale.

Safety reporting is limited to counts by sequence. Serious adverse-event counts, other adverse-event counts, and deaths are reported, but event descriptions and comparative analyses are not. These omissions constrain interpretation of both the clinical outcome and safety record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Sequential Therapy for the Treatment of Severe Bipolar Depression. — NCT02974010.
• Registry record last updated 2021-05-25; status last verified 2021-05.
• Results first posted 2021-05-25.
• 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.

18. Record Summary

The registry describes NCT02974010, a completed, randomized, quadruple-masked Phase 2 sequential study sponsored by NeuroRx, Inc., with an enrollment of 22. The study concerns severe bipolar depression with acute suicidal ideation or behavior. Its brief summary describes ketamine infusion followed, among satisfactory responders, by random allocation to NRX-101 or lurasidone; the adverse-event record lists four sequences involving ketamine or saline followed by either oral treatment.

The primary endpoint is change from baseline in BISS-derived MADRS score at Day 42, with higher scores indicating worse severity. The registry identifies use of LOCF but does not report a numerical primary endpoint result or its uncertainty. Serious adverse events and deaths are recorded as 0 in each sequence, while other adverse events are listed by group. The record does not provide enough information to make a comparative treatment-effect or safety conclusion.