Biosignatures of Latuda for Bipolar Depression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Biosignatures of Latuda for Bipolar Depression (NCT02239094), a na na study. The primary outcome is Montgomery-Asberg Rating Scale for Depression (MADRS).

2. Clinical Question

3. Trial Design

NCT02239094, titled “Biosignatures of Latuda for Bipolar Depression,” is listed as completed. The sponsor is the New York State Psychiatric Institute. The record lists a start of 2015-01 and primary completion of 2016-09. Enrollment is listed as 20, and the condition is bipolar depression.

The study is classified as a single-group design, with no allocation and no masking. The arm description says that all participants would receive open-label Latuda. There is therefore no concurrent control arm in the registry description. The phase is listed as NA. Further details about recruitment, eligibility criteria, treatment schedule, dose, adherence, and follow-up are not reported in the registry record.

The brief summary describes a pilot study of biological predictors of lurasidone response in bipolar depression. That objective is distinct from the listed primary outcome: a baseline MADRS measure. The record lists two outcome measures but no analyses. It does not report the biological measurements, a prespecified definition of response, or an analysis comparing biological features between response groups.

Because the study has one open-label group, changes over time would not have a randomized control comparison. Even if outcomes were available at more than one time point, the design alone could not establish that a change was caused by treatment. The registry record does not describe outcome results beyond the safety counts summarized in the safety information.

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:

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 Montgomery-Asberg Rating Scale for Depression (MADRS), measured over Baseline.

The listed primary endpoint is the mean MADRS score at baseline, measured in units on a scale. MADRS is a clinician-administered measure of depression symptoms. Its ten items are intended to assess depression, and the description notes that the scale was designed to be particularly sensitive to antidepressant treatment effects in people with major depression.

Nine items are based on patient report, while one is based on the rater’s observation during the interview. Each item is rated on a 0–6 continuum, with 0 indicating no abnormality and 6 indicating severe symptoms. The registry entry lists the outcome’s summary measure as a mean and its dispersion measure as a standard deviation.

A baseline score describes symptoms at the start of the assessment period; by itself, it does not measure improvement or response to treatment. The registry record does not report a numerical baseline mean or standard deviation, a later MADRS assessment, or a change from baseline. It also does not describe how a clinically meaningful response would be defined. Thus, the endpoint description explains the intended measurement, but the posted information does not provide a numerical result for judging symptom change or linking change to biological predictors.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The methods information in the registry record is limited. It identifies a single-group, open-label study, a baseline continuous outcome summarized as a mean with a standard deviation, two outcome measures, and zero listed analyses. Details of a statistical analysis plan, missing-data methods, analysis populations, and inferential procedures are not reported in the registry record.

What does a mean and standard deviation describe?

In general, a mean is a measure of the center of a set of numerical observations. A standard deviation describes how dispersed the observations are around that mean. They answer different questions: the mean describes the average score, while the standard deviation describes variation among scores. Neither quantity alone gives the full distribution of scores or indicates how precisely an estimate has been measured.

For the listed primary outcome, the score is MADRS at baseline. The record identifies the intended summary statistics but does not give their numerical values. It is therefore not possible to assess the baseline average or the spread of participants’ scores from the registry entry. A baseline mean and standard deviation, even if reported, would describe the group at baseline rather than a treatment response.

What would an uncertainty interval add?

In general, a confidence interval expresses the statistical uncertainty around an estimated quantity under the assumptions of the method used. Its width depends on factors such as the amount of information and variability in the data. A confidence interval does not resolve design limitations, bias, or missing data; it quantifies uncertainty within an analysis. No confidence interval or other uncertainty estimate for a clinical outcome is reported in the registry record.

Why does the single-group design matter?

In general, a randomized comparison can help distinguish differences associated with an assigned intervention from other influences, because the comparison group provides a reference. This study is listed as single-group, with no allocation and no masking, and all participants were to receive open-label lurasidone. There is no like-for-like control group in the record. A change over time, if one were reported, would not by itself establish that the intervention caused the change. The registry entry does not report such a change.

Open-label means the intervention was not masked in the study description. In general, awareness of treatment can affect participant reports, clinical ratings, and study conduct. The record does not provide information about safeguards against measurement or assessment bias, so their use cannot be assumed.

How should the planning illustration be understood?

The record describes a planning illustration based on an observed effect, 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 a stated fraction of the observed effect. Such an illustration is a hypothetical planning exercise, not an observed result and not evidence that the study had a particular probability of detecting an effect. The record’s study itself is single-group; the two-arm allocation assumption in the illustration does not turn it into a two-arm trial. The numerical illustration values are not reported in the registry record.

In general, a sample-size calculation depends on assumptions about the outcome, variability, effect to detect, significance threshold, power target, allocation, and analysis. If those assumptions differ from the eventual study or the scientific question, a planning result may not apply. The record does not report the trial’s own sample-size justification, and no such rationale can be inferred from the planning illustration.

What do the listed analyses tell us?

The registry record lists zero analyses. This means the entry does not report analysis results for the stated predictor question or the primary endpoint. It does not establish whether analyses were conducted outside the registry record. Results such as p-values, effect estimates, confidence intervals, adjustments, and model specifications are not reported in the registry record. No recalculated outcome estimates are available to interpret.

In general, a statistical analysis should match the outcome and study design, define the analysis population, state how missing observations are handled, and describe any adjustment or model assumptions. If data are more variable than a model assumes, uncertainty estimates and tests may be misleading; this is one reason model assumptions should be checked. The registry entry does not describe these methods, so no claim about their use or adequacy is possible.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Lurasidone (Latuda)")
n   <- c(18)
m   <- c(24.3)
sp  <- c(6.4)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Lurasidone (Latuda)  18  24.300  6.400  21.117  27.483

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists two outcome measures and zero analyses. It does not report a multiplicity plan, identify which outcomes were included in a formal testing hierarchy, or describe any adjustment for examining multiple outcomes or predictors. Without those details and without analysis results, the chance of false-positive findings across multiple tests cannot be assessed. No claim about a prespecified primary analysis beyond the listed baseline MADRS outcome can be made.

Missing-data procedures are not reported in the registry record. The entry does not describe how missing outcome values, withdrawals, incomplete assessments, or missing biological measurements were handled. It also does not state whether an analysis used all enrolled participants or a defined subset. Enrollment of 20 should not be treated as proof that every participant contributed to every measure.

The record lists 20 as enrollment, while the safety information lists 18 participants at risk for serious adverse events and 18 at risk for other adverse events. The reason for the difference between those counts is not reported in the registry record. The counts should not be used to infer withdrawals, exclusions, or missing outcome data.

Analysis populations, imputation methods, sensitivity analyses, and rules for handling protocol deviations are not reported in the registry record. These omissions limit assessment of how robust any clinical or biomarker findings might be, and the absence of reported analyses means those findings cannot be evaluated from this entry.

12. 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 record lists 18 participants at risk for serious adverse events, with 0 affected. It lists 18 at risk for other adverse events, with 8 affected, and 0 deaths. These figures describe the counts recorded for the lurasidone group; there is no comparator group in the study design.

In general, an at-risk count is the number of participants included in the registry’s accounting for that category, while an affected count records participants with at least one event in that category. These counts do not describe event severity beyond the serious-versus-other distinction, the number of episodes per participant, or when events occurred. The registry record does not report the specific other adverse events, their timing, their duration, or whether investigators considered them related to treatment.

The safety denominator is 18, whereas enrollment is listed as 20. The reason for this difference is not reported in the registry record. The counts therefore should not be interpreted as a complete account of outcomes for every enrolled participant. With no control group, these figures also cannot establish whether an event count is higher or lower than it would have been under another treatment or no treatment.

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

The registry record describes a completed, open-label pilot study in which all participants were to receive lurasidone for bipolar depression. It identifies an intended research focus—biological predictors of response—and lists baseline MADRS as the primary outcome. The available entry does not report a post-baseline depression result, a response definition, or an analysis of biological predictors.

Accordingly, the record does not establish whether lurasidone changed depression scores in this study, whether a specific biological measure predicted response, or how precisely any treatment or predictor association was estimated. The listed baseline endpoint is not a measure of treatment effect. The study’s single-group design also lacks a control comparison, so any change over time, if reported elsewhere, would need cautious interpretation.

The safety counts describe zero participants with a serious adverse event among 18 listed at risk, eight with other adverse events among 18 listed at risk, and zero deaths. They do not provide a comparative safety estimate, identify the other events, or establish attribution. The reason that safety counts are based on 18 when enrollment is 20 is not reported in the registry record.

These limits concern what can be concluded from the registry entry; they do not show that analyses or additional results do not exist elsewhere. The entry records the study as completed, but completion status is not a substitute for posted outcome estimates. Statistical interpretation requires the underlying estimates, their uncertainty, outcome definitions, analysis methods, and information about missing observations. Those details are not reported in the registry record.

14. Why This Trial Matters Statistically

This study is statistically notable because its stated scientific aim and its listed primary outcome are not the same kind of question. The aim concerns biological predictors of response, while the recorded primary outcome is a baseline symptom measure. A predictor analysis requires a defined response outcome and measurements of candidate predictors; neither the response definition nor the relevant analysis is reported in the registry record.

The design also makes the distinction between description and comparison important. A single-group study can describe participants who receive an intervention, but without a control group it cannot provide a direct randomized estimate of how outcomes compare with an alternative. Open-label assessment can add further uncertainty when outcomes rely partly on reports or ratings. These considerations do not invalidate descriptive information; they define what that information can support.

The record also illustrates why enrollment, outcome availability, and analysis counts should not be treated as interchangeable. Enrollment is listed as 20, safety counts use 18 participants at risk, and no analyses are listed. The reason for the difference in counts is not reported, and the record does not say which participants contributed to which outcomes. Clear reporting of analysis populations and missing-data handling would be needed to interpret estimates fully.

Finally, a planning illustration is not evidence of achieved precision or power. The record describes hypothetical two-arm assumptions even though the study design is single-group. Keeping planning assumptions separate from observed results prevents a sample-size calculation from being mistaken for an empirical finding. The posted information is sufficient to identify these design and reporting issues, but not to assess the proposed biomarker question statistically.

15. 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 central limitation for interpreting this registry entry is that it reports no outcome analyses. Although it lists a baseline MADRS endpoint and describes a study of biological predictors, it provides no numerical MADRS result, follow-up symptom measure, response definition, or predictor analysis. The effect and its precision therefore cannot be assessed.

The design is single-group and open-label, with no comparator. This prevents a like-for-like treatment comparison in the registry description and limits causal interpretation of any change that might be reported elsewhere. Masking is listed as none. Details of assessment procedures and safeguards against measurement bias are not reported in the registry record.

The record lists enrollment of 20, while the safety counts list 18 at risk. It does not explain the difference or describe participant flow, withdrawals, exclusions, or outcome availability. Missing-data methods and analysis populations are not reported. The listed zero analyses further means there is no reported statistical model, uncertainty estimate, or adjustment procedure to evaluate.

Additional study details—including eligibility criteria, treatment dose and duration, biological measures, follow-up schedule, sample-size justification, and multiplicity procedures—are not reported in the registry record. The phase is listed as NA. These omissions make it difficult to judge the study’s planned scope or the applicability of its findings. They should not be filled in by inference from the study title or its completed status.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Biosignatures of Latuda for Bipolar Depression — NCT02239094.
• Registry record last updated 2017-12-12; status last verified 2017-11.
• Results first posted 2017-11-13.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT02239094 is a completed, single-group, open-label pilot study sponsored by the New York State Psychiatric Institute. The record lists bipolar depression as the condition, enrollment of 20, and lurasidone (Latuda) as the intervention. All participants were to receive open-label Latuda. The listed study dates are a start of 2015-01 and primary completion of 2016-09.

The stated aim is to investigate biological predictors of lurasidone response. The listed primary outcome is mean MADRS at baseline, with standard deviation as the dispersion measure. Two outcome measures and zero analyses are listed. Numerical clinical outcome estimates and a biomarker-response analysis are not reported in the registry record.

Safety information lists 0 serious adverse events among 18 at risk, 8 affected by other adverse events among 18 at risk, and 0 deaths. The record does not explain why the safety count differs from enrollment or detail the other events. The study has no control group, and no masking is listed.