Antidepressant Effects of TS-161 in Treatment-Resistant Depression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Antidepressant Effects of TS-161 in Treatment-Resistant Depression (NCT04821271), a randomized phase 2 study. The primary outcome is Change in Montgomery-Asberg Depression Rating Scale (MADRS) Score.

2. Clinical Question

3. Trial Design

This was a phase 2, randomized, quadruple-masked, crossover study sponsored by the National Institute of Mental Health (NIMH). The registry lists enrollment of 11 and a status of terminated. The listed start date is 2021-06-10, and the primary completion date is 2024-05-23. The reason for termination is not reported in the registry record.

The study description outlines an initial period in which participants tapered off psychiatric medicines, followed by a two-week drug-free period. During the treatment phase, participants took TS-161 for three weeks and placebo for three weeks, with a two- to three-week drug-free interval between periods. Randomization assigned participants to TS-161 followed by placebo or placebo followed by TS-161. The registry describes participation as possibly lasting 12-16 weeks.

Study procedures described in the record included interviews, physical exams, psychological tests and symptom surveys, blood draws, urine samples, and possible mood and thinking tests. MRI, functional MRI, and MEG procedures are also described. These are design details in the study summary; the posted results do not establish which procedures each participant completed.

A crossover design exposes participants to both treatment conditions in sequence. In general, this can allow comparisons that account for characteristics shared by the same participant, but sequence, period, and carryover effects can matter. The statistical plan for handling these issues is not reported in the registry record. The record also does not report a sample-size justification or the reason the study was terminated.

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.

Period 1

Reasons for not completing this period:

Period 2

Analysis population for the primary outcome, as defined in the registry: Analyses based on the intent-to-treat (ITT) population and included all randomized participants who had at least one post-baseline assessment.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Change in Montgomery-Asberg Depression Rating Scale (MADRS) Score, measured over Baseline and day 21 (week 3).

The primary endpoint was change in the Montgomery-Asberg Depression Rating Scale (MADRS) total score from baseline to day 21 (week 3). MADRS is a 10-item clinician-rated questionnaire used to evaluate depressive symptoms in adults and assess changes in those symptoms. Each item is scored from 0 to 6, with a total score from 0 to 60; a higher score indicates worsening depression.

For each crossover period, the description says MADRS was completed 60 minutes before intervention and at 230 minutes, 1, 2, 3, 7, 14, and 21 days after the first dose. The primary analysis focused on the change between baseline and day 21, calculated using estimated marginal MADRS total score means from a linear mixed model regression.

In general, a symptom scale provides a structured way to quantify observations and compare measurements over time. The endpoint is useful for examining change in reported symptom severity, but a mean score does not show the range of individual participant experiences. The registry does not report confidence intervals for the arm means or for their contrast, so the precision of the treatment comparison cannot be read from the means alone.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the crossover design mean for the comparison?

Participants were randomized to one of two sequences: TS-161 followed by placebo, or placebo followed by TS-161. Each treatment period lasted three weeks, and a two- to three-week drug-free interval separated the periods. Thus, the design differs from a parallel-group trial in which each participant remains in only one assigned condition. In general, crossover studies can compare measurements from the same participants under different conditions, while also requiring attention to treatment order, period effects, and possible carryover.

The registry describes the primary outcome analysis as a linear mixed model regression used to calculate estimated marginal MADRS means. In general, a mixed model can account for correlated measurements and can estimate means adjusted for terms included in the model. The terms used in this trial's model, its covariance structure, and how sequence or period effects were handled are not reported in the registry record. No additional model assumptions should be inferred from the brief description.

What are estimated marginal means?

Estimated marginal means are model-based averages, rather than necessarily simple arithmetic averages of the recorded measurements. In general, they summarize estimated outcomes under a model and can be used to express results for treatment conditions. Here, the registry lists values of 28.02 for TS-161 and 27.79 for placebo. The arm-specific standard errors are 1.9 and 1.62. The record does not give further detail about model adjustments or an interval around the treatment contrast.

A standard error describes the estimated sampling uncertainty of an estimate under the analysis model. It is not the same as the spread of individual participants' scores. Because the registry reports standard errors for the arm estimates but not a standard error or confidence interval for the contrast, those two arm-level values do not by themselves give the uncertainty of the treatment comparison. For a crossover design, the relationship between measurements from the same participants can affect that comparison.

How should the reported test be understood?

The sponsor reports a t-test of estimated marginal means, with p = 0.91. In general, a t-test assesses a specified contrast relative to its estimated uncertainty under the test assumptions. The registry does not report the test statistic, degrees of freedom, exact contrast estimate, or confidence interval. The p-value alone cannot be used to recover those quantities or to judge whether a clinically meaningful difference has been ruled out.

A confidence interval, in general, gives a range of values compatible with the data and analysis at a stated confidence level. It communicates uncertainty about an estimate more directly than a p-value alone. No confidence interval for the treatment contrast is reported in the registry record. Accordingly, readers cannot use the posted result to determine which positive or negative contrast values remain compatible with the study under the sponsor's analysis.

What can recalculation add?

A recalculated estimate, when available, is a simple unadjusted calculation from posted summary statistics. It is not a substitute for a trial's own model-based analysis, particularly where the design is crossover and the participant-level relationship between conditions matters. No recalculated two-arm comparison is available for this continuous endpoint. The posted arm means and standard errors should not be used to invent an unadjusted contrast without the information needed to support it.

Planning illustrations based on an observed effect answer a different question from the trial's observed result. In general, such calculations explore how many participants per arm might be needed to detect an effect under stated assumptions. They are not observed power and do not show what the trial was designed to detect. The stated planning assumptions use a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a fraction of the observed effect. The trial's own design justification is not reported in the registry record.

Why do missingness and multiplicity matter?

The stated intent-to-treat population includes all randomized participants with at least one post-baseline assessment. In general, that definition identifies who contributed to the primary analysis, but it does not explain how missing follow-up measurements were handled. The model description does not specify an imputation method, a missing-data assumption, or sensitivity analyses. Those details are not reported in the registry record.

Multiplicity arises when many outcomes or analyses are examined. In general, repeated testing can make a nominal p-value harder to interpret as evidence for a single prespecified question unless the analysis plan addresses the testing strategy. The registry lists 19 posted secondary outcomes, 20 outcome measures, and 7 analyses. It does not report whether any multiplicity adjustment was used, how outcomes were prioritized beyond the listed primary endpoint, or whether analyses were prespecified. The result should therefore be interpreted as the reported test for the primary endpoint, without assuming a broader adjustment or analysis strategy.

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("TS-161", "Placebo")
n   <- c(10, 9)
m   <- c(28.02, 27.79)
sp  <- c(1.9, 1.62)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
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  TS-161  10  28.020  6.008  23.722  32.318
CARM  Placebo  9  27.790  4.860  24.054  31.526

11. Analyses Reported by the Sponsor

The sponsor's posted analysis is a t-test of estimated marginal means, with p = 0.91. The registry reports arm estimates of 28.02 for TS-161 and 27.79 for placebo, with standard errors of 1.9 and 1.62. It does not report the sponsor's estimated contrast, its confidence interval, or the test statistic. The p-value is therefore the main inferential result available for the primary comparison.

No unadjusted recalculation of the two-arm continuous comparison is available from the posted summary statistics. The registry record does not provide a contrast and uncertainty estimate that would allow an independent recalculation of the treatment comparison. A recalculation, when possible, would be a simple unadjusted analysis and would not necessarily reproduce the sponsor's mixed-model analysis of this crossover study.

The record does not flag a discrepancy between the posted result and a recalculation. The sponsor's result remains the reported analysis, but its limited detail constrains independent evaluation. In particular, the absence of a confidence interval prevents direct assessment of the range of treatment effects compatible with the sponsor's analysis. No more detailed sponsor analysis plan or model specification is reported in the registry record.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists 19 posted secondary outcomes, 20 outcome measures, and 7 analyses. The primary endpoint is change in MADRS total score from baseline to day 21 (week 3). The record does not describe a multiplicity adjustment, a hierarchy for testing outcomes, or how the analyses were selected. In general, examining multiple outcomes or analyses can increase the chance of at least one apparently unusual result when each is interpreted without accounting for the broader testing context. The posted p-value should be read as the reported test for its stated primary comparison, not as evidence that every secondary result was controlled for multiplicity.

The primary analysis population is described as the intent-to-treat population, including randomized participants who had at least one post-baseline assessment. The outcome counts are 10 for TS-161 and 9 for placebo, while enrollment is 11. The registry does not explain the difference between these counts, identify which assessments were missing, or report the timing or reasons for missing data.

The analysis description says change was calculated using estimated marginal means from a linear mixed model. It does not report an imputation procedure, missing-data assumptions, sensitivity analyses, or details of the model's handling of incomplete observations. In general, a mixed model can use available measurements under its assumptions, but its use alone does not establish that missingness was harmless or that results would be unchanged under other assumptions.

The record also does not report a statistical analysis plan, interim analyses, or the sample-size justification. These absences limit assessment of prespecification, missing-data robustness, and the intended interpretation of multiple outcomes. They do not establish that any particular procedure was or was not used outside what the registry record reports.

13. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports zero serious adverse events among 11 participants at risk for TS-161 and zero among 9 participants at risk for placebo. It also reports zero deaths in each condition. No serious adverse event terms are listed. These are the serious-event counts recorded for the posted results.

In general, a zero count means that no events of the specified type were recorded in the listed group during the reported observation, not that the underlying risk is zero. Small numbers of participants limit how much can be inferred about uncommon events. The registry does not provide event timing, exposure-adjusted summaries, or further details about serious-event ascertainment in the results.

The record separately lists other adverse events among 7 participants at risk for TS-161 and 6 participants at risk for placebo. Those counts are distinct from the serious-event counts. They should not be combined with serious events or treated as a measure of severity. The crossover sequences and the listed at-risk denominators also mean that the safety counts should be interpreted as registry-reported group summaries, not as a direct estimate of comparative risk.

These safety figures describe recorded events and do not establish that the treatments have identical safety profiles. They also do not support conclusions about events that were not captured or about effects beyond the recorded study period.

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

The registry record documents a randomized, quadruple-masked phase 2 crossover study of TS-161 and placebo, with MADRS change as the primary endpoint. The posted estimates are 28.02 for TS-161 and 27.79 for placebo; their standard errors are 1.9 and 1.62. The sponsor-reported test has p = 0.91. These are the available primary-outcome summaries and should be interpreted with the limits of their reporting in mind.

The result does not establish equivalence between TS-161 and placebo. A non-significant or large p-value is not, by itself, proof that two treatments have the same effect. The record does not report an equivalence margin, a confidence interval for the contrast, or a direct estimate of its precision. It therefore does not show how large a difference remains compatible with the analysis.

The result also does not establish that either treatment improves symptoms. The endpoint is a change score, and higher MADRS scores indicate worsening depression, but the reported information does not include the sponsor's treatment contrast with uncertainty. A conclusion about improvement relative to placebo would require an adequately described comparison and consideration of the study's design and analysis.

Enrollment was 11, and the primary outcome lists 10 participants for TS-161 and 9 for placebo. With this study size and without a reported confidence interval, precision is difficult to judge. A p-value cannot replace information about the magnitude and uncertainty of an effect. The result should not be treated as evidence that clinically meaningful effects have been excluded.

The registry lists zero serious adverse events in both conditions and zero deaths. Those counts are descriptive, and the number of participants at risk was limited. They do not establish equal safety or rule out uncommon harms. Overall, the record supports a narrow account of the posted estimates and event counts, rather than a definitive conclusion about efficacy, equivalence, or safety.

15. Why This Trial Matters Statistically

This trial is statistically notable because it uses a randomized crossover structure for a continuous symptom endpoint. In general, each participant's measurements across treatment periods can carry information about within-person change, but analysis must account for the relationship between those measurements and for the ordering of periods. The registry identifies a linear mixed model for the estimated marginal MADRS means, while leaving key model details unreported. That makes clear reporting of the contrast and its uncertainty especially important.

The record gives arm-level estimated means and standard errors, along with a sponsor-reported p-value, but no confidence interval for the treatment contrast. This illustrates why a p-value and two separate standard errors are not a complete description of a comparative result. A contrast estimate and its interval would help readers assess both direction and precision, while the reported p-value alone cannot establish equivalence or exclude meaningful differences.

The study also illustrates the difference between a trial's observed results and planning calculations based on those results. In general, a planning illustration is conditional on its assumptions and is not observed power. It cannot replace a prospective sample-size justification or explain why a study ended. The registry does not report the trial's own sample-size justification or termination reason.

Finally, enrollment was 11, while the primary outcome lists different counts for the treatment conditions. The registry defines the analysis population but does not fully describe missing assessments or sensitivity analyses. Transparent reporting of analysis populations, missing-data handling, model terms, multiplicity decisions, and uncertainty would make a small crossover study easier to evaluate and reproduce.

16. 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 study enrolled 11 participants, and the primary outcome lists 10 for TS-161 and 9 for placebo. The registry does not explain these analysis counts in relation to enrollment or describe missing assessments. This limits evaluation of the completeness of the primary outcome data and whether the included observations represent all randomized participants as intended by the stated analysis population.

The reported comparison lacks an estimated treatment contrast and confidence interval. The sponsor's p-value is reported, but the test statistic and further model details are not. Without these details, readers cannot assess the range of plausible treatment differences or independently reproduce the sponsor's comparison from the registry summaries.

The design used treatment sequences and a drug-free interval, but the registry does not report how the model handled sequence, period, or carryover effects. It also does not report the model's covariance structure, missing-data assumptions, or sensitivity analyses. These omissions are relevant when interpreting a crossover analysis.

The record lists multiple outcome measures and analyses without describing a multiplicity strategy or identifying whether the analyses were prespecified. A sample-size justification and the reason for termination are not reported in the registry record. The safety summaries list event counts but do not provide additional detail about timing or exposure-adjusted analyses. These reporting limits do not determine the direction of bias, but they narrow the conclusions that can be drawn from the posted results.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Antidepressant Effects of TS-161 in Treatment-Resistant Depression — NCT04821271.
• Registry record last updated 2025-06-10; status last verified 2024-07.
• Results first posted 2025-06-10.
• 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.

20. Record Summary

NCT04821271 is a phase 2 randomized, quadruple-masked crossover study of TS-161 and placebo in major depressive disorder and treatment-resistant depression. Enrollment was 11; the registry status is terminated. Participants were assigned to receive the treatments in different sequences, with each treatment period lasting three weeks and a two- to three-week drug-free interval.

The primary endpoint was change in MADRS total score from baseline to day 21 (week 3). The posted estimated marginal means were 28.02 for TS-161 and 27.79 for placebo, with standard errors of 1.9 and 1.62. The sponsor-reported t-test p-value was 0.91. A confidence interval for the treatment contrast is not reported in the registry record.

The registry reports zero serious adverse events and zero deaths in both conditions. Important details, including the model specification, missing-data handling, multiplicity strategy, sample-size justification, and reason for termination, are not reported in the registry record.