A Study to Evaluate the Effect of Food on the Pharmacokinetics of TAK-831 Tablet Formulation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Effect of Food on the Pharmacokinetics of TAK-831 Tablet Formulation (NCT03101293), a randomized phase 1 study. The primary outcome is Cmax: Maximum Observed Plasma Concentration for TAK-831.

2. Clinical Question

3. Trial Design

This was a randomized, open-label, single-dose, 2-period crossover Phase 1 study of TAK-831 T2 tablet formulation. The registry record lists healthy volunteers as the condition and Neurocrine Biosciences as the sponsor. The study status is completed. Its start date was 2017-04-04, and its primary completion date was 2017-05-19.

Each participant's assigned sequence included both dosing conditions: TAK-831 400 mg in the fasted state and TAK-831 400 mg in the fed state. One sequence began with fasted dosing in Intervention Period 1 and continued with fed dosing in Intervention Period 2. The other sequence reversed that order. Each administration was once on Day 1 of its period, and the sequences included a 7 days washout period between periods. The registry record does not report how many participants were assigned to each sequence.

A crossover design compares conditions within participants, rather than assigning each participant to only one condition. In general, this can help account for stable differences between participants when estimating a treatment-condition contrast. It also means that interpretation depends on how period, sequence, and any carryover effects are handled. The registry record identifies ANOVA as the analysis method but does not report the model terms or details of how these design features were handled.

The primary pharmacokinetic analysis set included participants from the safety set who had at least one measurable postdose TAK-831 plasma concentration. The posted Cmax summaries list 15 participants for the fasted condition and 14 for the fed condition. The reason for the difference between those condition-specific counts is not reported in the registry record. The trial also lists one secondary outcome concerning participants who experienced at least one treatment-emergent adverse event.

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: Healthy participants were enrolled in this 2-way crossover study and randomized in 1 of the 2 treatment sequences which determined the order to receive TAK-831 400 milligram (mg) film-coated tablets under fasted conditions and TAK-831 400 mg tablet under fed conditions (nutritional drink).

Intervention Period 1

Washout Period (at Least 7 Days)

Reasons for not completing this period:

Intervention Period 2

Analysis population for the primary outcome, as defined in the registry: The pharmacokinetic (PK) analysis set included all participants from the safety set who had at least 1 measurable postdose TAK-831 plasma concentration.

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 safety analysis set included all participants who were randomized and received at least 1 dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Cmax: Maximum Observed Plasma Concentration for TAK-831, measured over Day 1 pre-dose and at multiple time points (up to 72 hours) post-dose.

The primary endpoint was Cmax, the maximum observed plasma concentration of TAK-831. It describes the highest measured concentration in blood plasma after dosing. Cmax is a pharmacokinetic endpoint: it characterizes drug exposure over the measurement period rather than directly measuring symptoms, function, or another clinical outcome.

The registry lists the unit as nanogram per milliliter (ng/mL), summarizes Cmax using the geometric mean, and reports geometric coefficient of variation as its dispersion measure. Concentrations were assessed on Day 1 before dosing and at multiple time points up to 72 hours post-dose. The endpoint therefore depends on the observed concentration measurements and their timing; the registry record does not report the specific sampling times.

Comparing Cmax under fed and fasted conditions addresses whether food is associated with a change in the peak observed plasma concentration following the same stated dose and tablet formulation. In general, this comparison is useful for describing how administration conditions relate to exposure. It does not by itself establish effects on other pharmacokinetic features, such as total exposure over time, or on clinical benefits or harms. Those are different questions and require their own reported endpoints and evidence.

The geometric mean is used for measurements that are often analyzed on a multiplicative scale. The geometric coefficient of variation describes relative spread on that scale. It is not an ordinary standard deviation, so it should not be read as a fixed additive distance around the geometric mean. The registry record does not report individual concentration values or additional detail about the endpoint's derivation.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports an ANOVA-based least-squares mean ratio for the comparison between fasted and fed administration. The analysis is presented as an equivalence analysis. The key reported values are a ratio of 1.343, a 90% confidence interval from 1.146 to 1.574, and p = 0.006. The record does not give the complete statistical analysis plan or the model specification, so the meaning of every model term cannot be reconstructed from the registry entry.

What does the ratio represent?

A ratio compares quantities on a multiplicative scale. A ratio of 1 corresponds to equality between the conditions; a ratio above 1 indicates a larger value in the numerator condition, and a ratio below 1 indicates a smaller value. The posted arm summaries list the fasted geometric mean as 650.0 ng/mL and the fed geometric mean as 865.5 ng/mL. The model-reported least-squares mean ratio is 1.343 for the fed-versus-fasted comparison. A least-squares mean is a model-based estimate, so it can differ from a simple comparison of descriptive summaries depending on the model and data used.

Ratios are commonly used for pharmacokinetic measures because changes are often interpreted proportionally rather than as fixed additive amounts. In general, statistical models may analyze log-transformed values and then express a contrast as a ratio after returning to the original scale. The registry record does not say whether that specific transformation was used, so it should not be assumed as a documented detail of this analysis.

How should the confidence interval be read?

The 90% confidence interval gives an uncertainty range for the model-based ratio under the stated analysis. It runs from 1.146 to 1.574. In repeated use of a valid procedure under its assumptions, confidence intervals at a given confidence level would cover the underlying parameter at that rate. For an individual analysis, the interval does not mean that there is a 90% probability that the fixed underlying parameter lies inside these particular limits.

For an equivalence assessment, an interval is usually judged against two prespecified bounds defining differences considered acceptably small. The bounds are not reported in the registry record. As a result, the reported interval can be discussed as an estimate and measure of uncertainty, but the registry entry does not permit independent confirmation that it falls within an equivalence region. The label for the analysis does not replace the missing bounds.

What does the p-value add?

The reported p-value is 0.006. In general, a p-value summarizes the compatibility of the observed data with a null hypothesis under a statistical model. It is conditional on the model and its assumptions. It does not measure the probability that the finding occurred by chance, the probability that a hypothesis is true, or the practical importance of the estimated contrast. The registry does not identify the exact null hypothesis associated with this p-value, so it should be read as a reported ANOVA result rather than as a complete account of the inferential procedure.

Why does crossover structure matter?

In a crossover study, participants receive both conditions in separate periods. A within-participant comparison can reduce the impact of stable participant-level differences, because each participant contributes information under both conditions. In general, such an analysis must also account appropriately for period and sequence, and consider whether effects from an earlier period could persist into a later one. A washout period is part of the reported design, but its presence alone does not demonstrate that carryover is absent. The registry record does not report the ANOVA terms, sequence-specific results, or tests and assumptions concerning period or carryover.

The listed condition-specific Cmax summaries include 15 participants in the fasted condition and 14 in the fed condition. In a crossover design, these condition counts should not automatically be interpreted as separate, independent groups: participants may contribute observations under both conditions. The record describes the pharmacokinetic analysis set eligibility rule but does not report the participant-level paired data or the exact number included in each analysis.

What does the dispersion measure describe?

The geometric coefficient of variation is reported as 47.7 for fasted administration and 53.5 for fed administration. In general, it expresses relative variability on a multiplicative scale. It does not describe variability in the same way as an arithmetic standard deviation, and it should not be read as a margin around the geometric mean. The registry does not state the formula used for this measure or provide the underlying observations.

More generally, over-dispersion means that observed variability is greater than a model expects. If present and unaccounted for, it can make standard errors and confidence intervals misleading. The registry entry does not report a diagnosis of over-dispersion or a method used to address it, so no claim about its presence or handling can be made for this analysis.

10. Analyses Reported by the Sponsor

The registry-reported comparison is an ANOVA analysis with a least-squares mean ratio of 1.343, a 90% confidence interval from 1.146 to 1.574, and p = 0.006. These are the analysis results available for the fed-versus-fasted Cmax comparison. The posted geometric mean summaries are descriptive values for each condition; they are not a substitute for the model-based contrast.

A separate unadjusted re-calculation of the comparison is not reported in the registry record. Therefore, there is no second estimate or interval to set alongside the registry-reported ANOVA result. The record also does not provide sufficient model details to reproduce the least-squares mean ratio from participant-level data. Interpretation should consequently center on the analysis reported in the registry rather than treating the descriptive summaries as an independent analysis.

The registry labels the analysis as an equivalence analysis, but its equivalence bounds are not reported. The ratio, interval, and p-value can be read as the stated model results; they do not reveal the prespecified bounds or the full analysis plan. The registry record does not report an alternative analysis that would resolve those omissions.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one secondary outcome: the percentage of participants who experienced at least one treatment-emergent adverse event. It also reports four outcome measures and three analyses. The record does not describe a multiplicity adjustment, a testing hierarchy, or whether any outcomes were designated as exploratory. It is therefore not possible to determine from the registry entry how the possibility of multiple statistical tests was handled.

Multiplicity matters when several outcomes, contrasts, time points, or analyses are tested. In general, repeated testing can increase the chance of at least one result appearing unusually strong under null hypotheses. Appropriate approaches depend on the study's objectives and analysis plan. No adjustment or rationale should be inferred when it is not reported. The Cmax comparison is the primary outcome described in the record, but its relationship to all other analyses is not detailed.

The record defines the pharmacokinetic analysis set as participants from the safety set with at least one measurable postdose TAK-831 plasma concentration. This rule indicates that having a measurable postdose concentration was required for inclusion in that analysis set. It does not explain whether any participants lacked measurable concentrations, why any observations might be missing, or how missing measurements were handled. The condition-specific Cmax summaries list 15 participants for fasted dosing and 14 for fed dosing, but the reason for the difference is not reported in the registry record.

The record does not report imputation methods, sensitivity analyses for missing data, or details on the analysis population beyond the stated eligibility description. The randomized enrollment count is 15, but that should not be treated as the count analyzed for every endpoint. Nor should the fasted and fed condition counts be added to infer the number of participants, because the study used a crossover design and the counts describe conditions rather than necessarily separate groups.

The registry record also does not report interim analyses, sample-size justification, or a complete statistical plan. Their absence from the entry limits assessment of prespecification and analysis choices; it does not establish that these procedures were or were not used.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports no serious adverse events in the fasted condition among 15 participants at risk, and no serious adverse events in the fed condition among 14 participants at risk. It reports no deaths in either condition. These figures describe the recorded serious-event experience in the stated groups; they do not establish that serious events are impossible or rule out risks that a study of this size could not characterize.

For other adverse events, the record reports none among 15 participants at risk in the fasted condition and one among 14 participants at risk in the fed condition. The event term is not reported in the registry record. These are condition-specific counts in a crossover study, not necessarily counts from independent groups of participants. The record does not state whether the other event was considered treatment-related.

In general, adverse-event counts should be interpreted in light of the observation period, event definitions, and the number of participants exposed. The listed primary endpoint concerns pharmacokinetics, while the adverse-event information is a separate safety description. The registry record does not provide enough detail to compare event severity beyond the serious-versus-other categories or to draw conclusions about risks outside this study.

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

The posted results describe Cmax after a single 400 mg TAK-831 tablet dose under fed and fasted conditions in healthy participants. The fed-condition geometric mean was 865.5 ng/mL, compared with 650.0 ng/mL in the fasted condition. The registry-reported model estimate for the fed-versus-fasted comparison was a least-squares mean ratio of 1.343, with a 90% confidence interval from 1.146 to 1.574.

This is evidence about a pharmacokinetic measure and the conditions under which that measure was observed. Cmax is the highest measured plasma concentration during the stated sampling period, which extended to 72 hours post-dose. It does not summarize every aspect of drug exposure, and it is not itself a measure of clinical benefit or patient experience. No such clinical outcome is reported as the primary endpoint.

The interval indicates uncertainty around the model-based estimate, subject to the analysis assumptions. The registry does not report enough detail about the model to evaluate all of those assumptions, and it does not list equivalence bounds. Thus, the ratio and interval alone cannot confirm whether a formal equivalence criterion was met. The reported p-value should not be interpreted as a measure of clinical importance.

The findings apply to the studied single-dose tablet conditions and the healthy-participant population. The registry record does not establish whether the same results would apply to repeated administration, another formulation, other populations, or outcomes beyond Cmax. It also does not provide the details needed to understand every difference between participants included in the fasted and fed summaries.

The safety information records no serious adverse events and no deaths in either condition, along with one other adverse event in the fed condition and none in the fasted condition. These observations describe what was recorded in this study, not a complete account of possible harms. They should be kept separate from the pharmacokinetic comparison: a difference in Cmax does not by itself answer a safety or clinical-outcome question.

14. Why This Trial Matters Statistically

This trial illustrates how a crossover design can address a focused pharmacokinetic question. Each sequence included both fed and fasted dosing, allowing a comparison of conditions within participants. In general, this structure can make a condition contrast less dependent on differences between separate participant groups, although the analysis still needs to account for features such as period and sequence. The registry identifies an ANOVA analysis but does not provide enough detail to assess its full implementation.

The result also shows why descriptive group summaries and a model-based contrast should be distinguished. Geometric mean Cmax values and their geometric coefficients of variation describe each condition, while the least-squares mean ratio summarizes the reported model comparison. These quantities answer related but not identical statistical questions. The confidence interval adds information about the precision of the contrast, whereas the p-value alone does not communicate its magnitude or practical importance.

Equivalence labels require particular care. A ratio and confidence interval can be interpreted relative to equality, but a formal equivalence conclusion depends on prespecified bounds. Since the bounds are not reported in the registry record, readers cannot check that criterion from the listed result alone. This is a useful example of how an analysis label and a numerical result may not provide all the information required to evaluate the intended inferential claim.

The record also highlights the importance of analysis-set definitions and condition-specific counts in a crossover trial. Counts associated with each dosing condition should not be mistaken for independent parallel groups or combined into an overall participant total. Explicit reporting of the analysis population, missing observations, model terms, and equivalence bounds would make the comparison easier to evaluate.

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 registry record does not report the complete statistical analysis plan, the ANOVA model terms, or the precise null hypothesis associated with the p-value. It does not report the equivalence bounds, so the formal equivalence assessment cannot be checked from the available result. Details of how period and sequence were handled, and whether carryover was assessed, are also not reported.

The condition-specific Cmax summaries list 15 participants for fasted dosing and 14 for fed dosing, but the reason for the different counts is not reported. The record defines the pharmacokinetic analysis set by safety-set membership and at least one measurable postdose concentration, yet does not provide a separate total for the analysis set, participant-level paired observations, or a missing-data account. The distribution of participants across the two randomized sequences is not reported.

The study is described as a single-dose investigation in healthy participants using the TAK-831 T2 tablet formulation. The registry result therefore has a limited scope: it does not report repeated-dose findings, findings in other populations, or other formulation comparisons. The primary endpoint is Cmax, and the record does not provide other pharmacokinetic results in this comparison. The specific sampling times within the stated period are not listed.

Safety reporting in the record is limited to counts for serious adverse events, other adverse events, and deaths. The other event term and its relationship to treatment are not reported. The registry does not give a more detailed account of event timing or duration.

Finally, the record does not report a sample-size justification, interim analyses, or details of multiplicity control. These omissions constrain assessment of the study's planning and broader inferential framework. They do not, on their own, establish that any particular procedure was absent.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Evaluate the Effect of Food on the Pharmacokinetics of TAK-831 Tablet Formulation — NCT03101293.
• Registry record last updated 2021-06-14; status last verified 2021-06.
• Results first posted 2019-03-18.
• 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

NCT03101293 is a completed randomized, open-label, single-dose, 2-period crossover Phase 1 study sponsored by Neurocrine Biosciences. It enrolled 15 healthy participants and evaluated TAK-831 400 mg tablets under fed and fasted conditions, with a 7 days washout period between intervention periods.

The primary endpoint was Cmax, measured from Day 1 pre-dose through multiple post-dose time points up to 72 hours. The registry reports geometric mean summaries and an ANOVA least-squares mean ratio of 1.343 for fed versus fasted administration, with a 90% confidence interval from 1.146 to 1.574 and p = 0.006. Equivalence bounds and full model details are not reported.

The record lists no serious adverse events or deaths in either condition and reports one other adverse event in the fed condition. The event term is not reported in the registry record.