Pivotal Bioequivalence Study to Qualify Manufacturing Site Transfer for Prazosin Hydrochloride Capsules: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pivotal Bioequivalence Study to Qualify Manufacturing Site Transfer for Prazosin Hydrochloride Capsules (NCT04967443), a randomized phase 1 study. The primary outcome is Area Under the Plasma Concentration-Time Curve From Time Zero to Last Quantifiable Concentration (AUClast) of Prazosin.

2. Clinical Question

The study was intended to support a manufacturing-site transfer and process changes. It was not designed to assess blood-pressure control, heart-failure outcomes, or other clinical benefits. Its primary question concerns pharmacokinetic similarity: whether the amount of prazosin measured in plasma over the specified period was comparable between formulations made at the two sites.

The registry describes three formulation contrasts across two cohorts: Ascoli 2 mg against Barceloneta 2 mg; two Ascoli 1 mg capsules against Barceloneta 2 mg; and Ascoli 5 mg against Barceloneta 5 mg. Results are reported as adjusted geometric-mean ratios with confidence intervals. The registry does not report the prespecified acceptance bounds used to determine bioequivalence, so the numerical estimates alone do not establish whether a comparison met the study’s decision criteria.

3. Trial Design

This was a completed Phase 1, randomized, open-label, single-dose crossover study in healthy adults under fasted conditions. The registry lists 72 participants enrolled. Its description specifies two cohorts analyzed separately: Cohort 1 used a three-treatment, three-period, six-sequence crossover, while Cohort 2 used a two-treatment, two-period, two-sequence crossover. The record says approximately 36 participants would be enrolled in each cohort; it does not report the final cohort-specific enrollment in the registry record for this analysis.

The planned comparisons addressed the manufacturing-site transfer at two capsule strengths and a formulation comparison involving two 1 mg Ascoli capsules versus one 2 mg Barceloneta capsule. Participants received treatments in randomized sequences, and the crossover structure means that treatment comparisons were made within a study designed for participants to receive more than one treatment. This can reduce the influence of stable differences between participants, although it does not remove the need to account for period, sequence, and other relevant design features in analysis.

The registry reports that pharmacokinetic and statistical analyses were conducted separately for the two cohorts. The primary endpoint was AUClast, with assessments at the listed post-dose sampling times on Day 1 of each period. The record also describes AUCinf, Cmax, Tmax, and terminal-phase elimination in its study summary, and lists secondary outcomes including AUCinf, Tmax, half-life, and adverse events.

The study ran from 2021-09-22 to primary completion on 2022-02-15 and is marked completed. It was sponsored by Pfizer. Detailed information about allocation concealment, blinding (the registry identifies the study as open-label), sequence generation, protocol deviations, and the rationale for the sample size is 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.

Before assignment: A total of 72 participants were enrolled in the study and both cohorts (2 cohorts in total) enrolled 36 particpants.

Only the first 6 of 8 groups are shown.

Screening

Open Label Treatment

Reasons for not completing this period:

Follow-Up

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The AUClast analysis population is defined as all participants randomized and treated who had at least 1 of the PK parameters of primary interest in at least 1 treatment 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 Area Under the Plasma Concentration-Time Curve From Time Zero to Last Quantifiable Concentration (AUClast) of Prazosin, measured over 0 (pre-dose), 0.25, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 3.5, 4, 6, 8, 10, 12, 24 hours post-dose on Day 1 of each period.

AUClast is the area under the plasma concentration-time curve from dosing to the last quantifiable prazosin concentration. It summarizes measured drug exposure across the sampling period rather than relying on a single concentration reading. The registry specifies the endpoint in nanogram*hour per milliliter and describes measurements from pre-dose through 24 hours post-dose on Day 1 of each period.

In a bioequivalence study, comparing exposure is relevant because a manufacturing change should not materially alter the amount of drug reaching the bloodstream. The endpoint is therefore suited to the question of whether the proposed manufacturing site produces a pharmacokinetic profile comparable to the current site. AUClast is a pharmacokinetic measure, not a direct measure of blood pressure, heart-failure status, symptoms, or clinical benefit.

The registry summarizes AUClast using geometric means and geometric coefficients of variation. A geometric mean is suited to positive measurements that may be distributed asymmetrically, and a geometric coefficient of variation describes relative variability on a multiplicative scale. The sponsor’s reported comparisons use ratios of adjusted geometric means, which express one formulation’s estimated exposure relative to its comparator.

The record does not state the acceptable ratio bounds or other decision rules used for the bioequivalence conclusion. Without those prespecified criteria, the reported estimates and confidence intervals can be described, but a definitive determination against the study’s formal acceptance standard cannot be made from the registry record alone.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the geometric-mean ratio represent?

The primary endpoint is summarized by geometric means rather than arithmetic means. This is common for pharmacokinetic measurements because exposure values are positive and often vary proportionally: a multiplicative change can be more meaningful than an equal-sized additive change. A ratio compares the estimated exposure under one formulation with exposure under its comparator. The ratio is an estimate, not an exact measure of the difference between every participant’s responses.

The geometric coefficient of variation describes dispersion relative to the geometric mean. It should not be read as a standard deviation in the original measurement units. For interpretation, the arm summaries show both typical exposure and variability, while the adjusted ratio and confidence interval describe the modeled comparison. These summaries answer related but distinct questions.

Why use a crossover design?

In a crossover study, participants receive multiple treatments in different randomized sequences across periods. Since each participant can contribute observations under more than one treatment, comparisons can account for some stable between-participant variation. This is useful in pharmacokinetic studies, where exposure can differ substantially across people. The design still requires suitable analysis of the repeated treatment-period data; observations from one participant cannot simply be treated as independent observations.

The registry describes separate crossover arrangements for its two cohorts and says their data were analyzed separately. It reports mixed-model analysis for the primary contrasts, but does not report the full model specification, including the terms used for sequence, period, participant, or other design factors. Those details are not reported in the registry record, so the model cannot be independently reconstructed from the posted summary values.

How should the confidence intervals be read?

A confidence interval conveys statistical uncertainty around an estimate under the assumptions of the model and analysis. A narrower interval indicates greater precision than a wider interval, all else equal. For an equivalence question, the interval must be compared with acceptance bounds set in advance. The registry identifies the analyses as equivalence analyses and reports 90% intervals, but does not give the bounds or the exact decision rule. Consequently, the available information supports interpretation of the estimates and intervals, not an independent pass-or-fail classification.

Failure to establish a difference is not the same as establishing equivalence. Equivalence asks whether the data are sufficiently precise to rule out differences beyond prespecified limits. Conversely, a point estimate near the comparator does not by itself demonstrate that the interval is sufficiently narrow. This distinction is especially important when a study’s objective is a manufacturing comparison rather than a superiority test.

What can be learned from the secondary analyses?

The registry lists secondary pharmacokinetic outcomes and adverse-event outcomes. Half-life is summarized as a mean with a standard deviation, and the posted recalculations for its pairwise comparisons are unadjusted. A mean difference expresses an additive contrast in hours, whereas the standardized effect size expresses that difference relative to pooled variability. These are not the same estimand as the primary geometric-mean ratio for AUClast.

The registry also includes illustrative sample-size calculations based on observed half-life effects. They assume a two-sided significance level of 0.05, equal allocation, and a normal approximation, with planning scenarios tied to stated fractions of the observed effect. These are planning illustrations only, not observed power and not evidence that the original study was adequately powered. They also do not serve as sample-size justifications for the primary endpoint; the protocol’s sample-size rationale is not reported in the registry record.

Unadjusted secondary comparisons do not account for the possibility of chance findings across multiple outcomes or contrasts. They are useful for describing patterns and generating questions, but they carry less confirmatory weight than a prespecified primary analysis. The registry does not report a multiplicity strategy that would establish how secondary tests were to be interpreted together.

Why can dispersion affect interpretation?

Greater variability makes an estimate less precise when other design features are similar. In general, over-dispersion—variability beyond what a model expects—can make standard errors too small and confidence intervals too narrow if the analysis does not accommodate it. This is a general statistical point, not a finding that over-dispersion occurred in this study. The record’s geometric coefficients of variation indicate that exposure varied within treatment groups, but the summary information does not establish whether model assumptions were assessed or whether any adjustment was required.

Finally, a recalculation from posted summary statistics is not a replacement for the sponsor’s design-aware analysis. In crossover data, the within-participant relationship between treatment measurements is important, and arm summaries alone may not capture it. The primary comparison should therefore be read from the sponsor’s reported mixed-model results; simple unadjusted recalculations on other outcomes have a more limited role.

10. Analyses Reported by the Sponsor

The primary AUClast comparisons in the registry are the sponsor’s mixed-model analyses. They report ratios of adjusted geometric means and 90% confidence intervals for the three formulation contrasts. No recalculated primary comparison is reported alongside them. The registry does not flag a discrepancy between sponsor and recalculated primary results.

The recalculated analyses described for half-life are different: they use posted summary statistics to make simple, unadjusted comparisons. They do not reproduce the sponsor’s mixed-model approach for AUClast, and they do not substitute for the crossover analysis. In a crossover design, the relationship between measurements from the same participant is relevant; summaries that do not retain that relationship cannot fully reconstruct a design-aware treatment comparison.

Accordingly, the sponsor’s own analysis takes precedence for the primary endpoint. The adjusted ratios and confidence intervals should be interpreted as the registry’s reported primary results, with the important qualification that the formal acceptance bounds and bioequivalence decision rule are not reported in the registry record. Recalculated secondary summaries are descriptive checks, not a competing primary analysis.

11. Multiplicity, Missing Data and Analysis Populations

The registry describes three primary formulation comparisons and separate analyses for the two cohorts. It does not report how multiplicity across those comparisons was handled. For example, the record does not state whether there was a hierarchy among contrasts, whether each comparison had a separate decision rule, or whether any adjustment was made to control the chance of an incorrect conclusion across several comparisons. These details matter because interpreting multiple intervals or tests jointly differs from interpreting one prespecified comparison in isolation.

The record also does not describe an adjustment strategy for the secondary outcomes. Since the study considers several pharmacokinetic measures and adverse-event summaries, unadjusted secondary findings should be treated as exploratory or descriptive unless a prespecified analysis plan says otherwise. A formal hierarchy, endpoint-specific decision criteria, and any planned handling of multiplicity are not reported in the registry record.

The primary analysis population is defined as participants who were randomized and treated and had at least one primary-interest pharmacokinetic parameter in at least one treatment period. This definition explains who could contribute to the primary analysis, but it does not fully describe how missing measurements were handled. The record does not report whether missing values were imputed, how incomplete periods or treatment sequences were treated, or whether sensitivity analyses assessed the effect of missing data.

Details of exclusions, protocol deviations, adherence, and the number of participants contributing to each primary contrast are also not reported in the registry record beyond the posted group summaries. The record does not describe interim analyses or stopping rules. These omissions do not invalidate the reported estimates, but they limit independent assessment of the analysis population, the robustness of the findings, and the familywise interpretation of multiple results.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: road traffic accident (2.9%).

The registry reports serious adverse-event counts by treatment group. One participant in the Barceloneta 2 mg group was reported as affected by a serious adverse event, and one death is reported in that group. No serious adverse events or deaths are reported in the listed Ascoli groups or in the Barceloneta 5 mg group. The registry’s listed top serious-event term is Road traffic accident, reported as 2.9%.

These figures are descriptive counts from a single-dose study in healthy adults. The record does not establish that the serious event or death was caused by prazosin or by the manufacturing site. The event term alone does not establish causality, and the registry information does not provide the clinical narrative needed to assess it.

Other adverse events are also reported by group, but group-level counts do not by themselves establish comparative safety. This study’s design and size are not a basis for ruling out uncommon harms, and the registry does not report a formal comparative safety analysis in the registry record for this interpretation. Safety findings should therefore be read as recorded events, not as proof of a causal treatment effect or as a definitive comparison between formulations.

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

The results address a focused pharmacokinetic question: how prazosin exposure after a single dose compared between capsules manufactured at Ascoli and Barceloneta in healthy adults under fasted conditions. The primary endpoint, AUClast, captures measured exposure through the last quantifiable concentration. The reported adjusted geometric-mean ratios and confidence intervals describe the estimated comparisons and their precision.

They do not establish that one formulation is clinically better, safer, or more effective than another. The study was not described as a clinical outcome trial, and AUClast is not a measure of blood-pressure control, heart-failure outcomes, or patient benefit. Nor do single-dose results establish the effect of repeated use, use in people with hypertension or heart failure, or results under fed conditions.

The registry labels the primary comparisons as equivalence analyses, but does not report the acceptance bounds or decision rule. Thus, the estimates can be stated, but the formal conclusion that a comparison met the study’s bioequivalence criteria cannot be independently verified from the registry record. A ratio near equality is not sufficient by itself; the interval must satisfy the prespecified criterion.

The available results also do not establish that all potential differences between manufacturing sites are absent. They quantify uncertainty for specified pharmacokinetic contrasts under this study’s conditions. The secondary outcomes and adverse-event counts offer additional context, but the registry does not report a complete multiplicity strategy or all analysis details needed to treat those findings as confirmatory. The most defensible reading is bounded: these are pharmacokinetic estimates from a randomized crossover study, not broad evidence about clinical outcomes or every setting in which prazosin may be used.

14. Why This Trial Matters Statistically

This trial illustrates why bioequivalence studies require a different statistical interpretation from trials that test whether one treatment improves a clinical outcome. The central question is whether exposure under a proposed manufacturing change is sufficiently similar to exposure under the current process, according to prespecified limits. Point estimates alone are not enough; the uncertainty interval and the decision criteria must be considered together.

The crossover design is also central to interpretation. It permits comparisons across treatments within a study designed for participants to receive multiple formulations, potentially reducing the impact of stable differences between participants. At the same time, valid inference depends on appropriate modeling of the repeated, period-based observations. The registry reports mixed-model analyses for the primary comparisons, but does not provide enough model detail to reconstruct them independently.

Another important issue is that several comparisons are made across different formulations and cohorts. Their estimates are informative for the manufacturing questions posed, but multiplicity and the prespecified hierarchy influence how confidently a collection of results can be interpreted. Those plans are not reported in the registry record. This is a reminder that confidence intervals and formal decision rules must be understood in the context of the protocol, not read as stand-alone signals.

Finally, the study separates pharmacokinetic evidence from clinical outcomes. The results can inform whether measured prazosin exposure was comparable under the tested conditions, but they do not answer whether a treatment changes health outcomes. This distinction helps keep conclusions aligned with the design, endpoint, population, and analysis actually represented in the registry.

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 provides adjusted primary estimates and intervals but does not report the acceptance bounds or the exact rule used to conclude bioequivalence. This prevents an independent determination of whether each primary comparison met its formal criterion. It also does not provide the full mixed-model specification, so the primary analysis cannot be fully reproduced from the posted summary data.

The study involved a single dose in healthy adults under fasted conditions. Results therefore have limited scope for settings involving repeated dosing, fed conditions, or people with the conditions listed in the registry. The study’s stated objective concerns pharmacokinetics and a manufacturing-site transfer, not clinical efficacy.

Further limitations include unreported details about missing-data handling, protocol deviations, analysis-set exclusions, and the number of participants contributing to each contrast. The registry does not report whether interim analyses were planned or conducted, or how multiplicity across the primary and secondary comparisons was handled. The sample-size rationale is also not reported; the separate planning illustrations based on observed secondary effects do not establish the original design’s justification or power.

Safety interpretation is limited by the descriptive event reporting and the absence of information establishing causality. The posted serious-event counts cannot determine whether an event was related to a formulation, and the study is not a basis for excluding uncommon risks. Finally, secondary analyses are incomplete in the available registry information and include unadjusted recalculations, so they should not be treated as a comprehensive confirmatory assessment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pivotal Bioequivalence Study to Qualify Manufacturing Site Transfer for Prazosin Hydrochloride Capsules — NCT04967443.
• Registry record last updated 2024-07-12; status last verified 2024-01.
• Results first posted 2024-07-12.
• 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

The registry describes a completed, randomized, open-label, single-dose Phase 1 crossover study of prazosin hydrochloride capsules in healthy adults under fasted conditions. The study enrolled 72 participants and compared formulations manufactured at Ascoli with those manufactured at Barceloneta across two separately analyzed cohorts.

The primary endpoint was prazosin AUClast. The sponsor reported mixed-model ratios of adjusted geometric means and 90% confidence intervals for three formulation comparisons. The registry does not state the bioequivalence acceptance bounds or decision rule, so formal equivalence conclusions cannot be independently confirmed from the record. Secondary outcomes include additional pharmacokinetic measures and adverse events; their analysis and multiplicity details are incomplete. The results concern pharmacokinetic exposure under the study conditions and do not establish clinical benefit or comparative safety.