Study to Characterize the Pharmacokinetics of a Single Dose of SC Abatacept 125 mg Using the BD Autoinjector or the Prefilled Syringe: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Characterize the Pharmacokinetics of a Single Dose of SC Abatacept 125 mg Using the BD Autoinjector or the Prefilled Syringe (NCT01890473), a randomized phase 1 study. The primary outcome is Adjusted Geometric Mean of Maximum Observed Serum Concentration (Cmax) of a Single Dose of Subcutaneous (SC) Abatacept - PK-Evaluable Analysis Population.

2. Clinical Question

The question is therefore focused on pharmacokinetics following a single dose, rather than on whether abatacept changes a health outcome. Random allocation supports a comparison between the device groups, while the open-label design means participants and study personnel were not masked to the assigned device. The registry record describes the study as multicenter and parallel-group. It does not report the number or locations of study centers, or additional details about randomization procedures. The primary outcome’s stated timeframe, Day 1 to Day 71, reflects the period over which blood concentrations were assessed after administration.

3. Trial Design

This was a Phase Ib, randomized, open-label, parallel-group study in participants with rheumatoid arthritis. The study compared a single subcutaneous dose of abatacept 125 mg delivered through a BD autoinjector with the same dose delivered using a prefilled syringe. Participants were assigned to one of the two device groups; the registry record describes the allocation as randomized and masking as none. The registered enrollment was 356. The number included in a particular analysis can differ from the overall registered enrollment because the primary outcome was summarized in the PK-evaluable population, defined as randomized and treated participants with adequately evaluable pharmacokinetic parameters.

The primary outcome was adjusted geometric mean Cmax, the maximum observed serum concentration after the dose. Serum abatacept was measured using a validated enzyme-linked immunosorbent assay. Blood samples were scheduled on Day 1 before dosing and at 1, 2, and 8 hours after dosing, followed by collection on Days 2, 3, 5, 8, 15, 29, 43, 57, and 71. The registry record also lists pharmacokinetic outcomes such as time to reach maximum concentration, terminal phase elimination half-life, total body clearance, and volume of distribution, along with safety, immunogenicity, and laboratory outcomes.

The primary PK-evaluable summaries include 105 participants in the autoinjector group and 111 in the prefilled-syringe group. The terminal phase elimination half-life summaries include 104 and 111 participants, respectively. These analysis counts are not the same as the registered enrollment. The registry record does not report why participants were absent from particular endpoint analyses, how missing samples were handled, or whether the same participants contributed to every outcome. Although the study is described as multicenter, the number of sites is not reported in the registry record. The registry record also does not report the planned sample-size justification or further details of the allocation procedure.

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: 356 participants enrolled and 224 randomized. 132 not randomized and reasons for non-randomization: 106 no longer met study criteria, 15 other, 9 withdrew consent, and 2 lost to follow-up.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: PK-evaluable analysis population: All randomized and treated participants with adequately evaluable PK parameters were summarized.

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: All randomized participants were summarized.

6. Primary Endpoint

The registry lists the primary outcome as Adjusted Geometric Mean of Maximum Observed Serum Concentration (Cmax) of a Single Dose of Subcutaneous (SC) Abatacept - PK-Evaluable Analysis Population, measured over Day 1 to Day 71.

The primary endpoint is the adjusted geometric mean of maximum observed serum concentration, or Cmax, after a single subcutaneous abatacept dose. Cmax is the highest measured serum concentration during the sampling period. It is a pharmacokinetic measure: it describes the concentration of drug observed in blood, rather than a change in symptoms, function, or another clinical outcome.

Geometric means are commonly used for concentration measures because pharmacokinetic values can have a skewed distribution. On a logarithmic scale, a geometric mean summarizes the central value in a way suited to multiplicative variation; when results are converted back to the original scale, comparisons can be expressed as ratios. The registry reports adjusted geometric mean concentrations of 11.5 μg/mL for the autoinjector group and 12.6 μg/mL for the prefilled-syringe group. The reported comparison is a ratio of geometric means with a 90% confidence interval.

The sampling schedule matters to the endpoint: the highest concentration observed is the maximum among the scheduled measurements, not necessarily a continuously observed peak between sampling times. The registry specifies blood collection from the pre-dose assessment on Day 1 through Day 71. That extended schedule also supports estimation of other pharmacokinetic characteristics, including the time to maximum concentration and terminal elimination measures. The endpoint is used to characterize and compare drug exposure by delivery device. It should not be interpreted as a direct measure of clinical benefit, and the registry record does not report a clinical endpoint as the primary outcome.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Interpretation begins with the kind of outcome being measured. Cmax is a continuous pharmacokinetic outcome measured in μg/mL. The primary summaries are geometric means, rather than arithmetic means, and the device comparison is expressed as a ratio. A ratio is naturally interpreted relative to 1: a value of 1 corresponds to equal geometric means. The reported ratio is 0.908, and its 90% confidence interval extends from 0.815 to 1.01. Because the interval includes 1, it does not provide a clear separation from equal geometric means at the level of uncertainty represented by that interval.

A confidence interval conveys the precision of an estimate under its statistical assumptions. A narrower interval indicates greater precision than a wider interval, but interval width is not a direct measure of clinical importance. The interval does not mean that there is a 90% probability that the fixed true ratio lies within its endpoints. Rather, confidence intervals are constructed by a procedure that, under its assumptions and repeated use, has the stated coverage. The registry reports a 90% interval for the primary geometric-mean summaries and comparison. The record does not report the detailed model specification, covariates used for adjustment, or the method used to construct that interval.

Why use geometric means for concentrations?

Pharmacokinetic concentrations often vary multiplicatively: a change may be more naturally described as a fold difference than as an additive difference. General statistical practice is to analyze such outcomes on a logarithmic scale when that is appropriate, then express group summaries or comparisons on the original scale as geometric means and ratios. This approach can make a skewed concentration distribution more suitable for modeling. That is a general explanation, not a claim about undocumented analysis steps in this trial. The registry labels the primary endpoint as an adjusted geometric mean and reports a geometric-mean ratio, but the full statistical model is not reported in the registry record.

What does the comparison estimate?

The primary estimate concerns the relative maximum observed serum concentration after administration by the two devices in the PK-evaluable analysis population. It is not an estimate of the proportion of participants who respond, nor is it a treatment effect on rheumatoid arthritis. The ratio is a group-level comparison: it does not show how much concentrations varied among individual participants or whether one participant’s concentration changed relative to another’s. The reported arm-specific intervals and ratio interval are also not interchangeable. Each answers a different question: an arm-specific interval describes uncertainty in that arm’s geometric mean, while the ratio interval describes uncertainty in the comparison.

The registry reports 105 participants in the autoinjector primary analysis and 111 in the prefilled-syringe primary analysis. Those endpoint counts define the scale of the analyzed groups, but they do not by themselves explain who was excluded from the PK-evaluable population or why. The population definition requires randomized and treated participants to have adequately evaluable PK parameters. The criteria for adequacy, handling of incomplete concentration profiles, and details of any exclusions are not reported in the registry record. It is therefore important not to assume that the endpoint analysis includes every randomized participant.

How should the secondary recalculation be read?

The terminal phase elimination half-life outcome is presented as a mean with a standard deviation. The simple recalculation compares the two posted means without adjustment. It estimates a mean difference of 17.0 h, with a 95% confidence interval from -8.317 to 42.32 and an unadjusted p-value of 0.188. The interval includes zero, the value corresponding to no mean difference, and spans differences in either direction. The p-value is unadjusted, so it addresses this comparison without accounting for the fact that many secondary outcomes were considered. It should not be read as a definitive test of the entire study or as evidence that the groups are equivalent.

The standard deviation describes spread among participant measurements, not uncertainty in the group mean. It is different from a standard error or confidence interval. Greater variability generally makes a group mean less precisely estimated when the group size is held constant. In PK data, variability may arise from differences between participants, measurement variation, and other sources; the registry does not apportion these sources. General statistical caution is warranted when comparing means from distributions that may be skewed, although the posted half-life summary is a mean and standard deviation.

Why can over-dispersion matter?

Over-dispersion occurs when observed variability is greater than assumed by a statistical model. In general, a model that understates variability can produce confidence intervals that are too narrow and tests that overstate precision. Models for counts or event data may need to accommodate over-dispersion; for continuous pharmacokinetic data, the distributional assumptions and transformation may also affect uncertainty estimates. The registry does not report evidence of over-dispersion or the diagnostics used for the primary analysis. This is therefore a general consideration, not a reported problem in this trial.

What do planning illustrations mean?

Sample-size illustrations based on an observed effect are planning exercises, not observed trial outcomes and not retrospective proof that the study had adequate power. The posted assumptions describe a two-sided significance level of 0.05, equal allocation, and a normal approximation. The illustrations vary the effect to be detected as a fraction of the observed effect, and state the number of participants per arm for 80% or 90% power. These values show how required sample size changes with the assumed target effect under those assumptions. They are not the trial’s original design justification, and the registry record does not report the original sample-size justification. They also should not be confused with observed power, which is not a useful substitute for interpreting the confidence interval and estimate actually observed.

More generally, the registry record does not report the complete statistical analysis plan, whether interim analyses were planned or conducted, or any multiplicity adjustment. A reader should distinguish what is directly posted—the endpoint definitions, group summaries, confidence intervals, and one recalculated secondary comparison—from analysis choices that are not documented. That distinction is particularly important when several outcomes are listed and when the primary endpoint is a pharmacokinetic surrogate rather than a clinical outcome.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a ratio of geometric means of 0.908, with a 90% confidence interval from 0.815 to 1.01. The corresponding adjusted geometric mean Cmax values are 11.5 μg/mL for the autoinjector group and 12.6 μg/mL for the prefilled-syringe group. The registry does not report a p-value for the primary comparison. The ratio and interval are the sponsor-reported analysis; they should be distinguished from the separate recalculation of the secondary half-life outcome.

The half-life recalculation is a simple, unadjusted comparison of the posted group means. It gives a mean difference of 17.0 h, a 95% confidence interval from -8.317 to 42.32, and an unadjusted p-value of 0.188. This recalculation does not reproduce or replace the sponsor’s primary analysis: it concerns a different endpoint, uses a different summary scale, and is not adjusted. No discrepancy between a sponsor-reported estimate and a recalculation is flagged in the registry record. The sponsor’s own analysis therefore takes precedence for the primary endpoint, while the recalculated secondary result is best treated as an additional descriptive check rather than as a competing primary conclusion.

The registry record does not report the full primary model, its adjustment variables, or the detailed method used to obtain the primary interval. That limits independent reconstruction of the sponsor’s analysis from the posted summary statistics. The sponsor comparison remains the relevant posted result for the primary endpoint; the recalculation cannot fill in undocumented features of that analysis.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and multiple secondary outcomes, but it does not report whether statistical testing was adjusted for multiplicity. Multiplicity arises when several hypotheses or outcomes are examined: even if each individual test uses a conventional threshold, the chance of at least one apparently unusual result across a collection of tests can increase. General approaches include prespecifying a testing sequence, adjusting thresholds, or treating secondary findings as exploratory. The record does not identify which, if any, approach was used. The half-life recalculation is explicitly unadjusted and should therefore be read as one exploratory comparison, not as a multiplicity-controlled finding.

The primary analysis population is defined as randomized and treated participants with adequately evaluable pharmacokinetic parameters. The primary endpoint summaries include 105 participants in the autoinjector group and 111 in the prefilled-syringe group. For half-life, the group counts are 104 and 111. The differences in analysis counts indicate that the endpoint summaries do not all use identical numbers of participants, but the reasons are not reported in the registry record. The record also does not report how missing concentration samples, incomplete profiles, or other unavailable endpoint data were handled, or whether any imputation was performed.

The PK-evaluable definition is relevant to interpretation because it is not simply described as all randomized participants. Requiring adequate PK data can make a pharmacokinetic analysis possible, but it also means the estimate pertains to the analyzed population. Without details about exclusions or missingness, it is not possible to assess whether the omitted data could have changed the result or whether the analysis population remained comparable across assigned groups.

The registry record does not report an intention-to-treat analysis for the primary outcome, sensitivity analyses for missing data, interim analyses, or a complete statistical analysis plan. Nor does it report a prespecified equivalence margin for the device comparison. These omissions do not establish that such procedures were absent from study conduct; they mean only that the registry record does not document them. The confidence interval and endpoint counts are therefore the most direct basis for interpreting the posted primary result, while conclusions about multiplicity control and robustness to missing data remain limited.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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

Serious events with the highest arm-level rate: rheumatoid arthritis (0.9%).

The posted serious adverse event counts are 0 among 111 participants at risk in the autoinjector group and 1 among 113 participants at risk in the prefilled-syringe group. The most frequently listed serious-event term is rheumatoid arthritis, reported as 0.9%. The record also reports 0 participants with other adverse events in each of the two safety groups. Death counts are not reported in the registry record.

These counts are descriptive observations during the study, not estimates of comparative risk with strong precision. In particular, a small number of serious events gives little information about uncommon harms, and zero recorded events in a group does not demonstrate that the underlying risk is zero. The safety denominators differ from the primary PK analysis counts; the registry does not explain the reasons for those differences. The time period and detailed event-collection procedures for the safety summaries are not reported in the registry record.

The serious-event term and percentage should be read as registry-reported safety information, not as evidence that the device caused the event. This study compared delivery devices for the same drug and dose, and the record does not report an analysis establishing causal attribution. The small number of reported serious events also means that the safety summaries cannot establish that the devices have identical safety profiles.

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

The result describes serum exposure after a single subcutaneous dose of abatacept 125 mg administered by one of two devices. The primary estimate is a ratio of geometric mean Cmax values, with a 90% confidence interval that includes 1. This means the posted comparison does not clearly distinguish the groups on the primary pharmacokinetic measure at the precision reported. The interval also does not prove equality: it allows uncertainty around the estimated ratio, and the registry record does not report a margin for declaring equivalence.

This is not a trial of abatacept versus placebo, no treatment, or another medication. Both groups received the same named drug and stated dose, while the delivery device differed. The primary endpoint is drug concentration, not improvement in rheumatoid arthritis. The registry record does not report a primary clinical outcome such as symptoms or disease activity, so the Cmax result cannot be interpreted as a comparison of clinical effectiveness.

The study concerns a single dose and pharmacokinetic observations extending through Day 71. It does not by itself answer how devices compare during repeated administration, whether exposure remains similar over longer periods, or whether device choice changes clinical outcomes. The posted safety counts are also limited: the registry reports one serious adverse event in one group and none in the other, while deaths are not reported. These observations do not determine the likelihood of rare or longer-term events.

The secondary half-life estimate is an unadjusted recalculation and has an interval that includes no mean difference. It is useful as supporting descriptive information, but it does not repair the limitations of the primary comparison or establish a broad conclusion across all secondary outcomes. The registry does not report the full analysis plan, missing-data procedures, multiplicity approach, or equivalence criteria. Accordingly, the defensible interpretation is narrow: the record reports a comparison of peak serum concentration between two delivery devices after one dose, with uncertainty that does not establish either a clear difference or formal equivalence.

14. Why This Trial Matters Statistically

This study illustrates why a statistical question must be matched to the outcome actually measured. A pharmacokinetic trial can compare concentration profiles and features of drug disposition without directly testing clinical outcomes. Here, the principal endpoint is Cmax, a measure of peak observed serum concentration. The interpretation therefore concerns exposure after administration, not whether participants experienced a clinical benefit.

It also illustrates the importance of distinguishing a ratio from a difference. For geometric-mean concentrations, a ratio provides a natural comparison, and 1 represents equal group means on that relative scale. The reported ratio and its confidence interval communicate both the estimated comparison and its uncertainty. Because the interval includes 1, the data do not clearly separate the groups on the primary PK measure. At the same time, absence of a clear difference is not equivalent to proof of equivalence. An equivalence conclusion needs a prespecified margin and an analysis built around that question; the registry record does not report such a margin.

Analysis populations matter as well. The registered enrollment is 356, whereas the primary PK summaries include 105 participants in the autoinjector group and 111 in the prefilled-syringe group. The endpoint population requires evaluable PK parameters, and the record does not explain the exclusions or missing-data handling. Readers should interpret the estimate as applying to the posted analysis population rather than assume it represents every enrolled participant without qualification.

Finally, this record shows why secondary findings and sample-size illustrations require context. An unadjusted secondary recalculation is not the same as a prespecified primary test, and a sample-size illustration based on an observed effect is a planning aid rather than observed power or proof that the original study was adequately sized. Clear distinctions among endpoint, population, uncertainty, and analysis purpose make it possible to describe what the trial reports without overstating what it establishes.

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 study’s primary question is narrow: it characterizes Cmax after a single dose and compares the two delivery devices. It does not establish how the devices compare for clinical outcomes, repeated-dose use, or longer-term exposure. The registry record describes observations through Day 71 but does not provide a primary clinical endpoint. The result should therefore remain within its pharmacokinetic scope.

The primary analysis is based on a PK-evaluable population rather than a stated all-randomized population. The registry gives the definition in general terms but does not report the detailed criteria for adequate evaluability, the reasons for missing or excluded measurements, or how incomplete profiles were handled. The primary arm counts are lower than the registered enrollment, and the record does not explain the difference. Endpoint-specific counts also vary, including for the half-life outcome. This limits assessment of how analysis-population selection may affect interpretation.

The registry does not report the full statistical model for the adjusted geometric means, adjustment variables, interval-construction method, or a complete statistical analysis plan. It also does not report a multiplicity procedure, an equivalence margin, interim analyses, or the study’s original sample-size justification. These are documentation limits; they do not show that the procedures were not conducted. However, they prevent independent evaluation of those aspects using the registry record alone.

The secondary evidence is also limited. The half-life comparison is an unadjusted recalculation, and the registry does not report a sponsor analysis for it. Multiple outcome domains are listed, but the record does not support a single combined secondary conclusion. Safety counts are sparse, and deaths are not reported in the registry record. The open-label design is explicit, but the registry does not provide further information about how knowledge of device assignment may have affected study procedures or outcome assessment. Finally, the study is described as multicenter, but the number and locations of centers are not reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study to Characterize the Pharmacokinetics of a Single Dose of SC Abatacept 125 mg Using the BD Autoinjector or the Prefilled Syringe — NCT01890473.
• Registry record last updated 2015-11-26; status last verified 2015-10.
• Results first posted 2015-11-26.
• 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 record describes NCT01890473, a completed, Phase Ib, randomized, open-label, parallel-group study sponsored by Bristol-Myers Squibb. It enrolled 356 participants with rheumatoid arthritis and compared a single subcutaneous dose of abatacept 125 mg delivered using an autoinjector or a prefilled syringe.

The primary outcome was adjusted geometric mean serum Cmax in the PK-evaluable population over Day 1 to Day 71. The posted means were 11.5 μg/mL for the autoinjector and 12.6 μg/mL for the prefilled syringe; the reported geometric-mean ratio was 0.908, with a 90% confidence interval from 0.815 to 1.01. The interval includes 1 and does not establish formal equivalence.

Secondary outcomes included additional pharmacokinetic measures and safety, immunogenicity, and laboratory outcomes. The terminal half-life recalculation is unadjusted and should be regarded as secondary context. The registry record does not report the full analysis plan, multiplicity approach, or detailed missing-data procedures.