A Study to Evaluate Biomarkers to Predict Efficacy of Abatacept in Rheumatoid Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate Biomarkers to Predict Efficacy of Abatacept in Rheumatoid Arthritis (NCT03882008), a na phase 4 study. The primary outcome is Number of Participants With American College of Rheumatology (ACR) 20 Response at Week 14.

2. Clinical Question

3. Trial Design

This is a phase IV, open-label, single-group study of abatacept in people with rheumatoid arthritis. The registry record lists University of Washington as the sponsor and identifies the condition as rheumatoid arthritis. Enrollment is listed as 25. The study began on 2019-05-23, its primary completion date is 2023-06-30, and its status is completed.

Participants received abatacept as a 125mg subcutaneous injection weekly for 24 weeks. Because the record lists a single group, there is no separate untreated, placebo, or active-treatment comparator in the design described. Allocation is listed as not applicable, and masking as none; the official title also describes the study as open-label. The record does not report a randomized treatment allocation or a blinded comparison.

The stated primary objective is to evaluate whether baseline levels of T cell-associated biomarkers predict abatacept efficacy during treatment. The primary outcome is the number of participants with an ACR20 response at Week 14. The outcome description explains the response criteria, but does not report the observed response count or the biomarker analysis. Secondary outcomes are ACR20, ACR50, and ACR70 responses and good or moderate EULAR response at Week 24. These are listed as outcome measures, not as reported results in the registry record for this analysis.

A single-group biomarker study can describe outcomes among participants receiving the study treatment and investigate whether baseline measurements are associated with later response. Without a concurrent comparison group, it cannot isolate treatment effects from other explanations for changes over time. The registry record does not report further protocol details, the planned statistical model, or how biomarker levels were measured or classified.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With American College of Rheumatology (ACR) 20 Response at Week 14, measured over 14 Weeks.

The primary endpoint is the number of participants with an American College of Rheumatology (ACR) 20 response at Week 14. ACR20 is a composite response definition: it requires improvement of 20% in both tender and swollen joint counts, along with 20% improvement in three of five additional criteria. Those criteria include patient global assessment, physician global assessment, pain, functional ability measure, and erythrocyte sedimentation rate or C-reactive protein.

This kind of endpoint is used to classify whether a participant meets a defined threshold of improvement across joint counts and other aspects of rheumatoid arthritis assessment. It combines multiple clinical measures rather than relying on a single symptom or laboratory measure. The registry’s stated question is not only whether participants meet the response definition, but whether baseline T cell-associated biomarker levels predict that response during abatacept treatment.

The endpoint title specifies a participant count, but the primary response count is not reported in the registry record. The record also does not report an estimate of the association between biomarker levels and response, or the precision of such an estimate. A response definition alone cannot establish whether a biomarker predicts response: that requires reported data and an analysis connecting baseline biomarker measurements to participant outcomes. The Week 24 ACR and EULAR outcomes are secondary measures, distinct from the primary ACR20 endpoint at Week 14.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a binary response count describe?

ACR20 response is a binary outcome at the participant level: each participant either meets the defined response criteria at the assessment or does not. A count reports how many participants met the definition. If a count is accompanied by a denominator, the pair can be used to describe the observed response frequency in that group. A count by itself does not describe uncertainty, compare groups, or explain why participants had different outcomes.

For this study, the primary outcome is the number of participants with ACR20 response at Week 14. The registry record does not report that count. It also does not report a response proportion, confidence interval, or inferential test for the primary endpoint. Therefore, the endpoint definition can be explained, but its observed efficacy result cannot be interpreted from the record.

What would a biomarker prediction analysis address?

Prediction asks whether information measured before treatment is useful for anticipating a later outcome. Here, the stated candidate information is baseline T cell-associated biomarker levels, and the outcome is ACR20 response during abatacept treatment. A suitable analysis would need to connect the baseline measurements with participants’ response status and describe how that relationship was estimated. For example, the analysis plan would need to clarify how continuous or categorized biomarker values are represented, which outcomes and covariates are included, and how uncertainty is quantified. These details are not reported in the registry record, so no specific model or predictive performance measure can be attributed to this study.

An association is not automatically evidence that a biomarker can predict accurately in new participants. Prediction performance concerns how well a model or marker distinguishes outcomes and how reliable its predictions are beyond the data used to develop it. Assessment of that performance depends on methods and validation information. Neither the model, validation approach, nor prediction metrics are reported in the registry record. The study objective alone does not show that a predictive relationship was found.

How should the single-group structure affect interpretation?

The record lists a single-group study with no masking and no allocation to a comparator. In this setting, descriptions of outcomes refer to participants treated with abatacept. A within-group response count does not estimate a contrast against another treatment or against no treatment. In general, a concurrent comparison helps distinguish treatment effects from changes over time, background variation, and other influences. Because this study does not have such a comparator in the registry description, its design does not support a randomized treatment-effect estimate.

The absence of a comparator does not prevent an examination of whether baseline biomarkers are associated with outcomes among treated participants. However, it limits the interpretation of that association: the design described does not determine whether a biomarker predicts response specifically to abatacept rather than response patterns that might also occur under another treatment or in the absence of treatment. The registry record does not describe a design feature that would address that distinction.

What does precision mean?

A confidence interval is a range of values compatible with the data under the assumptions of the statistical method. Its width conveys precision: a wider interval reflects greater uncertainty about the estimated quantity, while a narrower interval reflects less uncertainty under that method. A point estimate without an interval can conceal substantial uncertainty, especially when the available study group is limited. No confidence interval for the primary endpoint or biomarker relationship is reported in the registry record.

For a binary endpoint, precision depends on the amount of outcome information and on the number of participants in each outcome category, as well as the chosen analysis. For a prediction question, precision also depends on how many response and nonresponse observations inform the estimated relationship and on the complexity of the analysis. The registry record does not report the response distribution or the details needed to evaluate these features. No precision estimate should therefore be inferred from enrollment alone.

What are the sample-size illustrations?

The planning illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They describe hypothetical planning for a stated fraction of an observed effect; they are not the study’s own design. The stated method gives participants per arm for non-time-to-event outcomes. The actual study is single-group, so an equal-allocation, two-arm planning scenario does not describe its treatment allocation.

Such sample-size calculations are planning illustrations, not observed power. Observed power is not a substitute for reporting an effect estimate and its uncertainty after the study. The registry record does not report a sample-size justification for the trial, and the planning assumptions should not be mistaken for the protocol’s design or evidence about the precision of a biomarker result. The calculation details do not establish that the study was designed to test any particular biomarker association.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Abatacept")
n      <- c(23)
events <- c(9)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Abatacept  23  9  39.1  19.7  61.5  22.2  59.2  22.1  59.4

11. Multiplicity, Missing Data and Analysis Populations

The study lists a primary outcome and several secondary outcomes. The primary outcome is ACR20 response at Week 14. The secondary outcomes are ACR20, ACR50, and ACR70 response and good or moderate EULAR response at Week 24. Examining multiple outcomes can raise a multiplicity issue: when multiple hypotheses are tested, the chance of at least one apparently notable result can increase unless the testing strategy accounts for the number and relationship of the analyses. The registry record does not report a multiplicity adjustment, a testing hierarchy, or a prespecified approach for interpreting multiple outcomes.

The stated objective also concerns whether baseline T cell-associated biomarkers predict response. The record does not identify the biomarkers, state how many were evaluated, or report how multiple biomarker tests were handled. It does not report whether any analysis was adjusted for other participant characteristics. These details are not reported in the registry record, so the risk of selective emphasis or chance findings across multiple analyses cannot be assessed from the registry entries.

Missing-data methods and analysis populations are also not reported in the registry record. The record does not describe how missing Week 14 or Week 24 assessments would be handled, whether participants with incomplete data were included in an analysis, or which population was used for efficacy analyses. No imputation method or sensitivity analysis is reported. These omissions matter because incomplete follow-up can affect response counts and the estimated relationship between baseline biomarkers and outcomes. The registry record lists no analysis entries; that does not establish what methods may have been used outside the registry record.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

For the abatacept group, the registry record lists 25 participants at risk for serious adverse events and 0 affected. It lists 25 at risk for other adverse events and 8 affected, along with 0 deaths. These are counts as recorded, not estimates of comparative risk: the study has no concurrent control group in the design described. No serious adverse-event terms are listed.

The record does not give event-level descriptions, timing, duration, severity details beyond the serious-event category, or an assessment of whether events were considered related to abatacept. The figures therefore do not explain the clinical character or course of the events. They also cannot be used to compare safety with another treatment or with an untreated group. A count of 0 recorded serious events does not demonstrate that serious events are impossible or rule out events not captured in the record. The safety information should be read as the registry’s descriptive reporting for this group, rather than as a comparative safety conclusion.

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

The record establishes that a phase IV, open-label, single-group study was completed to evaluate whether baseline T cell-associated biomarkers predict abatacept response in rheumatoid arthritis. It specifies ACR20 response at Week 14 as the primary outcome and identifies additional response outcomes at Week 24. It also reports enrollment and a limited set of safety counts. These details describe the study’s intended question, design, and recorded safety information.

The record does not report the primary ACR20 response count, an estimate of biomarker prediction, or the uncertainty around such an estimate. It therefore does not show whether any biomarker was associated with response, how strong such an association might have been, or whether it could predict outcomes reliably. The stated objective is not itself a finding. Nor do the listed secondary outcomes establish results for those measures.

The single-group design also limits what can be concluded about treatment effects. Outcomes among participants receiving abatacept are not a randomized comparison against another treatment or no treatment. Even if outcome changes were available, the design described would not by itself establish that abatacept caused them. Likewise, a relationship between a baseline marker and response among treated participants would not necessarily show that the marker predicts a response specific to abatacept.

The safety counts are descriptive and do not provide a comparative assessment. The absence of a reported serious adverse event in this group does not establish that such events cannot occur, and the count for other adverse events does not provide details about their nature or attribution. Overall, the registry entries support a description of the study design and reporting, but not a conclusion about treatment efficacy, biomarker performance, or comparative safety.

14. Why This Trial Matters Statistically

This study is statistically interesting because its central question is predictive rather than simply descriptive. It asks whether a baseline biological measurement can help anticipate a later clinical response. Answering that question requires more than reporting how many participants meet ACR20 criteria: the baseline marker data must be connected to each participant’s outcome, and the estimated relationship must be reported with enough detail to judge its uncertainty and predictive usefulness.

The design also illustrates why the target of inference matters. A single-group study can provide information about outcomes among people receiving abatacept and can explore relationships between baseline measurements and those outcomes. It cannot offer the same treatment comparison as a randomized, controlled design. A marker associated with response in a treated group is not necessarily a marker that identifies who benefits specifically from that treatment. That distinction is important when a study’s objective uses the language of predicting treatment efficacy.

Endpoint definition and timing matter as well. The primary outcome is a composite ACR20 response at Week 14, while the listed ACR and EULAR secondary outcomes are assessed at Week 24. These outcomes summarize different response thresholds or criteria and are not interchangeable. Clear reporting of the prespecified outcome, analysis population, missing-data approach, and biomarker methods would help readers understand the evidence and its precision.

In this record, the objective and endpoints are described, but the primary result and biomarker analyses are not reported. This makes the registry entry useful for understanding the question and structure of the study, while leaving the central predictive question unanswered by the registry entries.

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 principal limitation for interpreting the efficacy question is that the registry record does not report the primary ACR20 response count or an analysis relating baseline biomarker levels to response. Biomarker identities, measurement procedures, analysis methods, effect estimates, and uncertainty estimates are also not reported in the registry record. Without these, the predictive objective cannot be evaluated from the posted results.

The single-group, open-label design has no concurrent comparator described. It can characterize outcomes within the abatacept group, but cannot support a randomized comparison with another treatment or no treatment. The design also does not establish that a baseline biomarker predicts a treatment-specific benefit, rather than a response pattern among participants receiving abatacept.

Enrollment is listed as 25, but the registry record does not report the number contributing to each efficacy analysis, the response distribution, or the analysis population. Missing-data handling, follow-up completeness, and sensitivity analyses are not reported in the registry record. These omissions prevent assessment of how incomplete observations might influence the primary endpoint or biomarker association.

The record lists secondary outcomes at Week 24 but does not report their results or explain how multiple outcomes or biomarker tests were addressed. It also does not report a trial-specific sample-size justification or the statistical analysis plan. The registry’s safety reporting gives counts but not event-level terms or attribution; no serious adverse-event terms are listed. These limits mean the available record supports a careful description of the study, not a complete assessment of efficacy, prediction, or safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Evaluate Biomarkers to Predict Efficacy of Abatacept in Rheumatoid Arthritis — NCT03882008.
• Registry record last updated 2024-07-30; status last verified 2024-07.
• Results first posted 2024-07-30.
• 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

NCT03882008 is a completed phase IV, open-label, single-group study sponsored by University of Washington. It enrolled 25 participants with rheumatoid arthritis and evaluated abatacept, described as a 125mg subcutaneous injection weekly for 24 weeks. The stated objective was to assess whether baseline T cell-associated biomarkers predict efficacy in people with moderate-to-severe active rheumatoid arthritis who had an inadequate response to cDMARDs.

The primary outcome was the number of participants with ACR20 response at Week 14. Secondary outcomes were ACR20, ACR50, and ACR70 response and good or moderate EULAR response at Week 24. The primary outcome count and biomarker analysis results are not reported in the registry record. Safety reporting lists 0 participants with serious adverse events, 8 with other adverse events, and 0 deaths in the abatacept group. The record lists no analysis entries.