Long-Term Safety Study Of Tofacitinib In Patients With Juvenile Idiopathic Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Long-Term Safety Study Of Tofacitinib In Patients With Juvenile Idiopathic Arthritis (NCT01500551), a na phase 2, phase 3 study. The primary outcome is Number of Participants With Laboratory Test Abnormalities.

2. Clinical Question

3. Trial Design

This is a long-term, open-label follow-up study of tofacitinib in people with juvenile idiopathic arthritis who had previously participated in tofacitinib JIA studies. The registry identifies the design as single group, with no masking and no allocation to different treatment arms. All participants were assigned to the tofacitinib treatment group. The record lists Phase 2 and Phase 3, an enrollment of 302, and a completed status.

The study began on 2013-03-18, and the primary completion date is 2025-02-12. The primary outcome timeframe is during study treatment, with a maximum of up to 118 months. Those details establish a long follow-up window, but do not by themselves show how long any particular participant remained under observation. Individual follow-up durations and the distribution of time in the study are not reported in the registry record.

The registry’s stated purpose is to evaluate long-term safety and tolerability in participants previously involved in tofacitinib JIA studies. The primary endpoint counts participants with at least one laboratory abnormality. The record also lists secondary outcomes describing changes in physician global evaluation of disease activity and number of joints with active arthritis, at specified follow-up months and in named analysis sets. These listings indicate that clinical measures were included, but numerical results for those outcomes are not reported in the registry record in the information available for interpretation.

Because the study has one treatment group, changes during follow-up cannot be contrasted with outcomes in a concurrent untreated or active-control group. Open-label conduct also means masking was absent. These design features do not invalidate descriptive reporting, but they limit what can be inferred about comparative treatment effects. The registry record does not report a more detailed protocol, a sample-size justification, or additional design features beyond those stated above.

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 302 eligible participants who had previously completed any of the 3 qualifying/index Juvenile Idiopathic Arthritis (JIA) clinical studies of tofacitinib (A3921103 \[NCT01513902\], A3921104 \[NCT02592434\], or A3921165 \[NCT03000439\]) were enrolled in the current extension study A3921145 (NCT01500551).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety analysis set (SAS) included all enrolled participants with at least one dose of tofacitinib during the current study.

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 Laboratory Test Abnormalities, measured over During study treatment (maximum up to 118 months).

The primary endpoint is the number of participants with at least one laboratory test abnormality during study treatment, with follow-up described as a maximum of up to 118 months. It is a participant-level count: a participant is counted if at least one listed laboratory parameter is abnormal. The listed parameters span hematology, clinical chemistry, and urinalysis. They include blood-cell measures, liver and kidney-related tests, lipids, electrolytes, glucose, urine characteristics, and other specified tests.

General statistical context: Laboratory outcomes are commonly used in safety monitoring because they can identify changes in measured biological markers during treatment. A summary based on participants with one or more abnormalities answers a broad occurrence question. It does not, by itself, describe the number of abnormal tests per person, how severe or persistent an abnormality was, when it first occurred, or whether it was considered related to treatment. Those details are not reported in the registry record for this endpoint.

The registry defines the primary-outcome population as the safety analysis set: enrolled participants with at least one dose during the current study. The numerical endpoint result is not reported in the registry record. Accordingly, the endpoint definition can be described, but the posted record does not establish how many participants met it or how precisely that frequency was estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a single-group study and describes the primary endpoint as the number of participants with at least one laboratory test abnormality. The posted safety summaries are counts, rather than treatment-effect estimates from a randomized comparison. No inferential analysis results for the primary endpoint are reported in the registry record. The interpretation should therefore remain descriptive.

What does a participant count represent?

General statistical context: A count of participants with an event treats each participant as the unit of summary. For an endpoint defined as at least one abnormality, each person is classified according to whether any qualifying abnormality occurred. Multiple abnormal tests in one person do not necessarily add multiple participants to that count. This summary is useful for communicating how many people were recorded as meeting the endpoint, but it does not retain the number, type, duration, or severity of abnormalities for each person.

The registry’s endpoint description lists many tests across hematology, clinical chemistry, and urinalysis. A broad “at least one” endpoint groups those different measurements together. Its meaning depends on the criteria used to label an individual result abnormal, the timing and frequency of testing, and the handling of repeated results. Specific thresholds, rules for repeated abnormalities, and details of test-level data handling are not reported in the registry record. The endpoint count therefore cannot be interpreted as a count of clinically important events unless that classification is separately described.

Why is a comparator important?

General statistical context: A treatment comparison estimates how outcomes differ between groups that received different interventions. Random allocation can help make groups comparable on average, while a concurrent control can provide information about outcomes over the same study period in the absence of the study treatment. Here, the registry describes one open-label tofacitinib group, with allocation listed as not applicable. There is no concurrent control arm in the record. A count from this group can describe what was recorded among participants, but it cannot estimate a between-group difference.

Single-group follow-up can be informative for describing events over extended observation, especially when the objective is safety and tolerability. At the same time, interpreting events requires care: events can occur in the condition being studied, during background care, or for other reasons. The record does not provide a comparator that would separate these possibilities, nor does the posted information provide causal attribution for each event. The design itself should guide the strength of any conclusion drawn from the observed counts.

What can uncertainty estimates add?

General statistical context: A confidence interval expresses the uncertainty around an estimated quantity under a stated statistical method. It is not a guarantee that the true value lies in the interval, and its width depends on the amount and structure of information and on the method used. For a participant-level event proportion, an interval can help show how much precision a count affords. No confidence interval or other uncertainty interval for the primary laboratory endpoint is reported in the registry record. No primary endpoint estimate is reported there either, so its precision cannot be assessed from this record.

For safety counts, the number of affected participants is a useful descriptive starting point. Interpretation also depends on the observation time and completeness of follow-up. The primary outcome allows a maximum timeframe of up to 118 months, but participant-level follow-up durations and the approach to incomplete observation are not reported in the registry record. Without those details, counts alone do not describe event rates over person-time or make participants with different observation durations directly comparable.

How should the secondary outcome schedule be read?

The registry lists physician global evaluation of disease activity and number of joints with active arthritis at repeated follow-up months. The listed analysis sets include sJIA1, sJIA2, pJIA, enthesitis related arthritis, and psoriatic arthritis. The pJIA listings extend to Month 78 for the stated outcomes, while other listed sets have their own schedules. These are outcome definitions and planned assessment points in the registry, not evidence that every measure has a posted result at every time.

General statistical context: Repeated measurements from the same participant are correlated; they are not independent observations. Analyses of change over time often need methods that account for within-person correlation and for participants who do not have measurements at every visit. The registry record does not report the analysis method, model assumptions, or handling of missing repeated measurements. It also does not report numerical estimates for these listed secondary outcomes. No particular longitudinal method should be attributed to the study on the basis of the outcome schedule alone.

What do the planning illustrations mean?

General statistical context: A sample-size illustration asks what group size might be needed under stated assumptions to detect a specified effect with a chosen statistical procedure. It is a planning exercise, not an observed finding and not a measure of the probability that a completed study detected an effect. The record describes illustrations based on the observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is a stated fraction of the observed effect. Such assumptions describe the illustration, not the trial’s own design. Since this study is single group and has no like-for-like comparator, those hypothetical two-arm calculations do not convert its observed data into a controlled treatment comparison.

For non-time-to-event outcomes, the stated planning convention is participants per arm; time-to-event calculations use a total number of events. No specific calculated sample-size values are reported in the information available for this interpretation. The registry record does not report a sample-size justification for the actual study. These planning conventions should not be confused with observed enrollment, observed outcomes, or observed power.

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("Tofacitinib 5 mg BID (Open-label)")
n      <- c(302)
events <- c(294)
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  Tofacitinib 5 mg BID (Open-label)  302  294  97.4  94.8  98.8  94.9  98.7  94.9  98.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists numerous secondary outcome measures and repeated assessment months. Physician global evaluation of disease activity is listed for sJIA1, sJIA2, pJIA, enthesitis related arthritis, and psoriatic arthritis analysis sets. Active-joint counts are listed for several of those sets as well. The record reports 139 outcome measures and 0 analyses. A list of multiple outcomes does not itself establish which outcomes were analyzed, which were emphasized, or what inferential procedures were applied.

General statistical context: When many outcomes, subgroups, or time points are tested, the chance of obtaining at least one apparently notable result can increase if each is interpreted without regard to the full set of comparisons. Statistical approaches may control or describe this multiplicity, but the appropriate approach depends on the intended questions and analysis plan. A multiplicity strategy, hierarchy of endpoints, or adjustment method is not reported in the registry record. The posted outcome list therefore should not be treated as a set of independently confirmed findings.

The registry names the primary safety analysis set as enrolled participants who received at least one dose during the current study. It does not report the number included in that set as a separate primary-endpoint denominator, the number with endpoint data, or the number excluded from an analysis. The event summaries list 302 participants at risk in each adverse-event category, but those risk-set counts do not fill in missing details about the laboratory endpoint or secondary analyses.

Rules for missing laboratory measurements, missed visits, withdrawal, and incomplete follow-up are not reported in the registry record. Nor are methods for handling missing repeated secondary-outcome measurements, or details of the analysis populations for each measure. These omissions matter because observation time and availability of measurements can affect descriptive summaries and longitudinal estimates. No conclusion about the effect of missing data can be made from the outcome schedule alone.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: still's disease (2.6%), suicidal ideation (1.7%), herpes zoster (1.3%), pyrexia (1.0%), appendicitis noninfective (0.7%).

The adverse-event summary reports 48 participants with a serious adverse event among 302 at risk. It also reports 242 participants affected in the other adverse-event summary, among 302 at risk. These figures are counts of participants, not counts of every episode. A participant may have experienced more than one event; the registry record does not provide a total number of episodes or the individual-level overlap between the summaries.

The record reports 2 deaths. It does not report the causes, timing, or treatment attribution of those deaths in the information. The serious-event term list gives Still’s disease at 2.6%, suicidal ideation at 1.7%, herpes zoster at 1.3%, pyrexia at 1.0%, appendicitis noninfective at 0.7%, and haemophagocytic lymphohistiocytosis at 0.7%. Those percentages belong to the posted term-level summary; the registry information does not provide additional context about event recurrence or attribution.

General statistical context: A serious adverse event is a category used in safety reporting, while seriousness is not the same concept as intensity or severity. Event counts describe what was recorded among participants under observation; without a control group and details of exposure time, they do not establish comparative risk or causation. The study is open-label and single group, so these figures should be read as descriptive safety data from the tofacitinib group.

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

The record establishes that this was a completed, open-label, single-group follow-up study involving 302 enrolled participants with JIA who had previously participated in tofacitinib JIA studies. The study’s stated purpose was to evaluate long-term safety and tolerability. Its primary outcome was defined as participants with at least one laboratory test abnormality during treatment, with a maximum timeframe of up to 118 months. These design and endpoint details describe the question the registry set out to address.

The posted information also gives adverse-event counts: 48 participants with a serious adverse event among 302 at risk, 242 participants affected in the other adverse-event summary among 302 at risk, and 2 deaths. It lists several serious adverse-event terms with percentages. These observations describe events reported in the treatment group. They do not establish that tofacitinib caused any particular event, nor do they show whether the frequency was higher or lower than it would have been under another treatment or without treatment.

The primary laboratory endpoint’s numerical result is not reported in the registry record. Consequently, the record does not show the number of participants who had at least one laboratory abnormality, and it does not permit assessment of the precision of that primary outcome. The registry’s secondary outcome listings similarly do not supply numerical findings in the information available for interpretation. Listing repeated assessment months and analysis populations is not the same as reporting changes or their uncertainty.

General statistical context: Safety data from long-term follow-up can describe the types of events observed and the participants in whom they were recorded. Their interpretation is strengthened by clear denominators, observation time, definitions, missing-data procedures, and suitable comparative information when the question is comparative. Here, a control group and posted inferential results are absent. The study therefore contributes descriptive follow-up information, but the posted record does not establish a comparative treatment effect, a causal relationship, or a complete numerical result for the primary endpoint.

14. Why This Trial Matters Statistically

Long-term follow-up raises statistical questions that differ from those in a brief controlled comparison. Participants may be observed for different amounts of time, repeated measurements from the same person are related, and safety outcomes may occur more than once. A clear account of denominators, observation windows, and analysis populations helps readers understand what a count represents. This record states a maximum primary-outcome timeframe of up to 118 months and defines the safety analysis set, but does not report individual follow-up durations or detailed missing-data methods.

The single-group design is also central to statistical interpretation. It permits description of events observed in the tofacitinib group but does not provide a concurrent estimate of what would have occurred in a comparison group. That distinction is important in safety research, where an observed event count can be clinically relevant as a description without answering whether the treatment changed the event frequency. The record’s counts should therefore be read in light of the design rather than as comparative results.

The primary endpoint combines a wide range of laboratory parameters into a participant-level “at least one abnormality” outcome. Such a summary can provide a broad safety measure, but it cannot convey the pattern of abnormalities by test, duration, severity, or participant without further results. The registry lists many secondary measures and visit times, yet reports 0 analyses. This makes the difference between an outcome being listed and an outcome having an interpretable posted estimate especially important.

General statistical context: Transparent reporting of analysis methods, uncertainty, missingness, and multiplicity allows readers to judge both the strength and limits of results. The information in this record supports a careful account of study structure and selected safety counts. It also illustrates why descriptive findings, planned outcomes, and comparative evidence should be kept distinct when interpreting a long-term clinical study.

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 most consequential limitation for interpreting the primary endpoint is that its numerical result is not reported in the registry record. The endpoint definition is detailed, but there is no posted count of participants with at least one laboratory abnormality or corresponding uncertainty estimate. The record therefore cannot support a quantitative conclusion about that outcome.

The design is single group, open label, and has no like-for-like control comparison. The adverse-event counts describe observations among participants in the tofacitinib group; they cannot show how those counts compare with another intervention or background occurrence. The record does not provide treatment attribution for the listed events or deaths. The two reported deaths have no cause or timing details in the available information.

Although the primary endpoint allows follow-up for a maximum of up to 118 months, individual observation durations and person-time are not reported. Missing-data rules, withdrawal handling, laboratory abnormality thresholds, and analysis details are also not reported in the registry record. Those gaps limit assessment of how complete and comparable the observed follow-up was.

The secondary outcomes span multiple measures, analysis sets, and follow-up months, but numerical results and analysis methods are not reported in the registry record. A multiplicity approach is not reported either. The record also does not report a sample-size justification for the actual study. These limitations do not erase the value of the descriptive safety counts, but they constrain precision, causal interpretation, and the ability to evaluate the full outcome set.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Long-Term Safety Study Of Tofacitinib In Patients With Juvenile Idiopathic Arthritis — NCT01500551.
• Registry record last updated 2026-03-23; status last verified 2026-03.
• Results first posted 2026-03-23.
• 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

NCT01500551 is a completed, open-label, single-group follow-up study of tofacitinib in juvenile idiopathic arthritis. The registry lists Phase 2 and Phase 3 and an enrollment of 302. Participants had previously taken part in tofacitinib JIA studies, and the stated purpose was long-term safety and tolerability evaluation.

The primary endpoint counts participants with at least one laboratory abnormality during treatment, over a maximum timeframe of up to 118 months. Its numerical result is not reported in the registry record. The adverse-event summary reports 48 participants with a serious adverse event and 242 participants affected in the other adverse-event summary, each among 302 at risk, along with 2 deaths. The single-group design supports descriptive reporting, not a controlled treatment comparison.