A Study to Evaluate the Safety and Effectiveness of Ustekinumab or Golimumab Administered Subcutaneously (SC) in Patients With Sarcoidosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Safety and Effectiveness of Ustekinumab or Golimumab Administered Subcutaneously (SC) in Patients With Sarcoidosis (NCT00955279), a randomized phase 2 study. The primary outcome is Change From Baseline in Percent-predicted Forced Vital Capacity (FVC) at Week 16.

2. Clinical Question

The question is about a measured change in lung function, not a direct measure of every aspect of sarcoidosis or of overall health. FVC is the volume of air forcibly exhaled after a full breath in the upright position. Expressing it as a percentage of predicted FVC places the observed value in relation to an estimate based on sex, age and height. The registry specifies that change was calculated as the Week 16 value minus baseline, so a positive change represents a higher value at Week 16 than at baseline.

The study also included participants with skin involvement and posted secondary measures of walking distance, respiratory questionnaire score, skin assessment and FVC at Week 28. Those outcomes address different aspects of the condition and were measured at a different time point from the primary endpoint. The registry record does not report the original sample-size justification or enough detail to establish the full prespecified analysis plan.

3. Trial Design

This was a Phase 2, multicenter, randomized, double-blind, parallel-group, placebo-controlled study of ustekinumab and golimumab in chronic sarcoidosis. Quadruple masking is recorded in the registry. The trial enrolled 173 participants, and its status is listed as completed. The registry lists the start as 2009-11 and primary completion as 2012-05. It describes approximately 40 sites globally.

Participants were allocated to placebo, golimumab or ustekinumab. All treatments were administered subcutaneously. The golimumab schedule began with 200 mg at Week 0, followed by 100 mg every 4 weeks through Week 24. The ustekinumab schedule began with 180 mg at Week 0 and continued with 90 mg at Weeks 8, 16 and 24; matching placebo was administered at Weeks 4, 12 and 20. Matching placebo was administered every 4 weeks through Week 24 in the placebo group.

The registry distinguishes a primary pulmonary population from a secondary skin population, while noting that one participant could be included in both. The primary endpoint was assessed in the modified intent-to-treat population, defined as randomized participants who received at least one dose of study medication. The arm sizes posted for that endpoint are 44 for placebo, 42 for golimumab and 46 for ustekinumab. The skin responder outcome uses smaller groups, consistent with its focus on skin involvement; the reason for each endpoint’s analysis-set size is not reported in the registry record.

The primary endpoint was change in percent-predicted FVC from baseline to Week 16. Secondary outcomes were assessed at Week 28 and included change in six-minute walk distance, change in the St. George’s Respiratory Questionnaire total score, skin physician’s global assessment responders and change in percent-predicted FVC. The registry record does not report details of the randomization sequence, allocation concealment, original sample-size justification, interim analyses or a complete statistical analysis plan.

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:

Analysis population for the primary outcome, as defined in the registry: Modified intent-to-treat (mITT) population included all the participants who were randomized and who received at least 1 dose of study medication.

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 Change From Baseline in Percent-predicted Forced Vital Capacity (FVC) at Week 16, measured over Baseline (Day 1) and Week 16.

The primary endpoint is change from baseline in percent-predicted forced vital capacity at Week 16. FVC is the volume of air a person can forcibly exhale after a full inspiration, measured in liters. The percent-predicted measure compares the observed FVC with an estimate of healthy lung capacity based on sex, age and height. This standard pulmonary function measure is used to quantify respiratory muscle weakness and gives a numerical way to track lung function over time.

The registry defines change as the Week 16 value minus the baseline value. A positive change therefore means that the measured percent-predicted FVC was higher at Week 16 than at baseline; a negative change means it was lower. The outcome is expressed in percent of predicted FVC, not as a percentage change from baseline.

For this endpoint, the registry posts least-squares means and standard errors for each group. Least-squares means are model-based group summaries, rather than simple averages alone; the specific covariates used for the reported primary analysis are not reported in the registry record. The endpoint population included randomized participants who received at least one dose of study medication. FVC offers a standardized lung-function measure, but it does not by itself describe symptoms, walking ability, skin disease or all possible effects of sarcoidosis. Those domains appear among the secondary outcomes and should be considered as distinct measures, not substitutes for the primary endpoint.

7. Results Posted in the Registry

• Golimumab: mean 1.15 versus 2.02 on control; difference -0.87 (95% CI -4.70 to 2.96), standardized effect d = -0.10; z-test p = 0.656.
• Ustekinumab: mean -0.15 versus 2.02 on control; difference -2.17 (95% CI -5.81 to 1.47), standardized effect d = -0.25; z-test p = 0.243.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in 6-minute Walk Distance at Week 28

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Golimumab: mean 12.53 versus 14.52 on control; difference -1.99 (95% CI -42.39 to 38.41), standardized effect d = -0.02; z-test p = 0.923.
  • Ustekinumab: mean -13.22 versus 14.52 on control; difference -27.74 (95% CI -66.29 to 10.81), standardized effect d = -0.30; z-test p = 0.158.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Golimumab and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Ustekinumab and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in the St. George's Respiratory Questionnaire (SGRQ) Total Score at Week 28

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Golimumab: mean -6.86 versus -9.50 on control; difference 2.64 (95% CI -5.28 to 10.56), standardized effect d = 0.14; z-test p = 0.513.
  • Ustekinumab: mean -4.25 versus -9.50 on control; difference 5.25 (95% CI -2.31 to 12.81), standardized effect d = 0.29; z-test p = 0.173.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Golimumab and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Ustekinumab and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Responders With a Score of Less Than or Equal to 1 on Skin Physician's Global Assessment (SPGA) Scale

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Golimumab: 52.9% (9/17) versus 30.0% (6/20) on control; risk difference 22.9 percentage points (95% CI -8.1 to 54.0), risk ratio 1.76 (0.79–3.95); unadjusted two-proportion z-test p = 0.157.
  • Ustekinumab: 14.3% (3/21) versus 30.0% (6/20) on control; risk difference -15.7 percentage points (95% CI -40.8 to 9.3), risk ratio 0.48 (0.14–1.65); unadjusted two-proportion z-test p = 0.224.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Golimumab and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Ustekinumab and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes extend the assessment beyond the primary Week 16 lung-function measure. They include walking distance, respiratory health status, skin response and a later FVC assessment. Their results do not all point in one consistent direction across measures or treatment groups. This variation is unsurprising because the outcomes represent different domains, have different analysis populations and were measured at Week 28 rather than the primary Week 16 time point.

These measures should be read as complementary evidence, not as repetitions of the primary test. Walking distance captures functional performance, the respiratory questionnaire captures reported health status, and the skin assessment concerns the subgroup with relevant skin involvement. The later FVC outcome is related to the primary endpoint but uses a different follow-up time. An apparent pattern on one secondary measure does not establish a corresponding effect on the primary endpoint or on other aspects of disease.

The secondary comparisons are exploratory in the sense that the registry record does not report a multiplicity plan or a testing hierarchy. When several outcomes and treatment comparisons are examined, the chance of at least one apparently notable result can increase. Unadjusted secondary p-values do not account for that possibility. Smaller analysis groups, particularly for the skin outcome, also limit the precision of estimates. The results can inform questions for further study, but should carry less inferential weight than a prespecified primary analysis whose role and method are clearly identified.

The registry posts secondary outcomes and analyses, but does not report all model details or the rationale for the analysis populations. The overall pattern is therefore best described cautiously: secondary measures show varied estimates, and they do not remove the uncertainty in the primary comparison. They also cannot establish whether the treatments differ in subgroups or whether a change on one measure corresponds to another outcome.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline in 6-minute Walk Distance at Week 28
• Change From Baseline in the St. George's Respiratory Questionnaire (SGRQ) Total Score at Week 28
• Percentage of Responders With a Score of Less Than or Equal to 1 on Skin Physician's Global Assessment (SPGA) Scale
• Change From Baseline in Percent-predicted Forced Vital Capacity (FVC) at Week 28

9. Statistical Interpretation

10. Statistical Methods Explained

For the primary continuous endpoint, the registry reports least-squares mean change and a standard error for each treatment group. A least-squares mean is a model-based estimate of a group’s mean outcome. Such estimates can account for terms included in a statistical model, but the registry record does not identify the covariates used for the posted ANCOVA. The standard error describes the estimated mean’s precision: a smaller standard error generally indicates a more precisely estimated mean, while it does not describe how much individual participants vary around that mean.

What does the analysis compare?

The primary comparisons are each active-treatment group against placebo. The registry reports that the sponsor used ANCOVA, or analysis of covariance, for these comparisons. In general, ANCOVA compares group outcomes while adjusting for specified covariates. Adjustment can improve precision or account for baseline differences when the model and covariates are appropriate. The registry posts p-values for these analyses but does not post the estimated adjusted differences or their intervals. The specific model terms and assumptions are not reported in the registry record.

Alongside the sponsor-reported analyses, an unadjusted recalculation compares the posted group means. It uses the difference between the active-treatment and placebo means and an approximate pooled standard deviation derived from the reported standard errors and group sizes. That approximation is useful for a simple summary-statistic comparison but does not recreate the sponsor’s ANCOVA. It also cannot incorporate covariate adjustments that are not available in the record. Since the discrepancy flag is false, there is no flagged conflict between the sponsor result and the recalculation; they remain analyses based on different methods and should be interpreted in that context.

How should a p-value be read?

A p-value measures how compatible the observed data are with a null hypothesis under the assumptions of a particular statistical test. It is not the probability that the null hypothesis is true, nor a measure of the size or clinical importance of a difference. The sponsor’s primary p-values come from ANCOVA, while recalculated p-values are unadjusted. A p-value should be interpreted with the effect estimate, its interval, the analysis method and the endpoint’s role in the study, rather than used as a stand-alone verdict.

Why do intervals matter?

An interval estimate gives a range of values compatible with the data and the method used, at a stated confidence level. For the recalculated primary comparisons, the posted intervals include zero. This means the summary-statistic analysis does not rule out no group difference, and it also allows differences in either direction within the displayed range. The confidence level for these recalculated intervals is not reported in the registry record. The sponsor’s posted ANCOVA results do not include intervals, so the precision of its adjusted comparisons cannot be directly assessed from those entries.

Why does over-dispersion matter?

For continuous outcomes, greater variability among individual observations makes group means harder to estimate precisely, all else equal. For binary outcomes such as a responder proportion, the key issue is the number of participants and events in each group: with few responders, the estimated proportion and comparison can be uncertain. General statistical principle: if a binary outcome varies more than a model expects, the model may understate uncertainty unless it accounts for that extra variability. Over-dispersion is not identified as a feature of the primary continuous endpoint in the registry record, and no specific over-dispersion assessment is reported.

What methods appear for secondary outcomes?

The registry lists linear contrast tests for the six-minute walk and respiratory questionnaire comparisons, and Fisher exact tests for the skin responder comparisons. In general, linear contrasts test specified combinations of model estimates, while Fisher exact tests assess association in small categorical samples without relying on a large-sample approximation. The full models, covariates and missing-data handling for the secondary analyses are not reported in the registry record. The displayed recalculated secondary comparisons are unadjusted and should not be confused with the sponsor’s named tests.

These methods serve different outcome types and answer different questions. Continuous outcomes are reported as mean changes and mean differences, while the skin endpoint is a proportion meeting a specified threshold. A numerical estimate and its uncertainty are more informative than a significance label alone. The methods recorded in the registry do not reveal whether any hierarchy, gatekeeping procedure or correction for multiple comparisons was used.

11. Confidence Intervals and Precision

Precision concerns how narrowly an estimate identifies the range of values compatible with the data and analysis. For the primary endpoint, the recalculated golimumab-versus-placebo mean difference is -0.87, with an interval from -4.7 to 2.96. The ustekinumab-versus-placebo mean difference is -2.17, with an interval from -5.815 to 1.475. Both intervals span zero and include values on either side of it. They therefore do not resolve whether the underlying difference is below or above zero.

The intervals are wide relative to the point estimates, indicating substantial uncertainty about the size and direction of the unadjusted differences. This does not mean every value in the interval is equally likely, and it does not demonstrate equivalence. Demonstrating equivalence or noninferiority requires an appropriate prespecified margin and analysis; such a margin is not reported in the registry record.

These intervals accompany the summary-statistic recalculations, not the sponsor’s ANCOVA estimates. The registry’s own primary analyses post p-values but no adjusted effect estimates or intervals. The confidence level for the recalculated intervals is not reported in the registry record, so no confidence percentage should be attached to them. They are based on approximate standard deviations reconstructed from the posted standard errors and group sizes, rather than individual participant data.

Precision also depends on the amount of information available for each endpoint. The primary endpoint groups contain 44, 42 and 46 participants, while the skin responder outcome has smaller groups. The standard errors for the primary group means describe uncertainty in those estimates; they are not standard deviations of participant outcomes. In general, larger samples can improve precision, but variability and the analysis method also matter. The registry record does not provide enough information to assess precision under alternative model specifications or to quantify the impact of missing outcome data.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations use the observed effect as a starting point and show planning scenarios that target the full observed effect, 75% of it or 50% of it. The entries give participants per arm for 80% power and participants per arm for 90% power. They are not total sample sizes across all arms and are not counts of total events, because these outcomes are not time-to-event outcomes.

The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms and a normal approximation. Each row asks how many participants per arm a hypothetical two-group study would need to have the specified probability of detecting the chosen effect, if that effect were real and the assumptions held. The 80% column represents a design intended to detect that effect in 80% of repeated studies; the 90% column represents a higher target probability of detection and generally requires more participants.

Smaller effects require more participants to distinguish them from random variation under the same assumptions. This is why rows that plan for a smaller fraction of the observed effect have larger sample-size figures. The figures are planning illustrations based on the trial’s observed effects, not the trial’s original design calculations. The registry record does not report the original sample-size justification, the planning effect used at design or the original power target. This is not observed power: it does not measure the probability that the completed trial detected an effect after the results were known.

These calculations are simplified. The continuous outcomes use approximate variability derived from posted summary statistics, and the binary outcome calculations use the observed proportions. A real design would require assumptions and decisions specific to the research question, endpoint, analysis method, missingness and multiplicity plan. The illustrations help show the relationship between effect size and sample size; they do not establish what sample size would be appropriate for a future trial.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.870; pooled standard deviation σ = 9.054.
2. n = 2 × (1.96 + 0.842)² × 9.054² / -0.870² = 1,700.2, rounded up to 1,701 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.87                      # observed difference in means
sigma <- 9.05424250598264                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(955279)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 1,701 (80% power), 2,277 (90% power)
 75% of observed effect: n per arm = 3,023 (80% power), 4,047 (90% power)
 50% of observed effect: n per arm = 6,801 (80% power), 9,105 (90% power)
power.t.test: 1701.2 
Simulated power at n = 1,701 per arm: 0.804 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

For the primary endpoint, the registry reports sponsor analyses using ANCOVA. The posted p-values are 0.543 for golimumab versus placebo and 0.126 for ustekinumab versus placebo. The registry does not provide the adjusted mean differences or intervals for these analyses, so their effect sizes and precision cannot be compared directly from the sponsor entries.

The recalculated comparisons use the posted least-squares means, standard errors and group sizes to make simple unadjusted estimates. They give mean differences of -0.87 for golimumab versus placebo and -2.17 for ustekinumab versus placebo, with intervals and unadjusted p-values recorded for each comparison. These calculations do not reproduce ANCOVA, because the covariates and model details are not reported in the registry record. The two sets of p-values can differ without either being a transcription error: they arise from analyses that need not make identical adjustments or use identical calculations.

The sponsor’s named analysis is the registry’s own inferential analysis and should be kept distinct from the recalculated summary-statistic results. Since no discrepancy is flagged, there is no recorded warning that the two analyses conflict. However, the recalculations remain useful as a transparent, unadjusted view of the posted group summaries. Neither set of p-values alone describes the magnitude or precision of the treatment contrast. The sponsor’s analysis would be easier to interpret if adjusted estimates, intervals, covariates and model details were reported.

15. Multiplicity, Missing Data and Analysis Populations

The study has multiple treatment comparisons and multiple outcomes. The primary endpoint compares each active-treatment group with placebo, and the registry also posts several secondary outcomes. Testing many hypotheses can increase the chance of obtaining at least one small p-value by chance. A prespecified hierarchy, family-wise error adjustment or another multiplicity strategy can address this issue, but the registry record does not report whether any such plan was used. The secondary p-values should therefore be treated as unadjusted evidence unless the registry states otherwise.

Missing-data handling is also important. The primary population is defined as randomized participants who received at least one dose, but the registry entries do not explain how missing Week 16 measurements were handled, how many measurements were missing or whether any imputation was used. The endpoint-specific group sizes are posted, but those counts alone cannot show why participants were not included in a particular analysis or whether missingness differed between groups. The registry record does not report sensitivity analyses for missing data.

The listed modified intent-to-treat population is not the same as an unrestricted analysis of every randomized participant, because it requires receipt of at least one dose. The record does not state whether all randomized participants received treatment or describe additional exclusions. Without details of the analysis set and missing-data strategy, it is not possible to assess how robust the results are to alternative population definitions or assumptions about unavailable outcomes.

The original statistical analysis plan, sample-size justification, interim-analysis plan, covariates in the primary ANCOVA and multiplicity procedure are not reported in the registry record. These omissions limit independent assessment of whether the reported tests follow a prespecified plan and how the sponsor handled the full family of comparisons. They do not prove that a plan was absent; they mean its details are unavailable in the registry record.

16. 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

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (5.0%), bronchitis (1.8%), nasopharyngitis (1.8%), sepsis (1.8%), septic shock (1.8%).

The registry posts serious adverse event counts among 58 participants at risk in the placebo group, 55 in the golimumab group and 60 in the ustekinumab group. Serious adverse events affected 9, 7 and 10 participants, respectively. These are counts over the recorded safety groups; they do not establish that an event was caused by study treatment. The registry record does not report the deaths field, so deaths are not reported in the registry record.

Other adverse events affected 52 participants in the placebo group, 49 in the golimumab group and 52 in the ustekinumab group, using the same respective at-risk counts. These figures describe participants with events, not the total number of event episodes. The registry lists pneumonia at 5.0% among the most common serious event terms and bronchitis, nasopharyngitis, sepsis, septic shock and tuberculosis at 1.8% each. The registry record does not provide enough detail in these entries to reconstruct event timing, severity beyond the serious classification or individual treatment attribution.

Safety counts should be interpreted descriptively. Group sizes are limited, and the figures alone cannot establish comparative safety or rule out uncommon harms. The registry’s adverse-event entries are not a substitute for a full clinical safety analysis.

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

The primary result describes the observed change in percent-predicted FVC at Week 16 among participants in the modified intent-to-treat population. The posted group estimates show a positive mean change for placebo and golimumab and a small negative mean change for ustekinumab. The recalculated active-treatment versus placebo differences have intervals that include zero, while the sponsor’s ANCOVA p-values are 0.543 and 0.126. Together, these results leave considerable uncertainty about the size and direction of the group differences.

That uncertainty should not be turned into a claim of equivalence, proof of no effect or evidence of a clinically important difference. A non-significant p-value does not show that groups are identical; it indicates that the data and test do not provide strong evidence against the null hypothesis at the chosen test’s threshold. Conversely, a point estimate in one direction is not by itself proof that the underlying effect has that direction or magnitude. The registry does not report a clinical threshold for interpreting the FVC change.

The endpoint captures a particular measure of pulmonary function at a particular follow-up time. It does not answer every question about symptoms, daily functioning, skin involvement, long-term outcomes or safety. Secondary outcomes address some other domains, but they are also uncertain and their multiplicity plan is not reported. The study’s results cannot be extended beyond the studied population, schedules and follow-up based on this record alone.

The trial was Phase 2 and enrolled 173 participants. The posted evidence is most useful as a record of observed estimates, analyses and uncertainty that can help frame further research questions. It is not an individualized prediction, a guarantee of future trial results or a basis for medical advice. The registry record does not report enough protocol and analysis detail to answer all questions about design choices, missing data or subgroup effects.

18. Why This Trial Matters Statistically

This trial illustrates why a point estimate should be read together with uncertainty and analysis method. The primary group means are not the same as treatment effects: the comparisons with placebo are the quantities that address the relative difference. Those differences are imprecisely estimated in the summary-statistic recalculation, and the registry’s sponsor analysis posts p-values without adjusted estimates or intervals. The record therefore illustrates the value of reporting both effect sizes and their precision.

It also shows how analysis choices matter. The sponsor used ANCOVA, while the simple recalculations use unadjusted summary statistics. An estimate from a model-adjusted analysis and an estimate from an unadjusted comparison can differ because they answer the comparison using different information and assumptions. Reporting model terms and adjusted effects would make it easier for readers to understand those differences.

The secondary outcomes demonstrate the interpretive challenge of multiple endpoints and populations. Lung function, walking distance, respiratory questionnaire scores and skin response do not measure the same construct. A varied pattern across measures is not resolved by selecting the most favorable result; multiplicity, endpoint role, sample size and precision all matter. The skin responder groups are smaller than the primary endpoint groups, underscoring that different outcomes can have different information content.

Finally, the sample-size illustrations help convey how effect size and precision are linked, while not describing the original trial design. A small trial can produce estimates that remain uncertain even when a numerical difference is observed. Distinguishing planning calculations from observed power prevents a retrospective calculation from being mistaken for evidence that the design was adequate or inadequate.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record leaves important design and analysis details unreported. These include the original sample-size justification, interim analyses, randomization-sequence details, allocation concealment, the covariates used in ANCOVA, the prespecified multiplicity approach and the handling of missing outcome data. Their absence from the record limits independent assessment but does not establish that the trial team did not address them.

The sponsor’s primary analysis entries contain p-values but not adjusted effect estimates or intervals. The recalculated intervals are based on group-level summaries and approximate standard deviations rather than individual participant data, and do not reproduce the sponsor’s ANCOVA. The confidence level for those recalculated intervals is not reported in the registry record. As a result, the adjusted effect size and its precision cannot be directly evaluated from the posted sponsor entries.

The modified intent-to-treat definition includes participants who were randomized and received at least one dose. The registry record does not say how many randomized participants did not meet that definition or why endpoint-specific counts vary. Nor does it report the number of missing measurements or sensitivity analyses. Those omissions make it difficult to judge the potential influence of exclusions and missingness.

The primary endpoint is limited to FVC at Week 16, and the secondary outcomes cover different measures at Week 28. The posted figures cannot establish longer-term patterns or subgroup-specific effects. The safety counts are descriptive, with deaths not reported in the registry record. These limitations mean the findings should be interpreted as trial-specific estimates with incomplete reporting detail, rather than as a complete account of efficacy, safety or the study’s full analytical process.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate the Safety and Effectiveness of Ustekinumab or Golimumab Administered Subcutaneously (SC) in Patients With Sarcoidosis — NCT00955279.
• Registry record last updated 2014-07-17; status last verified 2014-06.
• Results first posted 2014-07-17.
• 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.

23. Record Summary

NCT00955279 is a completed Phase 2 randomized, quadruple-masked, parallel-group, placebo-controlled study of subcutaneous ustekinumab and golimumab for chronic sarcoidosis. The registry lists enrollment of 173 and describes pulmonary involvement as the primary analysis population and skin involvement as secondary, with possible overlap.

The primary endpoint was change from baseline in percent-predicted FVC at Week 16. The registry posts group least-squares means and standard errors, sponsor ANCOVA p-values and secondary outcomes at Week 28. Unadjusted recalculations provide additional estimates and intervals, but do not reproduce the sponsor’s adjusted analysis. The primary comparisons remain uncertain, and the record does not report key elements including the original sample-size justification, multiplicity plan and missing-data handling.