A Study to Evaluate the Effectiveness and Safety of SKI-O-703 in Patients Experiencing Active Rheumatoid Arthritis Despite Treatment With Conventional Therapies.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Effectiveness and Safety of SKI-O-703 in Patients Experiencing Active Rheumatoid Arthritis Despite Treatment With Conventional Therapies. (NCT04057118), a randomized phase 2 study. The primary outcome is Change in Disease Activity Score.

2. Clinical Question

The primary outcome is change in the Disease Activity Score for 28 joints using high-sensitivity C-reactive protein (DAS28-hsCRP). It combines tender and swollen joint counts with hsCRP into a score. The registry describes higher values as indicating greater disease activity; a value greater than 5.1 indicates relatively high disease activity, while a value less than 3.2 indicates lower disease activity. No theoretical full range is available.

The question is dose-specific: the 100 mg, 200 mg, and 400 mg groups are each compared with placebo. The posted primary outcome summaries report mean changes and standard errors by group, along with unadjusted comparisons. They do not establish whether any observed difference is statistically distinguishable from zero with precision, nor do they by themselves establish a treatment effect beyond this study’s population, duration, and outcome definition.

Secondary outcomes listed in the registry include ACR20, ACR50, and ACR70 response percentages, HAQ-DI, adverse events, and serious adverse events. Numerical secondary efficacy results are not reported in the registry record. The primary question therefore remains centered on the Week 12 change in DAS28-hsCRP rather than conclusions drawn from the listed secondary outcomes.

3. Trial Design

This was a phase 2, multicenter, randomized, double-blind, placebo-controlled, parallel-dose study sponsored by Oscotec Inc. The registry lists rheumatoid arthritis as the condition and records the study as completed. The study began on 2019-03-20, and primary completion was recorded as 2020-10-13. The planned study enrollment was 163.

Participants were randomized to one of four groups: placebo or SKI-O-703 at 100 mg, 200 mg, or 400 mg. The study summary describes oral administration twice daily for 12 weeks. Randomization and double masking are stated in the design record, but the allocation ratio, the method used to generate the random sequence, and the practical details of masking are not reported in the registry record. Arm-specific descriptions beyond the treatment labels are also not reported in the registry record.

The parallel design assigns each participant to a single treatment group for comparison with the other groups. The placebo group provides the reference for each dose comparison. The posted primary outcome uses baseline and Week 12 measurements, and the reported endpoint is mean change from baseline in DAS28-hsCRP. The outcome summaries list 37 participants for placebo and 36 for each SKI-O-703 dose. These are the arm sizes shown for this outcome; they should not be confused with the planned enrollment figure.

The registry record does not report the original sample-size justification, the intended statistical analysis plan, interim analyses, or detailed analysis-population definitions. It also does not state how missing Week 12 measurements were handled. Those omissions limit assessment of how the posted analyses relate to the study’s original protocol and planned methods.

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 Change in Disease Activity Score, measured over Baseline and Week 12.

The primary endpoint is mean change from baseline to Week 12 in the Disease Activity Score for 28 joints using high-sensitivity C-reactive protein. DAS28-hsCRP combines tender joint count, swollen joint count, and hsCRP into a continuous score. The registry states that higher values indicate more severe disease activity. It describes values greater than 5.1 as relatively high activity and values less than 3.2 as lower activity; no theoretical full range is available.

Change from baseline expresses how the score differs at follow-up from each participant’s starting value. Reporting a group mean change provides a summary of the average direction and size of change in that group. In the posted outcome, the group means are least-squares means, with standard errors as their dispersion measure. The comparison of interest is the difference in mean change between each dose group and placebo.

General: a continuous endpoint can preserve more information about the magnitude of changes than a threshold-only response measure. A composite score can also bring several clinical and laboratory components into a single outcome. Its interpretation depends on how the score is defined and on the analysis population and missing-data methods. Those details are only partly available in the registry record. In particular, no theoretical full range is specified, and the record does not report the original analysis plan or missing-data approach.

7. Results Posted in the Registry

• SKI-O-703 100 mg: mean -0.91 versus -0.95 on control; difference 0.04 (95% CI -0.39 to 0.47), standardized effect d = 0.04; z-test p = 0.854.
• SKI-O-703 200 mg: mean -1.06 versus -0.95 on control; difference -0.11 (95% CI -0.54 to 0.32), standardized effect d = -0.12; z-test p = 0.612.
• SKI-O-703 400 mg: mean -1.15 versus -0.95 on control; difference -0.20 (95% CI -0.63 to 0.23), standardized effect d = -0.22; z-test p = 0.357.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports least-squares mean changes in a continuous outcome, standard errors for those means, and dose-versus-placebo mean differences. General: a mean summarizes the central value of a set of measurements, while a standard error describes the estimated uncertainty in a mean. A standard error is not the spread of individual participants’ measurements. It generally becomes smaller when the estimate is based on more information, but its interpretation depends on the analysis and data structure.

What does a mean difference represent?

For each dose, the mean difference compares the reported mean change in that group with the reported mean change in placebo. A difference of 0 would represent equal group means for this outcome. A positive or negative estimate indicates the direction of the contrast according to the group ordering used in the registry’s comparisons. The reported estimates are 0.04 for 100 mg, -0.11 for 200 mg, and -0.2 for 400 mg. These are differences in score change, not percentages or ratios.

The group results are least-squares means, a type of model-based mean. The registry record does not report the model specification, covariates, or any adjustment details. It is therefore not possible to reconstruct the sponsor’s intended model from the posted information. The comparisons displayed are unadjusted recalculations from the posted summary statistics, rather than a substitute for an unreported prespecified analysis. The record does not flag a discrepancy between those comparisons and a sponsor analysis.

How should standard errors and confidence intervals be read?

General: a confidence interval gives a range of values that are reasonably compatible with the data and the method used to calculate the interval. It communicates uncertainty about an estimate, rather than the range of individual participant outcomes. The intervals for all three dose-versus-placebo differences include zero. That means the intervals include no mean difference, while also including values on either side of zero. The interval width indicates that the precise size and direction of the comparison remain uncertain.

The reported arm-level standard errors are 0.152 for placebo and 0.155 for each SKI-O-703 dose. The recalculated comparison uses an approximate pooled standard deviation derived from the posted summary statistics. General: pooling estimates of variability can be useful when comparing independent groups under an assumption of similar underlying variation. An approximation based on summary statistics may not reproduce a model-based analysis that uses participant-level data, covariate adjustment, or a different variance method. The registry record does not supply enough detail to assess those alternatives.

What do the p-values say?

The unadjusted p-values are 0.854 for 100 mg, 0.612 for 200 mg, and 0.357 for 400 mg. General: a p-value measures how unusual data at least as incompatible with a null model would be if that model and the test assumptions were correct. It is not the probability that the null hypothesis is true, nor a measure of clinical importance. A larger p-value does not prove that groups are equivalent. Equivalence would require a prespecified margin and an analysis designed to evaluate whether the uncertainty lies within that margin; such a margin and analysis are not reported in the registry record.

The p-values are unadjusted, and the registry lists three dose comparisons with placebo. General: when several hypotheses are tested, the overall chance of a false-positive finding can differ from the chance associated with each individual test. Methods for controlling multiplicity may include adjusting the testing threshold or using a planned testing sequence. The registry record does not report whether such a strategy was used in the original protocol or analysis.

Why can over-dispersion matter?

General: over-dispersion means that observed variability is greater than a model assumes. If a method underestimates variability, standard errors may be too small and confidence intervals too narrow; if it overestimates variability, the reverse can occur. The primary endpoint is continuous and the posted arm summaries provide standard errors, but the registry record does not report a diagnostic assessment of the model’s variance assumptions. Over-dispersion is therefore a general consideration, not a documented problem in this study.

Further details needed to reproduce or fully evaluate the original analysis—including model terms, the analysis population, missing-data handling, and the prespecified multiplicity approach—are not reported in the registry record. The recalculated summaries are useful for describing the posted group results, but do not resolve those omissions or establish what methods were originally planned.

10. Confidence Intervals and Precision

The primary outcome summaries are based on 37 participants in the placebo group and 36 in each SKI-O-703 group. The arm-level standard errors range from 0.152 to 0.155. These values describe uncertainty around the reported group mean changes; they are not measures of the spread of individual scores.

The mean-difference estimates are 0.04 for 100 mg, -0.11 for 200 mg, and -0.2 for 400 mg. Each posted confidence interval crosses zero. This is an important precision feature: the estimated differences are compatible with no difference, and also with values on either side of zero. The intervals therefore do not pin down a single direction or a narrow estimate of the between-group contrast. Exact interval endpoints are not repeated in this explanation.

General: a confidence interval is most useful when read together with its point estimate and the range of plausible values it contains. A broad interval can reflect limited information, substantial variability, or both. A confidence interval that includes zero does not prove that there is no difference; it indicates that the data and method do not exclude zero at the interval’s confidence level. Conversely, an interval that excludes zero would not by itself establish clinical importance.

The registry record does not report the original analysis model, the missing-data procedure, or whether any covariate adjustment was used. Those choices can affect both estimates and precision. The recalculated comparisons use approximate variability derived from posted summaries, so they should be read as unadjusted descriptions rather than as a complete reconstruction of an analysis based on participant-level data.

11. 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 planning figures are participants per arm for each two-group dose-versus-placebo comparison, not total events. They illustrate the number of participants that would be needed in each of the two arms under the stated planning assumptions. The figures are based on the observed comparison effect or a stated fraction of that effect. The smaller-effect scenarios require more participants because a subtler difference is harder to distinguish from random variation.

The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each scenario, the effect to detect is the stated fraction of the observed effect. The 80% power column describes the planned sample size that, under those assumptions, would give an 80% chance of detecting that assumed effect using the specified test. The 90% power column uses a higher chance of detection and therefore calls for a larger sample size under the same assumptions.

For example, the rows distinguish planning for the full observed effect from planning for 75% or 50% of that effect. These are hypothetical planning choices, not additional observed trial findings. The figures are not the original trial’s design calculation: the registry record does not report the original sample-size justification or its design assumptions. This is not observed power. It is a planning illustration based on the observed effect and the stated assumptions, and should not be interpreted as evidence about the trial’s probability of success.

Because the observed comparisons are uncertain, planning directly from their point estimates can produce unstable sample-size illustrations. A separate protocol-based target effect and variance estimate may lead to different planning figures; those original inputs are not reported in the registry record.

12. 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 = 0.040; pooled standard deviation σ = 0.927.
2. n = 2 × (1.96 + 0.842)² × 0.927² / 0.040² = 8,435.6, rounded up to 8,436 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 <- 0.0399999999999999                      # observed difference in means
sigma <- 0.927255742105491                     # 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(4057118)
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 = 8,436 (80% power), 11,293 (90% power)
 75% of observed effect: n per arm = 14,997 (80% power), 20,077 (90% power)
 50% of observed effect: n per arm = 33,743 (80% power), 45,172 (90% power)
power.t.test: 8436.6 
Simulated power at n = 8,436 per arm: 0.801 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists three primary dose-versus-placebo comparisons and a set of secondary outcomes: ACR20, ACR50, ACR70, HAQ-DI, adverse events, and serious adverse events. It lists six secondary outcomes as posted and reports seven outcome measures overall. The registry record does not provide statistical analyses for these outcomes, and the outcome listings do not give numerical secondary efficacy estimates in the information available for interpretation.

The three primary comparisons create a multiplicity question. General: when several comparisons are conducted, interpreting each p-value as if it were the only test can increase the chance of at least one false-positive result across the family of tests. A prespecified adjustment or testing strategy can address that issue. The registry record does not report whether the study used an adjustment, a hierarchy among doses, or another multiplicity procedure. The displayed p-values are unadjusted, so they should not be treated as multiplicity-controlled evidence.

Missing-data handling is also not reported in the registry record. The endpoint is defined using baseline and Week 12 measurements, but the record does not say how participants without an observed Week 12 value were treated, whether values were imputed, or whether a repeated-measures model was used. General: different approaches to incomplete follow-up rely on different assumptions and can affect estimates and uncertainty. Without the planned method and supporting information, the impact of missing measurements cannot be assessed.

The analysis populations are not defined in the registry record. It is therefore unclear whether the primary results correspond to all randomized participants, participants who received treatment, or another population. The posted outcome-specific group sizes are 37, 36, 36, and 36, but the record does not explain their relationship to planned enrollment or identify exclusions. These reporting gaps limit reproducibility and make it difficult to judge whether the comparisons follow the intended analysis plan.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: rheumatoid arthritis (2.4%), endometrial adenocarcinoma (2.4%), myocarditis (2.4%), pneumonia viral (2.4%).

The registry reports serious adverse events for 41 participants at risk in placebo, 41 in the 100 mg group, 40 in the 200 mg group, and 41 in the 400 mg group. The corresponding numbers recorded with serious adverse events were 1, 1, 0, and 1. Deaths were recorded as 0 in each group. These are counts, not rates calculated for a common exposure time.

The registry also lists other adverse events for 8 participants in placebo, 6 in the 100 mg group, 10 in the 200 mg group, and 13 in the 400 mg group, with the same group-specific numbers at risk. The record lists rheumatoid arthritis, endometrial adenocarcinoma, myocarditis, and viral pneumonia among the serious adverse event terms, each with a reported percentage of 2.4%. The available information does not identify an arm for those listed terms, so they should not be attributed to a particular dose or comparator.

General: adverse-event counts describe events recorded during a study, but do not by themselves establish that treatment caused them. Comparisons can also be affected by exposure duration, event definitions, and how events were collected. Those details are not reported in the registry record. The counts should be read as descriptive safety information for this study, not as a complete estimate of risk or a basis for individual medical decisions.

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

The registry’s primary outcome concerns average change in DAS28-hsCRP from baseline to Week 12. All four group means are negative: -0.95 for placebo, -0.91 for SKI-O-703 100 mg, -1.06 for 200 mg, and -1.15 for 400 mg. The dose-versus-placebo mean differences are 0.04, -0.11, and -0.2, respectively. The confidence intervals for all three comparisons include zero, and the unadjusted p-values are 0.854, 0.612, and 0.357.

These results describe the observed summaries and the uncertainty around the estimated comparisons. They do not establish that the groups are equivalent, because no equivalence margin or equivalence analysis is reported. They also do not show that a particular dose produces a clinically important change: the registry record does not provide a threshold for a clinically important difference in this outcome. The score’s clinical interpretation is limited to the registry’s description of higher and lower disease activity.

The findings are specific to the primary endpoint, the Week 12 timeframe, and the participants represented in the posted analysis. The record does not report enough detail about analysis populations or missing-data handling to determine how those choices affected the estimates. It also does not report the original analysis plan or whether multiplicity was addressed. Those limitations matter when interpreting the set of dose comparisons.

The listed safety counts provide descriptive information about serious adverse events, other adverse events, and deaths recorded in the groups. They do not establish causality, and the registry record does not provide the detail needed to compare exposure-adjusted risk. Secondary outcomes are listed, but numerical results for them are not reported in the registry record. No broader conclusion about outcomes beyond the recorded measures and study period follows from the posted results alone.

16. Why This Trial Matters Statistically

This study illustrates why a treatment comparison should be read through both its point estimate and its uncertainty. The three dose groups have mean changes that differ from the placebo mean, but the estimated between-group differences are small in relation to the uncertainty represented by the confidence intervals. All three intervals include zero. A point estimate alone can make a comparison look more definite than the data support; interval estimates show that several underlying values remain compatible with the results.

The parallel, randomized, double-blind, placebo-controlled design sets up direct dose-versus-placebo comparisons. But design labels are not a substitute for details of implementation and analysis. The registry record does not report the allocation ratio, analysis-population definitions, missing-data methods, model specification, or multiplicity procedure. These elements affect how confidently results can be interpreted and reproduced.

The study also illustrates the distinction between an observed estimate and a planning target. Sample-size illustrations based on the observed effect vary according to the assumed fraction of that effect. Smaller target effects call for more participants per arm under the stated planning assumptions. These projections are not observed power and do not reveal the original trial’s design rationale, which is not reported in the registry record.

Finally, the endpoint combines joint counts and hsCRP into a continuous score. Understanding the measure’s direction and composition helps readers interpret the negative mean changes, while the lack of a reported clinically important-difference threshold limits conclusions about practical importance. Clear statistical interpretation requires keeping these points together: the design, the endpoint, the estimates, their precision, and what the registry does not document.

17. 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 primary endpoint summaries are available by group, but the registry record does not report the original statistical analysis plan, model specification, covariates, or analysis populations. The comparisons described are unadjusted recalculations based on posted summary statistics and approximate variability; they cannot reproduce an analysis using participant-level information or unreported model terms.

Although the outcome is defined at baseline and Week 12, missing-data handling is not reported in the registry record. The record does not state how absent follow-up measurements were handled or whether the analysis included all randomized participants. The outcome-specific arm sizes are smaller than the planned enrollment figure, but the registry record does not explain that difference. No explanation should be inferred from the available counts.

Three dose comparisons are reported, and their p-values are unadjusted. A multiplicity strategy is not reported, so the nominal p-values should not be read as adjusted evidence across the family of comparisons. The record also does not report a threshold for a clinically important difference in DAS28-hsCRP, limiting interpretation of the practical meaning of the estimated contrasts.

Safety counts are descriptive. Exposure time, detailed event-collection methods, and arm attribution for the serious-event terms listed with percentages are not reported in the registry record. Finally, numerical results for the listed secondary efficacy outcomes are not reported in the registry record. These gaps constrain assessment of both efficacy and safety beyond the primary outcome summaries.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Study to Evaluate the Effectiveness and Safety of SKI-O-703 in Patients Experiencing Active Rheumatoid Arthritis Despite Treatment With Conventional Therapies. — NCT04057118.
• Registry record last updated 2024-07-31; status last verified 2024-04.
• Results first posted 2024-07-31.
• 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.

21. Record Summary

NCT04057118 is a completed phase 2, multicenter, randomized, double-blind, placebo-controlled, parallel-dose study sponsored by Oscotec Inc. It enrolled a planned 163 participants with active rheumatoid arthritis despite conventional therapies. Participants were assigned to placebo or oral SKI-O-703 at 100 mg, 200 mg, or 400 mg, twice daily for 12 weeks, according to the study summary.

The primary outcome was mean change from baseline to Week 12 in DAS28-hsCRP. Posted group means were -0.95 for placebo, -0.91 for 100 mg, -1.06 for 200 mg, and -1.15 for 400 mg. The three unadjusted comparisons with placebo have confidence intervals that include zero. The registry record does not report the original analysis plan, missing-data method, or multiplicity strategy.

Serious adverse events were recorded for 1 participant in placebo, 1 in the 100 mg group, 0 in the 200 mg group, and 1 in the 400 mg group; deaths were recorded as 0 in each group. These safety counts are descriptive and do not establish causality.