Phase IIb Study of STA-2 in Patients With Chronic Stable Angina: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase IIb Study of STA-2 in Patients With Chronic Stable Angina (NCT01239511), a randomized phase 2 study. The primary outcome is Change in Total Exercise Time (Seconds).

2. Clinical Question

3. Trial Design

This was a phase IIb, randomized, double-blind, placebo-controlled, parallel-group study sponsored by Sinphar Pharmaceutical Co., Ltd. The registry lists an enrollment of 186 and a completed status. The recorded start was 2010-11, and primary completion was 2012-04. The study title and summary describe a dose-response evaluation of STA-2 in people with chronic stable angina.

The intervention groups received daily STA-2 doses of 900 mg, 1800 mg, or 2700 mg for 42 days. Each group was compared with a placebo group. The treatment descriptions state that capsules were taken three times daily; the placebo description also specifies capsules taken three times daily. The record identifies double masking but does not report which roles were masked, how the masking was maintained, or whether its success was evaluated.

The primary outcome was change in total exercise time, measured in seconds. The registry describes this as the time difference from V2 to V5 and gives the timeframe as 6 weeks after the first exercise tolerance test. It identifies the analysis population as ITT, expanded in the record as intention-to-treat. The primary outcome entries report 46 participants for placebo, 43 for Treatment Group A, 42 for Treatment Group B, and 37 for Treatment Group C.

The registry lists secondary outcomes concerning exercise-test measures, diary-recorded angina frequency, short-acting nitrate use, pharmacological and lipid measures, and the dose-response relationship. Although it lists 9 outcome measures, only 8 secondary outcomes are posted in the record. The protocol, original statistical analysis plan, and reasons for this difference are not reported in the registry record.

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: In order to ensure the consistency, the difference of total exercise time between V1 and V2 should be less than 20%.

Overall Study

Analysis population for the primary outcome, as defined in the registry: ITT (intend-to-treat) population will be used for analysis

5. Baseline Characteristics

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

Population: For subjects without the primary efficacy variable due to early discontinuaton and those with protocol violation were excluded from the analysis group.

6. Primary Endpoint

The registry lists the primary outcome as Change in Total Exercise Time (Seconds), measured over 6 weeks after the first exercise tolerance testing is conducted.

Total exercise time is the duration of exercise achieved during an exercise tolerance test. The primary endpoint is the change in this duration, recorded in seconds. The registry defines the comparison as the time difference from V2 to V5 and gives the assessment timeframe as 6 weeks after the first exercise tolerance test. It describes the endpoint as a least-squares mean and states that the intention-to-treat population was to be used.

As a continuous measure, exercise time retains information about the size of a change rather than reducing it to a category such as improved or not improved. The exercise test also provides a standardized setting for recording performance. In general, an endpoint is useful when it directly represents an outcome of interest and can be measured consistently. These are general considerations, not evidence about how the test was conducted in this trial.

The registry does not give the full measurement protocol, the handling of participants without a final assessment, or the model used to obtain the stated least-squares means. Those details are not reported in the registry record. The posted arm results are means with standard deviations, while the comparison estimates and confidence intervals are approximate unadjusted recalculations from those summaries. They should not be mistaken for the sponsor’s full model-based analysis.

7. Results Posted in the Registry

• Treatment Group A: mean 18.20 versus 24.30 on control; difference -6.10 (95% CI -9.78 to -2.42), standardized effect d = -0.69; z-test p = 0.001.
• Treatment Group B: mean 33.90 versus 24.30 on control; difference 9.60 (95% CI 5.88 to 13.32), standardized effect d = 1.08; z-test p = <0.001.
• Treatment Group C: mean 11.90 versus 24.30 on control; difference -12.40 (95% CI -16.41 to -8.39), standardized effect d = -1.35; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What is being compared?

The posted comparisons use the placebo group as the reference and calculate an unadjusted difference between the mean change in each STA-2 group and the mean change in placebo. The endpoint is continuous and measured in seconds. A mean difference keeps the outcome in its original unit, which makes the estimate interpretable as a difference in average change in exercise time. The reported estimates are recalculations from the arm summaries, rather than output from a posted sponsor analysis. No sponsor analyses are listed in the registry.

The sign of a mean difference depends on the order of subtraction. In these comparisons, a negative estimate indicates that the treatment-group mean is lower than the placebo-group mean; a positive estimate indicates that it is higher. The estimates should be interpreted alongside the arm means and their uncertainty, rather than in isolation. A difference in means is not a statement that every individual in one group had a larger or smaller change than every individual in another group.

What does a standard deviation describe?

The arm summaries include standard deviations of 8.73 seconds for placebo, 8.95 seconds for Treatment Group A, 9.03 seconds for Treatment Group B, and 9.69 seconds for Treatment Group C. A standard deviation describes the spread of individual observations around a group mean. It is not the same as a confidence interval and does not, by itself, measure the precision of the estimated mean. Precision also depends on how many observations contribute to the estimate.

The posted group summaries identify the reported arm values as least-squares means, while the comparison calculations use approximate standard deviations and standard errors reconstructed from summary data. This distinction matters: an adjusted least-squares mean from a model can differ from a simple mean, and reconstructing an unadjusted comparison cannot reproduce a model that includes covariates or other design details. The registry does not post enough analysis detail to assess such a model or reconcile it with these approximate calculations.

How should the p-values be understood?

A p-value describes how compatible the observed data are with a specified null hypothesis under the assumptions of the statistical test. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. The displayed p-values are unadjusted. The registry does not report whether the sponsor used a prespecified testing hierarchy or another multiplicity procedure. When several comparisons are examined, unadjusted p-values do not reflect the increased opportunity for chance findings across those comparisons.

The registry’s comparison estimates are recalculated, so the associated p-values should be treated as approximate summaries rather than as sponsor-reported confirmatory results. The record does not provide the original statistical analysis plan, the test specification, or any planned handling of multiple comparisons. These details are not reported in the registry record.

Why does randomization matter?

In general, random allocation aims to make treatment groups comparable on average with respect to both measured and unmeasured characteristics at the start of a study. Double masking can reduce the influence of knowledge of treatment assignment on behavior, care, and assessment. These are general design principles. The registry reports randomized allocation and double masking, but it does not give the sequence-generation method, allocation concealment details, or information on masking success. Those specific procedures cannot be evaluated from this record.

What can the intention-to-treat label tell us?

The registry says the intention-to-treat population will be used for primary analysis. In general, intention-to-treat analysis seeks to retain participants in the groups to which they were randomized, regardless of treatment adherence. The label alone does not establish how the investigators handled missing outcome measurements, participants who discontinued, or deviations from the assigned regimen. The record does not report those details or identify the precise number analyzed for every outcome beyond the arm counts posted for the primary measure.

Why can variability affect interpretation?

General statistical principle: over-dispersion means that the observed variation is greater than a statistical model assumes. When a model underestimates variation, standard errors can be too small and confidence intervals too narrow, making evidence appear more precise than it is. This trial’s primary endpoint is continuous, and the registry posts standard deviations; it does not report an over-dispersion assessment or describe a model that would require such an assessment. The concept is relevant as a general caution, not a reported feature of this trial.

10. Confidence Intervals and Precision

The approximate 95% confidence intervals express the range of differences in mean change that is compatible with the posted summaries under the assumptions used for the recalculation. They are not ranges containing most individual participant results, and they are not guarantees about the true effect. In general, a narrower interval indicates greater statistical precision, while a wider interval indicates more uncertainty.

For Treatment Group A versus placebo, the recalculated mean difference is -6.1 seconds, with a 95% confidence interval from -9.777 to -2.423. For Treatment Group B versus placebo, it is 9.6 seconds, with an interval from 5.882 to 13.32. For Treatment Group C versus placebo, it is -12.4 seconds, with an interval from -16.41 to -8.386. Each interval lies entirely on one side of zero under this approximate unadjusted calculation. This does not address multiplicity, clinical importance, or whether the sponsor’s model-based analysis would produce the same estimates.

The intervals are approximate because the posted comparison calculations rely on summary statistics and approximate standard deviations. The registry does not report the sponsor’s confidence intervals, the full analysis model, or the covariance structure of any adjusted estimates. The posted arm values are identified as least-squares means, another reason not to treat the recalculated intervals as a reproduction of the sponsor’s analysis.

Precision also depends on the outcome’s variability and the number of observations contributing to an estimate. The arm standard deviations range from 8.73 to 9.69 seconds, and the primary-outcome group counts differ. The record does not explain missingness or provide participant-level results, so the effect of incomplete observations on precision cannot be assessed. Confidence intervals quantify sampling uncertainty under their assumptions; they do not capture every source of uncertainty, including possible bias from missing data or analysis choices.

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 sample-size figures are planning illustrations based on fractions of the observed comparison effects, not the trial’s original enrollment plan. For these continuous-outcome comparisons, the numbers are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the number of participants per arm estimated to provide an 80% chance of detecting the specified effect under the stated planning assumptions, if that effect is real and the assumptions hold. The 90% power column uses a higher target chance of detecting it and therefore generally calls for more participants per arm. For example, planning for 75% of the observed effect gives 59 participants per arm for 80% power and 79 participants per arm for 90% power for Treatment Group A. The corresponding entries are 24 and 32 for Treatment Group B, and 16 and 21 for Treatment Group C.

Each comparison has its own observed effect and its own planning figures, so figures should be read within the relevant treatment-versus-placebo comparison. Planning for a smaller effect requires more participants per arm than planning for a larger effect. These figures are not observed power: they are hypothetical planning calculations, not a measure of the probability that this completed trial detected an effect. The planning calculations also do not establish that the original study was designed with these assumptions.

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 = -6.100; pooled standard deviation σ = 8.837.
2. n = 2 × (1.96 + 0.842)² × 8.837² / -6.100² = 32.9, rounded up to 33 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 <- -6.1                      # observed difference in means
sigma <- 8.83689073888743                     # 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(1239511)
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 = 33 (80% power), 45 (90% power)
 75% of observed effect: n per arm = 59 (80% power), 79 (90% power)
 50% of observed effect: n per arm = 132 (80% power), 177 (90% power)
power.t.test: 33.9 
Simulated power at n = 33 per arm: 0.787 (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 lists three dose groups compared with placebo for the primary continuous outcome. The displayed p-values are unadjusted recalculations. The registry record does not report whether the original analysis controlled the chance of false-positive findings across these comparisons, whether a hierarchy or gatekeeping procedure was planned, or whether the dose-response objective had a separate prespecified test. No multiplicity adjustment should be assumed from the available record.

This matters because repeated testing can increase the chance of at least one apparently notable result even when there is no true difference. General methods for addressing multiple tests include prespecified ordering, adjustment of significance thresholds, or interpretation within a broader model. The choice should depend on the study’s aims and protocol; the registry does not report which approach, if any, was planned. The approximate comparisons are therefore best read as individual unadjusted comparisons, not as a complete multiplicity-controlled analysis.

The primary outcome description identifies an intention-to-treat population, but does not define the exact analysis set or explain how missing measurements were handled. The record does not report the number of missing primary outcomes, reasons for missingness, imputation methods, sensitivity analyses, or how treatment discontinuation was addressed. It also does not report whether participants were analyzed according to assigned group in the sponsor’s final analysis.

These gaps prevent evaluation of whether the posted arm counts represent all randomized participants with available data, a modified intention-to-treat set, or another analysis population. Missing-data methods can affect both estimates and their uncertainty, especially if the chance of an unavailable outcome is related to treatment or health status. That is a general concern; the registry does not establish that missing data biased these results.

The registry does not post the statistical analysis plan or a complete sponsor analysis, and it lists no analyses. Consequently, details such as covariate adjustment, model diagnostics, prespecified contrasts, and sensitivity analyses are not reported in the registry record. The recalculated results provide a transparent approximate comparison of the posted summaries but cannot fill those omissions.

14. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: gastrointestinal haemorrhage (2.1%), coronary artery disease (2.1%), angina pectoris (2.1%), pyrexia (2.1%), hepatitis acute (2.1%).

The adverse-event record reports serious adverse events for 0 of 47 participants at risk in the placebo group, 0 of 45 in Treatment Group A, 2 of 47 in Treatment Group B, and 1 of 47 in Treatment Group C. These are the reported numbers affected in each group, with the corresponding numbers at risk stated alongside them. The registry does not provide further detail about event timing, severity beyond the serious classification, relationship to study treatment, or whether any participant had more than one event.

For other adverse events, the record reports 0 of 47 in placebo, 2 of 45 in Treatment Group A, 2 of 47 in Treatment Group B, and 1 of 47 in Treatment Group C. The deaths field is not populated; deaths are not reported in the registry record. The record also lists serious-event terms with 2.1% each, but does not specify in that listing which treatment group each term pertains to. The term listing should not be assigned to a particular group without that information.

These figures are descriptive safety counts, not a formal comparison of risk. The registry does not report statistical testing of adverse events or enough detail to interpret differences between groups. Sparse counts also make estimates of comparative safety uncertain. The counts should be considered in the context of the study’s duration of 42 days and the reporting limitations, without inferring that an event was caused by or unrelated to STA-2.

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

The posted primary-outcome summaries show different mean changes in total exercise time across the placebo and STA-2 groups. The approximate unadjusted comparison estimates are negative for Treatment Groups A and C and positive for Treatment Group B. The associated confidence intervals quantify sampling uncertainty under the recalculation assumptions, but they do not account for every possible source of uncertainty or replace the sponsor’s analysis.

The findings do not establish a smooth dose-response pattern. The means do not rise or fall in sequence across the three doses, and the registry does not post a formal dose-response test. Nor does a difference between group averages show how each participant responded. The registry does not report a threshold for a clinically meaningful change, so the statistical estimates alone cannot determine whether the observed differences matter in clinical terms.

The record describes a randomized, double-blind design, but important methodological details remain unavailable, including the original statistical analysis plan, missing-data handling, and multiplicity approach. The reported p-values are unadjusted recalculations rather than sponsor-posted model results. These limitations mean the findings should be treated as a description of the posted summaries and their approximate precision, not as a complete evaluation of the trial’s prespecified analysis.

The study was conducted in people with chronic stable angina and assessed a particular exercise-test outcome after the described treatment period. The registry does not establish how results apply to people outside the study population, other treatment durations, or outcomes not reported in the posted results. It also does not establish why the group means differ. The data should not be interpreted as medical advice or as a definitive statement about treatment benefit or lack of benefit.

16. Why This Trial Matters Statistically

This trial is statistically notable for presenting a dose-ranging question alongside a placebo comparison. Evaluating several dose groups raises more than one analytical question: how each group compares with control, whether there is evidence of a dose-response relationship, and how uncertainty should be controlled when multiple comparisons are considered. The registry’s objective names dose-response evaluation, but its posted results do not include a formal dose-response analysis or a reported multiplicity strategy.

The group summaries also illustrate why a single headline estimate can be misleading. The observed mean changes vary in direction across the doses, and the approximate comparisons have distinct magnitudes and uncertainty intervals. A dose-response claim requires an analysis suited to the ordered dose question; separate treatment-versus-placebo comparisons alone do not establish a trend. The pattern is a reason to examine the prespecified analysis and full results, not a basis for inferring a mechanism.

The record offers enough summary information for approximate unadjusted comparisons, but not enough to reproduce the sponsor’s primary analysis. It identifies an intention-to-treat population and least-squares means, yet does not post model details, missing-data procedures, or the analysis plan. This creates an important distinction between transparent recalculation from available summaries and confirmation of the original statistical conclusions.

For clinical researchers, the trial underscores the value of reporting the planned contrasts, analysis population, treatment of missing outcomes, multiplicity control, and confidence intervals alongside group means. Those details make it possible to assess both statistical precision and the relation between a stated objective and the analysis used to address it. Their absence in this registry record limits the strength of conclusions that can be drawn from the posted results.

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.

Interpretation is limited by the registry’s summary-level reporting. The primary outcome is described as a least-squares mean, but the posted comparison calculations are approximate unadjusted recalculations from summary statistics. The registry does not give the model specification, any covariates, the sponsor’s confidence intervals, or the full sponsor analysis output. The recalculations therefore cannot verify the original analysis.

The record states that the intention-to-treat population will be used, but does not report the detailed analysis population or explain the difference between the enrollment figure and the arm counts reported for the primary outcome. The reasons for any unavailable observations, withdrawals, or exclusions are not reported in the registry record. The effect of missing data on the estimates cannot be evaluated.

The registry does not report a prespecified multiplicity procedure for the dose comparisons or a formal dose-response test. This is particularly relevant because the study objective includes dose-response evaluation, while the posted comparisons are separate unadjusted contrasts against placebo. The statistical analysis plan and the original sample-size justification are also not reported in the registry record.

The record gives limited detail about randomization and masking procedures, baseline characteristics, the exercise-test protocol, and the clinical importance threshold for a change in exercise time. The adverse-event listing does not fully attribute the serious-event terms to treatment groups, and deaths are not reported in the registry record. These limits restrict evaluation of the study’s internal validity, clinical interpretation, and safety comparisons.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Phase IIb Study of STA-2 in Patients With Chronic Stable Angina — NCT01239511.
• Registry record last updated 2014-08-27; status last verified 2014-08.
• Results first posted 2014-08-27.
• 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

The registry describes a completed phase IIb randomized, double-blind, placebo-controlled, parallel-group study of STA-2 in people with chronic stable angina. The recorded study start was 2010-11, and primary completion was 2012-04. The study objective was to evaluate three daily doses of STA-2 against placebo over 42 days.

The primary outcome was change in total exercise time, measured in seconds, with the comparison described from V2 to V5 and an assessment timeframe of 6 weeks after the first exercise tolerance test. The registry identifies an intention-to-treat population. Posted group means and standard deviations support approximate unadjusted comparisons, but the registry lists no analyses and does not provide the full statistical analysis plan.

Secondary outcomes concern exercise-test measures, angina frequency, short-acting nitrate use, pharmacological and lipid measures, and dose response. Safety entries report serious and other adverse-event counts by group; deaths are not reported in the registry record. Multiplicity handling, missing-data methods, and the original sample-size justification are not reported.