Study to Evaluate Safety, Tolerability and Efficacy of Saroglitazar Mg in Patients With Primary Biliary Cholangitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Evaluate Safety, Tolerability and Efficacy of Saroglitazar Mg in Patients With Primary Biliary Cholangitis (NCT03112681), a randomized phase 2 study. The primary outcome is Effect of a 16-week Treatment Regimen of Saroglitazar Magnesium 2 mg and 4 mg on Alkaline Phosphatase (ALP) Levels in Patients With Primary Biliary Cholangitis..

2. Clinical Question

3. Trial Design

This was a phase 2, prospective, multicenter, randomized, double-blind, placebo-controlled, parallel-group study. The registry record also describes masking as triple. Participants were assigned to saroglitazar magnesium 2 mg, saroglitazar magnesium 4 mg, or placebo. The treatment descriptions specify once-daily tablets for 16 weeks.

The registry lists enrollment as 37. Its brief summary says that a total of 36 subjects would be enrolled in a 1:1:1 ratio. These are different enrollment figures within the record; an explanation for the difference is not reported in the registry record. The posted primary outcome results give group sizes of 14 for the 2 mg group, 13 for the 4 mg group, and 10 for placebo. The reported analysis population is modified intent-to-treat.

The study is marked completed. Its start date is 2017-08-18 and its primary completion date is 2020-08-07. The sponsor is Zydus Therapeutics Inc. The registry record reports one outcome measure and two analyses. It does not report the original sample-size justification, the planned analysis population details, or the exact statistical analysis plan. The randomization sequence and allocation-concealment procedures are not reported in the registry record.

The primary endpoint concerns change in ALP at Week 16, with baseline as the reference measurement. The posted results show a mean change for each group and a mean difference for each saroglitazar magnesium dose compared with placebo. These results describe the recorded trial outcome; they do not, on their own, resolve questions about longer-term outcomes or the broader clinical meaning of a laboratory change.

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

Analysis population for the primary outcome, as defined in the registry: Modified intent-to-treat population

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 Effect of a 16-week Treatment Regimen of Saroglitazar Magnesium 2 mg and 4 mg on Alkaline Phosphatase (ALP) Levels in Patients With Primary Biliary Cholangitis., measured over Baseline and Week 16.

The primary endpoint was change in alkaline phosphatase (ALP) levels from baseline to Week 16. ALP is a continuous laboratory measurement, and the registry reports the group results as means with standard deviations, in U/L. A change score describes how the measured value at follow-up differs from its baseline value; a negative mean change in the posted results indicates a decrease over the measurement period.

ALP is used in the assessment of biliary and liver-related conditions. As a general statistical and clinical measurement principle, a laboratory endpoint can provide a measurable way to compare groups over a defined follow-up period. It is not automatically equivalent to a patient-important outcome, and a change in a laboratory measure does not alone establish how a patient feels, functions, or fares over the longer term.

The stated outcome description says the analysis concerns change after 16 weeks of treatment. The registry identifies the modified intent-to-treat population for this outcome but does not define that population or explain how missing baseline or follow-up measurements were handled. No secondary outcomes are posted in the registry record. The primary endpoint therefore gives a focused comparison of mean biochemical change at a specified time, while leaving those additional interpretive details unreported.

7. Results Posted in the Registry

• Saroglitazar Magnesium 2 mg: mean -163.82 versus -11.15 on control; difference -152.67 (95% CI -213.66 to -91.68), standardized effect d = -1.74; z-test p = <0.001.
• Saroglitazar Magnesium 4 mg: mean -162.31 versus -11.15 on control; difference -151.16 (95% CI -215.40 to -86.92), standardized effect d = -1.72; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: each participant contributes a numerical ALP change, and the registry reports group means and standard deviations. A mean summarizes the center of the group’s observed values. A standard deviation describes their spread around that mean. The standard deviation is not the uncertainty of the mean itself; uncertainty in a group estimate also depends on the number of observations and the analysis method.

What does the mean difference represent?

The reported comparisons subtract the placebo mean change from the mean change in one saroglitazar magnesium group. The recalculated mean difference is -152.7 U/L for 2 mg versus placebo and -151.2 U/L for 4 mg versus placebo. The negative sign follows from the reported order of groups and the negative mean changes: the average change in each active-dose group was more negative than the average placebo change. This is an unadjusted contrast based on posted summary statistics, not an adjusted estimate from a model incorporating participant characteristics.

A group contrast is more informative for a randomized comparison than looking at one group’s change alone, because it takes the observed placebo change into account. However, the posted summaries do not provide enough information to assess baseline balance, individual trajectories, or whether any adjustment was made in the sponsor’s analysis. Such details are not reported in the registry record.

What does the t-test do?

A t-test evaluates whether a mean contrast is compatible with a null model in which the relevant mean difference is zero, subject to the assumptions of the particular test. The registry identifies the sponsor’s method as a paired t-test and reports p values of < 0.001 and < 0.001 for the two active-dose comparisons with placebo. It does not explain what observations were paired, how the pairing was defined, or whether the test was applied to individual change scores or another quantity. The method label alone does not resolve those details.

The recalculated comparisons use posted summary statistics to produce unadjusted mean differences, confidence intervals, and p values. The calculations use pooled standard deviations reported for each comparison and the group sizes listed with the outcome. They are not a substitute for the sponsor’s full analysis, which may use participant-level data or other specifications not described in the registry record. The record marks no discrepancy between the recalculated and sponsor results; nevertheless, the methods and reported outputs are not identical descriptions and should be kept distinct.

Why are standard deviations relevant?

Variation in participant outcomes affects how precisely a group mean and the difference between group means can be estimated. Greater dispersion generally makes an estimate less precise when the number of observations is held fixed. The reported standard deviations are notably different across the active and placebo groups, so a simple equal-variance approach needs care. A pooled standard deviation is a summary used in the posted recalculations; it is not evidence that the underlying group variances are truly equal.

General statistical context: when variability differs between groups, methods that do not require equal variances can be considered. The registry does not report a variance assessment or enough detail to determine which assumptions were examined in the sponsor analysis. The available summary statistics also cannot reveal skewness, outliers, or the shape of the participant-level changes.

How should p values be read?

A p value is calculated under a specified null model and test procedure. It describes the compatibility of the observed data with that model; it is not the probability that the null hypothesis is true. The sponsor p values are < 0.001 and < 0.001, while the recalculated results are each reported as p < 0.001. These values indicate strong statistical evidence against a zero mean difference under the respective tests, but they do not quantify clinical importance or eliminate uncertainty about the size of the contrast.

The registry does not report whether the sponsor adjusted for multiple comparisons. Two active-dose comparisons are posted, and multiplicity can affect how a collection of tests is interpreted. Without a stated adjustment strategy or prespecified testing hierarchy, the family-wise interpretation of the p values cannot be determined from the registry record.

What does the analysis population imply?

The primary outcome is identified as using a modified intent-to-treat population. The record does not define which participants were included or excluded, nor does it state how missing Week 16 measurements were handled. In general, the exact analysis population and missing-data method matter because they determine which observed data contribute to the estimates and how incomplete outcomes are treated. The group sizes reported for the endpoint should be read as the analysis sizes associated with the posted results, not as a complete account of participant flow.

These statistical methods describe the posted contrasts and their uncertainty. They cannot recover unreported protocol details, model specifications, or participant-level diagnostics. The registry record does not report interim analyses or the original statistical analysis plan.

10. Confidence Intervals and Precision

The recalculated 95% confidence interval for the mean difference between saroglitazar magnesium 2 mg and placebo extends from -213.7 to -91.68 U/L. For saroglitazar magnesium 4 mg versus placebo, it extends from -215.4 to -86.92 U/L. Both intervals are entirely below zero, consistent with the negative mean differences in the unadjusted calculations.

A confidence interval gives a range of values compatible with the observed data under the calculation’s assumptions and procedure. In repeated use of the same procedure under its assumptions, intervals at the stated confidence level would cover the target parameter at the corresponding frequency. It is not a range containing a stated probability that this particular fixed effect lies within it. The interval also does not include uncertainty from every possible design or analysis choice.

The two intervals overlap substantially, but overlap alone is not a formal test of whether the two active doses differ. The registry does not report a direct active-dose comparison. The intervals are for each dose versus placebo, not for the difference between the doses.

The width of an interval reflects statistical precision: narrower intervals generally indicate more precise estimates, while wider intervals indicate greater uncertainty. The active groups and placebo group have limited reported analysis sizes, and the standard deviations differ across groups. The registry record does not report confidence intervals for the sponsor’s paired t-tests or enough analysis details to reconcile those with the recalculated intervals. These intervals therefore describe the specified unadjusted recalculations, not every uncertainty relevant to interpretation.

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 illustrations report participants per arm for 80% power and participants per arm for 90% power. They do not report total events: the outcome is continuous, so the listed figures are participants per arm. For the 2 mg comparison, planning for 100% of the observed effect gives 6 participants per arm for 80% power and 7 for 90% power; planning for 75% gives 10 and 13, and planning for 50% gives 21 and 28. For the 4 mg comparison, the corresponding figures are 6 and 8 at 100%, 10 and 13 at 75%, and 22 and 29 at 50%.

These are planning illustrations based on the observed effect, not the original design assumptions of the trial. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. The registry record does not report the trial’s original sample-size justification or design assumptions.

The 80% power column means the illustrated design would have an 80% chance of detecting the specified effect under its assumptions; the 90% column represents a higher chance under those same assumptions. Seeking higher power requires more participants in these illustrations. Smaller target effects also require more participants, as reflected by the larger figures for 50% of the observed effect than for 100%. This is not observed power; it is a planning calculation using an effect estimated from the posted results.

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 = -152.670; pooled standard deviation σ = 87.735.
2. n = 2 × (1.96 + 0.842)² × 87.735² / -152.670² = 5.2, rounded up to 6 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 <- -152.67                      # observed difference in means
sigma <- 87.735019234055                     # 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(3112681)
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 = 6 (80% power), 7 (90% power)
 75% of observed effect: n per arm = 10 (80% power), 13 (90% power)
 50% of observed effect: n per arm = 21 (80% power), 28 (90% power)
power.t.test: 6.3 
Simulated power at n = 6 per arm: 0.778 (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. Analyses Reported by the Sponsor

The registry reports sponsor analyses using paired t-tests. The reported p value is < 0.001 for saroglitazar magnesium 2 mg versus placebo and < 0.001 for saroglitazar magnesium 4 mg versus placebo. The record does not provide a sponsor-reported mean difference, confidence interval, test statistic, or further detail about the paired-test implementation.

The unadjusted recalculations from posted summary statistics estimate mean differences of -152.7 U/L and -151.2 U/L, respectively, and give p < 0.001 for both comparisons. These are simple recalculations, not a reproduction of the sponsor’s participant-level analysis. The record marks no discrepancy, but the two sources of results differ in the detail they report: the sponsor provides the test method and p values, while the recalculations provide effect estimates and confidence intervals.

The sponsor results should be interpreted according to the method as recorded, while recognizing that the registry does not describe the test’s pairing structure or analysis specifications. The recalculated estimates help express the magnitude and precision of the group contrasts, but they do not replace the sponsor’s analysis or resolve the missing methodological detail.

14. Multiplicity, Missing Data and Analysis Populations

The registry record posts two primary-outcome comparisons: saroglitazar magnesium 2 mg versus placebo and saroglitazar magnesium 4 mg versus placebo. It reports a sponsor p value for each and gives no secondary outcomes. The record does not state whether the two comparisons were treated as a family for inference, whether a multiplicity adjustment was applied, or whether testing followed a prespecified hierarchy. Without that information, the stated p values should be read as reported for each comparison rather than assumed to control an overall error rate across comparisons.

The primary analysis population is described as modified intent-to-treat. The record does not define the modification, identify exclusions from the randomized population, or provide a participant-flow account for the primary endpoint. The outcome group sizes are reported, but the reasons for any difference between enrollment and endpoint analysis are not reported in the registry record.

Missing-data handling is also not reported. The record does not say how absent Week 16 measurements were treated, whether any values were imputed, or what assumptions underlay an incomplete-data analysis. These choices can affect a mean change estimate and its uncertainty, especially when outcomes are not observed for all participants. The posted summary statistics do not allow those choices to be reconstructed.

The registry identifies a paired t-test as the sponsor’s method but does not explain the pairing, test assumptions, or whether covariates were included. It does not report interim analyses, a prespecified statistical analysis plan, or the trial’s original sample-size justification. These omissions limit assessment of whether reported tests followed a prespecified plan and how robust conclusions may be to alternative analysis choices. They do not negate the posted estimates, but they constrain the detail with which the statistical evidence can be evaluated.

15. 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: oedema peripheral (7.1%), appendicitis (7.1%), dyspnoea (7.1%).

The registry reports serious adverse events affecting 2 participants in the saroglitazar magnesium 2 mg group, none in the 4 mg group, and none in the placebo group. The groups at risk for these serious adverse event summaries were 14, 13, and 10, respectively. No deaths are reported in any group. The record also reports other adverse events affecting 12 participants in the 2 mg group, 11 in the 4 mg group, and 8 in the placebo group.

These are counts of affected participants as recorded, not a measure of event severity beyond the serious-event classification and not an estimate of comparative risk adjusted for follow-up or other factors. No causal relationship between an event and study treatment is established by the counts alone. The registry lists peripheral oedema, appendicitis, and dyspnoea among the top serious adverse event terms, each reported as 7.1%; it does not associate those listed terms with a particular group in this record.

Safety counts should be read alongside the limited group sizes and the short, 16-week treatment period. The registry record does not provide detailed event narratives, timing, treatment relatedness assessments, or a full safety analysis plan. These figures describe the reported record and are not sufficient to characterize uncommon or longer-term events.

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

The posted primary outcome results show negative mean ALP changes in all three groups, with a more negative mean change in each saroglitazar magnesium group than in placebo. The unadjusted mean differences versus placebo are also negative, and their confidence intervals exclude zero. The sponsor’s reported paired t-test p values are < 0.001 and < 0.001. These results describe the measured Week 16 endpoint in the registry’s modified intent-to-treat population.

They do not, by themselves, establish the clinical importance of the size of the laboratory difference. The registry does not report a threshold for a clinically meaningful ALP change, patient-reported outcomes, or secondary clinical outcomes. It also does not post a direct comparison of the two active doses, so similarity of their observed group means is not evidence that the regimens are equivalent.

The record concerns a phase 2 study with a placebo comparator and a 16-week treatment period. It does not establish what happens beyond that period, in populations not represented in the analysis, or on outcomes not reported. The small analysis groups and absent details about missing data, analysis populations, and multiplicity add uncertainty to a full appraisal.

Statistical evidence against a zero mean difference is not the same as proof of a clinically meaningful effect, and a p value does not describe the probability that a result will recur or the chance of an outcome for an individual. The posted results support a description of group-level biochemical changes and their calculated uncertainty. Decisions about clinical meaning require context beyond these registry results, including relevant outcomes and fuller trial methods.

17. Why This Trial Matters Statistically

This trial presents a useful statistical contrast between a continuous laboratory outcome in two active-dose groups and a placebo group. Looking at the mean change in each group alongside the mean difference makes clear why a within-group decrease alone is not the complete comparison: the placebo group also had a mean change, and the randomized comparison concerns the difference between groups.

The results also illustrate the value of reporting uncertainty alongside point estimates. The recalculated confidence intervals communicate a range of plausible mean differences under the stated calculation, while the p values answer a different, more limited question about compatibility with a null model. Neither summary alone expresses clinical importance.

Several common interpretation issues are visible in the record. The sponsor’s paired-test method is named but not fully described; multiplicity handling and missing-data methods are not reported; and a modified intent-to-treat population is named without a definition. These gaps show why a numerical result needs accompanying analysis details to be fully assessed. A small p value cannot substitute for understanding the estimand, analysis population, test assumptions, and handling of incomplete observations.

The record also separates planning illustrations from the trial’s original design rationale. Sample-size figures based on the observed effect are useful for illustrating how smaller target effects and higher power alter planning requirements, but they are not evidence of the study’s original rationale or of achieved power. Keeping these distinctions clear helps prevent post hoc planning calculations from being mistaken for design details.

Overall, the statistical importance of the record lies in its explicit group-level outcome, comparative estimates, and visible limits on what can be inferred. The results can be interpreted at the level of the posted Week 16 ALP outcome while remaining cautious about unreported methods and broader clinical implications.

18. 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 contains an enrollment inconsistency: it lists enrollment as 37, while the brief summary says a total of 36 subjects would be enrolled. The record does not explain this difference. The primary outcome group sizes are reported as 14, 13, and 10, but a participant-flow account explaining endpoint inclusion is not reported.

The study is described as triple-masked in its design fields and double-blind in its title and summary. The record does not explain the masking terminology or identify which roles were masked. Random sequence generation and allocation concealment are not described. The original sample-size justification, statistical analysis plan, and any interim analyses are not reported in the registry record.

The primary endpoint is a Week 16 change in ALP in a modified intent-to-treat population, but the record does not define that population or describe how missing outcome data were handled. It also does not specify whether or how multiplicity was addressed for the two dose-versus-placebo comparisons. The sponsor’s paired t-test is named, but details about the pairing and model implementation are absent.

The posted record gives group means and standard deviations, but not participant-level distributions, baseline ALP summaries, adjusted estimates, or a direct comparison of the active doses. The duration of treatment is 16 weeks; longer-term outcomes and secondary outcomes are not posted. These limitations affect the scope and precision of interpretation, even though the primary comparison results are reported.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study to Evaluate Safety, Tolerability and Efficacy of Saroglitazar Mg in Patients With Primary Biliary Cholangitis — NCT03112681.
• Registry record last updated 2024-09-19; status last verified 2024-09.
• Results first posted 2024-09-19.
• 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.

22. Record Summary

The registry record describes a completed phase 2, randomized, parallel-group, placebo-controlled study of saroglitazar magnesium in patients with primary biliary cholangitis. The treatment groups received 2 mg, 4 mg, or placebo once daily for 16 weeks. The primary endpoint was change in ALP from baseline to Week 16 in a modified intent-to-treat population.

Posted mean changes were -163.8 U/L, -162.3 U/L, and -11.15 U/L for the 2 mg, 4 mg, and placebo groups. The unadjusted recalculated mean differences versus placebo were -152.7 U/L and -151.2 U/L, with p < 0.001 for each. Sponsor-reported paired t-test p values were < 0.001 and < 0.001. The registry does not report secondary outcomes, missing-data methods, or multiplicity procedures. Safety summaries report 2 serious adverse events in the 2 mg group and none in the other groups; no deaths were reported.