Use of Low Dose Pioglitazone to Treat Autosomal Dominant Polycystic Kidney Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Use of Low Dose Pioglitazone to Treat Autosomal Dominant Polycystic Kidney Disease (NCT02697617), a randomized phase 2 study. The primary outcome is Safety: Total Body Water.

2. Clinical Question

3. Trial Design

This was a randomized, double-masked phase 2 trial with a crossover model. Indiana University is listed as the sponsor, and the trial is marked completed. The recorded start date is 2016-01-26, and primary completion is recorded as 2019-10. Enrollment was 18. The registry lists pioglitazone and placebo as the interventions, with one arm described as receiving pioglitazone and the other as receiving placebo.

A crossover trial differs from a parallel-group trial because treatment comparisons can be made within participants across treatment periods. That design can account for stable differences between participants when the analysis appropriately uses the paired observations. The registry identifies a paired t-test for the primary comparison, which is consistent with treating observations as paired. However, it does not report the treatment sequence, period order, washout arrangements, or how period effects were handled. Those details are not reported in the registry record.

The primary-outcome population is described as all patients randomized. The primary outcome timeframe is an average of 4 measures in each 12 month arm. The record does not explain the timing of those measurements, how the average was formed, or what happened if a measurement was unavailable. Those details are not reported in the registry record.

The registry lists five secondary outcomes: hypoglycemia, elevated liver function tests, glomerular filtration rate, blood pressure, and bone marrow fat. It also reports seven outcome measures and five analyses. The analysis methods for the secondary outcomes and the relationship between the listed outcome measures and analyses are not detailed in the registry record. A sample-size justification and the trial's planned statistical analysis document 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: Patients who fulfilled the initial screening underwent further screening with a baseline magnetic resonance imaging (MRI) and randomized to pioglitazone or placebo providing the total kidney volume (TKV) was ≥675 ml (18-25 years old), ≥ 900 ml (26-35 years old), and ≥ 1350 ml (36-55 years old).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All patients randomized

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: 18 patients randomized to either pioglitazone for 12 months and then placebo for 12 months and then cross over to the other arm. So baseline data represents both arms

6. Primary Endpoint

The registry lists the primary outcome as Safety: Total Body Water, measured over average of 4 measures in each 12 month arm.

The primary endpoint was total body water assessed using bioimpedance analysis, with the measure reported in Ohms. The registry specifies that an increase in BIA in Ohms indicates a decrease in total body water. This direction is important: a higher reported BIA value should not be read as a higher amount of body water under the registry's stated interpretation.

The outcome was summarized as a mean, with a 95% confidence interval for each arm. The timeframe was an average of 4 measures in each 12 month arm, and the stated population was all patients randomized. Repeated measurements can describe the outcome over the treatment period more fully than a single observation, but the registry does not explain how the repeated values were combined beyond specifying an average.

As a general statistical point, a continuous measure such as BIA can capture variation across a range rather than grouping participants into categories. It is useful to distinguish the measurement itself from its clinical interpretation: the registry defines the direction of the BIA scale, but it does not report a threshold for a clinically important change in total body water. Such a threshold is not reported in the registry record.

The record reports arm-level means and confidence intervals, as well as a paired test of the comparison. It does not report the mean within-participant treatment contrast or a confidence interval for that contrast. Without those contrast statistics, the arm summaries alone do not convey the size and precision of the paired treatment comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: BIA in Ohms, summarized as a mean. A mean describes the center of the observed measurements, while a confidence interval expresses uncertainty around that estimate under the method used to calculate it. The registry reports a 95% confidence interval for each arm mean. These arm-level intervals are not the same as an interval for the treatment contrast.

Why does pairing matter?

In a crossover design, observations from the same participant under different treatments are related. A paired analysis works with within-participant comparisons, rather than treating the observations as though they came from unrelated groups. This can remove some variation due to stable participant characteristics. As a general statistical principle, that benefit depends on the design and analysis being appropriate, including adequate consideration of order or period effects.

The registry reports a paired t-test for the comparison of pioglitazone and placebo. In general, a paired t-test evaluates whether the mean of the within-pair differences is distinguishable from zero, under assumptions about those differences. The registry does not report the mean paired difference, its standard error, or a confidence interval for it. The p-value therefore indicates the reported test result but does not show the estimated contrast's size or precision.

What does the p-value say?

The posted p-value is 0.024. In general, a p-value describes how unusual data at least as incompatible with the null model would be if that model and the test assumptions held. It is not the probability that the null hypothesis is correct, and it does not measure clinical importance. Its interpretation also depends on the analysis having been specified and applied appropriately. The planned analysis details are not reported in the registry record.

The registry labels the paired comparison as a superiority analysis. A superiority test asks whether the data support a difference in the direction specified by the analysis, rather than testing whether two treatments are equivalent. The record does not report a prespecified superiority margin or a clinically important threshold for this endpoint. Those items are not reported in the registry record.

How should the confidence intervals be read?

The pioglitazone arm mean is 45.78, with a 95% confidence interval of 39.19 to 52.36. The placebo arm mean is 44.17, with a 95% confidence interval of 37.96 to 50.39. As a general explanation, a confidence interval gives a range of values compatible with the observed data and the interval method, under its assumptions. It is not a range containing a specified proportion of individual participant measurements, and it is not a direct probability statement about this particular interval after it has been calculated.

Because these intervals belong to separate arm means, they do not supply the uncertainty for a paired contrast. In a crossover analysis, the variability of the within-participant differences is relevant. The registry does not report that variability or a confidence interval for the treatment contrast. It is therefore not possible to infer the contrast's precision from the two arm intervals alone.

What assumptions are relevant?

General considerations for a paired t-test include the quality of the pairing, the distribution of the within-pair differences, and whether the outcome observations represent the intended treatment periods. In crossover trials, sequence, period, and carryover effects may also matter. The registry record does not report how these issues were assessed or addressed. Their absence from the record does not establish that they were ignored; it means the registry record does not provide the details.

The primary outcome averages 4 measures in each 12 month arm. Repeated measurements can be correlated within a participant. As a general matter, the analysis should account for the structure of repeated observations or define clearly how those observations are reduced to the analysis value. The registry states that an average was used, but the calculation procedure, measurement schedule, and approach to incomplete measurements are not reported in the registry record.

Over-dispersion is a concern for some count outcomes when observed variability exceeds that expected under a simple count model. It is not the central issue for this continuous mean endpoint, although appropriate modeling of variability remains important. For the listed safety and efficacy outcomes, outcome-specific methods would be needed; those methods are not reported in the registry record.

What can planning calculations establish?

Any sample-size figures based on an observed effect are planning illustrations, not observed power. General planning calculations depend on assumptions, including the effect chosen for detection, significance level, allocation, and statistical approximation. The registry's stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such planning values do not show the probability that this completed trial detected an effect, and they do not replace a sample-size justification made before the trial. The sample-size justification is not reported in the registry record.

The registry also does not report whether interim analyses were planned or conducted, how multiplicity was controlled, or whether the primary test was adjusted for any design features beyond pairing. These details are not reported in the registry record. The posted p-value should therefore be interpreted as the result of the reported paired test, without assuming additional planned procedures.

10. Analyses Reported by the Sponsor

The registry's own reported primary comparison is a paired t-test, with p = 0.024. The posted arm means are 45.78 for pioglitazone and 44.17 for placebo, with separate 95% confidence intervals. The test result relates to the paired comparison; it should not be recreated by treating the arm means as independent or by comparing their separate intervals.

A paired treatment-effect estimate and its confidence interval are not reported in the registry record. A separate recalculated paired contrast is also not reported in the registry record, so there is no recalculated estimate to compare with the registry analysis. The record does not flag a discrepancy between a registry analysis and a recalculation. The sponsor's reported paired test is therefore the available analysis for this outcome, while the missing contrast estimate limits interpretation of its size and precision.

As a general statistical principle, a test result and an effect estimate answer different questions: the test describes evidence against a null model, while the estimate describes the contrast's magnitude, and its interval describes precision. The posted p-value should not be treated as an effect size or as evidence that the endpoint change is clinically important. A clinical-importance threshold is not reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five secondary outcomes: hypoglycemia, elevated liver function tests, glomerular filtration rate, blood pressure, and bone marrow fat. It reports seven outcome measures and five analyses, but does not give enough detail to map every measure to an analysis or specify which were confirmatory. A prespecified hierarchy of outcomes is not reported in the registry record.

When several outcomes or analyses are examined, the chance of at least one nominally unusual result can increase. As a general statistical concept, multiplicity procedures adjust interpretation or testing to account for multiple opportunities to find a result. The registry does not report whether any multiplicity adjustment was planned or used, whether any analyses were exploratory, or whether interim analyses were planned or conducted. Those details are not reported in the registry record. The primary p-value should consequently be read as the result of its reported test, not assumed to be adjusted or unadjusted beyond what the registry states.

The primary population is described as all patients randomized. The registry does not report whether this means every randomized participant contributed to the primary analysis, how deviations from assigned treatment were handled, or whether another analysis population was used. Those details are not reported in the registry record.

The primary measure averages 4 observations in each 12 month arm. The number of participants with missing measurements, the reasons for missingness, and any imputation or other missing-data method are not reported in the registry record. In general, missing outcome data can affect an estimate when the reason data are missing is related to the outcome or treatment. Without the missing-data approach and completeness information, the reader cannot assess how sensitive the posted result may be to missing observations.

The record also does not report a protocol-based statistical analysis plan, a sample-size justification, or details of sensitivity analyses. These omissions do not establish that such procedures were absent; they mean they are not reported in the registry record. They limit assessment of whether the posted analyses followed a prespecified plan and how robust the results are to alternative assumptions.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization (5.6%), pyelonephritis (5.6%), surgery (5.6%).

The safety results list 18 participants at risk in each arm. The record reports 0 affected by serious adverse events in the pioglitazone arm and 3 in the placebo arm. It reports 13 affected by other adverse events in the pioglitazone arm and 15 in the placebo arm. Deaths are reported as 0 in both arms.

These are descriptive counts in the registry record. They do not, on their own, establish whether a treatment caused an event or whether the groups differed reliably in safety. In a crossover trial, interpretation of adverse-event counts also depends on how events were attributed to treatment periods and how exposure time was handled; those details are not reported in the registry record.

The registry lists hospitalization, pyelonephritis, and surgery among the top serious-event terms, each with a reported percentage of 5.6%. Those terms should be read as listed event categories, not as proof of a causal relationship with either treatment. The record does not provide further event-level context, timing, severity detail, or attribution for these terms. Such details are not reported in the registry record.

The safety figures are not a substitute for the efficacy analysis, and the efficacy p-value does not answer a safety question. They describe the adverse-event information posted for this trial; they do not establish overall safety or rule out uncommon harms.

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

The posted primary outcome is a BIA-based measure of total body water, reported in Ohms. Under the registry's definition, an increase in BIA in Ohms indicates a decrease in total body water. The pioglitazone arm mean was 45.78, and the placebo arm mean was 44.17. A paired t-test comparing the arms is reported with p = 0.024.

These results describe the reported measurement and test in this completed phase 2 crossover trial. They do not establish that pioglitazone slows cyst growth or changes the course of autosomal dominant polycystic kidney disease. Although the study summary describes that as a motivation, the primary outcome presented is total body water, not a direct measure of cyst growth. The record lists other efficacy outcomes, including glomerular filtration rate and blood pressure, but the results and analysis details for those outcomes are not included in the stated primary-outcome information.

The p-value does not tell readers how large the paired treatment difference was. The registry does not report a paired mean contrast or its confidence interval, and the arm-specific confidence intervals cannot stand in for that contrast. A threshold for clinical importance is also not reported in the registry record. Consequently, the result cannot by itself show whether the measured difference was large enough to matter clinically.

The trial's design, enrollment, and available reporting also limit broader conclusions. The record identifies a crossover trial with enrollment of 18, but does not report the treatment order, washout arrangements, missing-data handling, or sample-size justification. The safety counts are descriptive and do not establish treatment causation. These considerations leave uncertainty about precision, design-related effects, and generalizability.

The result should therefore be interpreted narrowly: the registry reports a paired statistical comparison of the BIA endpoint with p = 0.024, alongside arm means and their confidence intervals. It does not establish a clinical benefit, prove a treatment effect on kidney disease progression, or demonstrate that no such effect exists.

14. Why This Trial Matters Statistically

This trial is statistically instructive because its crossover design makes the unit of comparison important. In a crossover study, treatment-period observations from the same participant are paired. The registry reports a paired t-test, which recognizes that structure, but does not provide the paired effect estimate or its confidence interval. That gap highlights why a p-value alone is not a complete account of a treatment comparison.

The endpoint also illustrates the need to understand measurement direction before interpreting an arm mean. The registry explicitly states that higher BIA in Ohms indicates lower total body water. A reader who assumes that a higher value means more of the measured quantity would reverse the endpoint's stated interpretation. Statistical summaries acquire meaning from the outcome definition, not merely from the scale or the direction of a numerical comparison.

Another important distinction is between a biomarker-like measurement and a clinical outcome. The study's summary frames pioglitazone as a possible way to slow cyst disease, while the primary endpoint is total body water measured by BIA. The registry record does not report a clinical-importance threshold linking a change in this measure to disease progression. That makes it essential to separate a statistical finding on the selected measure from a conclusion about patient-important effects.

Finally, the record demonstrates the value of transparent reporting for interpretation. Treatment sequence, missing-data methods, a paired contrast interval, and multiplicity procedures are not reported in the registry record. Their absence does not determine what the trial did, but it restricts what readers can verify from the registry entry. Clear reporting of effect estimates, uncertainty, and design details is particularly important when a small crossover study is interpreted.

15. Limitations

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

The primary contrast is reported through a paired t-test p-value, but the paired mean difference and its confidence interval are not reported in the registry record. This limits assessment of both the contrast's magnitude and its precision. The arm means and their separate confidence intervals do not provide that paired estimate.

Enrollment was 18, and the registry describes the primary population as all patients randomized. The sample-size justification and the basis for choosing enrollment are not reported in the registry record. The record also does not report the number of participants contributing complete measurements to the primary analysis or whether all randomized participants were included in that analysis.

Although the trial is identified as a randomized, double-masked crossover study, the treatment sequence, period order, washout arrangements, and approach to period or carryover effects are not reported in the registry record. These are relevant features for interpreting a crossover comparison. The record describes an average of 4 measures in each 12 month arm, but does not report the measurement schedule, the method used to calculate that average, or the handling of incomplete measurements.

The record lists secondary outcomes, but outcome-specific estimates, analysis methods, and multiplicity procedures are not reported in the primary-outcome information. Missing-data methods, sensitivity analyses, and the planned statistical analysis document are also not reported in the registry record. A threshold for a clinically important BIA change is not reported, limiting translation of the statistical result into clinical terms.

These limitations concern what can be assessed from the registry record; they do not establish that the trial failed to address these issues. They do mean that the posted primary result cannot be fully evaluated for the size and precision of the paired effect, sensitivity to missing data, or the handling of key crossover design features.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Use of Low Dose Pioglitazone to Treat Autosomal Dominant Polycystic Kidney Disease — NCT02697617.
• Registry record last updated 2021-01-15; status last verified 2020-12.
• Results first posted 2021-01-15.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry describes a completed, randomized, double-masked phase 2 crossover trial sponsored by Indiana University, with enrollment of 18. It compared pioglitazone with placebo in people with polycystic kidney disease. The primary endpoint was mean BIA in Ohms, an indicator interpreted by the registry such that an increase corresponds to a decrease in total body water. The timeframe was an average of 4 measures in each 12 month arm.

The pioglitazone arm mean was 45.78, with a 95% confidence interval of 39.19 to 52.36. The placebo arm mean was 44.17, with a 95% confidence interval of 37.96 to 50.39. The reported paired t-test p-value was 0.024. A paired treatment-effect estimate and its confidence interval are not reported in the registry record. Safety counts are descriptive, and details on missing data, crossover periods, and multiplicity are not reported in the registry record.