Trial Evaluating the Efficacy and Safety of Oral Vadadustat Once Daily (QD) and Three Times Weekly (TIW) for the Maintenance Treatment of Anemia in Hemodialysis Subjects Converting From Erythropoiesis-Stimulating Agents (ESAs): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial Evaluating the Efficacy and Safety of Oral Vadadustat Once Daily (QD) and Three Times Weekly (TIW) for the Maintenance Treatment of Anemia in Hemodialysis Subjects Converting From Erythropoiesis-Stimulating Agents (ESAs) (NCT04313153), a randomized phase 3 study. The primary outcome is Change From Baseline in Hb to the Average Over the Primary Evaluation Period (PEP) (Weeks 20 to 26).

2. Clinical Question

The clinical question is about the average change in hemoglobin (Hb), measured in grams per deciliter, over a defined evaluation period. The registry describes the intended purpose as demonstrating the efficacy and safety of vadadustat compared with darbepoetin alfa. The posted primary endpoint is continuous: it reports a least-squares mean change for each group, rather than a count of participants meeting a response threshold.

The two vadadustat schedules allow separate comparisons of once-daily and three-times-weekly dosing with the active control. The registry record does not report a direct comparison between the vadadustat schedules as a sponsor analysis. The posted summary therefore supports reading the arm-level results and the available recalculated comparisons, not assuming a head-to-head conclusion between the two schedules.

3. Trial Design

This was a Phase 3b, randomized, open-label, active-controlled, parallel-group trial. The registry reports an enrollment of 319 and a completed status. Akebia Therapeutics is identified as the sponsor. Participants were assigned to vadadustat QD, vadadustat TIW, or darbepoetin alfa. Masking is listed as none, so participants and study personnel were not blinded according to the registry design description.

The study concerned maintenance treatment of anemia in hemodialysis participants who were converting from existing ESA therapy. The primary endpoint compared change in Hb from baseline with the average Hb during Weeks 20 to 26. A secondary Hb-change endpoint used the average during Weeks 46 to 52. These periods represent different points in the study and should be kept distinct when considering whether results form a consistent pattern over time.

For the primary endpoint, the posted group sizes are 105 for vadadustat QD, 106 for vadadustat TIW, and 108 for darbepoetin alfa. The secondary endpoint uses those same posted group sizes. These are outcome group counts and do not by themselves explain whether every randomized participant contributed measurements at each evaluation period. The record identifies the primary analysis population as the Randomized Population but does not give further details about participant flow in the endpoint description.

The registry record does not report the original sample-size justification, the planned treatment-effect target, or the trial's own power assumptions. It also does not report interim analyses or detailed allocation procedures. Planning illustrations calculated from the posted effects are a separate exercise and should not be read as a description of how the trial was designed.

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: A total of 319 participants were enrolled in the study. Data was collected by the arm to which participants were randomized and not by dose received in vadadustat QD or vadadustat TIW arms.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomized 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.

Population: Randomized Population is defined as all participants randomized.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Hb to the Average Over the Primary Evaluation Period (PEP) (Weeks 20 to 26), measured over Baseline; Weeks 20 to 26.

The primary endpoint is change from baseline in Hb to the average over the PEP, Weeks 20 to 26, measured in grams per deciliter. Baseline Hb is defined as the average of the last two central laboratory Hb values taken on or before the first dose date. The PEP value is the average of central laboratory Hb measurements within the three visit windows during Weeks 20 through 26, regardless of intercurrent events. Change is defined as the PEP value minus the baseline value.

Hb is a continuous measure, so the endpoint summarizes the average change rather than counting how many participants crossed a specified threshold. This approach retains information about the magnitude and direction of measured change. The registry reports least-squares means, which are model-based group averages, and standard errors as the measure of dispersion. The primary model is described as an analysis of covariance (ANCOVA), with baseline Hb and randomization stratification factors as covariates, and multiple imputation for missing data.

Using an evaluation-period average rather than a single measurement is a way to characterize Hb over a defined period. In general, averaging repeated measurements can make the endpoint less dependent on any one visit, although the registry record does not give further details about the visit windows or how measurements within them were distributed. The endpoint is intended to address maintenance after conversion from ESA treatment; it does not, by itself, describe every aspect of anemia control or safety.

7. Results Posted in the Registry

• Vadadustat TIW: mean -0.19 versus 0.07 on control; difference -0.26 (95% CI -0.59 to 0.07), standardized effect d = -0.21; z-test p = 0.126.
• Darbepoetin Alfa: mean 0.34 versus 0.07 on control; difference 0.27 (95% CI -0.06 to 0.60), standardized effect d = 0.22; z-test p = 0.112.

8. Secondary Efficacy Endpoints

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

Change From Baseline in Hb to the Average Over the Secondary Evaluation Period (SEP) (Weeks 46 to 52)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Vadadustat TIW: mean 0.03 versus 0.04 on control; difference -0.01 (95% CI -0.43 to 0.41), standardized effect d = -0.01; z-test p = 0.962.
  • Darbepoetin Alfa: mean 0.44 versus 0.04 on control; difference 0.40 (95% CI -0.02 to 0.82), standardized effect d = 0.26; z-test p = 0.059.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vadadustat TIW and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Darbepoetin Alfa and the control. The interval gives the range of effects that remain compatible with the data.

The secondary endpoint measures change in Hb from baseline to the average over the Secondary Evaluation Period (SEP), Weeks 46 to 52. It uses the same measurement unit as the primary endpoint and gives a later view of average Hb change. The broad pattern is that the two vadadustat arm means are close to one another, while the darbepoetin alfa mean is higher. This is directionally similar to the ordering of the primary-period arm means, where the darbepoetin alfa mean is also higher than the vadadustat TIW mean.

The secondary-period recalculated comparisons provide context for the primary result but do not independently settle the treatment question. Their intervals cross zero, and the registry record contains no sponsor-reported analyses for this endpoint. The later endpoint is not a replacement for the primary endpoint: it evaluates a different period and may reflect a different set of observed measurements and missing-data patterns. The registry record does not report the secondary endpoint's analysis method in comparable detail to the primary endpoint.

Unadjusted secondary results carry less inferential weight when several outcomes or contrasts are considered. General explanation: examining multiple endpoints increases the chance that at least one result may appear notable by chance, unless the analysis plan addresses multiplicity. The registry record does not report whether the trial adjusted for multiple comparisons, how any testing hierarchy was defined, or whether the secondary endpoint was intended to support a confirmatory claim. Its p-values should therefore be read as unadjusted descriptive evidence rather than as proof of a prespecified confirmatory finding.

The secondary findings can be considered alongside the primary result as a check on whether the arm-level pattern looks broadly consistent over the two evaluation periods. They cannot establish that the pattern is stable for every participant, that the treatments are equivalent, or that a difference is clinically important. The posted group averages and approximate contrasts describe the evidence available in the registry record; unreported analysis details and multiplicity handling remain important qualifications.

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

• Change From Baseline in Hb to the Average Over the Secondary Evaluation Period (SEP) (Weeks 46 to 52)

9. Statistical Interpretation

10. Statistical Methods Explained

The registry describes the primary endpoint analysis as an analysis of covariance (ANCOVA) with multiple imputation for missing data. The reported endpoint statistic is a least-squares mean, and the accompanying dispersion measure is a standard error. Separately, the posted comparison estimates and intervals are recalculations from summary statistics; they are unadjusted and are not the sponsor's ANCOVA comparisons. The registry record does not report sponsor-reported analyses for the outcomes.

What does ANCOVA contribute?

General explanation: ANCOVA compares group outcomes while accounting for selected covariates. In this trial's primary endpoint description, those covariates are baseline Hb and randomization stratification factors. Adjusting for baseline can account for differences in starting measurements and may improve precision when baseline is associated with the later outcome. The registry does not name the stratification factors or give the model's full specification, so their identity and coding are not reported in the registry record.

A least-squares mean is a model-based estimate of a group's average outcome, adjusted according to the covariates in the model. It is not necessarily the raw arithmetic mean of each participant's observed changes. The estimates of 0.07, -0.19, and 0.34 g/dL are the posted least-squares mean changes for the primary endpoint. The standard error of 0.12 for each arm describes uncertainty in estimating its model-based mean; it is not the standard deviation of individual participant outcomes.

Why is missing-data handling important?

The primary analysis description says that multiple imputation was used for missing data. General explanation: multiple imputation represents missing values using multiple plausible completed datasets, analyzes each, and combines the results to reflect uncertainty about the missing values. This method depends on assumptions about the missingness process and the imputation model. The registry record does not report those assumptions, the variables used in imputation, the number of imputations, or sensitivity analyses. The fact that multiple imputation is named therefore does not establish how robust the results are to different missing-data assumptions.

The endpoint description also says that the PEP average used central laboratory measurements within three visit windows during Weeks 20 through 26, regardless of intercurrent events. That definition identifies the measurements used to form the period average. The registry record does not explain how every intercurrent event was handled beyond that statement or how participants with incomplete measurements were represented in the imputation procedure.

How should the recalculated comparisons be understood?

General explanation: a mean difference expresses the contrast between two group means on the original measurement scale. The posted primary recalculations compare each non-control group with vadadustat QD, the listed control group for these comparisons. They are not adjusted for the baseline Hb and stratification factors in the sponsor's described ANCOVA model. The recalculated values also rely on an approximate standard deviation derived from the posted standard errors, as indicated by the comparison calculations. That approximation can affect the resulting intervals and p-values.

Consequently, the recalculated comparisons are useful as a transparent, approximate reading of the summary statistics, but they should not be presented as the trial's definitive adjusted treatment estimates. The record does not report sponsor contrast estimates, their standard errors, or their confidence intervals. There is no discrepancy flagged between the available results and the recalculations, but that does not turn the recalculations into a sponsor analysis.

What does a confidence interval communicate?

General explanation: a confidence interval describes a range of values compatible with the observed data under the analysis assumptions and method. A narrower interval indicates greater precision, while a wider interval indicates greater uncertainty. An interval crossing zero for a mean difference includes the possibility of no difference, as well as values on either side. It does not prove that the true difference is zero, and it does not show whether all clinically important differences have been ruled out.

The posted recalculated intervals for both primary comparisons cross zero. Their interpretation is limited by their unadjusted nature and the approximate standard deviations used in the recalculation. The record does not identify a confidence level for these recalculated intervals, so a particular confidence level is not reported in the registry record.

What is not specified?

The registry record does not report the original sample-size calculation, analysis-set rules beyond the primary population label, a multiplicity adjustment, or interim analyses. It also does not provide a detailed statistical analysis plan or sponsor-reported treatment contrasts. These omissions matter because the planned estimand, missing-data strategy, and approach to multiple comparisons help determine how confirmatory results should be interpreted. General statistical principles can explain those issues, but they cannot fill in the trial-specific details that are not reported.

11. Confidence Intervals and Precision

Precision concerns how tightly an estimate is determined by the data and analysis. The posted standard error for each primary arm mean is 0.12 g/dL. A standard error measures uncertainty around an estimated mean; it is distinct from the amount of variation among individual participants. The recalculated comparisons have intervals that cross zero, indicating that the posted summaries do not pinpoint the direction of the contrast with high precision.

For vadadustat TIW compared with vadadustat QD, the recalculated mean difference is -0.26 g/dL. For darbepoetin alfa compared with vadadustat QD, it is 0.27 g/dL. The intervals accompanying both estimates extend across zero. The unadjusted p-values are 0.126 and 0.112, respectively. These are approximate, unadjusted recalculations, not intervals and tests from the described sponsor ANCOVA model.

General explanation: an interval gives a range of estimates compatible with the observed data under its statistical assumptions. An interval crossing zero does not establish equality, because it may also include nonzero differences. Conversely, an estimate away from zero does not by itself establish a meaningful treatment difference. The clinical importance of a difference depends on a meaningful threshold, and such a threshold is not reported in the registry record.

The secondary endpoint's recalculated intervals also cross zero. The registry record does not state the confidence level used for the recalculated intervals. The estimates should therefore be read as approximate displays of uncertainty, not as formal sponsor-reported confidence intervals. Their precision is further limited as an interpretation by the absence of reported sponsor contrasts and by the fact that the calculations are unadjusted.

12. Planning a Similar Trial: Sample Size

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

The planning illustrations are participant counts per arm for these continuous outcomes. The trial does not include a time-to-event endpoint in the posted results, so the total-event convention does not apply; the counts are participants per arm, not total events and not totals across both arms.

The planning calculation uses 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 registry record does not report the original trial's design assumptions or sample-size justification; these values are planning illustrations based on observed effects, not a reconstruction of the trial's design.

For example, planning for 100% of the observed vadadustat TIW comparison effect gives 353 participants per arm for 80% power and 473 participants per arm for 90% power. Planning for 75% of that observed effect gives 628 participants per arm for 80% power and 840 participants per arm for 90% power. The other rows use 50% of the observed effect. The darbepoetin alfa comparison has its own observed effect and corresponding planning values, so its counts should be read separately.

General explanation: 80% power means that, if the assumed effect and statistical assumptions were true, the design would have an 80% chance of detecting that effect using the specified test. The 90% power column describes a higher target chance of detection under the same kind of assumptions. Smaller effects require more participants to distinguish them from random variation with the same power target. These values are not observed power; they are planning illustrations, and they do not establish that the completed trial had the assumed effect or achieved a particular probability of detecting it.

13. How the Sample Size Is Derived

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

Formula

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

Worked calculation for this trial

1. Observed difference Δ = -0.260; pooled standard deviation σ = 1.233.
2. n = 2 × (1.96 + 0.842)² × 1.233² / -0.260² = 352.8, rounded up to 353 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 <- -0.26                      # observed difference in means
sigma <- 1.23257228987219                     # 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(4313153)
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 = 353 (80% power), 473 (90% power)
 75% of observed effect: n per arm = 628 (80% power), 840 (90% power)
 50% of observed effect: n per arm = 1,412 (80% power), 1,890 (90% power)
power.t.test: 353.8 
Simulated power at n = 353 per arm: 0.799 (20000 replicates)

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

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when an analysis considers more than one treatment comparison, endpoint, or time period. In general, repeated testing can increase the chance of finding at least one apparently notable result even when there is no true difference. A trial's analysis plan may address this through a prespecified hierarchy, adjustment, or another strategy. The registry record does not report a multiplicity strategy, so it is not possible to establish how the primary and secondary comparisons were intended to be interpreted together.

The recalculated primary p-values and intervals are unadjusted. The secondary comparisons are also unadjusted, and no sponsor-reported analyses are listed for the primary or secondary outcomes in the record. As a result, the recalculated comparisons do not reveal whether the trial's intended inferential plan controlled the overall error rate or assigned different roles to the primary and secondary endpoints. General principles about multiplicity explain why this information matters but cannot determine what plan the investigators used.

Missing-data handling is more fully described for the primary endpoint: the registry says the ANCOVA used multiple imputation. However, the imputation assumptions, model details, and sensitivity analyses are not reported in the registry record. The secondary endpoint's missing-data method is also not reported. Missing measurements can affect both the group estimates and their uncertainty, and the stated method alone is not enough to assess sensitivity to unobserved outcomes.

The primary endpoint identifies the Randomized Population as its analysis population. Further details about the analysis-set definition, participant exclusions, or handling of participants who did not contribute measurements are not reported in the registry record. The posted group sizes are outcome counts; they do not explain all aspects of participant flow or establish that every randomized participant contributed to every endpoint.

The registry record also does not report interim analyses or the original statistical analysis plan. Without those details, it is not possible to verify whether the posted endpoints, comparisons, missing-data procedures, and multiplicity approach correspond to a prespecified confirmatory framework. This uncertainty is separate from the numerical estimates themselves and should be kept in view when assigning evidential weight.

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: covid-19 pneumonia (7.7%), acute myocardial infarction (5.8%), anaemia (4.8%), hypervolaemia (3.8%), covid-19 (3.7%).

The registry reports participants affected by serious adverse events and other adverse events, together with the numbers at risk in each group. The serious adverse event counts are 47 among 105 at risk for vadadustat QD, 47 among 104 for vadadustat TIW, and 47 among 108 for darbepoetin alfa. Death counts are also reported by group: 12, 9, and 7, respectively. These are counts, not adjusted comparisons of risk.

The record lists serious-event terms including COVID-19 pneumonia at 7.7%, acute myocardial infarction at 5.8%, anaemia at 4.8%, hypervolaemia at 3.8%, COVID-19 at 3.7%, and myocardial infarction at 2.9%. The registry record does not clarify the denominator for these term percentages in the listed summary, so they should be read as reported rather than converted into comparative rates.

Safety counts require context: group size, exposure time, event definitions, and whether an event is considered treatment-related can affect interpretation. The listed counts do not by themselves establish causation or a comparative safety effect. The registry record does not report adjusted safety comparisons or further detail on exposure-adjusted event rates in the material.

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

The results describe average Hb changes in three randomized groups during the primary and secondary evaluation periods. The posted primary arm means are 0.07 g/dL for vadadustat QD, -0.19 g/dL for vadadustat TIW, and 0.34 g/dL for darbepoetin alfa. The simple unadjusted recalculations show contrasts with intervals crossing zero. Together, these registry entries describe the observed group-level pattern and its uncertainty; they do not establish that the groups had identical outcomes or that a specific difference is confirmed.

The estimates do not describe each participant's outcome. A group mean can conceal variation among participants, and the reported standard errors describe precision of the means rather than the spread of individual Hb changes. The registry does not report a threshold for a clinically important difference. Therefore, the numerical magnitude cannot be classified as clinically important or unimportant using a trial-specific threshold in this record.

The primary result also should not be treated as a formal conclusion from the sponsor's adjusted analysis. The registry describes an ANCOVA with multiple imputation, but it does not post sponsor-reported comparison analyses. The available contrasts are unadjusted recalculations from summary statistics using an approximate standard deviation. Their p-values and intervals are useful for understanding the posted summaries, but they are not a substitute for the missing sponsor contrasts or analysis details.

The secondary endpoint offers a later-period perspective, but it does not remove uncertainty in the primary analysis. Its results are also unadjusted recalculations, and no sponsor-reported secondary analyses are listed. The registry record does not report how multiplicity was handled, so findings across endpoints and treatment comparisons should not be treated as independent confirmatory tests.

Safety counts provide descriptive information on serious adverse events, other adverse events, and deaths during the reported study groups. They cannot establish that an event was caused by a treatment or quantify comparative risk without additional information and analysis. Overall, the registry record supports a careful description of group averages, approximate contrasts, and event counts, while leaving important inferential and clinical questions unresolved.

17. Why This Trial Matters Statistically

This trial illustrates a common interpretive distinction in clinical biostatistics: a posted arm-level estimate is not the same thing as a fully reported treatment contrast. The registry gives least-squares mean changes and standard errors, and it describes an adjusted ANCOVA model. However, it does not report sponsor comparison estimates for the outcomes. Recalculating contrasts from summary statistics makes the numerical relationship more visible, but those unadjusted values do not reproduce the described model.

The example also shows why uncertainty should be read with the point estimate. Both primary recalculated intervals cross zero, and the unadjusted p-values are not especially informative without a stated multiplicity plan. An estimate alone cannot indicate how precisely a contrast is determined, while an interval that includes zero does not prove equality. The absence of a reported clinical-importance threshold further limits interpretation of the magnitude.

Missing-data methods matter for continuous outcomes based on repeated measurements. The primary endpoint description names multiple imputation, but does not provide the assumptions or sensitivity analyses needed to judge how missing outcomes could affect the estimates. The secondary endpoint has less reported analysis detail. A reader should distinguish what is explicitly documented from what would ordinarily be useful in a complete statistical analysis plan.

Finally, planning illustrations based on observed effects are not a retrospective measure of whether the trial was adequately designed. They depend on the selected effect, allocation, test, and approximation. Keeping these planning calculations separate from observed results helps prevent a common error: treating a post hoc sample-size exercise as evidence that a study did or did not have sufficient power. The registry record is informative about outcomes, but its omissions constrain stronger statistical conclusions.

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 most important trial-specific limitation in interpreting the posted results is that the registry record does not report sponsor treatment contrasts for the primary or secondary endpoints. The recalculated comparisons are unadjusted, use an approximate standard deviation derived from posted standard errors, and do not incorporate the covariate adjustment described for the primary analysis. Their intervals and p-values should consequently be treated as approximate summaries rather than the trial's definitive adjusted inference.

The record does not report the original sample-size justification or design assumptions. Although planning illustrations are available based on observed effects, those calculations do not establish the trial's intended detectable effect or confirmatory power. The record also does not report the multiplicity plan, interim analyses, or detailed analysis plan. Those omissions make it difficult to assess whether the posted comparisons align with prespecified testing priorities.

Multiple imputation is named for the primary endpoint, but the imputation model, assumptions, and sensitivity analyses are not reported. The secondary endpoint's missing-data method and detailed model are not reported either. This matters because the endpoints use measurements over different study periods, and incomplete observations could affect the averages and their precision.

The record identifies the Randomized Population for the primary endpoint but does not give detailed participant flow or clarify how many individuals contributed measurements at each period. The open-label design is documented; the registry does not report how awareness of assigned treatment was addressed in measurement or study conduct. The safety information is descriptive and does not include adjusted comparative analyses, exposure-adjusted rates, or a detailed causal assessment in the summary. These limitations do not negate the posted outcomes, but they narrow the conclusions that can be drawn from them.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Trial Evaluating the Efficacy and Safety of Oral Vadadustat Once Daily (QD) and Three Times Weekly (TIW) for the Maintenance Treatment of Anemia in Hemodialysis Subjects Converting From Erythropoiesis-Stimulating Agents (ESAs) — NCT04313153.
• Registry record last updated 2025-05-22; status last verified 2025-05.
• Results first posted 2025-05-22.
• 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

NCT04313153 is a completed, randomized, open-label, active-controlled, parallel-group Phase 3b trial sponsored by Akebia Therapeutics. It enrolled 319 participants with anemia receiving hemodialysis and converting from ESA therapy. The groups received vadadustat QD, vadadustat TIW, or darbepoetin alfa.

The primary endpoint was change from baseline in Hb averaged over Weeks 20 to 26; the secondary endpoint used Weeks 46 to 52. The registry describes an ANCOVA with multiple imputation for the primary analysis, but does not report sponsor treatment contrasts. The available comparisons are unadjusted recalculations with intervals crossing zero. Multiplicity strategy, original sample-size assumptions, and several missing-data details are not reported in the registry record.

Serious adverse event and death counts are reported by group. They are descriptive and do not establish causation or comparative risk.