Anemia Study in Chronic Kidney Disease (CKD): Erythropoiesis Via a Novel Prolyl Hydroxylase Inhibitor (PHI) Daprodustat-Blood Pressure (ASCEND-BP): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Anemia Study in Chronic Kidney Disease (CKD): Erythropoiesis Via a Novel Prolyl Hydroxylase Inhibitor (PHI) Daprodustat-Blood Pressure (ASCEND-BP) (NCT03029247), a randomized phase 2 study. The primary outcome is Average of Systolic Blood Pressure (SBP) Measured by Ambulatory Blood Pressure Monitoring (ABPM) Over 6-hour Post Dosing on Day 57.

2. Clinical Question

3. Trial Design

The registry record describes a completed, randomized, open-label, parallel-group phase 2 study. Enrollment was 105. Participants had anemia associated with chronic kidney disease and were dependent on hemodialysis. The study compared daprodustat with epoetin alfa after participants switched from a stable erythropoiesis-stimulating-agent dose. Because treatment was open-label, masking was not used.

The recorded study period began on 2017-07-27, and primary completion was recorded as 2020-07-09. Before treatment, participants underwent screening and a 2-week erythropoiesis-stimulating-agent washout. On Day 1, randomization was stratified by prior erythropoiesis-stimulating-agent dose. Participants then underwent an acute challenge: a single dose of daprodustat or epoetin alfa, followed by blood-pressure assessment. The highest planned once-daily maintenance dose of daprodustat was 24 mg, and the highest starting dose of epoetin alfa was 100 U/kg.

Following the first challenge, participants entered an 8-week hemoglobin-maintenance period. Doses of either study treatment were administered and adjusted during that period. On Day 57, the acute challenge was repeated using the same treatment dose given for the Day 1 challenge. The primary blood-pressure comparison concerns the 6-hour post-dose period on Day 57, after maintenance treatment. A follow-up visit was planned within 14+/-3 days after treatment completion.

Enrollment describes the study overall; the primary outcome analysis included 27 participants in the daprodustat group and 28 in the recombinant human erythropoietin group with data at the specified time points. The registry record does not report the original sample-size justification or the full details of the trial's initial statistical design assumptions.

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 105 participants were enrolled of which 88 were randomized to receive study treatment. 17 did not receive study treatment as they were withdrawn prior to randomization due to adverse events (2); protocol deviation (2); protocol specified withdrawal criteria (3); lost to follow-up (1); physician decision (2) and withdrawal by participant (7).

Acute Challenge 1 (Day 1)

Hgb Maintenance Phase (8 Weeks)

Reasons for not completing this period:

Acute Challenge 2 (Day 57)

Follow-up Period (14 Days)

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT)-Population comprised of all randomized participants who received at least one dose of study treatment. Only those participants with data available at the specified time points were analyzed.

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 Average of Systolic Blood Pressure (SBP) Measured by Ambulatory Blood Pressure Monitoring (ABPM) Over 6-hour Post Dosing on Day 57, measured over Up to 6 hours post dose on Day 57.

The primary endpoint was the average of systolic blood pressure measured by ambulatory blood-pressure monitoring during the 6 hours after dosing on Day 57. It is expressed in millimeters of mercury. Day 57 followed the 8-week hemoglobin-maintenance period and included the second acute challenge, using the same treatment dose as the first challenge. The endpoint therefore captures a defined post-dose window after maintenance treatment rather than a single clinic blood-pressure reading or a measure over the full treatment period.

Ambulatory monitoring records blood pressure outside a single measurement encounter. Averaging measurements over a stated time window can summarize the blood-pressure level during that period while reducing reliance on an individual reading. The primary question was whether this average differed between the daprodustat and epoetin alfa groups.

For the primary analysis, the registry record reports least-squares means and their standard errors, along with an adjusted treatment difference. Least-squares means are model-based group estimates adjusted for the covariates and terms included in the analysis; they are not simply the raw arithmetic averages. The stated analysis used analysis of covariance and included baseline-related blood-pressure information, prior erythropoiesis-stimulating-agent dose, and specified differences in blood pressure between the acute challenges. The endpoint was chosen to assess the blood-pressure comparison after the maintenance period, but the registry record does not report the clinical threshold for a meaningful difference or a separate justification for selecting this endpoint.

7. Results Posted in the Registry

• Recombinant Human Erythropoietin: mean 143.03 versus 142.87 on control; difference 0.16 (95% CI -8.20 to 8.52), standardized effect d = 0.01; z-test p = 0.970.

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.

Average of Heart Rate (HR) Measured by ABPM Over 6 Hour Post Dosing on Day 1

• 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:
  • Recombinant Human Erythropoietin: mean 73.72 versus 70.77 on control; difference 2.95 (95% CI -1.10 to 7.00), standardized effect d = 0.35; z-test p = 0.154.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Recombinant Human Erythropoietin and the control. The interval gives the range of effects that remain compatible with the data.

AUEC of HR Measured by ABPM Over 24-hour Post Dosing on Day 1

• 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:
  • Recombinant Human Erythropoietin: mean 1,727 versus 1,653 on control; difference 74.40 (95% CI -26.61 to 175.41), standardized effect d = 0.38; z-test p = 0.149.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Recombinant Human Erythropoietin and the control. The interval gives the range of effects that remain compatible with the data.

Average of HR Measured by ABPM Over 6 Hour Post Dosing on Day 57

• 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:
  • Recombinant Human Erythropoietin: mean 70.62 versus 70.49 on control; difference 0.13 (95% CI -3.02 to 3.28), standardized effect d = 0.02; z-test p = 0.935.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Recombinant Human Erythropoietin and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists blood-pressure and heart-rate outcomes measured during the Day 1 acute challenge and after the Day 57 challenge, including averages over a post-dose period and area-under-the-effect-curve measures over a longer post-dose period. These outcomes add context to the primary Day 57 systolic blood-pressure question by examining other cardiovascular measurements, time windows, and the earlier acute challenge. They do not replace the primary endpoint or change its adjusted analysis.

For the posted secondary results with detailed estimates, the reported ANCOVA confidence intervals include zero. Their point estimates therefore do not provide clear evidence of a between-group difference for those particular outcomes. The intervals also allow effects in either direction, so a nonsignificant result should not be taken as proof that the groups are identical. The registry lists 21 posted secondary outcomes, but the secondary details in this record do not include a full set of estimates for every listed outcome; a conclusion about the entire secondary-outcome set is therefore not warranted.

Secondary results generally carry less weight than a prespecified primary result when they are considered alongside multiple outcomes, time windows, and comparisons. Unless multiplicity is addressed, repeated testing increases the chance of obtaining at least one apparently notable result by chance. The registry record does not report a multiplicity adjustment or a hierarchy for interpreting secondary outcomes. Their unadjusted p-values should therefore be viewed as exploratory evidence rather than as independent confirmation of the primary finding.

The secondary outcomes are also not all interchangeable: an average heart rate and an area-under-the-effect-curve measure have different units and meanings, and Day 1 and Day 57 assess different treatment stages. Interpretation should preserve those distinctions and avoid treating a pattern across unlike measures as a single definitive result.

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

• Average of SBP, Diastolic Blood Pressure (DBP) and Mean Arterial Blood Pressure (MAP) Measured by ABPM Over 6-hour Post Dosing on Day 1
• Average of Heart Rate (HR) Measured by ABPM Over 6 Hour Post Dosing on Day 1
• Area Under the Effect Curve (AUEC) of SBP, DBP and MAP Measured by ABPM Over 24-hour Post Dosing on Day 1
• AUEC of HR Measured by ABPM Over 24-hour Post Dosing on Day 1
• Average of DBP and MAP Measured by ABPM Over 6-hour Post Dosing on Day 57
• Average of HR Measured by ABPM Over 6 Hour Post Dosing on Day 57
• AUEC of SBP, DBP and MAP Measured by ABPM Over 24-hour Post Dosing on Day 57
• AUEC of HR Measured by ABPM Over 24-hour Post Dosing on Day 57

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: systolic blood pressure is measured on a numeric scale and summarized as an average over a defined post-dose period. The reported group values are least-squares means, and the reported dispersion is a standard error. A standard error describes the uncertainty in an estimated mean; it is not the spread of individual participants' blood-pressure readings. The group estimates and standard errors therefore describe precision of the estimated averages, not the range of individual responses.

What does the adjusted analysis do?

The registry record says the primary endpoint was analyzed using analysis of covariance, or ANCOVA. In general, ANCOVA compares group outcomes while accounting for selected covariates. Adjustment can improve the comparison when the included covariates explain outcome variation, and it can address imbalances in those measured variables. The record lists treatment, prior erythropoiesis-stimulating-agent dose category, post-hemodialysis/pre-challenge systolic blood pressure, the difference between post-hemodialysis/pre-challenge systolic blood pressure at the two challenges, and an interaction between treatment and that difference. These are the terms described for the primary model.

The resulting least-squares mean difference is model-adjusted. It should not be expected to equal a subtraction of the displayed group least-squares means after rounding, and it should not be conflated with a simple unadjusted estimate from summary values. The registry record does not provide the full model specification, diagnostic checks, or details of how the listed terms were coded beyond the description given.

How should the treatment contrast be read?

A treatment contrast expresses the estimated difference between groups on the outcome scale. Its sign depends on the order of subtraction used in the analysis. The reported analysis lists the groups as daprodustat and recombinant human erythropoietin and gives a least-squares mean difference of -0.17 mmHg. The registry record does not define the sign convention in a separate statement, so the estimate is best quoted with its group labels and confidence interval rather than interpreted as a direction of benefit.

A confidence interval gives a range of values compatible with the data and the analysis assumptions at the stated confidence level. It conveys the uncertainty around an estimated contrast. Here, the reported 95% interval extends from -8.80 to 8.47 mmHg and contains zero. This means the data do not distinguish the adjusted contrast from no difference at conventional significance thresholds. It does not mean that every value in the interval is equally plausible or that the interval is a probability statement about the fixed true effect.

The p-value is a measure of how compatible the observed data are with a specified null model, under the assumptions of the analysis. The reported p-value is 0.969. It is not the probability that the null is true, and it does not measure the clinical importance of the estimate. Interpretation should consider the estimate and interval as well as the p-value.

What is the role of the unadjusted re-calculation?

A simple unadjusted re-calculation uses the posted group summary statistics rather than the trial's full ANCOVA model. For the primary outcome, it estimates a mean difference of 0.16 mmHg, with an interval from -8.203 to 8.523 mmHg and p = 0.97. This is a check based on summary statistics, not a replacement for the reported adjusted analysis. The re-calculation cannot reproduce adjustment for prior dose, blood-pressure measurements, and the interaction term. Its interval also contains zero, but its interpretation remains distinct from that of the registry's ANCOVA result.

The re-calculation uses estimates of variability derived from the reported standard errors. Such a calculation is approximate because the summary data do not contain participant-level measurements or the covariance structure required to recreate the adjusted model. In general, over-dispersion means that observed variability exceeds what a simpler model assumes; accounting for it can affect standard errors and uncertainty. The registry record does not report whether over-dispersion was assessed or relevant to this analysis.

What can the analysis establish?

The reported analysis is a superiority comparison. A nonsignificant superiority test does not demonstrate that the groups are equivalent. Equivalence or noninferiority requires a prespecified margin and a design and analysis directed at that question. Such a margin is not reported in the registry record. Likewise, the record does not report the trial's original sample-size justification or whether a formal interim analysis was planned.

All statistical estimates apply to the analyzed endpoint, population, and time window. General statistical principles do not remove the limitations of the available measurements: an adjusted model depends on its specification, and estimates can be affected by incomplete outcome data. The registry record gives the endpoint population rule but does not report detailed missing-data methods, sensitivity analyses, or diagnostic results. These absent details limit how fully the model and its assumptions can be independently assessed.

11. Confidence Intervals and Precision

The primary adjusted estimate is -0.17 mmHg, and its 95% confidence interval ranges from -8.80 to 8.47 mmHg. The interval includes zero, so the data are compatible with no adjusted difference as well as with differences in either direction. Its width signals that the point estimate is more precise than a statement of direction alone would be, but still leaves room for a range of possible effects. The interval does not establish that clinically important differences are absent.

Precision is different from statistical significance and from clinical importance. A confidence interval is a measure of uncertainty under the analysis model; it is not a range containing a known proportion of individual participant measurements. Standard errors reported for group least-squares means describe uncertainty in those estimated means, while the confidence interval for the difference describes uncertainty in the contrast.

The unadjusted re-calculation produces a 95% interval from -8.203 to 8.523 mmHg. It is similar in broad interpretation to the reported adjusted interval, but it comes from a different calculation and should not replace the trial's ANCOVA result. Differences between the two estimates can arise because the trial analysis adjusts for listed covariates, whereas the re-calculation uses only summary statistics.

The registry record does not report a threshold defining a clinically important blood-pressure difference. Without such a threshold, the interval can be described statistically, but not classified against a trial-specific clinical margin. The confidence interval also pertains only to average systolic blood pressure during the specified Day 57 post-dose period in the analyzed population. It does not quantify uncertainty for other outcomes or time windows.

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 figures are participants per arm, not total events. They are illustrations based on the observed effect, not the trial's original design assumptions. The stated 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 the stated fraction of the observed effect. The trial's own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm under which a study would have an 80% chance of detecting the specified effect if that effect and the planning assumptions held. The 90% power column describes the corresponding planning target for a 90% chance. These are design calculations, not guarantees about what a future study will find. They are also not observed power for this completed trial.

For the primary endpoint, planning for 100% of the observed effect gives 153,469 participants per arm for 80% power and 205,451 participants per arm for 90% power. Planning for 75% of the observed effect gives 272,833 participants per arm for 80% power and 365,246 participants per arm for 90% power. Planning for 50% of the observed effect gives 613,873 participants per arm for 80% power and 821,802 participants per arm for 90% power. These are mathematical illustrations based on a very small observed contrast relative to the estimated variability, not proposed or recommended enrollment targets.

Smaller effects require more participants to detect under the same assumptions, which is why the illustrative requirements increase when the effect fraction is reduced. The calculations are sensitive to the observed effect and variability, so they can be unstable when the estimated effect is close to zero. They do not show that the trial had a particular probability of detecting an effect, and they do not replace the unreported original design rationale.

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 = 0.160; pooled standard deviation σ = 15.820.
2. n = 2 × (1.96 + 0.842)² × 15.820² / 0.160² = 153,468.2, rounded up to 153,469 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 <- 0.159999999999997                      # observed difference in means
sigma <- 15.8201415934794                     # 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(3029247)
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 = 153,469 (80% power), 205,451 (90% power)
 75% of observed effect: n per arm = 272,833 (80% power), 365,246 (90% power)
 50% of observed effect: n per arm = 613,873 (80% power), 821,802 (90% power)
power.t.test: 153469.2 
Simulated power at n = 153,469 per arm: 0.804 (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. Analyses Reported by the Sponsor

The registry's own primary analysis is the adjusted ANCOVA comparison. It reports a least-squares mean difference of -0.17 mmHg, a 95% confidence interval from -8.80 to 8.47 mmHg, and p = 0.969. This is the result to prioritize when describing the trial's primary inference because it reflects the stated model and its covariate adjustment.

The unadjusted re-calculation from posted summary statistics estimates a difference of 0.16 mmHg, with a 95% interval from -8.203 to 8.523 mmHg and p = 0.97. It is a straightforward approximation, not a reconstruction of the ANCOVA. The two estimates differ slightly in value and direction, which can occur because one is adjusted and the other is not, as well as because of rounding and the limits of summary-statistic calculations. Both intervals contain zero, but that shared feature does not make the two methods interchangeable.

The record does not flag a discrepancy between the posted results and the re-calculation. The sponsor-reported analysis remains the primary source for the adjusted treatment contrast; the re-calculation is useful as a transparent, unadjusted comparison and should be labeled as such. Neither result establishes equivalence, and neither provides a probability that a treatment effect is absent.

15. Multiplicity, Missing Data and Analysis Populations

The registry record reports a primary endpoint and multiple secondary outcomes, including measures of systolic and diastolic blood pressure, mean arterial pressure, and heart rate at different post-dose windows and study visits. When many outcomes are tested, the overall chance of at least one chance finding can increase. Statistical methods can control or describe this multiplicity, for example through an adjustment or a prespecified testing sequence. The registry record does not report whether such a method was used, nor does it report a hierarchy for the secondary tests. Accordingly, secondary p-values should not be interpreted as if each were the only test conducted.

The primary outcome analysis population is described as the intent-to-treat population comprising randomized participants who received at least one dose of study treatment. The record also states that only participants with data available at the specified time points were analyzed. This qualification means the endpoint analysis depends on available measurements; the number analyzed can differ by outcome and time point. The record does not report detailed reasons for missing measurements, the extent of missingness by reason, or the method used to address missing data beyond analysis of those with data available.

Missingness can matter if the availability of a measurement is related to the outcome or treatment. General statistical principle: analyses of observed cases can be biased when missing outcomes differ systematically from observed outcomes in ways not accounted for by the analysis. The registry record does not report sensitivity analyses or alternative missing-data assumptions, so the robustness of the primary estimate to missing measurements cannot be assessed from this record.

The record provides an ANCOVA description, but does not report the full statistical analysis plan, model diagnostics, or a detailed account of any interim analyses. It also does not report the original sample-size justification. These omissions do not negate the posted estimate, but they limit assessment of prespecification, multiplicity handling, and the sensitivity of results to analysis choices.

16. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fluid overload (6.7%), hypertensive urgency (4.4%), hypotension (4.4%), pneumonia (4.4%), anaemia (4.4%).

The registry reports serious adverse events by treatment group using an at-risk count and a count of participants affected. In the daprodustat group, 10 of 45 participants at risk were recorded as affected by a serious adverse event. In the recombinant human erythropoietin group, 5 of 43 participants at risk were recorded as affected. The record reports 0 deaths in each group. These figures describe the recorded serious-adverse-event counts; they do not establish that treatment caused an event or that a between-group difference is statistically reliable.

The listed serious-event terms include fluid overload at 6.7%, hypertensive urgency at 4.4%, hypotension at 4.4%, pneumonia at 4.4%, anaemia at 4.4%, and hyperkalaemia at 2.3%. The record does not assign these listed term percentages to a treatment group in the registry record recorded here, so they should not be attributed to one group. Nor should they be added to create a total: event terms can describe different events and the record does not state that they are mutually exclusive.

Serious adverse-event summaries are descriptive. Differences in affected counts do not by themselves account for event timing, severity, relatedness, or exposure, and the registry record does not provide those details in the information summarized. The figures are safety observations from this study, not a causal comparison or a basis for treatment advice.

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

The primary result describes average systolic blood pressure during the 6-hour period after dosing on Day 57, following the hemoglobin-maintenance period, among participants included in that endpoint analysis. The adjusted estimate was close to zero, and its 95% confidence interval included zero and extended in both directions. The result does not establish a between-group difference on that endpoint.

That finding is not the same as proof that daprodustat and epoetin alfa have identical effects. A superiority analysis that does not show a statistically significant difference cannot establish equivalence or noninferiority without an appropriate design, prespecified margin, and analysis. Such a margin is not reported in the registry record. The interval also leaves uncertainty about differences that could be compatible with the data.

The trial addressed a specific population: participants with anemia associated with chronic kidney disease who were receiving hemodialysis and were switched from a stable erythropoiesis-stimulating-agent dose. The primary measure was taken over a defined post-dose window at a specified visit. The result does not directly answer questions about other populations, other blood-pressure windows, longer-term outcomes, or outcomes not assessed in the study.

The posted results also include secondary blood-pressure and heart-rate measures, but those outcomes are distinct from the primary endpoint and are subject to the limitations of multiple comparisons and incomplete reporting. A pattern of nonsignificant findings does not demonstrate absence of effect, especially when confidence intervals remain compatible with effects in either direction.

Finally, the reported data are estimates with uncertainty, not individual predictions. They do not determine what an individual participant's blood pressure would be under either treatment. The registry record describes trial results and does not support medical advice or a conclusion about treatment choice.

18. Why This Trial Matters Statistically

This trial illustrates why a near-zero point estimate and a broad confidence interval should be read together. The adjusted primary estimate is close to zero, but the interval spans possible contrasts in either direction. A reader who considers only the point estimate could overstate certainty; a reader who considers only whether the p-value crosses a threshold could overlook the range of effects compatible with the data.

It also shows the distinction between an adjusted trial analysis and a simple re-calculation from group summaries. The ANCOVA accounts for listed covariates and an interaction, while the summary-statistic calculation does not. Both are useful for different purposes, but the adjusted result is the relevant reported primary inference and the approximate calculation is not a substitute for the analysis model.

The sample-size illustrations highlight how planning calculations depend on the effect being targeted. When the observed contrast is very small, calculations based on detecting that contrast can yield extremely large planning figures. Those figures are not observed power, and they are not the trial's original design justification. They show why effect size assumptions matter, rather than establishing that any particular sample size is appropriate.

Finally, this record emphasizes the importance of transparent reporting beyond a headline result. Details about missing data, multiplicity, model diagnostics, and the original sample-size rationale shape how confidently estimates can be interpreted. Their absence does not erase the posted findings, but it limits a full assessment of the analysis and the scope of conclusions.

19. Limitations

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

The primary endpoint analysis included participants with data available at the specified time points, and the registry record does not give detailed reasons for missing measurements or sensitivity analyses for missingness. This limits assessment of whether the available-case result could be influenced by incomplete outcome data. The endpoint analysis counts also differ from overall enrollment, making the analysis population important when interpreting precision.

The study was open-label, so masking was not used. The primary outcome was measured using ambulatory monitoring, but the record does not report implementation details or quality-control procedures for those measurements. The record gives the main ANCOVA terms but not complete model diagnostics, a full analysis plan, or enough detail to reproduce the adjusted estimate from participant-level data.

The registry record does not report a prespecified clinically important difference, an equivalence or noninferiority margin, the original sample-size justification, or whether multiplicity was addressed for secondary outcomes. This constrains conclusions about clinical importance, equivalence, and the confirmatory status of secondary findings. While the primary result is not statistically significant, the confidence interval remains compatible with differences in either direction.

Safety figures are reported as counts and listed terms, without the additional detail needed to establish causality or compare event timing, severity, or relatedness. The available record also describes a specific hemodialysis population and a defined post-dose blood-pressure window. Findings should not be generalized beyond the studied setting or interpreted as evidence about outcomes that were not measured.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Anemia Study in Chronic Kidney Disease (CKD): Erythropoiesis Via a Novel Prolyl Hydroxylase Inhibitor (PHI) Daprodustat-Blood Pressure (ASCEND-BP) — NCT03029247.
• Registry record last updated 2021-06-03; status last verified 2021-06.
• Results first posted 2021-06-03.
• 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.

23. Record Summary

NCT03029247 was a completed, randomized, open-label, parallel-group phase 2 study comparing daprodustat with epoetin alfa in hemodialysis participants with anemia associated with chronic kidney disease. Enrollment was 105. The primary endpoint was average systolic blood pressure measured by ambulatory monitoring during the 6 hours after dosing on Day 57, following an 8-week hemoglobin-maintenance period.

The reported adjusted ANCOVA difference was -0.17 mmHg, with a 95% confidence interval from -8.80 to 8.47 mmHg and p = 0.969. The interval includes zero and leaves uncertainty about the size and direction of a possible difference. This superiority result does not establish equivalence or noninferiority.

Interpretation is limited by the available-case endpoint analysis, unreported details on missing-data methods and multiplicity, and the absence of a reported original sample-size rationale. The record's serious adverse-event counts and listed event terms are descriptive and do not establish causality.