A Study of the Efficacy and Safety of Albiglutide in Subjects With Type 2 Diabetes With Renal Impairment.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Efficacy and Safety of Albiglutide in Subjects With Type 2 Diabetes With Renal Impairment. (NCT01098539), a randomized phase 3 study. The primary outcome is Change From Baseline in Glycosylated Hemoglobin (HbA1c) at Week 26.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, active-controlled, parallel-group, multicenter study. The registry record lists 507 participants as enrolled. Participants were assigned to weekly albiglutide or sitagliptin, with matching placebos intended to maintain masking. The treatment arms were albiglutide 30 mg and sitagliptin 100 mg.

The study concerned people with type 2 diabetes and renal impairment whose glycemia was inadequately controlled on diet and exercise or on specified oral antidiabetic therapy. The registry summary names metformin, a thiazolidinedione, a sulfonylurea, or combinations of these medicines as possible current regimens. More detailed eligibility criteria and the distribution of participants by baseline renal status are not reported in the registry record.

The registry record lists the study start as 2010-05 and primary completion as 2012-11, with a status of completed. The primary endpoint was change in HbA1c from baseline to Week 26. Other posted outcomes include fasting plasma glucose, body weight, HbA1c measurements at additional visits, HbA1c response thresholds, and safety outcomes.

The primary analysis population was defined as participants who received at least one dose and had at least one post-baseline HbA1c assessment. The endpoint description says that only participants available at the indicated time point were assessed. The numbers contributing to the primary endpoint were 242 for albiglutide and 236 for sitagliptin; the safety counts use different denominators. The reason these populations differ is not reported in the registry record.

The original sample-size justification, planned recruitment target, and original design assumptions are not reported in the registry record. The posted sample-size illustrations are based on the observed treatment contrast and are not a record of the study's original planning.

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: Eligible participants entered into 2 weeks of Pre-screening and Screening; 4 weeks of Run-in/stabilization; 52-week Treatment Period for evaluation of efficacy and safety and 8 weeks of post treatment Follow-up. A total of 771 participants were screened, 507 were randomized and 495 received at least one dose of study treatment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) Population: all participants who received at least one dose of study medication and who had at least one post-Baseline assessment of the primary endpoint, HbA1c. Only those participants available at the indicated time point were assessed.

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 Change From Baseline in Glycosylated Hemoglobin (HbA1c) at Week 26, measured over Baseline; Week 26.

The primary endpoint was change from baseline in glycosylated hemoglobin, or HbA1c, at Week 26. HbA1c is a form of hemoglobin measured to identify average plasma glucose over a period of 2 to 3 months. The registry record defines baseline as the last non-missing value before treatment began and defines change as the Week 26 value minus the baseline value. A negative change therefore represents a lower HbA1c measurement at Week 26 than at baseline.

HbA1c is useful as a clinical-trial endpoint because it summarizes glycemia over a period of time rather than reflecting only a single measurement. That general property makes it suitable for assessing change during a study of type 2 diabetes. The endpoint description states that the group estimates are least-squares means from an analysis of covariance model, with standard errors as the reported measure of dispersion.

The registry reports adjusted mean changes of -0.83 for albiglutide and -0.52 for sitagliptin. These are estimates for the primary analysis population under the stated missing-data procedure, not raw changes for every participant enrolled. The endpoint does not by itself describe individual responses, longer-term outcomes, or other aspects of health. Those questions require separate outcome data and interpretation.

7. Results Posted in the Registry

• Sitagliptin 100 mg: mean -0.52 versus -0.83 on control; difference 0.31 (95% CI 0.14 to 0.48), standardized effect d = 0.32; z-test p = <0.001.

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 Fasting Plasma Glucose (FPG) at Week 26

• 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:
  • Sitagliptin 100 mg: mean -0.22 versus -1.42 on control; difference 1.20 (95% CI 0.69 to 1.71), standardized effect d = 0.42; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Sitagliptin 100 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Body Weight at Week 26: LOCF

• 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:
  • Sitagliptin 100 mg: mean -0.19 versus -0.79 on control; difference 0.60 (95% CI 0.07 to 1.13), standardized effect d = 0.20; z-test p = 0.028.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Sitagliptin 100 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry record lists numerous secondary outcomes, including repeated HbA1c and fasting plasma glucose assessments, response thresholds for HbA1c, and other outcomes. Numerical results are posted for fasting plasma glucose and body weight at Week 26, as well as other endpoint records. The reported Week 26 group means for fasting plasma glucose and weight change are both more negative in the albiglutide group than in the sitagliptin group. This is directionally consistent with the primary result's larger estimated decrease in HbA1c, but these outcomes measure different characteristics and should not be treated as interchangeable evidence.

The registry identifies 15 secondary outcomes as posted, but the record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting them. When many endpoints or time points are examined, some comparisons can appear noteworthy by chance even if there is no underlying difference. Unadjusted p-values do not account for the full set of comparisons, so they carry less evidential weight than a result tied to a clearly prespecified primary endpoint and analysis.

Secondary outcomes can add context by showing whether the observed pattern extends to related measures or other aspects of the study. They do not independently establish the primary result, and interpretation is limited when the analysis plan, multiplicity strategy, and results for all outcomes are not fully described. The registry record does not report enough information to determine how the secondary outcomes were prioritized or whether the displayed set represents every analysis conducted.

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

• Mean Change From Baseline in HbA1c at Weeks 4, 8, 12, 16, and 20: LOCF
• Mean Change From Baseline in HbA1c at Weeks 4, 8, 12, 16, 20, 26, 36, 48, and Week 52: Observed Cases
• Change From Baseline in Fasting Plasma Glucose (FPG) at Week 26
• Mean Change From Baseline in FPG at Weeks 4, 8, 12, 16, 20, and 26: LOCF
• Mean Change From Baseline in Fasting Plasma Glucose (FPG) at Weeks 1, 2, 3, 4, 8, 12, 16, 20, 26, 36, 48, and Week 52: OC
• Number of Participants Who Achieved Clinically Meaningful HbA1c Response Levels of <6.5% and <7.0% at Week 26: LOCF
• Number of Participants Who Achieved a Clinically Meaningful Improvement in the HbA1c Response Level of >=1.0%, >=1.5%, and >=2.0% at Week 26: LOCF
• Number of Participants Who Achieved a Clinically Meaningful HbA1c Response Level of <6.5% and <7.0% at Week 52: OC

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: HbA1c is measured on a numeric scale, and the analysis compares average change between groups. For a continuous outcome, a mean summarizes the central tendency across participants. It does not imply that every participant had a change equal to the group mean, and it can conceal variation in individual changes. The registry reports least-squares means, which are model-based group estimates rather than simple unadjusted arithmetic averages.

What does the ANCOVA model do?

The primary endpoint description says that an analysis of covariance, or ANCOVA, was used. The model included treatment group, region, history of prior myocardial infarction, and age category as factors, with baseline HbA1c as a continuous covariate. In general, ANCOVA estimates a treatment-group contrast while accounting for specified baseline or design variables. Adjustment can improve comparability and precision when the model and its assumptions are suitable. The registry does not report the fitted coefficients or further diagnostics, so the reported adjusted estimates cannot be independently reconstructed from the registry record alone.

The baseline HbA1c covariate is particularly relevant because the outcome is change from baseline. Including baseline measurement in the model can account for differences in starting values when estimating the group contrast. The categories listed for age and prior myocardial infarction are model factors, not evidence that the groups were imbalanced on those characteristics. Their observed distributions are not reported in the registry record.

How should the mean difference be read?

The recalculation compares the posted least-squares means. Its positive mean difference of 0.31 is defined as sitagliptin relative to albiglutide. Since the endpoint is Week 26 minus baseline, a more negative group change represents a larger decrease. The sign of a contrast depends on the order in which groups are subtracted; it does not change the underlying group estimates. The confidence interval and p-value associated with the recalculated contrast address the uncertainty and compatibility with a no-difference value under that calculation. They are not a replacement for a sponsor analysis with a different estimand or method.

Standard errors are measures of the uncertainty of estimated means. They are distinct from standard deviations, which describe variation among observations. The primary results provide standard errors of 0.062 and 0.063. The recalculation uses those reported quantities to derive an approximate comparison. It is described as approximate because standard errors alone do not reproduce all details of the original fitted model or its covariance structure.

What does the missing-data method mean?

The primary analysis used last observation carried forward, abbreviated LOCF. The registry describes LOCF as using the last non-missing post-baseline on-treatment measurement to impute a missing measurement. If a participant had missing observation or observations immediately after baseline, the baseline value was not carried forward and remained missing. In general, LOCF fills in a missing later measurement with an earlier observed measurement. This approach makes an analysis possible for some participants with incomplete follow-up, but it relies on an assumption about what the outcome would have been after the last observed measurement. That assumption may not match the participant's unobserved course. The registry record does not report sensitivity analyses comparing LOCF with other missing-data approaches.

The stated primary population was the Intent-to-Treat population, defined in the endpoint description as participants who received at least one dose and had at least one post-baseline primary-endpoint assessment. This definition is narrower than every participant enrolled because it requires exposure and a post-baseline assessment. The endpoint text also says only those available at the indicated time point were assessed. The registry record does not fully describe how all missing observations, discontinuations, or deviations were handled in each analysis beyond the LOCF statement.

How does the sponsor analysis differ?

The sponsor's posted analysis is described as a one-sided t-test of median difference in final values and is labelled non-inferiority or equivalence. It reports a value of -0.32, a 95% confidence interval from -0.49 to -0.15, and p < 0.001. The recalculation instead uses the posted least-squares mean changes and their standard errors to estimate a mean difference. A median and a mean are different summaries: the mean uses the magnitude of all values, while the median identifies the middle of the ordered values. Final values and change from baseline are also different endpoint formulations. Therefore, agreement or disagreement between the numerical values should not be judged as if they were estimates of the same quantity. The registry record does not report the non-inferiority margin or further details needed to assess the sponsor's non-inferiority framework.

For the recalculated comparisons, planning illustrations are based on the observed effect and the stated normal approximation. Such illustrations answer a prospective design question under stated assumptions; they do not measure the probability that the observed result is true. Their interpretation depends on the assumed effect, variability, significance level, allocation, and analysis framework. The original study's sample-size rationale and design assumptions are not reported in the registry record.

11. Confidence Intervals and Precision

A confidence interval describes a range of values compatible with the data and the statistical model at the stated confidence level. It should be read together with the point estimate: the estimate locates the center of the observed contrast, while the interval expresses uncertainty around it. A narrower interval indicates greater precision on the outcome's scale, while a wider interval indicates less precision. Neither form of interval guarantees that the true effect lies within its endpoints in any single study.

For the sponsor's analysis, the registry reports a 95% confidence interval from -0.49 to -0.15 for the median difference in final values. The interval does not include zero. This is an interval for the sponsor's stated median-based analysis, not the recalculated mean change contrast. The sponsor's interval and test must be interpreted using the analysis definition that accompanies them.

The recalculated primary mean-difference interval is reported with the numerical comparison, and its limits are both positive. As a result, it does not include zero. The registry gives standard errors of 0.062 for albiglutide and 0.063 for sitagliptin. These quantify uncertainty in each least-squares mean, while a confidence interval for their contrast addresses uncertainty in the difference between them. Because the recalculation is approximate, it should not be mistaken for the original ANCOVA's full model-based interval.

Precision also depends on the number of participants contributing to an endpoint and on variation in the measurement. The primary analysis includes 242 participants in the albiglutide group and 236 in the sitagliptin group. These counts are endpoint-specific, and they are not the same as the safety denominators. A confidence interval does not resolve possible bias from missing data, model choices, or outcome selection. The registry record does not report sensitivity analyses that would show how robust the estimated contrast is to alternative missing-data assumptions.

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 give participants per arm, not a total number across both arms and not a total number of events. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as the stated fraction of the observed effect. The original trial's own sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed to have an 80% chance of detecting the assumed effect under the stated model, if that effect were the true one and the assumptions held. The 90% power column gives the corresponding planning figure for a higher chance of detection under the same assumptions. These are design probabilities, not guarantees about a future result. They are not observed power.

Rows planning for 100%, 75%, and 50% of the observed effect illustrate how the required sample size changes as the assumed effect changes. A smaller effect is harder to distinguish from random variation under the same conditions, so detecting a smaller effect generally requires more participants. For the primary comparison, the illustrations range from 153 to 610 participants per arm for 80% power, and from 205 to 817 participants per arm for 90% power, across those assumed effect fractions. These are planning examples calculated from the observed contrast, not evidence that the original study was designed around those values.

Planning figures depend on assumptions about the effect and statistical model. They do not account automatically for every issue that could affect a real trial, including missing data or departures from the planning assumptions. The registry record does not report whether the original study used these assumptions, a different target effect, or other design considerations.

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 = 0.310; pooled standard deviation σ = 0.966.
2. n = 2 × (1.96 + 0.842)² × 0.966² / 0.310² = 152.5, rounded up to 153 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 <- 0.31                      # observed difference in means
sigma <- 0.966139450066495                     # 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(1098539)
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 (80% power), 205 (90% power)
 75% of observed effect: n per arm = 272 (80% power), 363 (90% power)
 50% of observed effect: n per arm = 610 (80% power), 817 (90% power)
power.t.test: 153.4 
Simulated power at n = 153 per arm: 0.798 (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 record reports a sponsor analysis using a one-sided t-test of median difference in final values. It is labelled non-inferiority or equivalence and reports -0.32, a 95% confidence interval from -0.49 to -0.15, and p < 0.001. This is distinct from the recalculated comparison of least-squares mean change from baseline, which gives a mean difference of 0.31 and p <0.001.

The analyses use different summaries and different outcome formulations: a median difference in final values versus a mean difference in change from baseline. For that reason, their estimates should not be treated as direct numerical replications or expected to match in sign or magnitude. The registry record labels the sponsor analysis as non-inferiority or equivalence, but does not report the non-inferiority margin or enough detail to evaluate the full testing framework. The discrepancy flag is false, so the posted sponsor result is not identified as conflicting with the recalculation. The sponsor's analysis remains the trial's own reported analysis; the recalculation is an approximate, unadjusted comparison based on the posted summary statistics.

15. Multiplicity, Missing Data and Analysis Populations

The record identifies a primary HbA1c outcome and lists 15 posted secondary outcomes. The secondary outcomes include measurements at multiple visits and response definitions using different HbA1c thresholds. Examining multiple endpoints and time points raises a multiplicity issue: the more comparisons considered, the more opportunities there are for a nominally small p-value to arise by chance. In general, a prespecified testing sequence or statistical adjustment can address this issue, but the registry record does not report whether either approach was used.

The primary analysis description is relatively specific about the ANCOVA factors, baseline covariate, analysis population, and LOCF handling. The registry record does not report the complete statistical analysis plan, the original sample-size justification, interim analyses, or a full account of how all secondary outcomes were tested. It also does not report a non-inferiority margin for the sponsor's analysis. Those omissions prevent a complete assessment of the intended hypothesis-testing framework.

Missing data were handled for the primary endpoint using LOCF. As a general method, carrying the last observed value forward assumes that this earlier observation is an acceptable substitute for the missing later value. That assumption may not hold for every participant or for every reason for missingness. The record does not report sensitivity analyses using alternative assumptions, the amount or pattern of missing primary-endpoint data, or a detailed account of missingness for each secondary endpoint.

The primary population required at least one dose and at least one post-baseline HbA1c assessment. The registry record separately describes availability at the indicated time point. These definitions clarify who could contribute to the endpoint but do not provide a full participant flow or explain all exclusions from each analysis. The reason that the primary endpoint counts differ from the safety denominators is not reported in the registry record.

16. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: atrial fibrillation (1.6%), gastroenteritis (1.2%), ischaemic stroke (1.2%), pneumonia (0.8%), vitreous haemorrhage (0.8%).

The registry reports serious adverse events in 30 of 249 participants at risk in the albiglutide group and 33 of 246 participants at risk in the sitagliptin group. Other adverse events are reported for 174 of 249 and 162 of 246 participants, respectively. These are counts of participants with events against the listed group denominators. They are not measures of severity for every event or estimates of causal effects of treatment.

The listed serious-event terms and percentages include atrial fibrillation at 1.6%, gastroenteritis at 1.2%, ischaemic stroke at 1.2%, pneumonia at 0.8%, vitreous haemorrhage at 0.8%, and urinary tract infection at 0.8%. These are reported terms, not a complete description of each participant's clinical course. The record does not provide a full narrative of event timing, duration, relatedness, or treatment action.

Deaths are not reported in the registry record. Safety comparisons should be interpreted descriptively: the posted counts alone do not establish that treatment caused or prevented an event. The safety denominators also differ from the primary endpoint analysis counts, and the registry record does not explain the difference.

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

The primary results describe average change in HbA1c at Week 26 among participants in the stated analysis population. Both group estimates are negative, and the albiglutide estimate is more negative than the sitagliptin estimate. The recalculated mean contrast is 0.31 when expressed as sitagliptin relative to albiglutide, with p <0.001. The result is a comparison of group averages under the posted endpoint definition and analysis, not a prediction for an individual participant.

The confidence interval communicates uncertainty around an estimated contrast, and the p-value describes compatibility with a no-difference hypothesis under the recalculation's assumptions. Neither statistic measures clinical importance by itself. The registry record does not provide a threshold for a clinically important between-group difference in HbA1c, so the clinical importance of the estimated contrast cannot be assessed from these statistics alone.

The sponsor's analysis is also posted, but it uses a median difference in final values and a one-sided t-test labelled non-inferiority or equivalence. It is not the same estimand as the recalculated mean change contrast. Its reported interval and p-value should be understood in the context of that different analysis, for which the non-inferiority margin is not reported in the registry record.

The results do not establish outcomes beyond the measures and time points reported. The registry lists additional HbA1c, glucose, response, and safety outcomes, but multiplicity handling is not reported. Missing primary measurements were handled by LOCF, which carries assumptions about unobserved outcomes. These limitations matter when judging how broadly to interpret the findings. The statistics describe the posted trial results; they do not support individual treatment decisions or conclusions about outcomes not measured in the record.

18. Why This Trial Matters Statistically

This study offers a useful statistical example of why an endpoint definition matters. The primary result is a least-squares mean change from baseline, while the sponsor's own posted analysis is a median difference in final values. Both are comparisons between the randomized groups, but they answer different statistical questions and need not produce identical estimates. Reading an estimate without its outcome definition and method can lead to misleading comparisons.

The record also illustrates the distinction between a point estimate and its precision. The primary group estimates have standard errors, and the recalculated contrast has a confidence interval. These convey uncertainty in the estimates, not the range of individual participant responses. The analysis is model-adjusted for listed factors and baseline HbA1c, while a separate approximate recalculation uses summary statistics; the two approaches should not be conflated.

Finally, the outcome set includes multiple secondary measures and time points, while the record does not report a multiplicity strategy or complete statistical analysis plan. This limits how strongly secondary comparisons can be interpreted. The study therefore highlights several recurring issues in clinical biostatistics: estimands, model adjustment, missing-data assumptions, multiplicity, and the difference between planning calculations and observed trial evidence.

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.

Several trial-specific limitations constrain interpretation. The primary analysis used LOCF for missing measurements. The registry record does not report the amount or pattern of missing HbA1c data or sensitivity analyses using other missing-data assumptions. Because LOCF substitutes an earlier measurement for a missing later one, its validity depends on how well that earlier value represents the unobserved value.

The primary analysis population required both treatment exposure and a post-baseline HbA1c assessment. The endpoint description also states that only participants available at the indicated time point were assessed. A complete participant flow and reasons for missing data or exclusion are not reported in the registry record. The primary analysis denominators and safety denominators differ, and the reason is not reported.

The registry lists multiple secondary outcomes and assessments at multiple time points, but does not report whether statistical testing was adjusted for multiplicity or organized in a prespecified hierarchy. The original sample-size justification and the study's own design assumptions are also not reported. The sample-size illustrations based on observed effects cannot fill that gap.

The sponsor's analysis is labelled non-inferiority or equivalence, but its non-inferiority margin and fuller testing details are not reported. The sponsor analysis and recalculation also use different summaries and endpoint formulations. Finally, the record does not provide detailed renal-impairment categories, results by renal status, or enough information to assess whether results differ across such categories. These omissions limit assessment of the analysis framework and the applicability of results across the range of renal impairment represented in the study.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of the Efficacy and Safety of Albiglutide in Subjects With Type 2 Diabetes With Renal Impairment. — NCT01098539.
• Registry record last updated 2017-02-28; status last verified 2017-01.
• Results first posted 2014-05-16.
• 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

The registry record describes a completed Phase 3 randomized, double-blind, active-controlled, parallel-group study in participants with type 2 diabetes and renal impairment. It lists 507 enrolled participants and compares weekly albiglutide 30 mg with sitagliptin 100 mg. The primary endpoint was change in HbA1c from baseline to Week 26, analyzed with ANCOVA and LOCF for missing measurements.

The posted least-squares mean changes were -0.83 for albiglutide and -0.52 for sitagliptin. The recalculated mean difference was 0.31, with p <0.001. A separate sponsor analysis of median difference in final values reports -0.32, a 95% confidence interval from -0.49 to -0.15, and p < 0.001. These analyses use different definitions and should be interpreted separately.

The record also lists multiple secondary outcomes and safety results. Multiplicity handling, original sample-size justification, detailed missing-data patterns, and the non-inferiority margin are not reported in the registry record.