Safety and Efficacy Study of Albiglutide Liquid Drug Product in Type 2 Diabetes Mellitus: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy Study of Albiglutide Liquid Drug Product in Type 2 Diabetes Mellitus (NCT02683746), a randomized phase 3 study. The primary outcome is Change From Baseline in Glycated Hemoglobin (HbA1c) at Week 26.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group study of liquid and lyophilized albiglutide in participants with type 2 diabetes mellitus. The registry identifies GlaxoSmithKline as the sponsor and lists the study status as completed. The reported enrollment was 308. The study start date was 2016-03-16 and the primary completion date was 2017-04-03.

The registry describes a 26-week treatment period and a comparison of the two formulations. Participants assigned to the liquid arm received active liquid albiglutide through an auto-injector and matching placebo through the lyophilized-product pen. Participants assigned to the lyophilized arm received active lyophilized albiglutide and matching placebo through the liquid auto-injector. This double-dummy arrangement is consistent with the stated double masking: each group used both delivery devices, while active treatment differed by formulation.

The posted primary endpoint was change from baseline in HbA1c at Week 26, analyzed using a mixed-effect model with repeated measures. The primary analysis included scheduled HbA1c measurements from Week 4 through Week 26, including measurements collected after hyperglycemia rescue and after discontinuation of investigational product. The registry defines the analysis population as randomized participants who received at least one dose and had a baseline assessment.

The study also considered safety, tolerability, pharmacodynamic response, anti-albiglutide antibodies, injection site reactions, and fasting plasma glucose. The registry describes screening and follow-up periods in the brief summary, but the original sample-size justification and the full original statistical planning assumptions are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

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 and have a Baseline assessment.

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: One participant was randomized to receive Albiglutide and withdrawn from the study prior to receiving any study medication for reason 'withdrew consent'.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Glycated Hemoglobin (HbA1c) at Week 26, measured over Baseline and Week 26.

HbA1c is a measure of glycated hemoglobin, as a percentage of total hemoglobin. It reflects blood glucose exposure over time and is used in diabetes trials as an indicator of glycemic control. The endpoint was the change from baseline to Week 26; the registry defines change as the post-baseline value minus the baseline value. A negative change therefore represents a lower HbA1c measurement at follow-up than at baseline.

The posted group estimates are least-squares means with standard errors. Least-squares means are model-based estimates rather than simple arithmetic averages. The registry states that the primary analysis used a mixed-effect model with repeated measures (MMRM), incorporating scheduled measurements from Week 4 through Week 26. Repeated measurements can contribute information about the outcome trajectory while accounting for their within-participant relationship, subject to the model’s assumptions.

HbA1c was selected because the study’s stated primary hypothesis concerned glycemic control with the liquid formulation compared with the lyophilized formulation. The intended question was non-inferiority, not whether the formulations were identical. A non-inferiority analysis asks whether the evidence rules out a loss in effect larger than a prespecified acceptable margin. The margin and its justification are not reported in the registry record. Consequently, the formal interpretation depends on the sponsor’s reported analysis and cannot be reconstructed fully from group means alone.

7. Results Posted in the Registry

• Albiglutide Lyophilized: mean -1.18 versus -1.12 on control; difference -0.06 (95% CI -0.26 to 0.14), standardized effect d = -0.07; z-test p = 0.556.

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.

Number of Participants With Positive Result for Anti-albiglutide Antibody

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Albiglutide Lyophilized: 10.5% (16/152) versus 11.2% (17/152) on control; risk difference -0.7 percentage points (95% CI -7.7 to 6.3), risk ratio 0.94 (0.49–1.79); unadjusted two-proportion z-test p = 0.854.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Albiglutide Lyophilized and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With Injection Site Reactions (ISR)

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Albiglutide Lyophilized: 11.7% (18/154) versus 11.1% (17/153) on control; risk difference 0.6 percentage points (95% CI -6.5 to 7.7), risk ratio 1.05 (0.56–1.96); unadjusted two-proportion z-test p = 0.874.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Albiglutide Lyophilized and the control. The interval gives the range of effects that remain compatible with the data.

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:
  • Albiglutide Lyophilized: mean -1.88 versus -2.22 on control; difference 0.34 (95% CI -0.19 to 0.87), standardized effect d = 0.15; z-test p = 0.213.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Albiglutide Lyophilized and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover different aspects of the comparison, including antibody findings, injection site reactions, fasting plasma glucose, and safety measures. The outcomes with detailed posted results show broadly similar group summaries, while the unadjusted interval estimates are not precise enough to rule out potentially meaningful differences. The registry does not provide sponsor-reported comparisons for the antibody or injection site reaction outcomes, and the posted fasting plasma glucose comparison is not the primary endpoint analysis.

These secondary findings provide context for the primary HbA1c result but do not confirm or overturn the sponsor’s non-inferiority analysis. They address distinct questions: immunogenicity, local reactions, and another glycemic measure are not substitutes for the prespecified HbA1c endpoint. The registry lists multiple secondary outcomes, but a complete multiplicity strategy is not reported in the registry record.

Unadjusted secondary results carry less weight when many outcomes are examined because the chance of at least one apparently notable finding can increase with repeated testing. Without a reported adjustment plan, p-values or interval estimates from individual secondary comparisons should be treated as exploratory rather than as a collection of confirmatory tests. This does not mean the outcomes are uninformative; it means their interpretation should account for uncertainty, endpoint selection, and the absence of a stated multiplicity procedure. They are best read as supporting descriptive evidence rather than independent proof of equivalence or difference between formulations.

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

• Number of Participants With On-therapy Adverse Events (AEs) and Serious AEs (SAEs)
• Number of Participants With Clinical Chemistry Parameters of Potential Clinical Concern (PCC)
• Number of Participants With Hematology Parameters of PCC
• Number of Participants With Vital Signs of PCC
• Number of Participants With Electrocardiogram (ECG) Parameters of PCC
• Number of Participants With Positive Result for Anti-albiglutide Antibody
• Number of Participants With Injection Site Reactions (ISR)
• Change From Baseline in Fasting Plasma Glucose (FPG) at Week 26

9. Statistical Interpretation

10. Statistical Methods Explained

What does the primary model do?

The registry identifies a mixed-effect model with repeated measures (MMRM) for the primary endpoint. In general, an MMRM analyzes repeated outcome measurements from participants while allowing measurements from the same participant to be related. The model can estimate group differences over scheduled visits and use available measurements without reducing each participant’s repeated observations to a single value. Its estimates depend on the model specification and assumptions about the outcome and missingness. The complete model specification is not reported in the registry record.

The stated primary analysis included scheduled HbA1c values from Week 4 through Week 26. The registry says those values included measurements after hyperglycemia rescue and after discontinuation from investigational product. It also states that imputation under the non-inferiority null hypothesis for missing data would be incorporated. The exact imputation procedure and the details of the model are not reported in the registry record, so the analysis cannot be reproduced from the summary alone.

What is a least-squares mean?

A least-squares mean is a model-based group estimate. It can differ from a raw average because the model may account for repeated measurements or other specified factors. The posted primary group values are labeled least-squares means, with standard errors. A standard error describes the estimated uncertainty of a mean, rather than the spread of individual participant measurements. The registry does not give all model details needed to interpret the standard errors independently.

How does non-inferiority differ from a test of difference?

A conventional test of difference asks whether data provide evidence that groups differ. Non-inferiority instead evaluates whether the intervention is not worse than its comparator by more than an agreed margin. The margin is set before interpreting results and defines the largest loss considered acceptable for the trial’s purpose. A non-inferiority conclusion is therefore not the same as demonstrating equal effects or proving that there is no difference. The non-inferiority margin and its rationale are not reported in the registry record.

The sponsor’s reported result is labeled non-inferiority and includes a 95% confidence interval. A confidence interval expresses a range of values compatible with the model and data under its assumptions. In a non-inferiority analysis, the relation of that interval to the prespecified margin is central. Since the registry record does not state the margin, readers cannot independently check that comparison from the posted summary alone.

What does the unadjusted recalculation represent?

The recalculated continuous-outcome comparison uses the posted group summaries to form an unadjusted difference and its uncertainty. It is not the sponsor’s MMRM analysis. The posted summaries include estimates and standard errors, and the recalculation uses an approximate standard deviation. It does not reproduce the repeated-measures model, its covariate structure, its handling of missing values, or the sponsor’s non-inferiority imputation. The recalculated p-value is likewise an unadjusted comparison and should not be substituted for the sponsor’s formal test.

For binary secondary outcomes, the recalculations compare the posted event proportions between groups. A risk difference describes the difference between those proportions, while a risk ratio compares them as a ratio. These are general measures of association; the posted analyses are unadjusted and do not account for a prespecified multiplicity strategy. Small numbers of events can make estimates imprecise, and intervals spanning zero for risk differences or one for ratios are compatible with a range of possible group relationships.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variation is greater than a statistical model expects. For binary outcomes, a simple model may understate uncertainty if participants’ outcomes are more variable or clustered than assumed. For continuous outcomes, an incorrectly specified variance model can also affect standard errors and intervals. The registry does not report a specific over-dispersion assessment for these outcomes. This is a general statistical consideration, not a claim that over-dispersion occurred in this trial.

The registry posts outcome counts, group estimates, and selected sponsor analyses, but not a complete statistical analysis plan. The handling of all secondary endpoints, interim analyses, adjustment for multiple comparisons, and detailed model diagnostics is not reported in the registry record. Those omissions limit independent assessment of the full analysis strategy.

11. Confidence Intervals and Precision

Precision concerns how narrowly an estimate is determined by the data and analysis. A confidence interval communicates uncertainty around an estimate; a narrow interval indicates greater precision than a wide interval, under the model’s assumptions. It does not describe the range of individual participant responses, and it does not by itself establish clinical importance.

For the sponsor’s primary analysis, the reported 95% confidence interval for the mean difference (net) runs from -0.13 to 0.24. The interval includes values on both sides of zero. For a difference estimate, that means the reported interval is compatible with different directions of group contrast, subject to the model and its assumptions. Its relevance to non-inferiority depends on the prespecified margin, which is not reported in the registry record.

The unadjusted recalculation also has an interval spanning zero. That interval comes from a different analysis and is not a replacement for the sponsor’s model-based interval. The recalculated result is less tailored to the repeated-measures design and does not incorporate the sponsor’s stated missing-data approach. The discrepancy flag further cautions against treating the two estimates as interchangeable.

The secondary endpoint intervals are also broad relative to their point estimates. For antibody positivity and injection site reactions, the recalculated risk-difference intervals include zero. For fasting plasma glucose, the recalculated interval for the mean difference includes zero. These intervals indicate uncertainty about the magnitude and direction of the respective contrasts; they do not show equivalence. The posted summaries do not establish that small differences are excluded.

Precision should be considered alongside study design, measurement, missing data, and the clinical relevance of possible effects. The registry does not report the non-inferiority margin, a complete analysis plan, or detailed missing-data sensitivity analyses. Those omissions limit how fully the reported interval can answer the study’s stated question.

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 estimates are illustrations based on the observed effects, not the trial’s original design calculation. For these outcomes, the estimates are participants per arm; the general sample-size description says only time-to-event displays use total event counts. The planning assumptions listed in the registry 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 80% power column gives an illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if the assumptions hold. The 90% power column gives the corresponding planning target for a 90% chance. These are design calculations, not probabilities that the observed trial result is correct or that a treatment effect exists. This is not observed power.

The rows describe planning for 100%, 75%, and 50% of the observed effect. A smaller effect is harder to distinguish from random variation and therefore requires more participants. For example, the primary-outcome illustration lists 3,470 participants per arm for 80% power when planning for 100% of the observed effect, and 13,880 participants per arm when planning for 50% of that effect. These figures are not evidence that the trial was designed to those targets. The original sample-size assumptions and the trial’s own sample-size justification are not reported in the registry record.

These calculations are especially sensitive when the observed effect is small, as it is for the recalculated primary contrast. They should be read as hypothetical planning illustrations based on the posted estimates, not as a retrospective judgment of whether the trial was adequately designed.

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 = -0.060; pooled standard deviation σ = 0.892.
2. n = 2 × (1.96 + 0.842)² × 0.892² / -0.060² = 3,469.9, rounded up to 3,470 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 <- -0.0599999999999998                      # observed difference in means
sigma <- 0.892049605325094                     # 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(2683746)
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 = 3,470 (80% power), 4,646 (90% power)
 75% of observed effect: n per arm = 6,169 (80% power), 8,259 (90% power)
 50% of observed effect: n per arm = 13,880 (80% power), 18,581 (90% power)
power.t.test: 3470.8 
Simulated power at n = 3,470 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 sponsor’s primary analysis is reported as an MMRM-based mean difference (net) of 0.06, with a 95% confidence interval from -0.13 to 0.24 and a reported p-value for non-inferiority. The registry classifies the result as a non-inferiority analysis. This is the trial’s reported formal analysis and takes precedence over the unadjusted recalculation.

The recalculation from the posted group estimates gives a mean difference of -0.06 and an unadjusted p-value of 0.556. It uses posted summary statistics and an approximate standard deviation; it does not reproduce the sponsor’s repeated-measures model or its missing-data procedure. The registry explicitly flags a discrepancy. The values have opposite signs under the displayed comparison conventions, and their intervals and analysis methods are not identical. The difference should not be resolved by selecting whichever estimate seems more favorable.

The sponsor’s confidence interval and non-inferiority conclusion are the appropriate record of the sponsor’s own analysis, but the registry does not report the non-inferiority margin or its justification. This prevents an independent check of the margin-based interpretation. The recalculation remains useful as a transparent, simple comparison of the posted group summaries, not as an alternative formal result.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study tests several outcomes, time points, subgroups, or comparisons. Even when each test is interpreted using a conventional threshold, the chance of at least one apparently positive finding can increase as the number of tests grows. Statistical approaches can address this issue by controlling a family-wise error rate, controlling a false discovery rate, or specifying a hierarchy of primary and secondary tests. The registry record does not report a multiplicity adjustment strategy for the posted secondary outcomes.

The primary endpoint is distinct from the secondary outcomes, but the registry also lists multiple safety, immunogenicity, injection-site, and metabolic measures. Without a prespecified testing hierarchy or adjustment details, the secondary findings should not be treated as a set of independent confirmatory claims. The original statistical analysis plan and details of any interim analyses are not reported in the registry record.

Missing data are particularly important in repeated-measures analyses. The registry states that imputation under the non-inferiority null hypothesis for missing data would be incorporated into the primary analysis, and that scheduled measurements after rescue and discontinuation were included. However, the exact imputation method, assumptions, and sensitivity analyses are not reported in the registry record. Non-inferiority conclusions can be sensitive to missing-data assumptions, so these details matter for independent appraisal.

The defined primary analysis population comprised randomized participants who received at least one dose and had a baseline assessment. This is the registry’s stated analysis population; it is not simply every randomized participant. The registry does not report enough detail to assess the full flow of participants through the analysis, all reasons for missing outcomes, or whether additional analysis populations were used in the sponsor’s non-inferiority assessment. Those details would help readers evaluate the robustness of the result.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: angina pectoris (0.7%), breast cancer (0.7%), spindle cell sarcoma (0.7%), squamous cell carcinoma of skin (0.7%), asthma (0.7%).

The registry reports serious adverse events for each treatment group separately from other adverse events. The serious-event figures are 7 among 153 participants at risk in the liquid active-treatment group and 9 among 154 at risk in the lyophilized active-treatment group. No deaths were reported in either group. The registry also lists several serious-event terms at 0.7% each; those terms should be read as individual reported categories, not as a complete description of every safety outcome.

These counts are descriptive. They do not by themselves establish that one formulation is safer, because event counts may be influenced by chance, exposure, follow-up, and the characteristics of participants. The figures do not show event timing, severity beyond the serious-event classification, causality assessment, or a formal adjusted comparison. Those details are not reported in the registry record.

Other adverse events were reported for 101 participants in the liquid group and 94 in the lyophilized group, using the respective at-risk counts listed in the registry. A participant count does not convey the number of events per person or the duration of exposure. Safety interpretation therefore requires caution and should not be reduced to a simple comparison of counts.

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

The registry describes a randomized, double-blind comparison of two albiglutide formulations and reports a sponsor analysis classified as non-inferiority for the Week 26 HbA1c endpoint. The posted group estimates show reductions from baseline in both arms. The sponsor’s own analysis, rather than the unadjusted recalculation, is the appropriate result to cite for the trial’s formal non-inferiority assessment because the record flags a discrepancy.

The result does not mean that the formulations are identical. Non-inferiority is a specific statistical conclusion relative to an acceptable margin, not proof of equality. Since the margin and its rationale are not reported in the registry record, the basis for independently judging the formal conclusion is incomplete. The recalculated estimate and its interval also cannot replace the sponsor’s model-based analysis.

The findings do not establish that every participant had the same response, that outcomes would be the same in other settings, or that differences in safety are absent. Group averages and counts summarize outcomes across participants; they do not describe each person’s experience. Secondary outcomes are unadjusted and should be interpreted as contextual evidence rather than definitive confirmation.

Nor do these statistical results constitute medical advice or a recommendation about treatment. They describe the posted trial results and their uncertainty. A fuller assessment would require the non-inferiority margin, complete analysis methods, missing-data sensitivity analyses, and more detailed safety reporting, none of which is fully available in the registry record.

18. Why This Trial Matters Statistically

This trial raises a formulation-comparison question in which the intended claim is non-inferiority rather than superiority. That distinction is statistically important: failure to find a difference in an ordinary test is not evidence that a new formulation preserves an acceptable level of effect. Non-inferiority requires a prespecified margin and an analysis designed to evaluate that margin.

The record also illustrates why a posted group summary and a sponsor’s model-based analysis should not be treated as interchangeable. The primary outcome uses repeated measurements and a stated missing-data approach, while the simple recalculation is unadjusted and based on summary statistics. The registry flags a discrepancy between them. Recognizing this distinction helps prevent a basic recalculation from being mistaken for the trial’s formal analysis.

Finally, the secondary endpoints and safety counts demonstrate the limits of isolated outcome comparisons. Several distinct outcomes are relevant, but without a reported multiplicity plan or more complete uncertainty analyses, individual secondary results cannot carry the same confirmatory weight as the designated primary endpoint. The record is statistically instructive because it brings together a non-inferiority question, repeated measures, missing-data considerations, multiple outcomes, and incomplete reporting of key design assumptions.

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 limitations specific to this registry record constrain interpretation. The non-inferiority margin and its rationale are not reported, although they are central to evaluating the sponsor’s formal conclusion. The registry flags a discrepancy between the sponsor’s analysis and the recalculation from posted summaries; the underlying model output and complete analysis details are not available to reconcile it independently.

The primary model is identified as MMRM, and the record describes the measurement visits and a missing-data approach in general terms. The full model specification, exact imputation procedure, and sensitivity analyses are not reported in the registry record. These omissions matter because repeated-measures and non-inferiority results can depend on modeling and missing-data assumptions.

The original sample-size justification and the trial’s own design assumptions are also not reported. The displayed planning illustrations are derived from observed effects and should not be mistaken for the design calculation. The record does not report a complete multiplicity strategy for secondary endpoints, or sufficient detail to independently evaluate interim analyses and model diagnostics.

Finally, the posted secondary results are limited to selected outcomes and unadjusted recalculations, while the registry lists additional safety and laboratory measures. The serious-event summaries do not provide full event narratives, timing, exposure-adjusted rates, or causality assessments. These limitations do not invalidate the posted results, but they restrict the strength and scope of conclusions that can be drawn from the registry record alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Safety and Efficacy Study of Albiglutide Liquid Drug Product in Type 2 Diabetes Mellitus — NCT02683746.
• Registry record last updated 2019-07-23; status last verified 2019-07.
• Results first posted 2018-05-30.
• 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

NCT02683746 is a completed Phase 3, randomized, double-blind, parallel-group study sponsored by GlaxoSmithKline. It compared liquid albiglutide delivered by auto-injector with lyophilized albiglutide delivered by a dual-chamber cartridge pen injector in participants with type 2 diabetes. The registry reports enrollment of 308.

The primary outcome was change from baseline in HbA1c at Week 26, analyzed using MMRM. The sponsor reported a non-inferiority analysis, while a simple recalculation from posted summaries differs; the registry flags this discrepancy, so the sponsor’s analysis takes precedence for the trial’s formal result. The non-inferiority margin is not reported in the registry record. Secondary outcomes and safety summaries provide additional descriptive context, with limitations from unadjusted analyses, incomplete statistical reporting, and uncertainty around missing-data and multiplicity procedures.