Safety and Efficacy of Exendin 9-39 in Patients With Postbariatric Hypoglycemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Exendin 9-39 in Patients With Postbariatric Hypoglycemia (NCT03373435), a randomized phase 2 study. The primary outcome is Postprandial Glucose Nadir.

2. Clinical Question

The endpoint concerns the lowest plasma glucose value occurring during the stated post-meal observation window. The registry reports a mean for each regimen and placebo, alongside standard deviations, and gives mean-difference estimates with confidence intervals and unadjusted p-values for the two treatment-versus-placebo comparisons. These are descriptive and inferential summaries of the posted outcome; they do not by themselves establish how the crossover sequence, period, or within-person correlation was handled in the underlying statistical analysis.

The trial record identifies a phase 2, multicenter, randomized, single-blind, placebo-controlled crossover study. The stated enrollment is 18. The efficacy population description says that both treatment groups were included and that one participant was excluded from the efficacy analysis because of a major protocol deviation. The results list 17 participants for each outcome arm. Since the record describes a crossover, those arm counts should not be read as separate, non-overlapping sets of people.

3. Trial Design

The registry describes NCT03373435 as a phase 2, multicenter, randomized, single-blind, placebo-controlled crossover study. Its stated purpose was to assess the efficacy and safety of exendin 9-39 in patients with postbariatric hypoglycemia. Enrollment was 18, and the record lists the study as completed. The start date was 2018-03-19, and primary completion was 2019-01-30.

In a crossover design, participants receive more than one assigned condition in a sequence, so comparisons can involve measurements made in the same participants under different conditions. The registry describes two treatment groups that received the same three types of treatment in different sequences: two dose regimens of exendin 9-39 and one placebo. The outcome arms are labeled Avexitide 30 mg BID, Avexitide 60 mg QD, and Placebo. The record does not give the sequence details, period durations, washout information, or a description of how the analysis accounted for treatment period or sequence.

Randomization indicates that treatment sequences were assigned by chance, but the randomization procedure and allocation concealment details are not reported in the registry record. The study is described as single-blind; the identity of the blinded party and the procedures used to maintain blinding are not reported in the registry record. The record also does not report the original statistical design assumptions or the justification for the planned enrollment.

The primary outcome was postprandial glucose nadir, expressed as a mean in mg/dL. The measurement was the plasma glucose nadir occurring within 3 hours of mixed-meal tolerance testing. The registry’s efficacy-population description includes both treatment groups and notes that one participant was excluded because of a major protocol deviation. The outcome summaries list 17 participants for each arm. Those arm-specific counts are consistent with the posted summaries but do not describe independent parallel groups in this crossover study.

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.

Treatment Period 1 (14 Days)

Treatment Period 2 (14 Days)

Treatment Period 3 (14 Days)

Analysis population for the primary outcome, as defined in the registry: Analysis population are both treatment group 1 and treatment group 2. One participant was excluded from the efficacy analysis due to major protocol deviation.

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 Postprandial Glucose Nadir, measured over 3 hours following a liquid meal.

The primary endpoint is the postprandial glucose nadir: the lowest plasma glucose value occurring within 3 hours of a mixed-meal tolerance test. A mixed-meal test provides a defined meal challenge, and monitoring the nadir captures the lowest glucose measurement during the specified post-meal window. The registry expresses this continuous measure in mg/dL and summarizes it with a mean and standard deviation.

A mean gives the average observed value for a treatment condition, while the standard deviation describes how spread out individual measurements are around that average. The endpoint therefore summarizes both the typical nadir in the posted outcome group and the variation among its measurements. It does not show every participant’s glucose pattern or the timing of each individual nadir.

The posted means are 57.1 mg/dL for Avexitide 30 mg BID, 59.2 mg/dL for Avexitide 60 mg QD, and 47.1 mg/dL for placebo. The registry also reports each treatment-versus-placebo mean difference and its confidence interval. These summaries describe the measured outcome during the stated observation window; they do not, on their own, describe other glucose outcomes, longer-term patterns, or patient experiences. No secondary outcomes are posted in the registry record.

7. Results Posted in the Registry

• Avexitide 30 mg BID: mean 57.10 versus 47.10 on control; difference 10.00 (95% CI 0.10 to 19.90), standardized effect d = 0.68; z-test p = 0.048.
• Avexitide 60 mg QD: mean 59.20 versus 47.10 on control; difference 12.10 (95% CI 2.35 to 21.85), standardized effect d = 0.83; z-test p = 0.015.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is continuous: it is recorded on a numerical scale in mg/dL, and the registry defines the reported summary as a mean. A mean describes the average measurement within a treatment condition. The accompanying standard deviation describes dispersion among the measurements; a larger standard deviation indicates more spread around the mean. Standard deviation is not the same as uncertainty in the mean, and it should not be interpreted as a confidence interval.

The reported outcome means and standard deviations are 57.1 and 16.5 for Avexitide 30 mg BID, 59.2 and 16.1 for Avexitide 60 mg QD, and 47.1 and 12.7 for placebo. Each condition has a listed outcome count of 17. These are registry summaries. The record does not provide individual observations, a distribution plot description, or enough analytic detail to establish how those summaries were adjusted for the crossover design.

The registry reports two mean-difference comparisons against placebo. A mean difference expresses the estimated separation between the treatment and comparator means on the same measurement scale, so its unit is mg/dL. The reported value is 10.0 mg/dL for Avexitide 30 mg BID versus placebo and 12.1 mg/dL for Avexitide 60 mg QD versus placebo. These are the posted estimates; they are not newly calculated from the arm summaries in this explanation.

A confidence interval gives a range of values compatible with the estimate under the statistical method and assumptions used to calculate it. In repeated use of the same procedure, intervals constructed this way would contain the target parameter at the stated confidence level at the corresponding rate. A confidence interval is not a range that assigns a probability to the true value after observing the data. The 95% intervals reported for the comparisons are 0.102 to 19.9 and 2.352 to 21.85, respectively.

The unadjusted p-values are 0.048 for the 30 mg BID comparison and 0.015 for the 60 mg QD comparison. A p-value describes how compatible the observed data are with a specified null model, assuming the model and analysis conditions hold. It does not give the probability that the null hypothesis is true, and it does not quantify the clinical importance of a difference. Because the registry labels these values unadjusted, they do not incorporate a reported correction for making more than one comparison.

General: In a crossover study, measurements from the same participant under different conditions are related. An analysis can use this within-participant information, which may improve precision when the method appropriately accounts for that dependence. Treating observations from the same participant as if they came from unrelated people can misstate uncertainty. Period, sequence, or residual effects can also matter in crossover settings. The registry record does not report the statistical model or explain how these design features were handled, so the posted summary statistics cannot establish which approach was used.

General: Over-dispersion means that the observed variability is greater than a model expects. It can make uncertainty estimates too narrow if a model assumes less variability than the data contain. The reported standard deviations show outcome spread, but the registry does not report an over-dispersion assessment or identify a model for which this issue was evaluated. It would therefore be inappropriate to assert that over-dispersion occurred or that it was corrected.

The record lists one primary outcome measure and zero secondary outcomes. It also lists zero analyses, although summary results and treatment comparisons are posted. The record does not report a complete statistical analysis plan, a prespecified model, missing-data procedures, an interim-analysis plan, or how multiplicity was handled. The registry’s original design and sample-size assumptions are not reported in the registry record. These are limits on what can be inferred about the method; they are not evidence that a particular procedure was or was not used outside the information recorded.

The posted arm summaries are accompanied by comparison estimates, confidence intervals, and unadjusted p-values. No discrepancy between the reported comparison estimates and the recalculated summaries is flagged. A recalculation from group-level summaries cannot recover a crossover analysis that uses participant-level pairing or other adjustments. The registry’s own reported comparison is therefore the relevant result to describe, while recognizing that its full model specification is unavailable in the record.

10. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimate. The registry reports mean differences of 10.0 mg/dL for Avexitide 30 mg BID versus placebo and 12.1 mg/dL for Avexitide 60 mg QD versus placebo. Their 95% confidence intervals are 0.102 to 19.9 and 2.352 to 21.85, respectively. The interval endpoints indicate that estimates compatible with the data under the stated confidence procedure span a substantial range, particularly for the first comparison.

For the 30 mg BID comparison, the lower confidence limit is close to zero, while the upper limit is 19.9. For the 60 mg QD comparison, the interval extends from 2.352 to 21.85. Both intervals lie above zero as reported, but that feature should be considered alongside the unadjusted nature of the p-values and the absence of a reported multiplicity strategy. The registry record does not state a threshold for a clinically important difference, so the intervals cannot be classified as showing or excluding a meaningful clinical change.

Each condition has 17 observations in the posted outcome summaries. The trial is a crossover study, so precision depends not only on the number of observations but also on the degree of within-person association and on the analysis used. The record does not report the standard error method, the participant-level covariance, or a model accounting for treatment sequence and period. Those omissions limit an independent appraisal of how the confidence intervals were constructed.

Confidence intervals describe statistical uncertainty under a method and its assumptions. They do not account automatically for missing outcomes, deviations from protocol, bias, or unreported analytical choices. The record notes one efficacy exclusion due to a major protocol deviation but does not describe sensitivity analyses or missing-data handling. The intervals are informative about the reported estimates, but should not be treated as covering every source of uncertainty.

11. 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 event counts. They are illustrations based on the observed effect rather than the trial’s original design. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is a stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record.

For the comparison of Avexitide 30 mg BID with placebo, planning for 100% of the observed effect gives 35 participants per arm for 80% power and 46 participants per arm for 90% power. Planning for 75% of the observed effect gives 61 participants per arm for 80% power and 81 participants per arm for 90% power. Planning for 50% of the observed effect gives 137 participants per arm for 80% power and 183 participants per arm for 90% power.

For Avexitide 60 mg QD versus placebo, the corresponding figures are 23 and 31 participants per arm when planning for 100% of the observed effect; 41 and 54 when planning for 75%; and 91 and 121 when planning for 50%. Within each pair, the first figure is for 80% power and the second for 90% power. In plain language, power is the chance that a study would meet its specified statistical criterion if the assumed effect and planning conditions were true.

Smaller effects require more participants under these assumptions, because weaker separations are harder to distinguish from random variation. These calculations are planning illustrations, not observed power. They use a two-arm equal-allocation framework and should not be mistaken for the original crossover trial’s design calculation or a reconstruction of its analysis.

12. 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 = 10.000; pooled standard deviation σ = 14.723.
2. n = 2 × (1.96 + 0.842)² × 14.723² / 10.000² = 34.0, rounded up to 35 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 <- 10                      # observed difference in means
sigma <- 14.7231110842783                     # 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(3373435)
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 = 35 (80% power), 46 (90% power)
 75% of observed effect: n per arm = 61 (80% power), 81 (90% power)
 50% of observed effect: n per arm = 137 (80% power), 183 (90% power)
power.t.test: 35 
Simulated power at n = 35 per arm: 0.801 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure and zero secondary outcomes. It provides two treatment-versus-placebo comparisons for that primary outcome, with unadjusted p-values. The record does not report whether the comparisons were prespecified as separate primary contrasts, whether a hierarchy was used, or whether any adjustment for multiple comparisons was planned or performed. Without that information, the p-values should be read as unadjusted rather than as values that control an overall error rate across comparisons.

The registry lists zero analyses while also posting outcome summaries and comparison results. This does not explain the statistical procedures used to produce the posted estimates. The analysis plan, model specification, handling of period and sequence, and any interim analyses are not reported in the registry record. The original sample-size assumptions are likewise not reported in the registry record.

The efficacy-population description says that both treatment groups were included and that one participant was excluded from the efficacy analysis because of a major protocol deviation. The outcome summaries list 17 participants for each condition. The record does not report whether all scheduled measurements were available, how missing observations were handled, or whether the exclusion was assessed in sensitivity analyses. No imputation method or missing-data assumption is reported in the registry record.

These details matter because analysis populations and missing-data methods can affect both the estimated means and their uncertainty. In a crossover study, the available outcome for a participant may differ by treatment period, and the connection among repeated measurements is relevant to analysis. The registry does not explain whether the posted summaries use all available measurements, a complete-case set, or another approach. It also does not report how protocol deviations other than the stated exclusion were considered.

The absence of these details does not show that appropriate procedures were not used. It limits what can be evaluated from the registry record and makes it important to distinguish the posted numerical results from a fully specified statistical analysis.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: presyncope (5.6%).

The registry reports serious adverse event counts among 18 participants at risk for each listed condition. The count is 0 for Avexitide 30 mg BID, 1 for Avexitide 60 mg QD, and 0 for placebo. The record also lists 0 deaths in each condition. These are counts of participants affected, as recorded, not rates calculated for this explanation.

Presyncope is listed among the top serious-event terms at 5.6%. The record does not provide further detail about its timing, clinical course, relationship to study treatment, or the method used to summarize that term. The serious-event count for Avexitide 60 mg QD is one, but the record does not explicitly map the listed presyncope term to a particular treatment condition in the information.

Other adverse events are listed separately: 3 participants affected for Avexitide 30 mg BID, 9 for Avexitide 60 mg QD, and 14 for placebo, with 18 at risk for each condition. These counts should not be combined across conditions because the design is crossover and participants could contribute observations under more than one condition. Adverse-event counts are descriptive and do not, on their own, establish causation or comparative risk. The registry record does not report event-level details or a statistical comparison of safety outcomes.

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

The posted results describe postprandial glucose nadir during the 3 hours following a liquid meal. The mean was 57.1 mg/dL for Avexitide 30 mg BID, 59.2 mg/dL for Avexitide 60 mg QD, and 47.1 mg/dL for placebo. The reported mean differences versus placebo are 10.0 mg/dL and 12.1 mg/dL, respectively, with confidence intervals and unadjusted p-values as. These figures characterize one measured endpoint in this phase 2 crossover trial.

They do not establish how large a difference is important to patients or clinicians, because the registry does not report an important-difference threshold. Nor do they show effects on outcomes beyond the glucose nadir, since no secondary outcomes are posted. The registry record does not provide enough information to determine how the statistical analysis handled within-participant dependence, sequence, or period. That makes it difficult to independently assess the precision of the crossover comparisons from the posted summaries alone.

The findings also should not be extended beyond the study population and conditions recorded. The record identifies people with postbariatric hypoglycemia, but detailed eligibility criteria and broader participant characteristics are not reported in the registry record summarized for this trial. The results therefore do not describe every person with the condition or every treatment context.

Safety counts provide a separate descriptive view. Serious adverse events were recorded for 0 participants in Avexitide 30 mg BID, 1 in Avexitide 60 mg QD, and 0 in placebo; no deaths were reported in the listed conditions. These small counts and the limited event detail do not establish comparative safety or causation. Overall, the registry offers numerical estimates for a defined endpoint and safety counts, while leaving uncertainty about clinical importance, broader outcomes, analysis methods, and applicability.

16. Why This Trial Matters Statistically

This trial is statistically notable as a crossover study with repeated treatment conditions rather than a conventional parallel-group comparison. In a crossover design, each participant can contribute measurements under more than one condition. That structure can support within-person comparisons, but valid uncertainty estimates depend on recognizing that repeated measurements from the same person are related. The registry posts condition-level means and treatment-versus-placebo estimates, but does not describe the model used to account for that dependence.

The posted estimates are paired with confidence intervals, which helps readers assess uncertainty rather than focusing only on a point estimate or p-value. The reported intervals make clear that each estimate has a range of uncertainty. The unadjusted p-values also illustrate why a statistical result needs context: the registry reports two comparisons, but does not describe a multiplicity plan. A p-value alone cannot convey the size, precision, or clinical importance of an outcome difference.

The trial also illustrates the distinction between enrollment and the analysis summaries. Enrollment was 18, the efficacy-population description notes one exclusion for a major protocol deviation, and each outcome condition lists 17 observations. In a crossover design, those arm counts should not be interpreted as separate groups of independent participants. Understanding how exclusions and incomplete measurements were handled is important, but those procedures are not reported in the registry record.

Finally, planning calculations based on observed effects can be useful for illustrating how assumptions about effect size change required sample size. They are not the original design rationale and are not evidence about the power of the completed trial. Keeping those planning illustrations separate from the observed outcome estimates helps prevent a common misreading of statistical results.

17. 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 registry record provides limited detail about participant characteristics and eligibility beyond the condition of postbariatric hypoglycemia. It does not report demographic or clinical features that would allow readers to judge how closely the study population represents particular groups of interest. The enrollment is 18, and the efficacy description notes one participant excluded because of a major protocol deviation.

The crossover structure is relevant to interpretation, but sequence details, treatment-period duration, washout information, and assessment of period or carryover effects are not reported in the registry record. The statistical model and the method used to calculate the confidence intervals are also not reported. This prevents an independent check of how repeated measurements and within-person dependence were incorporated.

The record posts two unadjusted comparisons but does not report a multiplicity strategy. It does not describe missing-data handling, sensitivity analyses, interim analyses, or the original sample-size justification. These omissions leave uncertainty about the prespecified analysis and the robustness of the posted results to alternative assumptions.

The outcome set is narrow in the registry: one primary measure is listed and no secondary outcomes are posted. The glucose nadir captures a defined feature of the response to a mixed-meal test, but the record does not provide additional measures that would describe other aspects of glucose response or broader clinical outcomes. Safety reporting includes counts but limited event-level detail, and does not establish causality. The registry lists the study as completed, but the posted summary alone cannot resolve these design, analysis, and reporting limitations.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Safety and Efficacy of Exendin 9-39 in Patients With Postbariatric Hypoglycemia — NCT03373435.
• Registry record last updated 2022-07-18; status last verified 2022-06.
• Results first posted 2022-06-28.
• 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.

21. Record Summary

NCT03373435 is a completed phase 2, multicenter, randomized, single-blind, placebo-controlled crossover study of exendin 9-39 in patients with postbariatric hypoglycemia. The registry lists enrollment of 18, a start date of 2018-03-19, and primary completion on 2019-01-30.

The primary outcome was mean postprandial glucose nadir in mg/dL within 3 hours of a mixed-meal tolerance test. The posted condition summaries list 17 observations for Avexitide 30 mg BID, Avexitide 60 mg QD, and placebo. The record reports treatment-versus-placebo mean differences with confidence intervals and unadjusted p-values, as well as adverse-event counts. One participant was excluded from the efficacy analysis because of a major protocol deviation.

The record does not report the full statistical analysis plan, original sample-size assumptions, or details needed to evaluate how the crossover structure and missing data were handled. The numerical results should be read as registry-posted summaries with those limitations in mind.