Comparison of Oral Octreotide Capsules to Injectable Somatostatin Analogs in Acromegaly: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison of Oral Octreotide Capsules to Injectable Somatostatin Analogs in Acromegaly (NCT02685709), a randomized phase 3 study. The primary outcome is Proportion of Patients Who Are Biochemically Controlled Throughout the RCT Phase.

2. Clinical Question

The study objective also included safety and patient-reported outcomes. The primary endpoint is narrower than that overall objective: it represents sustained biochemical control under the registry definition, not every aspect of symptoms, severity, safety, or patient experience. The registry record does not report the full eligibility criteria, prior treatment requirements, or the exact process by which participants entered each phase.

3. Trial Design

This was a phase 3, randomized, active-controlled, parallel-group study with no masking reported. The registry record describes an oral octreotide run-in phase followed by a randomized controlled trial phase comparing oral octreotide capsules with injectable somatostatin analogs. It also lists a combination-phase sub-study in which octreotide capsules were used with cabergoline. These phases should not be treated as though they were all parts of the same randomized comparison: the primary comparison concerns the oral and injectable arms of the randomized phase.

The registry lists enrollment of 146 and a completed study. The randomized-phase primary endpoint counts are reported separately for the oral and injectable groups, with 55 and 37 participants, respectively. The enrollment figure describes the study overall and should not be read as the number analyzed in the primary randomized-phase comparison. The record does not explain in detail how participants moved from the run-in phase into randomization, how many were eligible for that transition, or how analysis populations were defined.

The primary outcome was assessed throughout the randomized phase, with a timeframe of 62 weeks. The endpoint was a binary classification: controlled or not controlled according to the IGF-1 time-weighted-average criterion. The registry lists additional outcomes concerning biochemical and clinical control, active symptoms, the Acromegaly Index of Severity, extension-phase entry, IGF-1, and growth hormone. Those measures broaden the study's clinical scope, but do not replace the primary endpoint.

The study started in 2016-02, had primary completion in 2020-10, and is listed as completed. The registry record does not report masking procedures beyond the designation of no masking, the randomization sequence method, allocation concealment, stratification factors for the randomization, or details of the protocol's sample-size justification. These omissions limit assessment of design safeguards and the intended statistical operating characteristics.

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: Prior to Run-in - all patients treated with SRLs (59.6% octreotide and 39.4% on lanreotide). 78.8% of patients were treated with medium to high doses of SRLs. Prior to RCT - all patients were on oral octreotide capsules.

Run-in Phase

RCT Phase

Combination Phase

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: Acromegaly patients who previously tolerated and demonstrated a biochemical control on SRLs

6. Primary Endpoint

The registry lists the primary outcome as Proportion of Patients Who Are Biochemically Controlled Throughout the RCT Phase, measured over 62 weeks.

The primary endpoint is the proportion of participants considered biochemically controlled throughout the randomized controlled trial phase. The registry definition classifies a participant as controlled when the IGF-1 time-weighted average during that phase is <1.3 ULN. The timeframe is 62 weeks, so the endpoint is intended to reflect control across the treatment period rather than a single isolated measurement.

IGF-1 is a biochemical measure relevant to acromegaly, and a threshold-based endpoint makes the outcome interpretable as the fraction of participants meeting a prespecified control criterion. A time-weighted average can represent measurements taken across a period rather than relying on only one value. In general, this approach may describe sustained biochemical status more fully than a measurement at one visit, although the exact sampling schedule and calculation details are not reported in the registry record.

The endpoint is binary once the criterion is applied: each participant is counted as controlled or not controlled. That makes group counts and proportions straightforward to describe, but it also discards information about how far an individual measurement lies from the threshold and how values vary over time. A participant just on one side of the cutoff and a participant farther from it can receive the same classification. The primary outcome therefore addresses biochemical control under this particular definition; it does not by itself establish symptom improvement, quality-of-life effects, safety, or overall clinical benefit.

7. Results Posted in the Registry

• RCT Phase - Injectables: 100.0% (37/37) versus 90.9% (50/55) on control; risk difference 9.1 percentage points (95% CI 1.5 to 16.7), risk ratio 1.10 (1.01–1.20); unadjusted two-proportion z-test p = 0.059.

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.

Proportion of Patients With Clinical and Biochemical Control at the End of the RCT Phase

• 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:
  • RCT Phase - Injectables: 73.0% (27/37) versus 65.5% (36/55) on control; risk difference 7.5 percentage points (95% CI -11.5 to 26.6), risk ratio 1.11 (0.85–1.47); unadjusted two-proportion z-test p = 0.447.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between RCT Phase - Injectables and the control. The interval gives the range of effects that remain compatible with the data.

Proportion of Patients Who Maintain or Reduce the Overall Number of Active Acromegaly Symptoms at the End of the RCT Phase

• 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:
  • RCT Phase - Injectables: 70.3% (26/37) versus 74.5% (41/55) on control; risk difference -4.3 percentage points (95% CI -23.0 to 14.4), risk ratio 0.94 (0.73–1.22); unadjusted two-proportion z-test p = 0.651.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between RCT Phase - Injectables and the control. The interval gives the range of effects that remain compatible with the data.

Proportion of Patients Who Maintain or Improve Their Overall Acromegaly Index of Severity (AIS) Score at the End of the RCT Phase

• 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:
  • RCT Phase - Injectables: 67.6% (25/37) versus 72.7% (40/55) on control; risk difference -5.2 percentage points (95% CI -24.3 to 14.0), risk ratio 0.93 (0.71–1.22); unadjusted two-proportion z-test p = 0.594.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between RCT Phase - Injectables and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the assessment beyond sustained biochemical control. They include clinical and biochemical control at the end of the randomized phase, maintenance or reduction in active symptoms, maintenance or improvement in the Acromegaly Index of Severity, entry into the study extension, and changes in IGF-1 and growth hormone. This range covers different dimensions of disease status and treatment experience, but each outcome has its own definition, timeframe, and potential sources of missingness.

The posted numerical secondary results concern clinical and biochemical control, active symptoms, and the Acromegaly Index of Severity. Their estimates are imprecise, with confidence intervals that include values on both sides of no difference. They do not yield a single consistent pattern that independently resolves the interpretation of the primary biochemical endpoint. The registry record does not report numerical results for extension-phase entry or changes in IGF-1 and growth hormone among the listed secondary measures.

Secondary results generally carry less confirmatory weight when many outcomes are examined without a reported multiplicity strategy. Repeated testing increases the chance of apparently notable findings arising by chance, and selection among outcomes can make isolated results difficult to interpret. The registry record does not report whether these secondary analyses were adjusted for multiplicity, which outcomes were designated confirmatory, or a hierarchy for interpreting them. Their estimates are therefore best viewed as supportive or exploratory evidence rather than as a set of independent definitive tests.

The planning illustrations based on observed secondary effects do not make those results more conclusive. They describe hypothetical participants per arm under stated assumptions and fractions of the observed effect. They are not observed power, do not establish that the observed effects are reproducible, and should not be used to select whichever outcome appears most favorable. The overall pattern is informative as a description of what was measured, but it does not substitute for a prespecified analysis strategy.

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

• Proportion of Patients With Clinical and Biochemical Control at the End of the RCT Phase
• Proportion of Patients Who Maintain or Reduce the Overall Number of Active Acromegaly Symptoms at the End of the RCT Phase
• Proportion of Patients Who Maintain or Improve Their Overall Acromegaly Index of Severity (AIS) Score at the End of the RCT Phase
• Proportion of Patients of Those Completing the RCT Phase Who Entered the Study Extension Phase
• Change in IGF-1 Levels in the RCT Phase
• Change in GH Levels in the RCT Phase

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary endpoint is binary, so the basic data consist of the number meeting the control definition and the number represented in each group. For this endpoint, the registry gives outcome counts of 50 among 55 in the oral arm and 37 among 37 in the injectable arm. A risk difference expresses the contrast between group proportions on an absolute scale. In general, it is often easier to interpret as a difference in event frequency than a relative measure, but its direction depends on the order used to define the comparison.

The recalculated risk difference is 0.091, with a 95% confidence interval from 0.015 to 0.167. The unadjusted p value is 0.059. These recalculated statistics are simple comparisons of the posted summary counts; they are not adjusted for stratification or other covariates. The record does not report an adjusted analysis for the primary endpoint in the posted results. A p value summarizes how compatible data are with a specified null model under the test's assumptions; it is not the probability that the null hypothesis is true, nor a measure of clinical importance.

A confidence interval describes a range of effect values compatible with the data and the method used to construct the interval. It reflects sampling uncertainty, but does not account automatically for bias, missing outcomes, measurement error, or choices made in the analysis. The primary recalculation's interval is wholly on one side of zero, while its unadjusted p value is 0.059. This apparent difference in threshold-based readings can arise when interval and test procedures are not mathematically identical, particularly with small group counts or boundary outcomes. The registry record does not state enough about the recalculation methods to resolve that distinction; neither summary should be isolated from the sponsor's reported analysis.

What does the sponsor analysis represent?

The registry lists a stratified Miettinen & Nurminen analysis for non-inferiority, with a reported value of -19.9 and a 95% confidence interval from -19.9 to 0.5. This is not the same estimand or necessarily the same method as a simple unadjusted risk-difference recalculation. Stratified methods account for strata in the analysis, but the strata and the detailed calculation are not reported in the registry record. The non-inferiority margin is also not reported. Without that margin and the analysis context, the numeric output cannot establish whether the prespecified non-inferiority criterion was satisfied.

Why does the endpoint definition matter?

Threshold classification simplifies a continuous biochemical measurement into a count of participants who meet or do not meet the criterion. This is useful when the scientific question concerns the probability of attaining a defined target. It can, however, hide variation within each category and sensitivity to measurements close to the cutoff. A time-weighted average uses measurements across the randomized phase, but the registry record does not report the measurement schedule, missing-measurement rules, or detailed computation of that average. Those details matter because endpoint classification can depend on both the definition and the available measurements.

What are the planning illustrations?

The sample-size illustrations in the posted material are planning calculations based on the observed secondary-outcome effects, not the trial's original design justification and not observed power. They assume a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect set at a stated fraction of the observed effect. Such calculations describe hypothetical future study sizes under those assumptions. They do not demonstrate that the current trial was adequately powered, and they can be unstable when the observed effect is imprecise. The registry record does not report the trial's own sample-size assumptions or justification.

More generally, statistical inference depends on the planned estimand, endpoint, analysis population, missing-data strategy, and any adjustments for multiple outcomes. These details are not all available in the registry record. The recalculations therefore help describe the posted counts but should not be mistaken for a replacement of the trial's complete statistical analysis plan or sponsor analysis.

11. Confidence Intervals and Precision

The primary recalculation has a 95% confidence interval from 0.015 to 0.167 for the risk difference. The interval indicates uncertainty in the estimated contrast, even though the endpoints are on the positive side of zero. The width reflects the limited information in the group counts and the method used for the recalculation. It does not indicate that every value inside the interval is equally likely, and it does not include uncertainty about trial conduct, endpoint measurement, or missing-data assumptions.

The primary outcome is especially vulnerable to imprecision when outcomes cluster near a boundary, including when all participants in a group meet the endpoint. In general, a group with no observed failures can still have uncertainty about the underlying event probability; an observed proportion at the boundary does not prove the population probability is exactly that value. The injectable arm's reported count is 37 of 37, so the estimate must be read with the finite group size in mind.

The recalculated confidence interval and unadjusted p value do not give the same simple threshold impression. They should be understood as outputs of statistical procedures rather than as independent proof for or against a treatment difference. The sponsor's stratified non-inferiority result uses a different analysis framework, and its interpretation also requires the non-inferiority margin, which is not reported in the registry record.

Precision is distinct from clinical importance. The registry record does not define a minimum clinically important difference for the primary endpoint, nor does it report the non-inferiority margin. Without those reference points, the size of the estimated contrast cannot be judged against a stated clinical or design threshold. The confidence interval also does not imply that a future study would necessarily reproduce the same estimate.

12. Analyses Reported by the Sponsor

The registry reports the sponsor's primary analysis as a stratified Miettinen & Nurminen non-inferiority analysis. Its reported value is -19.9, with a 95% confidence interval from -19.9 to 0.5. No p value is listed for that analysis. The non-inferiority margin and further details needed to interpret the scale and decision rule are not reported in the registry record.

The simple recalculation from the posted outcome counts is different in purpose and method: it gives a risk difference of 0.091, a 95% confidence interval from 0.015 to 0.167, and an unadjusted p value of 0.059. This recalculation does not reproduce the stratified non-inferiority analysis, and the two estimates should not be treated as interchangeable or compared as if they answered precisely the same statistical question.

The sponsor analysis is the reported trial analysis, while the recalculation is a descriptive check based on the posted counts. The record does not include the information required to determine why the estimates differ in scale and interval, or whether the sponsor result met a non-inferiority criterion. In particular, the margin is essential to that judgment. No definitive non-inferiority conclusion can be drawn from the reported numbers alone without the prespecified margin and analysis details.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple secondary outcomes and a total of 18 outcome measures, alongside 15 analyses. When several outcomes are tested, the chance of at least one apparently unusual result can increase if each is evaluated without adjustment. A multiplicity plan may control that risk through a hierarchy, adjusted testing, or another prespecified method. The registry record does not report whether multiplicity was controlled, which results were intended to be confirmatory, or how the primary and secondary hypotheses were ordered.

Missing-data handling is also important for a time-based biochemical endpoint. Participants may have measurements unavailable at some visits or may discontinue before the end of follow-up. The registry record does not report the amount or pattern of missing primary endpoint data, the method used to handle missing measurements, or whether sensitivity analyses were conducted. The counts should therefore be read as posted outcome results, not as evidence that every randomized participant had complete data under a common rule.

The analysis population is not described in enough detail to determine whether the reported randomized-phase counts use an intention-to-treat population, a per-protocol population, or another defined set. For a non-inferiority question, analysis population choices can matter because exclusions and adherence can affect the estimated contrast. The registry record also does not describe deviations from assigned treatment, crossover, or the rules for participants who entered later study phases.

These omissions do not invalidate the posted results, but they limit independent evaluation of how robust the inference is to analysis choices. The reported non-inferiority result cannot be fully assessed without the margin and analysis plan, while the unadjusted recalculations do not address missingness, multiplicity, or population-definition concerns. Those issues should remain explicit when weighing the evidence.

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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: acute myocardial infarction (2.7%), sinus bradycardia (2.7%), cholelithiasis (2.7%), pericarditis (1.8%), toxicity to various agents (1.8%).

Serious adverse events are reported separately by study phase. The registry lists 6 affected participants among 146 at risk in the run-in phase, 3 among 55 in the randomized oral phase, 3 among 37 in the randomized injectable phase, and 0 among 14 in the combination-phase sub-study. Deaths are listed as 0 in the run-in phase, 1 in the randomized oral phase, 0 in the randomized injectable phase, and 0 in the combination-phase sub-study. These are phase-specific counts and denominators; they should not be added together or interpreted as a single randomized comparison.

The record also lists other adverse events and serious-event terms with percentages, including acute myocardial infarction, sinus bradycardia, cholelithiasis, pericarditis, toxicity to various agents, and atrial flutter. A listed term does not by itself establish that the event was caused by treatment. The registry record does not provide enough detail to assess event timing, recurrence, severity beyond the serious classification, or treatment relatedness.

Safety comparisons require caution because the phases have different roles and participant groups, and the randomized-phase groups have different reported denominators. Counts alone do not capture exposure duration or all relevant clinical context. These figures describe reported safety events; they do not establish comparative safety or determine an individual's risk.

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

The primary result describes biochemical control as defined by the registry: an IGF-1 time-weighted average below 1.3 ULN throughout the randomized phase. The posted counts indicate that the endpoint was met by 50 of 55 participants in the oral arm and 37 of 37 in the injectable arm. The recalculated contrast and its confidence interval quantify uncertainty in a simple unadjusted comparison of those counts. They do not establish that the treatments are equivalent or that one treatment is superior in all relevant respects.

The endpoint is a threshold-based biochemical measure. It does not, on its own, show the amount of biochemical change, symptom burden, quality of life, treatment preference, or long-term outcomes. The registry's broader secondary outcome list recognizes some of these additional dimensions, but the posted data do not report numerical results for every listed measure.

The trial compares two active treatment approaches in a randomized, open-label design. Because masking was not reported, participant and investigator awareness could influence outcomes that involve judgment or reporting, although the primary endpoint is based on a laboratory-related measure. The record does not report details sufficient to assess measurement procedures, adherence, or missing outcomes.

The sponsor's non-inferiority analysis is reported separately from the simple recalculation. A conclusion about non-inferiority requires the margin and the prespecified decision framework; the margin is not reported in the registry record. The recalculated risk difference is not a substitute for those elements. Similarly, the presence of a confidence interval does not resolve all uncertainty about bias, endpoint definition, or generalizability.

These results therefore support a bounded statistical description of the posted randomized-phase outcomes. They do not justify medical advice, broad claims about treatment effectiveness, or assumptions about patients who differ from those enrolled. The registry record does not report enough population detail to establish how well the results generalize beyond the study participants.

16. Why This Trial Matters Statistically

This trial raises a common statistical challenge in active-controlled studies: a non-inferiority question is not answered by observing similar-looking group outcomes alone. The analysis depends on a prespecified margin, a suitable estimand, and methods that preserve the intended comparison. Although the registry reports a stratified Miettinen & Nurminen non-inferiority analysis, it does not report the margin or enough analysis detail to independently interpret the decision.

The endpoint also illustrates the trade-offs of binary classification. A threshold makes the outcome clinically legible as control achieved or not achieved, while compressing the underlying continuous IGF-1 information. The time-weighted-average definition aims to represent control across the treatment phase, but interpretation depends on measurement schedules and missing-data rules that are not reported in the registry record.

The trial's active comparator and randomized parallel design make the outcome comparison more informative than an uncontrolled observation, while the lack of masking and incomplete statistical-plan detail leave questions about analysis and interpretation. The posted recalculation makes the count-based contrast transparent, but it cannot recover omitted protocol information or reproduce the sponsor's stratified non-inferiority framework.

More broadly, the registry record shows why statistical conclusions need to distinguish the primary endpoint, secondary outcomes, safety counts, and planning illustrations. Each answers a different question and carries different inferential weight. Clear reporting of margins, analysis populations, missing-data handling, multiplicity, and endpoint computation is necessary for readers to assess how far the observed results support the study's intended claims.

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.

Several trial-specific limitations follow from information not reported in the registry record. The non-inferiority margin is absent, so the sponsor's reported non-inferiority estimate and interval cannot be mapped to the study's decision criterion. The record also does not give the randomization method, allocation concealment details, stratification factors, or the full statistical analysis plan. These gaps make it difficult to evaluate design protections and reproduce the reported analysis.

The primary endpoint depends on an IGF-1 time-weighted average across the randomized phase, but the registry does not describe the sampling schedule, computational details, missing-measurement rules, or sensitivity analyses. It does not specify the analysis population or report the extent of missing primary outcome data. Those omissions matter for judging whether the posted counts represent a consistent assessment across groups.

The trial was unmasked. That may be relevant to treatment behavior, adherence, reporting, and outcome assessment, particularly for patient-reported or symptom outcomes. The registry does not provide details that would establish the extent of any such influence. The active comparator means the result addresses the studied treatment strategies, not comparison with no treatment.

The outcome record is also incomplete for a broad evaluation of the study objective: numerical results are not reported for every listed secondary measure in the registry material. Safety counts are organized by phase, and phase-specific figures should not be combined into a single randomized comparison. Finally, the planning illustrations use observed secondary effects and stated assumptions; they do not explain the trial's original design and should not be taken as evidence of achieved power.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Comparison of Oral Octreotide Capsules to Injectable Somatostatin Analogs in Acromegaly — NCT02685709.
• Registry record last updated 2022-04-22; status last verified 2022-03.
• Results first posted 2022-04-22.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

The registry describes a completed phase 3 randomized, active-controlled, parallel study of oral octreotide capsules and injectable somatostatin analogs in acromegaly. The primary endpoint was biochemical control throughout the randomized phase, defined by an IGF-1 time-weighted average below 1.3 ULN over 62 weeks.

The posted primary counts are 50 of 55 in the oral arm and 37 of 37 in the injectable arm. A simple unadjusted recalculation gives a risk difference of 0.091, a 95% confidence interval from 0.015 to 0.167, and an unadjusted p value of 0.059. The registry separately reports a stratified Miettinen & Nurminen non-inferiority analysis, but the non-inferiority margin and further interpretation details are not reported.

Secondary outcomes cover clinical and biochemical control, symptoms, severity, extension entry, IGF-1, and growth hormone. Safety counts are reported by study phase. Missing-data handling, analysis populations, multiplicity procedures, and several design details are not reported in the registry record.