Study of Management of Pasireotide-induced Hyperglycemia in Adult Patients With Cushing's Disease or Acromegaly: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Management of Pasireotide-induced Hyperglycemia in Adult Patients With Cushing's Disease or Acromegaly (NCT02060383), a randomized phase 4 study. The primary outcome is Change in HbA1c From Randomization to Approximately 16 Weeks.

2. Clinical Question

The study addressed management after hyperglycemia arose during pasireotide treatment, rather than comparing pasireotide with another treatment. The registry describes a screening and metformin period before randomization: participants whose glucose remained above target on metformin within the first 16 weeks could be assigned to incretin-based therapy or insulin. Randomization was stratified by disease type and baseline glycemic status.

The primary endpoint was a continuous measure: change in HbA1c from randomization to the end of the core phase. The registry describes an analysis of the mean change in each group and the mean difference between groups. However, a numerical primary-outcome estimate and its confidence interval are not reported in the registry record, so the size and precision of the observed between-group difference cannot be characterized from the record.

3. Trial Design

This was a phase IV, multicenter, randomized, open-label, parallel-group study sponsored by Novartis Pharmaceuticals. The registry lists enrollment as 249 and the study status as completed. The study began on 2014-05-23, and primary completion was recorded as 2018-02-05. The record describes both a core phase and an optional extension phase.

Participants had Cushing's disease or acromegaly and received pasireotide in a formulation associated with their condition: subcutaneous pasireotide for Cushing's disease and pasireotide LAR for acromegaly. Participants who developed diabetes-range fasting blood glucose during pasireotide treatment started metformin. Those who continued to have glucose above target on metformin within the first 16 weeks could be randomized in a 1:1 ratio to incretin-based therapy or insulin for approximately 16 weeks.

The randomized comparison therefore concerns add-on approaches for hyperglycemia in a selected group: participants who were receiving pasireotide and whose hyperglycemia was not adequately controlled with metformin. It does not represent a randomized comparison of the original pasireotide regimens, nor does it directly compare the two glucose-lowering approaches in all participants enrolled in the study.

The trial was open label, so participants and investigators knew the assigned treatment. The registry describes no masking. In the incretin-based group, treatment started with sitagliptin, with a switch to liraglutide if hyperglycemia was not controlled. In the insulin group, treatment began with basal insulin, with investigator-directed titration and a possible switch to basal plus prandial insulin if glucose remained uncontrolled.

Participants who continued to receive clinical benefit after the core phase could enter an optional extension if pasireotide was not commercially available in their country or a local access program was unavailable. The extension continued until the last participant randomized in the core phase completed 16 weeks of post-randomization treatment. The registry does not report a sample-size justification or a detailed account of how many participants entered the randomized comparison.

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: 249 participants were included in study \& treated with pasireotide s.c. (59 with Cushing's disease) or pasireotide LAR (190 with acromegaly). Following pre-randomization period (up to 16 weeks), 81 participants were randomized to either incretin-based therapy or insulin (72 evaluable for the primary analysis) \& 168 not qualified for randomization.

Core Phase

Reasons for not completing this period:

Extension Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomized Analysis Set (RAS): all patients who received at least one dose of pasireotide and were assigned to either incretin based therapy or insulin by randomization. If the patient discontinued the study or used rescue treatment within 8 weeks after randomization, it was considered missing.

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: Full Analysis Set (FAS): all participants who received at least one dose of pasireotide.

6. Primary Endpoint

The registry lists the primary outcome as Change in HbA1c From Randomization to Approximately 16 Weeks, measured over Randomization, 16 weeks.

The primary endpoint was change in HbA1c from randomization to approximately 16 weeks. HbA1c is a continuous measure of average blood glucose over time; a change from the randomization value describes how that measure shifted during follow-up. The registry specifies an absolute change in HbA1c and a mean difference in change between the randomized treatment groups.

Using a continuous endpoint retains information about the size of each participant's measured change rather than reducing the outcome to a category such as improved or not improved. Comparing group means can estimate an average contrast between the assigned management strategies. The confidence interval is intended to show the precision of that estimated contrast: a narrower interval indicates greater precision, while a wider interval indicates greater uncertainty. The registry names a 95% confidence interval for the endpoint, but the numerical estimate and interval are not reported in the registry record.

The endpoint was assessed at the end of the core phase, approximately 16 weeks after randomization. For participants who discontinued or required rescue treatment before the primary assessment, the registry describes carrying forward the last HbA1c assessment collected 8 weeks after randomization and before or on the date rescue treatment began. If discontinuation or rescue occurred within 8 weeks after randomization, the endpoint was considered missing.

HbA1c was relevant to the study question because the comparison concerned approaches to managing elevated glucose during ongoing pasireotide treatment. The endpoint measures a laboratory outcome; it does not by itself establish differences in symptoms, long-term complications, quality of life, or other outcomes.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary endpoint represent?

HbA1c is measured on a continuous scale. A change score compares the endpoint measurement with the value at randomization, so the analysis focuses on movement over the follow-up period rather than on a single measurement alone. The registry describes reporting the mean change for each randomized group and the mean difference between those changes.

A mean summarizes the central value of a continuous outcome across a group. A mean difference expresses the contrast between group averages in the outcome scale. It does not indicate that every participant experienced a change equal to the group mean; individual responses can vary around that average. The registry identifies the outcome unit as HbA1c percentage, but does not provide the primary-outcome estimates needed to describe their observed values.

What is the role of the confidence interval?

A confidence interval expresses uncertainty around an estimate using the data and analysis method. In repeated applications of the same procedure under its assumptions, intervals constructed this way would cover the underlying parameter at the stated confidence level at the stated rate. It is not a range containing a known percentage of individual participant outcomes, nor does it assign a probability to a particular fixed treatment effect after the data have been observed.

The endpoint description specifies a 95% confidence interval. Because the registry record does not report the interval's endpoints or the corresponding estimated mean difference, its precision cannot be judged. A confidence interval is also not a replacement for considering study design, measurement, missing data, and the clinical context of the outcome.

Why use an analysis of variance model?

For a continuous outcome, an analysis of variance model can estimate adjusted group means and a treatment contrast while accounting for specified factors. The registry describes a model with treatment, disease type, and baseline glycemic status as fixed effects. Disease type distinguished Cushing's disease from acromegaly; baseline glycemic status used the HbA1c categories <7% and at least 7%.

Including the stratification factors in the model aligns the analysis with the factors used when assigning treatment. Adjustment can account for differences across those categories in the comparison. The registry does not report the model estimates, residual diagnostics, or other details needed to evaluate how the model fit the observed data. No further assumptions or model checks are reported in the registry record.

How should treatment assignment be understood?

Random assignment is intended to make treatment groups comparable, on average, with respect to measured and unmeasured factors at the point of assignment. It supports a causal interpretation of a between-group contrast when the trial is conducted and analyzed appropriately. Here, randomization occurred after participants had developed hyperglycemia on pasireotide and remained above target on metformin. The randomized comparison therefore concerns this eligible population and treatment stage.

The design was open label. Knowledge of assignment may affect behavior, co-interventions, reporting, or decisions such as rescue treatment. The endpoint is a laboratory measure, but open-label status can still matter to the surrounding conduct and interpretation. The registry does not report masking or independent blinded assessment; it explicitly lists masking as none.

What does the missing-data approach do?

The registry describes a specific rule for participants who discontinued or required rescue treatment before primary assessment. The last HbA1c assessment collected 8 weeks after randomization, if it occurred before or on the date rescue treatment started, was carried forward. If discontinuation or rescue happened within 8 weeks after randomization, the value was considered missing.

Carrying a previous observation forward substitutes an earlier measurement for a later one that was not used as planned. This approach can be easy to describe, but it depends on assumptions about what would have happened after the earlier measurement. A carried-forward value does not represent a newly observed endpoint value, and it may not capture changes that occurred after that measurement. Treating some cases as missing leaves the analysis dependent on how missing observations were handled. That additional handling is not reported in the registry record.

The registry identifies the randomized analysis set as patients who received at least one dose of pasireotide and were assigned to incretin-based therapy or insulin by randomization. It also specifies when a primary-endpoint value was considered missing. The record does not report the number included in the primary analysis or a fuller accounting of missing observations by group.

What can and cannot be inferred from the registry statistics?

The endpoint, model, and analysis population describe how the primary comparison was framed. They do not reveal the observed treatment effect. Since the registry record does not report a numerical estimate or confidence interval for the primary outcome, no conclusion about direction, size, or precision follows from the method description alone. The registry also does not report an interim-analysis plan, sample-size justification, or multiplicity strategy. Those items should be treated as not reported in the registry record, rather than inferred from the phase or enrollment information.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and several secondary outcomes, including changes in HbA1c and fasting plasma glucose, insulin rescue in the incretin-based group, and other HbA1c measures. When multiple outcomes or analyses are considered, the chance of at least one apparently notable result can increase if no adjustment or prespecified hierarchy is used. The registry record does not report a multiplicity strategy, such as an adjustment method or a testing hierarchy. It is therefore not possible to determine from the record how multiple statistical tests were intended to be interpreted.

The missing-data rule for the primary endpoint is specified in part. For participants who discontinued or required rescue treatment before endpoint assessment, an HbA1c value collected 8 weeks after randomization and before or on the rescue date was carried forward. If discontinuation or rescue occurred within 8 weeks after randomization, the value was considered missing. The record does not report the additional method used to handle missing values in the primary analysis, nor the amount of missingness by group.

The randomized analysis set includes patients who received at least one dose of pasireotide and were assigned to one of the two randomized therapies. This is the population description in the registry; the record does not state the number analyzed for the primary endpoint. A complete interpretation would distinguish participants randomized from those included in each outcome analysis and describe reasons for any exclusions or missing measurements.

The registry does not report an interim-analysis plan. This should not be interpreted as evidence that interim analyses did or did not occur; the information is not reported in the registry record. Likewise, the record does not report a sample-size justification, so the enrollment figure alone cannot establish the study's planned precision or power for the primary comparison.

11. 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: febrile neutropenia (5.3%), cellulitis (5.3%), sepsis (5.3%), glycosylated haemoglobin increased (5.3%), diabetes mellitus inadequate control (5.3%).

The registry lists serious adverse-event figures separately for each randomized and non-randomized group. In the incretin-based randomized group, 38 participants were at risk and 6 were reported as affected by serious adverse events. In the insulin randomized group, 43 were at risk and 3 were reported as affected. The record reports 0 deaths in each of these randomized groups.

For the non-randomized groups, the baseline insulin group lists 19 at risk and 4 affected by serious adverse events, with 1 death. The oral antidiabetic drugs group lists 46 at risk and 2 affected, with 0 deaths. The no oral antidiabetic drugs group lists 103 at risk and 7 affected, with 1 death. The registry also gives figures for other adverse events in each group; these are separate from the serious-event figures and should not be conflated with them.

The serious-event terms listed include febrile neutropenia, cellulitis, sepsis, increased glycosylated haemoglobin, inadequately controlled diabetes mellitus, and hyperglycaemia, each with a reported value of 5.3%. The record does not establish from these figures whether terms overlap within participants or specify a denominator for the term percentages. Group counts are descriptive, and the non-randomized groups are not suitable as randomized comparators. These figures alone do not establish whether a treatment caused an event or quantify a comparative safety effect.

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

The study's central statistical question is a comparison of average HbA1c change between two randomized management strategies after hyperglycemia remained above target on metformin during pasireotide treatment. The endpoint and model are described in the registry, but the numerical primary-outcome result is not reported in the registry record. As a result, the direction and size of the observed contrast and its precision cannot be stated.

A study design is not itself evidence of an outcome. Randomization and a planned adjusted analysis describe a framework for comparing groups; they do not establish that the groups differed, that they were similar, or that either strategy performed better. Those interpretations require the treatment-effect estimate and its uncertainty, alongside information about analysis conduct and missing data.

Even if a difference in HbA1c were available, it would describe the outcome and follow-up period studied. It would not automatically answer questions about long-term glucose control, complications, symptoms, treatment burden, or outcomes beyond the core phase. The registry's primary endpoint is a laboratory measure assessed at approximately 16 weeks after randomization, and the record does not provide evidence for broader outcomes beyond those listed.

The findings, when available, would apply most directly to the selected participants eligible for randomization: adults with Cushing's disease or acromegaly who were receiving pasireotide and whose hyperglycemia remained above target on metformin. The comparison is not a direct test of pasireotide against another treatment and does not describe all participants enrolled in the study. Participants entered different treatment pathways, including non-randomized groups, and those groups should not be interpreted as randomized comparisons.

The open-label design and the specified handling of some missing endpoint values are relevant to interpretation. Knowledge of treatment can affect care decisions, while carrying forward an earlier measurement may not represent the later outcome that would otherwise have been observed. The registry does not report enough detail to assess the consequences of these features. Accordingly, no conclusion about efficacy, equivalence, or lack of difference should be drawn from the information recorded about design alone.

13. Why This Trial Matters Statistically

This trial is statistically notable because it addresses a treatment decision that occurs after an initial management step has not brought glucose to target. The randomized comparison is therefore embedded in a sequence: pasireotide treatment, development of hyperglycemia, metformin treatment, and then possible randomization to incretin-based therapy or insulin. The target population for the randomized estimate is shaped by that sequence.

The endpoint is continuous, and the registry describes an adjusted mean comparison rather than a binary success threshold as the primary analysis. Such an analysis can express the average difference in HbA1c change while retaining the outcome scale. Interpretation still depends on the actual estimate and confidence interval, which are not reported in the registry record.

Stratification by disease type and baseline glycemic status recognizes that the randomized participants came from clinically different categories. Including these factors in the planned analysis makes the stated model correspond to those stratification factors. It does not, by itself, show whether treatment effects differed across categories; the registry record does not report subgroup estimates for the primary outcome.

The record also illustrates why analysis populations and missing-data rules matter. It defines a randomized analysis set and specifies when an earlier HbA1c value could be carried forward or treated as missing. These choices can affect the estimate and its precision, especially when outcomes are unavailable after discontinuation or rescue. The record does not report the extent of missing data or additional handling details.

Finally, a completed trial record can describe design, outcomes, and safety without providing a usable numerical result for every endpoint. In this case, the design and measurement question are clear, while the primary contrast is not reported as a single numerical estimate. Statistical interpretation must distinguish what the trial set out to estimate from what the registry record makes available for evaluation.

14. Limitations

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

The primary limitation for interpreting the principal comparison is that the registry record does not report a numerical estimate of the mean HbA1c difference between randomized groups or its confidence interval. Without those values, readers cannot assess the observed direction, magnitude, or precision of the primary outcome.

The registry does not report a sample-size justification or a detailed statistical analysis plan. Enrollment is recorded as 249, but that figure does not establish how many participants were randomized, analyzed for the primary endpoint, or included in each treatment group. A rationale for the selected enrollment and the intended precision is not reported in the registry record.

The study was open label. Participants and investigators knew the assigned treatment, which may be relevant to treatment changes, rescue decisions, and outcome follow-up. The endpoint is a laboratory measure, but open-label conduct can still influence the circumstances in which measurements are obtained or missing.

The primary endpoint description includes carrying forward a measurement collected 8 weeks after randomization under specified conditions, while some early discontinuations or rescue use were considered missing. The record does not report the extent of missingness, the number of carried-forward values, or other details of missing-data handling. These omissions limit assessment of how the rule may have influenced the estimate.

The study's randomized population was selected after hyperglycemia persisted above target on metformin during pasireotide treatment. Results from that comparison, if available, would not automatically apply to people with different conditions, treatment histories, or glucose control. The record also includes non-randomized groups, whose descriptive safety figures should not be interpreted as randomized treatment contrasts.

The registry does not report a multiplicity plan, interim-analysis plan, or detailed subgroup results for the primary endpoint. The serious-event term percentages are listed without a clear denominator in the recorded information, and the relationship among terms is not specified. These are additional reasons to avoid stronger claims than the record supports.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of Management of Pasireotide-induced Hyperglycemia in Adult Patients With Cushing's Disease or Acromegaly — NCT02060383.
• Registry record last updated 2019-05-29; status last verified 2019-05.
• Results first posted 2019-05-29.
• 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.

18. Record Summary

NCT02060383 is a completed, phase IV, multicenter, randomized, open-label, parallel-group study sponsored by Novartis Pharmaceuticals. It enrolled 249 participants with Cushing's disease or acromegaly and investigated management of hyperglycemia during pasireotide treatment.

Participants whose hyperglycemia remained above target on metformin could be randomized to incretin-based therapy or insulin for approximately 16 weeks. The primary endpoint was change in HbA1c from randomization to approximately 16 weeks, analyzed using an analysis of variance model with treatment, disease type, and baseline glycemic status as fixed effects.

The registry describes the endpoint and missing-data rule but does not report a numerical primary-outcome estimate or confidence interval. It also does not report the sample-size justification or multiplicity strategy. Safety figures are recorded for randomized and non-randomized groups and should be read as descriptive, not as direct evidence of comparative treatment effects.