Safety and Efficacy of Linagliptin (BI 1356) as Monotherapy or in Combination in Type 2 DM: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Linagliptin (BI 1356) as Monotherapy or in Combination in Type 2 DM (NCT00736099), a non-randomized phase 3 study. The primary outcome is Frequency of Patients With Adverse Events (AEs).

2. Clinical Question

The study record describes an extension of earlier trials, rather than a new randomized comparison of treatment options. Participants continued into different treatment modalities based on their previous trial: linagliptin as monotherapy, with pioglitazone, added to metformin, or added to metformin plus a sulphonylurea. The registry’s intervention list separately names linagliptin alone and linagliptin with pioglitazone; it does not give further detail about the enrollment or analysis of each prior-trial cohort.

The primary question is descriptive: how many participants in the reported groups experienced adverse events over the extension? The posted results distinguish serious events from other adverse events, and also list several serious event terms. They do not establish whether any difference between groups is attributable to treatment. The non-randomized allocation, open-label treatment, and differing treatment histories make a direct causal comparison especially difficult to interpret.

3. Trial Design

This completed phase 3 study, NCT00736099, is titled “Safety and Efficacy of Linagliptin (BI 1356) as Monotherapy or in Combination in Type 2 DM.” Its official title describes a 78 Week Open Label Extension to trials assessing BI 1356 (5 mg) as monotherapy or in combination with other antidiabetic medications in patients with type 2 diabetes. The registry lists enrollment as 2122.

The design is listed as parallel, non-randomized, and unmasked. The brief summary says the objective was to investigate the safety and tolerability of BI 1356 5 mg once daily over 78 weeks in different treatment modalities. The modalities were determined by the treatment participants had received in their earlier blinded trial. The listed histories include monotherapy, combination with pioglitazone, addition to metformin, and addition to metformin plus a sulphonylurea.

The registry lists linagliptin 5 mg and linagliptin 5 mg with pioglitazone 30 mg as drug interventions, with both arms described as open label. The adverse-event results use the group labels “Old Lina” and “New Lina.” The registry record does not explain those labels, so they should not be taken to identify treatment histories beyond the information explicitly recorded.

The study began in 2008-08 and had primary completion in 2010-12; its status is completed. The primary endpoint timeframe is 78 weeks. These design features describe a follow-up safety assessment, not a randomized treatment-effect study. In particular, treatment assignment was not randomized in this extension, and the registry cautions that the arms are not a like-for-like control comparison. The protocol, detailed statistical analysis plan, and rationale for the enrollment target 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: Treated Set: all screened patients who were documented to have taken at lease 1 dose of study drug.

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 Frequency of Patients With Adverse Events (AEs), measured over 78 weeks.

The primary endpoint is the frequency of patients with adverse events over 78 weeks. It is expressed as a participant count, rather than a continuous measurement. The endpoint description includes events detected during routine physical examination and electrocardiogram procedures, so it covers events identified through those assessments as well as other adverse-event reporting.

Safety endpoints of this kind are used to describe the occurrence of unwanted medical events during exposure to a study drug. They can show how many participants in the defined analysis population were recorded as having an event, and can distinguish serious events from other adverse events when the record reports that breakdown. Such counts are useful for describing the observed safety experience, but the number alone does not establish why an event occurred or whether it was caused by treatment.

The primary population is the Treated Set, defined in the record as screened patients documented to have taken at least one dose of study drug. The posted group counts are 158 participants with serious adverse events among 1532 at risk in “Old Lina,” and 52 among 589 at risk in “New Lina.” For other adverse events, the corresponding counts are 784 among 1532 and 285 among 589. The registry does not provide a randomized comparison, and the groups should not be interpreted as exchangeable treatment arms.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is a count of participants with adverse events. This is a binary outcome at the participant level: each person is classified according to whether an event was recorded for the endpoint. The count is not the same as the total number of event episodes, because one participant may experience more than one event. The registry reports participants affected and participants at risk for the two posted groups, but does not describe all event-counting rules.

What does a count describe?

A count records how many participants met the endpoint definition in the specified group. In this study, the primary endpoint concerns patients with adverse events over 78 weeks. The group results give the number affected and the number at risk. These quantities describe the observed safety experience in each group. They do not, by themselves, distinguish an event caused by study treatment from an event that occurred for another reason.

When groups are randomized and otherwise comparable, a difference in event frequency can be used to estimate a treatment contrast, subject to uncertainty and the analysis plan. Here, allocation is non-randomized, and the registry says that the arms are not a like-for-like control comparison. A simple comparison of the recorded counts would therefore not isolate a treatment effect. Randomization is important because it helps balance measured and unmeasured characteristics on average; without it, group differences may reflect who entered each group, their prior treatment, or other factors.

Why does the analysis population matter?

The primary population is the Treated Set, defined as screened patients documented to have taken at least one dose. This defines who is included in the reported safety assessment. A treated population is commonly used for safety summaries because adverse events are assessed among people exposed to study treatment. It differs conceptually from an intention-to-treat population, which generally retains participants according to randomized assignment. The record does not report a randomized assignment for this extension or describe additional analysis populations.

The registry gives group sizes at risk of 1532 for “Old Lina” and 589 for “New Lina.” It does not explain the group labels or provide further information about how the at-risk sets were constructed. The enrollment figure of 2122 is a separate registry entry and should not be treated as interchangeable with the posted at-risk counts. Reasons for any difference between enrollment and analysis counts are not reported in the registry record.

What do uncertainty measures add?

A confidence interval is a statistical range calculated from the observed data and an analysis model. It conveys uncertainty about an estimated quantity, such as an event frequency or a contrast between groups. A narrow interval indicates greater precision under the model; a wide interval indicates less precision. The registry information used for these results does not report confidence intervals or other uncertainty estimates for the adverse-event counts. Without them, readers cannot assess the precision of a group contrast from the posted results.

A hypothesis test asks whether data are inconsistent with a specified null hypothesis, under a chosen statistical model. A p-value is interpreted in relation to that model and does not measure the probability that a treatment works, nor does it establish causation. No analysis records are reported for this trial, and no p-value for the primary safety endpoint is included in the record. A test would also need to address the non-randomized design and the groups’ potentially different histories; a simple test of counts would not solve those design limitations.

How should serious-event terms be read?

The registry lists selected serious adverse-event terms with percentages. These entries identify event categories and their recorded frequencies. They are not automatically mutually exclusive: a participant could potentially have more than one event term, and the record does not describe whether or how that can occur. The denominator for these percentages and the selection procedure for the displayed terms are not reported. The terms therefore cannot be used to reconstruct a complete event distribution or to make an adjusted comparison.

For count outcomes, statistical modeling may be used when the question concerns event rates or group contrasts, but the appropriate model depends on the endpoint definition, follow-up, and data structure. The registry does not report a statistical analysis plan, model, covariate adjustment, or handling of repeated events. There is no basis for attributing a particular model to the sponsor. General statistical explanations do not replace those unreported design and analysis details.

What is the role of planning calculations?

A sample-size illustration based on an observed effect is a hypothetical planning exercise. It is not the number of participants actually enrolled, and it is not observed power. Such an illustration depends on assumptions about the target effect, allocation, significance level, and statistical approximation. The record describes the illustration as using a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect defined as a stated fraction of the observed effect. Those planning assumptions do not establish that this extension was designed using that approach. The trial’s own sample-size justification is not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 46 outcome measures, including changes in HbA1c from baseline at several follow-up times and the number of patients with HbA1c below 7.0% over time. The primary endpoint is the frequency of patients with adverse events. No analysis records are reported, and the registry record does not give a statistical analysis plan describing which outcomes were confirmatory, how comparisons were prioritized, or whether any adjustment for multiple testing was planned.

Multiplicity arises when several outcomes, time points, groups, or statistical tests are examined. The more opportunities there are to identify a result as notable, the more important it is to know which analyses were prespecified and how the overall error rate was handled. Without that information, results across multiple outcomes should not be treated as if they came from a single isolated test. Numerical results for the listed HbA1c outcomes are not reported in the information available for this account.

Missing-data handling is also not reported in the registry record. The record does not state how participants without follow-up measurements were handled, how incomplete adverse-event reporting was assessed, or whether any imputation or sensitivity analysis was planned. For a 78-week extension, the handling of differing observation time and discontinuation can affect how event summaries are interpreted; the registry does not provide those details.

The primary population is described as the Treated Set, consisting of screened patients documented to have taken at least one dose. That definition identifies the intended safety population, but it does not explain the treatment of missing observations or the exact analysis set for each outcome. Further detail about participant flow, exclusions, and endpoint-specific denominators is not reported in the registry record.

11. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: myocardial infarction (1.0%), acute myocardial infarction (0.7%), angina pectoris (0.5%), prostate cancer (0.5%), angina unstable (0.3%).

The serious adverse-event figures are participant counts: 158 among 1532 at risk in “Old Lina” and 52 among 589 at risk in “New Lina.” The figures for other adverse events are 784 among 1532 and 285 among 589, respectively. These describe the recorded experience within each group; they are not adjusted or randomized estimates of a treatment contrast. The group labels are not explained in the registry record, and the record cautions against treating the arms as a like-for-like comparison.

The serious-event term list includes myocardial infarction (1.0%), acute myocardial infarction (0.7%), angina pectoris (0.5%), prostate cancer (0.5%), angina unstable (0.3%), and atrial fibrillation (0.3%). The registry does not report the denominator used for these term percentages, whether the terms are mutually exclusive, or the rule for selecting the listed terms. Those percentages should not be combined or interpreted as a complete summary of all serious events.

Counts and term frequencies do not establish that treatment caused an event. The registry does not report comparative uncertainty estimates or an analysis adjusting for differences between the non-randomized groups. The figures support a descriptive reading of the recorded adverse-event experience only.

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

The result supports a descriptive statement about adverse events recorded during a 78-week open-label extension: the registry reports counts of participants with serious and other adverse events in two groups, along with selected serious adverse-event terms. The primary endpoint is a safety measure, and the reported counts make the observed event experience visible for the named groups.

The data do not show that one treatment strategy is safer or less safe than another. Allocation was non-randomized, the groups are explicitly not a like-for-like control comparison, and the registry does not explain the group labels. The participants entered the extension from different prior treatment histories, which further limits direct comparison. No adjusted estimate, confidence interval, hypothesis test, or sponsor analysis is reported in the registry record for the primary endpoint.

The figures also do not establish whether an adverse event was caused by linagliptin, a co-administered medication, an underlying condition, or another factor. An adverse event is an event recorded during a study; its presence in a safety listing does not by itself establish causation. The selected serious-event terms describe reported categories, but their denominators and selection method are not stated.

Likewise, the registry entries do not provide numerical results for the listed HbA1c outcomes in the information. The study title includes efficacy, but the stated primary endpoint is adverse-event frequency, and the available results do not support a conclusion about glycemic efficacy. The findings should be understood as registry-reported safety descriptions, not as medical advice or a definitive comparison of treatment options.

13. Why This Trial Matters Statistically

This trial is statistically notable because it illustrates the difference between a safety description and a comparative treatment-effect estimate. A count of participants with events can be informative about what was recorded during treatment exposure. But a causal comparison needs design and analysis features that support attributing differences to treatment. The non-randomized allocation and the registry’s warning that the groups are not like-for-like limit that interpretation.

The study also has an extension design linked to four prior blinded trials and multiple treatment histories. This makes the structure of the population important: participants did not enter with a single, uniform background treatment. The registry describes the treatment modalities as determined by the earlier trial, while giving two intervention entries and two unexplained adverse-event group labels. Those distinctions matter when readers consider who is being compared.

It also highlights why endpoint definitions and analysis populations should be read together. The primary endpoint concerns patients with adverse events over 78 weeks, and the population is defined as treated participants. That is a specific safety perspective, but the registry does not report the handling of missing data, detailed event rules, or analysis methods. These omissions limit assessment of precision and completeness.

Finally, a long list of outcome measures and time points can create multiplicity concerns, while the registry reports no analysis records. Knowing which outcomes were prespecified and how they were analyzed is essential for interpreting any efficacy findings. The record’s safety counts remain useful descriptive observations, but they do not substitute for a clearly reported comparative analysis.

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 main limitation for interpreting the adverse-event groups is the non-randomized design. The registry explicitly says the arms are not a like-for-like control comparison, so differences in observed counts cannot be attributed to treatment alone. Participants’ prior trial and treatment modality may differ, but group-specific histories and baseline characteristics are not reported alongside the adverse-event figures.

The labels “Old Lina” and “New Lina” are not explained in the registry record. Although the study summary describes several prior-treatment modalities and the intervention list names linagliptin alone and with pioglitazone, the record does not connect each safety group to a specific regimen or history. This limits interpretation of the group-level results.

The registry provides no confidence intervals, hypothesis tests, adjusted comparisons, or analysis records for these safety results. It does not report the statistical analysis plan, event attribution procedures, detailed definitions for serious and other adverse events, or how repeated events were counted. The percentages for selected serious-event terms lack a reported denominator and selection method.

The registry enrollment is 2122, while the adverse-event results list at-risk group sizes of 1532 and 589. The record does not explain the relationship between these entries, so reasons for the different figures should not be inferred. Missing-data procedures, participant flow details, and outcome-specific missingness are also not reported. Although HbA1c outcomes are listed, their numerical results are not included in the results. These limitations mean that the posted information supports descriptive reading, not a firm comparative or efficacy conclusion.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Safety and Efficacy of Linagliptin (BI 1356) as Monotherapy or in Combination in Type 2 DM — NCT00736099.
• Registry record last updated 2014-06-27; status last verified 2013-12.
• Results first posted 2012-03-08.
• 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

NCT00736099 is a completed phase 3, open-label, non-randomized, parallel extension study in type 2 diabetes. The registry lists enrollment as 2122, a start of 2008-08, primary completion in 2010-12, and a 78-week primary endpoint timeframe. The objective was to investigate safety and tolerability of BI 1356 (linagliptin) 5 mg once daily in different treatment modalities following earlier blinded trials.

The primary endpoint is the frequency of patients with adverse events in the Treated Set. The posted results report serious and other adverse-event counts for “Old Lina” and “New Lina,” plus selected serious-event terms. The registry does not explain those group labels, and it cautions that the groups are not a like-for-like control comparison. Statistical analysis details, missing-data methods, and the sample-size justification are not reported in the registry record.